En - Student Indebtedness

Fdration tudiante universitaire du Qubec
Student Indebtedness
Comprehensive study

Presented for the 151
st
meeting of the Board of Directors
(CAO-15111)
August 20-21, 2011
In Montral

The Fdration tudiante universitaire du Qubec (FEUQ) is an organization that brings
together 15 student associations with more than 125,000 students from all levels of and every
region of Quebec. Established since 1989, its main mandate has been to defend the rights and
interests of students with governments and education stakeholders. Throughout its twenty years
of existence, it has endeavored to defend a humanistic education as a societal choice. It focuses
particularly on defending its members before, during and after their passage in university by
demanding, above all, an accessible and quality education.


15, rue Marie-Anne Ouest
2
e
tage
Montral (Qubec)
H2W 1B6
Telephone: (514) 396-3380
Fax: (514) 396-7140

Supervision Ariane Campeau, vice-president of sociopolitical affairs
Analysis, writing,
linguistic revision and
page formatting
Louis-Philippe Savoie, contract researcher

All rights reserved FEUQ 2011
Student indebtedness : Comprehensive study
Fdration tudiante universitaire du Qubec - 2011
i
Summary
Student indebtedness is an undeniable dimension of many Western university systems.
Quebec is not alone: its loans and bursaries program, though much more generous than
those offered in the rest of Canada and the United States, is nonetheless often a high
source of indebtedness. The public debate on student indebtedness also tends to leave
aside two other forms of indebtedness, which are contracted from financial institutions
and relations (family and friends). This notice hopes to fill this gap.
The first step consists of giving oneself a conceptual framework for student
indebtedness. Following a review of the definitions currently employed, we will define
as all debt contracted for professional or postsecondary studies, independently of the issuer,
which allows the student to pay his school expenses as well as living expenses, with the exception
of a mortgage or the purchase of a vehicle. Subsequently, we will study and compare the
characteristics of public and private loans; it emerges that public loans are generally
more advantageous, despite the interdiction of bankruptcy seven years after graduation.
The sales techniques of financial institutions often turn out to be misleading.
Subsequently, we will focus on the impacts of indebtedness, as observed in the
literature. Overall, it emerges that students are not very informed on credit and its use.
Before studies, indebtedness acts as a barrier to entrance, especially for the poorest
students. This is explained, among other things, by a relation to risk that is different in
more disadvantages students. The theoretical models leave us to envisage an influence
on the choice of educational institution and field of studies, essentially for monetary
questions. During studies, indebtedness is an impediment: it generates dropping out,
tends in a direct and indirect manner, to motivate youth to work more outside the
university during studies. It also generates psychological stress and can even lead to the
abandonment of study projects in graduate studies. Finally, indebtedness has harmful
impacts after studies: a high debt increases the rate of default and bankruptcies, and the
finances of the new graduate are negatively affected.
After having explored the scholarly literature, we will proceed to a vast statistical study,
based essentially on descriptive analysis as well as bi-variable correlations. This detailed
analysis serves to feed a model of student indebtedness made up of six determinants
and twenty-four factors. Overall, it emerges that student indebtedness particularly
strikes populations that are in difficult social situations:
Students coming from more disadvantaged backgrounds are penalized: family
income contributes to decreasing the rate and level of indebtedness. Students
without a parental contribution are 1.5 times more indebted, to amounts that are
1.5 times higher.
Older students, parents, and those that do not live with their parents are all more
indebted.
Indebtedness predicts indebtedness: the fact of being indebtedness has a source
and is often accompanied by indebtedness to other sources, and to amounts that
are often higher. We offer the hypothesis of the existence of a spiral of
indebtedness, which will have to be confirmed by additional work.
ii
Financial information is not very available, which makes students more
vulnerable, especially as they learn a lot in this area by trial and error.
Various external and uncontrollable forces tend to increase the level of student
indebtedness: the cost of studies, governmental limits on loans, the level of aid
granted, the offer of credit by financial institutions and the cost of living increase
indebtedness.
Conversely, students seem adopt strategies to limit their indebtedness, such as
the increase in remunerated work during studies or reducing superfluous
expenses, such as recreation and transportation.
Quebec university students enrolled in undergraduate studies full-time seem to make
great efforts to limit their indebtedness. However, various levers escape them. A
generation of students are currently indebting themselves in a considerable manner, and
the new tuition fee policy of the Charest government, namely an increase of $1,625 in
the student bill in five years, will undoubtedly lead to an increase in student
indebtedness. We rather recommend the adoption of a strategy of a fight against student
indebtedness based on five axes: the tuition fee freeze, more generous loans and
bursaries, more numerous merit grants in graduate studies, better regulation of financial
institutions and a better distribution of financial information. These five intervention
axes will allow us to structure an effective intervention to limit and reduce the burden
that we are leaving on the shoulders of the next Quebec middle class.
iii
List of recommendations
1. That the government of Quebec develop and implement a strategy to fight
against student indebtedness that puts in place:
A tuition fee freeze as of 2012 accompanied by a better regulation of ancillary
fees;
Improvements to loans and bursaries and merit grants from granting agencies;
A better regulation of financial institutions;
Communication mechanisms to improve the financial skills of youth.
2. That the government of Quebec renounce the fee hikes announced in the 2011-
2012 Budget.
3. That the National Assembly of Quebec adopt a law to regulate the mandatory
institutional fees required by university institutions (as well as their components)
and stipulate that such fees cannot be imposed unless the nature, amount and
modalities of these fees are the object of an agreement between the institution and
the recognized student association as representatives of the students concerned.
(CAU-643)
4. That the maximum loan limit of student financial assistance not be increased.
5. That the special allowance covering the increase of tuition fees be paid out in the
form of a bursary to all recipients without exception, and that it not lead to any
increase in indebtedness.
6. That Student Financial Assistance increase the amount of allowable expenses for
recipients of student financial assistance, notably through an improvement of the
amounts for living expenses and transportation expenses for students that do not
have access to public transit. Transportation expenses for non-residents and
Internet expenses must also be included in allowable expenses.
7. That the Ministry of Education, Leisure and Sports introduce an automatic
annual indexation mechanism for all allowable expenses in the calculation of
Student Financial Assistance. That this indexation be equivalent to the Consumer
Price Index (CPI) for the year concerned.
8. That the exemption amounts for the maintenance of the family unit be
established at $45,000 and subsequently indexed.
9. That the government of Quebec adapt the loans and bursaries program with a
view to recognizing the diversity of academic paths and studies-work-family
integration.
10. That the government of Canada abolish the discriminatory provisions toward
students in the Bankruptcy and Insolvency Act.
11. That governments increase the level of financing for the different granting
agencies. (CNCS-426 [2.3.])
12. That the FEUQ support the provisions of bill 24 aiming at limiting the over-
indebtedness of consumers.
iv
13. That the government of Quebec forbid false representation in regard to student
indebtedness according to which credit can allow one to go through university
studies sheltered from financial troubles and on the future profitability of studies.
14. That the government of Quebec impose on financial institutions the duty of
clarifying financial products offered directly to students, among others, by stating
clearly the consequences of changing program or dropping out on the repayment
of the debt.
15. That financial institutions downwardly assess the maximum loan limits for
students.
16. That the government of Quebec forbid financial institutions from proposing to
students discounts on financial products that are not clearly or specifically
destined for them.
17. That the government of Quebec impose on financial institutions the obligation of
presenting on their Internet sites and informational folders presenting their
student products the student financial assistance program and its modalities..
18. That the Office de la protection du consommateur, in collaboration with student
financial assistance and university institutions, develop and distribute
information material on student indebtedness, with an emphasis on private
indebtedness and credit targeted toward students.

v
List of acronyms
SFA Student Financial Assistance
CCAFE Comit consultatif sur laccessibilit financire aux tudes
CNCS-FEUQ Conseil national des cycles suprieurs de la Fdration tudiante
universitaire du Qubec
CREPUQ Confrence des recteurs et principaux des universits du Qubec
CMSF Canada Millennium Scholarship Foundation
CFS Canadian Federation of Students
FEUQ Fdration tudiante universitaire du Qubec
MIF Mandatory Institutional Fees
MI Methodological individualism
MELS Ministre de lducation, du Loisir et du Sport
MEQ Ministre de lducation du Qubec
MESS Ministre de lEmploi et de la Solidarit sociale
OPC Office de la protection du consommateur
RCT Rational choice theory

vi
Table of contents
1. Introduction....................................................................................................................................................1
2. Conceptualoutline.......................................................................................................................................2
2.1. TheManyFormsofStudentDebt.............................................................................................................................2
2.2. StudentFinancialAssistancefromtheQubecGovernment........................................................................5
2.3. LoansfromFinancialInstitutions.........................................................................................................................10
3. Impactsofstudentindebtedness:whatthescientificliteraturehastosay...........................20
3.1. Methodology...................................................................................................................................................................20
3.2. Indebtedness:ageneraldiscussion......................................................................................................................22
3.3. Preventingthestartofuniversityeducation...................................................................................................25
3.4. Duringuniversity:abreak.......................................................................................................................................31
3.5. Afteruniversity..............................................................................................................................................................33
3.6. Modellingoftheimpactsofstudentindebtedness.........................................................................................36
4. Methodology................................................................................................................................................39
4.1. Statisticalmethods......................................................................................................................................................39
4.2. Samplingandrepresentativeness.........................................................................................................................40
4.3. Treatmentofthedata................................................................................................................................................40
5. Studentindebtedness:maincharacteristics....................................................................................42
5.1. Characteristicsoftherateofindebtedness.......................................................................................................42
5.2. Theimpactofalreadycontractedindebtedness............................................................................................46
5.3. Aspiralofindebtedness.............................................................................................................................................52
6. Theimpactofstudentcharacteristics................................................................................................53
6.1. Theimpactofsocioeconomiccharacteristics..................................................................................................53
6.2. SchoolCharacteristics................................................................................................................................................66
7. TheImpactofSourcesandMethodsofFunding..............................................................................78
7.1. TotalFunding................................................................................................................................................................78
7.2. Paidwork.........................................................................................................................................................................80
7.3. ParentalContribution................................................................................................................................................88
7.4. Meritgrants....................................................................................................................................................................92
7.5. Establishingthecomplexandmultifacetedlinksbetweenfundinganddebt...................................96
8. Theimpactofthelevelsofexpenses...................................................................................................99
8.1. Totalexpenses...............................................................................................................................................................99
8.2. Universityexpenses..................................................................................................................................................101
8.3. Livingexpenses...........................................................................................................................................................105
8.4. Otherexpenses............................................................................................................................................................113
8.5. Theleveloflivingexpensesincreasesthelevelofindebtedness...........................................................117
9. Thesituationingraduatestudies.....................................................................................................119
9.1. Methodologyandlimits..........................................................................................................................................119
9.2. Descriptionoflevelsofindebtedness................................................................................................................119
9.3. Roleofsourcesandmodesoffunding..............................................................................................................120
10. Analysisandrecommendations.........................................................................................................123
10.1. Studentdebt:Keyfactorsandimpact..............................................................................................................123
10.2. Characteristicsofdifferentloans.......................................................................................................................128
10.3. Studentdebtload:Amodel...................................................................................................................................129
10.4. Recommendations....................................................................................................................................................135
vii
11. GeneralConclusion................................................................................................................................141
Bibliography.....................................................................................................................................................145
Officialdocuments.................................................................................................................................................................145
Scholarlyarticles....................................................................................................................................................................145
Monographs..............................................................................................................................................................................147
Internetsites............................................................................................................................................................................147
Statisticalreferences............................................................................................................................................................147
AnnexeI- Caractristiquesdesprtsprivs..................................................................................149
AppendiceI-CodeSPSSpremiercycle.................................................................................................154
Crationdenouvellesvariables.......................................................................................................................................154
Codedegnrationdestableaux....................................................................................................................................172
AppendiceII-CodeSPSScyclessuprieurs........................................................................................182

List of figures
Figure3-1:Useofvariouscreditproductsamongstudents18to29yearsold,progressionfrom1994
to2004 __________________________________________________________________________________________________________23
Figure5-1:Rateofindebtednessaccordingtosource _________________________________________________________42
Figure5-2:Numberofsourcesofdebts___________________________________________________________________________43
Figure5-3:Variationofthelevelofdebtaccordingtothenumberofsourcesofdebts __________________43
Figure5-4:DistributionoftheamountsofdebtduetotheSFA_______________________________________________44
Figure5-5:Distributionofdebtamountsfromfinancialinstitutions _______________________________________45
Figure5-6:Distributionofamountsoffamilydebts____________________________________________________________46
Figure5-7:Distributionoftotalamountsofdebt_______________________________________________________________46
Figure6-1:Changeinaverageamountsofdebtbysourcebygender_______________________________________53
Figure6-2:Table6-2:Descriptionofstudentsage____________________________________________________________54
Figure6-3:Changeindebtratiobysourceaccordingtoage__________________________________________________55
Figure6-4:Changeinaverageamountsofdebtbysourceaccordingtoage________________________________56
Figure6-5:Changeindebtratiobysourcebasedonfamilyincome ________________________________________57
Figure6-6:Changeindebtratiobysourcebyplaceofresidence____________________________________________59
Figure6-7:Changeinaverageamountsofdebtbysourcebasedonthebasisofplaceofresidence ___59
Figure6-8:Changeindebtratiobysourceaccordingtothestudyarea_____________________________________60
Figure6-9:Changeindebtratiobysourcebasedonmovingaway__________________________________________62
Figure6-10:Changeinaverageamountsofdebtbysourcebasedonthechangeofregion______________62
Figure6-11:Changeindebtratiobysourceaccordingtothepresenceofdependentchildren________63
Figure6-12:Changeinaverageamountsofdebtbysourceaccordingtothepresenceofdependent
children _________________________________________________________________________________________________________64
Figure6-13:Changeindebtratiobysourcebasedonthegenerationofthestudent_____________________65
Figure6-14:Changeinaverageamountsofdebtbysourcebasedonthegenerationofthestudent___65
Figure6-15:Distributionofstudentsaccordingtotheirprogressintheirprogram_____________________68
Figure6-16:Changeindebtratiofromsourcedependingonyearofstudy _______________________________68
Figure6-17:Changeinaverageamountsofdebtbysourceaccordingtoyearofstudy__________________69
Figure6-18:Yearofstudyandnumberofsourcesofdebt____________________________________________________70
Figure6-19:Descriptionoftheexpectedlengthofthedegree_______________________________________________70
Figure6-20:Changeindebtratiobysourceaccordingtothedurationofthediploma__________________71
Figure6-21:Changeinaverageamountsofdebtbysourceaccordingtothedurationofthediploma 72
Figure6-22:Changeindebtratesbasedonextensionsofpastuniversitystudies _______________________73
Figure6-23:Changeindebtamountsbasedonextensionsofpastuniversitystudies____________________73
Figure6-24:Changeindebtratiofromsourcedependingonthedegreeofeconomicprofitability___75
Figure6-25:Changeinaverageamountsofdebtbysourcebasedontheprofitabilityofthediploma 76
viii
Figure 7-1: Description of Total Funding ___________________________________________________________________________78
Figure 7-2: Description of Number of Funding Sources ___________________________________________________________79
Figure 7-3 : Variation in debt percentage by sources as a function of total funding_____________________________79
Figure 7-4: Variation of average amounts of debt by source as a function of total funding ____________________80
Figure 7-5 : Variation in debt percentage by souce as a function of the presence of a job in autumn 2009 ___81
Figure 7-6 : Variation of average debt by source as a function of the employment rate ________________________82
Figure 7-7 : Numbers of hours worked per week, autumn 2009 __________________________________________________83
Figure 7-8 : Variation of the level of debt by source as a function of the number of hours worked per week per
job in autumn 2009 _____________________________________________________________________________________________84
Figure 7-9: Variation of average amounts of debt by source as a function of the number of hours worked in
2009_______________________________________________________________________________________________________________85
Figure 7-10: Variation in debt percentage by source as a function of job status_________________________________85
Figure 7-11: Variation of average debt amount by source as a function of job status___________________________86
Figure 7-12 : Description of gross annual income from a job______________________________________________________86
Figure 7-13 : Variation on debt percentage by source as a function of gross annual income___________________87
Figure 7-14 : Variation in average amounts of debt by source in function of gross annual income____________88
Figure 7-15: Variation of debt percentage by source as a function of the presence of a parental contribution89
Figure 7-16: Variation of average debt amounts by source as a function of the presence of a parental
contribution______________________________________________________________________________________________________90
Figure 7-17: Description of Levels of Parental Contribution______________________________________________________91
Figure 7-18 : Variation in debt percentage by source as a function of the level of parental contribution_____91
Figure 7-19: Variation of average debt amounts as a function of the level of parental contribution___________92
Figure 7-20: Description of amounts for merit grants and grants for internships_______________________________93
Figure 7-21: Variation of debt percentage by source as a function of the presence or absence of a merit
bursary___________________________________________________________________________________________________________93
Figure 7-22: Variation of average debt amount by source as a function of the presence or absence of merit
bursaries_________________________________________________________________________________________________________94
Figure 7-23: Variation of debt percentage by source in function of the size of institutional granted bursary_95
Figure 7-24: Variation of average debt amount by source as a function of the size of institutional bursaries 95
Figure8-1:Distributionoftotalexpenses________________________________________________________________________99
Figure8-2:Variationoftheindebtednessratebysourcebasedonthelevelofexpenses_______________100
Figure8-3:Variationintheaverageamountsofindebtednessaccordingtothelevelofexpenses____101
Figure8-4:Descriptionofamountsoftuitionfeespaidin2009bypermanentresidentsandstudentsof
Quebecorigins ________________________________________________________________________________________________102
Figure8-5:Variationofratesofindebtednessbysourceaccordingtotheleveloftuitionfeespaid__103
Figure8-6:Variationoftheaverageamountsofindebtednessbysourceaccordingtothelevelof
tuitionfeespaid_______________________________________________________________________________________________103
Figure8-7:Descriptionoftheamountspaidinschoolmaterial_____________________________________________104
Figure8-8:Variationofindebtednessratesaccordingtothelevelofexpensesonschoolmaterial___105
expensesinschoolmaterial_________________________________________________________________________________105
Figure8-10:Descriptionofrentexpenses______________________________________________________________________106
Figure8-11:Variationofratesofindebtednessbysourceaccordingtothelevelofexpensesonrent107
expensesonrent______________________________________________________________________________________________107
Figure8-13:Descriptionoffoodexpenses______________________________________________________________________108
Figure8-14:Variationoftherateofindebtednessbysourceaccordingtothelevelofexpensesonfood
___________________________________________________________________________________________________________________109
expensesonfood______________________________________________________________________________________________109
Figure8-16:Descriptionofexpensesontransportation _____________________________________________________110
Figure8-17:Variationoftheratesofindebtednessbysourceaccordingtothelevelofexpenseson
transportation_________________________________________________________________________________________________112
Figure8-18:Variationintheaverageamountsofindebtednessbysourceaccordingtoexpenseson
transportation_________________________________________________________________________________________________112
ix
Figure8-19:Descriptionofexpensesfordependentchildren_______________________________________________113
Figure8-20:Variationoftheratesofindebtednessbysourceaccordingtothelevelofexpensesfor
dependentchildren __________________________________________________________________________________________114
expensesfordependentchildren__________________________________________________________________________114
Figure8-22:Descriptionofrecreationalandotherexpenses________________________________________________115
Figure8-23:Variationofexpensesonrecreationaccordingtotherateofindebtednessbysource __116
Figure8-24:Variationinrecreationalexpensesandotherexpensesaccordingtotheaverageamounts
ofindebtednessbysource___________________________________________________________________________________117
Figure9-1:Descriptionofthelevelofindebtednessofgraduatestudents________________________________120
Figure10-1:Modellingofthesixkeyfactorsofstudentdebt. _______________________________________________130
Figure10-2:Classificationofkeyfactorsinrelationtothechoicetotakeondebtandtheimpactof
debt _____________________________________________________________________________________________________________130
Figure-3:Studenttuitionfeeratesfrom1994to2017(projected) ________________________________________135
Figure-4:Mandatoryundergraduateinstitutionalfeesbyuniversity(2010-2011)_____________________136
List of tables
Table3-1:Canadiansperceptionsofcostsandbenefitsofuniversityeducationin2003..........................27
Table3-2:Individualattitudestowardsthefinancialcostsandbenefitsofuniversityeducation...........27
Table3-3:EvolutionoftherateofemploymentforCanadianstudentsenrolledinfulltimestudies
between1976and2008,SeptembertoApril...................................................................................................32
Tableau3-4:Impactsofstudentindebtedness..........................................................................................................37
Table4-1:Exampleoftablesummarizingstatisticaltests.....................................................................................41
Table5-1:Characteristicsofdebtsbysource.............................................................................................................44
Table5-2:RelationbetweentherateofindebtednesstoSFAandtherateofprivateindebtedness......47
Table5-3:RelationbetweentherateofindebtednessatSFAandtherateoffamilyindebtedness........48
Table5-4:Relationbetweentherateofprivateindebtednessandtherateoffamilyindebtedness......48
Table5-5:RelationbetweentheamountoftheSFAdebtandthepresenceofothersourcesofdebt.....49
Table5-6:Variationoftheaverageamountsofdebtaccordingtotheamountofdebtsduetostudent
financialassistance....................................................................................................................................................50
Table5-7:Variationoftheindebtednessratebysourcebasedaccordingtothesizeoftheprivateloan
..........................................................................................................................................................................................50
Table5-8:Variationoftheaverageamountsofdebtspersourceaccordingtothesizeoftheprivate
loan..................................................................................................................................................................................51
Table5-9:Variationoftherateofindebtednessbysourcebasedonthesizeofthefamilyloan.............51
Table5-10:Variationoftheaverageamountofdebtsbysourcebasedonthesizeofthefamilyloan...51
Table5-11:Summaryofstatisticaltestsonindebtedness.....................................................................................52
Table6-1:Descriptionofstudentsage.........................................................................................................................54
Figure6-2:Table6-2:Descriptionofstudentsage.................................................................................................54
Table6-3:Descriptionofgrossfamilyincome...........................................................................................................57
Table6-4:Changeinaverageamountsofdebtbysourcebasedongrossfamilyincome...........................58
Table6-5:Descriptionoftheareaoforiginofstudents.........................................................................................60
Table6-6:Averagedebtaccordingtothestudyarea..............................................................................................61
Table6-7:Summaryofstatisticaltestsonsocioeconomiccharacteristics......................................................66
Tableau6-8:Profileofborrowersbasedonacademicprogress.........................................................................67
Table6-9:Classificationofthecurriculumaccordingtotheirprofitability....................................................74
Table6-10:Correlationcoefficientsbetweenfieldofstudyandcertaincharacteristics...........................75
Table6-11:Summaryofstatisticaltestsonschoolcharacteristics....................................................................76
Table 7-1 : Correlation between family income and family contribution..................................................................88
Table 7-2 : Summary of statistical testing between sources of funding and debt.....................................................97
Table8-1:Descriptionoftheamountsoftuitionfeesaccordingtocitizenshipstatus..............................102
Table8-2:Descriptionofthemodeoftransportationusedmostregularlyinautumn2009andaverage
annualexpenseassociated...................................................................................................................................111
x
Table8-3:Descriptionofthemodeoftransportationusedmostregularlyinautumn2009basedonthe
regionofstudies......................................................................................................................................................111
Table8-4:ComparisonofmeasuresofthecentraltrendofrecreationalexpensesforQuebec
householdsandstudents......................................................................................................................................116
Table8-5:Statisticalanalysisofthelinksbetweenthesourcesofexpensesandindebtedness............118
Table9-1:Numberofborrowersthathadtoassumeattheendoftheirstudiestherepaymentoftheir
loansobtainedaccordingtotheamountofthestudentdebt,2008-2009...........................................119
Table9-2:BelongingtoaresearchgroupandparticipationinSFA.................................................................120
Table9-3:Presenceorabsenceofmeritgrants.....................................................................................................121
Table9-4:Presenceorabsenceofagrantforconferenceorinternship.......................................................121
Table9-5:Presenceorabsenceofafamilycontribution.....................................................................................121
Table9-6:Presenceorabsenceofaninternaljobattheuniversity................................................................122
Table9-7:Presenceorabsenceofanexternaljobtotheuniversity...............................................................122
Table10-1:Descriptionoftherelationbetweendebtload,thesixkeyfactors,andstudiesonacademic
careerpaths...............................................................................................................................................................132
TableauI-1:Offredecartesdecrditsrguliresparinstitutionfinancire..............................................149
TableauI-2:Margesdecrditrgulires..................................................................................................................150
TableauI-3:Offredecartedecrditcibleparinstitutionfinancire...........................................................151
TableauI-4:Margesdecrditoffertesparinstitutionfinancireselonledomained'tude................152
1
1. Introduction
Student indebtedness is an inevitable dimension of the debate on the student
contribution and the financing of Quebec universities. We have been mandated by the
FEUQ to draw a portrait of the situation and propose recommendations emerging from
this portrait. We have thus proceeded to this analysis in various manners.
The study begins with a presentation of the conceptual framework. We will provide a
definition of indebtedness and explore the mechanisms associated with the different
forms of student indebtedness.
Subsequently, we will proceed to a review of the scholarly literature, by focusing on the
impacts of student indebtedness before, during and after studies. We will also explore
the situation of indebtedness in Quebec youth.
Next, we will proceed to a statistical analysis. We will begin with the methodology used
to then evaluate the sequence of individual characteristics, the sources and modes of
funding as well as the levels of student expenses, to see where the relations exist
between these student characteristics and the different forms of student indebtedness.
This analysis will concentrate on the situation of undergraduate university students
enrolled full-time. We repeat the same analysis, on a smaller scale, for students enrolled
in graduate studies full-time.
Subsequently, we summarize the effects observed of student indebtedness and issue
nineteen recommendations to limit the level of student indebtedness, better regulate
financial institutions and ensure better financial skills for students.

2
2. Conceptual outline
In this section, we will attempt to equip ourselves with the intellectual tools necessary to
understand the concept of student debt. We will begin by defining student debt so that
we may next describe the mechanisms of public lending and private lending from
financial institutions as it concerns the granting of loans, their particular features and
their repayment. We will conclude by comparing the key characteristics of government-
guaranteed loans and loans granted by financial institutions.
2.1. The Many Forms of Student Debt
Student debt can be defined in many ways. This study focuses on a specific kind of
student debt, which we will define after examining various common typologies.
2.1.1. The Various Forms of Debt
Before defining student debt, it is necessary to analyze the various forms any given debt
may take. We will therefore examine the typologies included in some recent studies on
the subject. Before we begin, it is worth noting that several studies deal only with debt
acquired through government programs (Allen and Vaillancourt, 2004; AFE, 2010a), do
not differentiate between different types of student loans (ISQ, 2010), or group together
all debts that do not come from a government program (AFE, 2010b).
Firstly, the survey into the origins and types of undergraduate financing (FEUQ, 2010a),
which will be our main source of information for this study, presents four major types of
debt:
Public debt, acquired mainly from student financial assistance programs
Credit card debt
Debt from a line of credit or a personal loan
Debt to family and friends.
It is, however, important to note that debt from the purchase of a house or vehicle was
excluded from the responses, by means of a specification in the questionnaire.
Other authors have opted for various typologies that are roughly equivalent. The main
one that we have retained for the purposes of this study comprises four points and was
developed by the Canadian University Survey Consortium (CUSC, 2009):
Government aid
Financial institutions
Parents and family
Other sources of debt
The two typologies are similar, although indebtedness to friends is here included in the
other sources of debt. For the purpose of this study, we will maintain a typology
divided into two themes.
Public debt
Acquired through a system of government-funded financial aid
3

Private debt
Acquired directly from a financial institution, including credit cards,
lines of credit, personal loans, or other forms of debt
Acquired from family members, friends or parents
Debt acquired from other sources
Due to a lack of conclusive data, we will not deal with other sources of debt. A quick
look at institutional practices does not allow for an analysis of loans within universities,
as opposed to student bursaries (see CNCS-FEUQ, 2007).
Household debt is a growing concern in Qubec society: the last few years have seen a
significant increase in debt levels, because of, among other reasons, increasingly easy
access to credit cards. Statistics Canadas Survey of Financial Security has established a
typology of the kinds of existing household debt (Statistics Canada, 2005, p. 15). In a
recent study, the Institut de la statistique du Qubec grouped this information into five
sub-categories (ISQ, 2010, p. 153).
Mortgages
Lines of credit and credit cards
Student loans
Automobile loans
Other debt
This information offers a satisfactory typology of debt according the source and the type
of debt.
2.1.2. Definition of Student Debt
But first, what do we mean by student debt?
It seems necessary to create this definition, given the lack of an easily usable one in the
literature. The concept of student debt may vary widely depending on authors and their
requirements, depending on what is included and excluded:
1. Debt may be acquired through government-funded student loans only (Allen
and Vaillancourt, 2004; AFE, 2010a)
2. Debt may exclusively include loans taken out directly for the completion of
studies (ISQ, 2010)
3. Debt may include all types of debt, excepting major acquisitions (FEUQ,
2010a; AFE, 2009)
4. Debt may include all types of debt (CUSC, 2009).
Our main source of data will be the survey on undergraduate students living
conditions (FEUQ, 2010a). Essentially, this survey uses the following definition of
student debt, which seems relevant to this study. It approaches most closely the third
definition by including most debts except mortgages, automobile loans and other
sources of debt.
4
Table 2.1: Sources and types of student debt
Source/Type Mortgage Line of
credit/credit
card
Student
loans
Automobile
loans
Other debt
Public X
P
r
i
v
a
t
e

Financial
Institutions
X X
Family, friends
and parents
X X
Other sources

Before continuing towards the creation of a definition, it is important to discuss some
definitions that were mentioned earlier, and why we believe they should be excluded.
We will not choose a complete financial portrait, such as that suggested by the fourth
category of definition. Indeed, it is wise, for information-processing purposes, to
exclude mortgages. These may be excessively high,
1
especially for young families,
falsifying the larger picture. Contrary to consumer or student debt, a mortgage is backed
by an asset that can be seized in the case of a defaulted payment. Additionally, the
backed asset, especially a house, has independent value (Lachance, Beaudoin and
Robitaille, 2005). In the framework of the survey on student living conditions, we would
run the risk of finding ourselves in a situation of double-counting, given that one of the
questions asked was regarding the amount spent each month on rent. Similar reasoning
can be applied to spending on vehicles and transportation, explaining the exclusion of
these types of loans. It seems, then, appropriate to keep some private loans that
correspond with overall consumer debt in the analysis.
It may be considered controversial to include debt that is not strictly from student
loans.
2
Some people believe that students live above their means. It is, of course, clear
that we do not share this opinion. Modern financial assistance programs for students
around the world (see OCDE 2010) all grant financial aid in the form of loans or
bursaries, even in the case of jurisdictions where education is free, which demonstrates
the importance of being able to pay living expenses while in school. The survey
Conditions de vie des tudiants de premier cycle (FEUQ, 2010a) and the Enqute sur les sources
et modes de financement des tudiants de cycles suprieurs (CNCS-FEUQ, 2008) have both
demonstrated that university students are frequently financially independent from their
parents, implying the necessity of using the financial resources available to provide for
their needs. Yet we see that consumer debt is common for students, sometimes at high
levels: the portrait of student debt would not be complete without private debt.

1
For families under the age of 35, the median mortgage rate was $90,000 in 2005 (Statistics Canada in ISQ,
2009, p. 149)
2
Student expenses, in particular, are often subject to debate in the general media
5
A more valid reason can be discussed. In order to analyze the functioning of public
policy, it is necessary to understand how it is used. This is the case for Student Financial
Assistance (SFA), which presents well-developed statistical reports. Such tools allow for
quick evaluation of the strengths and weaknesses of a given program, though this is not
the objective of our study. A complete financial portrait allows us to show not only
debts from public programs, but also needs expressed and fulfilled effectively in the
market. Given that, as we will see later, the method of calculating financial assistance
assumes that it is the main financial support for the beneficiarys studies (after taking
into account various contributions), setting up a wide portrait of student debt, including
private debt, gives an interesting indication of the flaws that are frequently denounced
in student assistance programs (see, among others, FEUQ, 2010b).
Our definition of student debt should therefore include three sources: public debt,
private debt and debt to family and friends. It should also exclude debt from mortgages,
in order to correspond approximately to debt for consumer and living expenses,
meeting real financial needs during the course of studies. We will propose, therefore, a
definition of student debt as follows:
all debt acquired as part of professional or post-secondary studies, regardless of the
issuer, that allows the student to pay school expenses as well as living expenses,
excepting mortgage payments and the purchase of a vehicle.
For the purposes of analysis, we will divide up student loans first and foremost based
on their source: public, financial institution, family/friends. The division by loan type
presents a significant methodological challenge, because it would be necessary to
distinguish loans from financial institutions serving to finance studies and those that
serve other more or less specified ends. However, we do not have access to this
information. The definition is not ideal, but it gives an idea of the issue. This definition
offers the best reflection of the financial situation of students, given, as we will see later,
the important role that private debt plays in financing university studies, which
excludes the first definition, based solely on debt from public sources.
2.2. Student Financial Assistance from the Qubec Government
Close to 45% of undergraduate university students registered in full-time studies expect
to leave university with debt from some kind of student assistance program (FEUQ,
2010A). It is important to understand the inner workings of aid programs established by
the government of Qubec: these include some significant details. The reader wishing to
gain additional knowledge should consult the Trousse sur laide financire aux etudes
(FEUQ, 2010b). We will spend more time, of course, on the analysis of the sources of
student debt.
2.2.1. The Loans and Bursaries Program
The student financial assistance system is made up of two main programs: the Loans
and Bursaries Program (which is targeted mainly to full-time students) and the Loan
Programs for part-time students.
6
The first program is based on contributory and auxiliary principles. It assumes a
contribution by the student, parent and/or spouse and compensates for a presumed
3
lack
of income. To do this, it takes into account contributions and allowable expenses, which
are shown in Table 2-2.
Table 2-2: Contributions and Allowable Expenses
Contributions Allowable Expenses
Student
Parent
Spouse
Tuition fees
Living costs
Transportation costs
Costs of dependent children
Other expenses

Source: FEUQ, 2010b

In short, this calculation provides a volume of assistance. This is then transformed into
loans and bursaries according to the lending ceiling, which is the annual debt
maximum. This amount varies depending on a variety of factors. For one thing, the
lending ceiling varies according to the level of study. Table 2-3 shows the lending ceilings for
the university level, as well as public debt scenarios according to various criteria
4
.

3
And not for a real lack of income, which is an important distinction. The Student Financial program is
based on forecasted expenses, which do not necessarily correspond to expenses actually incurred
4
Several other factors may influence the amount of debt incurred during the course of studies, including
debt incurred while studying as well as the duration of studies
7
Table 2-3: Lending Ceilings
Per
month
of
study
8
months
of study
12
months
of study
Undergraduate
(3 years)
Undergraduate
(4 years)
Masters
(16
months)
Ph.D (4
years)
University,
undergraduate
$305 $2,440 $3,660 $7,320 $9,760 - -
University,
graduate
studies or
possessing an
undergraduate
diploma
$405 $3,240 $4,860 $7,320 $9,760 $6,480 $12,960
Organization
known for
lending only
$950 $7,600 $11,400 - - - -
Source: Rglement sur laide financire aux tudes, RRQ, c A-13.3, r 1, art 51. Calculations by the author according to the
length of studies forecast by SFA
Yet this lending ceiling may been modified in some cases. Indeed, about a third of loan
and bursary beneficiaries receive only loans. These students benefit from a special
allowance, which is shown below and is added to the lending ceiling.
Table 2-4: Special Allowance
University year Special Allowance
2006-2007 $0
2007-2008 to 2011-
2012
$3.33/cumulative units $100 per year (30
units)
$16.65/full-term unit $500 per year
2012-2013 to 2016-
2017
$10.83/cumulative units $325 per year (30
units)
$54.15 $/cumulative units $1625 full-term
units

It should be noted that these modifications were adopted following the unfreezing of
tuition fees in 2007. They led to a significant increase in student debt, especially for
middle class beneficiaries as well as those who receive only a loan.
Finally, we will note the existence of debt limits, varying between $30,000 and $55,000
depending on the level of study (AFE 2010 in FEUQ, 2010b).
It should be noted that loans from the Loans and Bursaries Program have more
advantageous features than private loans. Students do not have to pay interest until a
month after the end of their studies (AFE, 2011a). Additionally, they have a period of
six months before repayment must begin.
8
2.2.2. Loans Program for Part-Time Studies
The Loans Program for Part-Time Studies differs in more than one respect from the
program for full-time students. It is only open to students who are registered part-time
for six to eleven university course units per year. It takes into account only educational
expenses and childcare costs. It is open to students who have annual financial resources
of less than $35,000 for single students and $50,000 for married students or those with a
parental contribution. The amounts are increased by $2,805 for each child, with a bonus
of $2,101 if the student is a single parent. In both cases, the student must live with the
child (RRQ, c. A-13.3, r. 1, art. 82).
In 2010-2011, allowable expenses were $105.23 per university unit for educational
expenses and $490 per semester for childcare costs (RRQ, c. A-13.3, r. 1, art. 86-7). A
student registered for 18 credits in a year (two semesters of three courses each), who
also had a dependent child, would receive a loan of $2380. The debt limit is $8,000. The
repayment terms are the same as for the Loans and Bursaries Program.
For the purpose of this study, we have excluded part-time students. They have different
characteristics than their full-time colleagues and are generally non-traditional
students. Their profile requires a detailed, dedicated analysis, which is not the purpose
of this study.
2.2.3. After Study: Repayment
Debt implies repayment. It is worthwhile to look into the specifics of repayment of a
debt obtained from a student financial assistance program. We will examine the specific
features of these debts, the repayment terms, and the advantages and disadvantages of
student loans.
Former students are required to start repaying their student debt at the end (or
abandonment) of full-time studies. They are given a grace period of six months on the
repayment and a one-month grace period for repaying interest. Former students must
also negotiate a repayment agreement with their financial institution. The interest rate
is set by student financial assistance by-laws at 0.5% higher than the base rate for
businesses, which was 3% in July 2011
5
(RRQ c. A-13.3, r. 1, art. 73). This rate is
generally better than that for personal loans. For example, the interest rate for student
lines of credit hovered around 4.5% (infra section 1.3). Similarly, only the length of the
repayment term can be negotiated with the financial institution.
There are three types of assistance for new graduates, which make student loans more
advantageous than personal loans.
The deferred repayment program is offered to former students with low
income for a period of more than four months.
6
In this case, the government of Qubec
takes over interest payments for a period of six months. Students may enrol in this
program for a total of twenty-four months (AFE, 2011b, p. 32).

5
See the Bank of Canada website: http://www.bankofcanada.ca/rates/daily-digest
6
Gross monthly income from $1,522 per month for students without dependent children to $2,458 per
month for the head of a single parent family with four children
9
The debt forgiveness program, established in 2000, decreases by 15% the
debt of students who have completed their studies without interruption and in the
expected timeframe, receiving a bursary from the Loans and Bursaries Program each
year (FEUQ, 2010b, p.33). The number of recipients is relatively low; approximately 998
out of the 27,276 students who took out loans in 2008-2009, or less than 4% of former
students from that year, participated in the debt forgiveness program.
The tax credit offered on interest paid on a student debt is another
incentive. Both levels of government offer a tax credit. It is for 20%, reportable, non-
transferrable and non-refundable at the provincial level (FEUQ, 2011a, p. 9) and 15% at
the federal level, with the same terms.
In addition, it is possible to temporarily stop repayment because of a
pregnancy of 20 weeks or more, following the birth or adoption of a child, because of a
temporary disability, or because of election as a permanent member of a national
student association.
It is, nevertheless, important to note that student loans, despite their advantages, also
have their disadvantages, the main one being the prohibition against filing for
bankruptcy. The Bankruptcy and Insolvency Act (R.S.C., 1985, c. B-3) was modified in
1997, 1998 and 2005 to prevent the inclusion of student debt in the bankruptcy process
for seven years (five years in special cases). These modifications were made following
an overhaul of bankruptcy law in 1992 that went about
the abolition of preferential debt for debt owed to the government. This change relegates
the Crown to the level of ordinary creditors that share in the prorated assets of the debtor after
secured creditors and preferential creditors. The Crown did not therefore have priority over
other creditors concerning its loans to students (Smith, 2002).
New regulations concerning bankruptcy are controversial, especially since they were
established at a particularly difficult time for Canadian students, as student debt
reached levels never before seen in recent history. It is nevertheless important to note
that bankruptcy has significant consequences on an individuals credit rating and
impacts the interest rate of future loans.
7

As a corollary to the impossibility of filing for bankruptcy, some loans become
irrecoverable debt, despite attempts at repayment. From the moment that repayment
difficulties are noticed, the financial institution can transfer the loan to the payment
department for Student Financial Assistance. In these cases, it falls on the Qubec
government to vouch for the loan and become the creditor, repaying the financial
institution. Failing to repay student debt entails serious consequences:
Ineligibility for the preferential repayment program
Tax reimbursements from Qubec as well as provincial sales tax credits will be applied to
repayment of student debt, in compliance with section 31 of the Act respecting the ministre du Revenu.
Ineligibility for the Loans and Bursaries Program if at least 50% of the student debt is not.

7
For more information, consult Raymond, Chabot Inc. La faillite personnelle,
http://www.avocat.qc.ca/public/iifailliteperson.htm [consulted June 1, 2011)
10
The financial institution may notify credit reporting companies.
8

Approximately 14% of outstanding loans since 1996 were repaid by the government of
Qubec (AFE, 2010a, p. 67). In addition, Qubecs education ministry receives between
9,000 and 13,500 claim requests from financial institutions concerning loans; close to
9,000 are repaid annually.
It appears, then, that loans granted by Student Financial Assistance have some
particularities. They have advantages and disadvantages, compared to regular loans
taken out directly from financial institutions.
2.3. Loans from Financial Institutions
Financial institutions offer a range of products. We have taken a survey of two main
financial products that affect student debt: credit cards and lines of credit. In each case,
two types of offers generally exist: those open to all students and those open to only
specific students in specific fields of study. We will note here that personal loans
targeted directly to students do not exist; lines of credit, being relatively flexible, fill this
role. Yet lines of credit imply less upstream planning than a personal loan, which grants
a fixed sum (similar to a student loan).
In order to establish this section, we have taken a survey of the information available on
the websites of eight financial institutions in Qubec, which will be designated
Financial institution 1, Financial institution 2,, Financial institution 8 in the
following analysis. The complete results will be present in appendix I. We will also
present, when pertinent, selling points mentioned by these lending institutions.
2.3.1. Lines of Credit and Credit Cards: Differences
Lines of credit and credit cards are two forms of consumer credit.
9
Both are revolving
credit: the line or card must be paid off before lending can begin again. Here, the
similarities end.
Credit cards generally have high interest rates (at least 19% annually) and a relatively
low credit limit. Nevertheless, consumers generally have twenty-one days after
receiving the balance to pay; if they do not pay, interest fees will be charged.
Lines of credit are linked to the consumers chequing account. The interest rate is
markedly lower than for a credit card; on the other hand, interest is charged from the
moment of lending. The credit limit is often much higher than for a credit card.
The Office of Consumer Affairs financial calculator
10
offers a practical illustration of the
differences between lines of credit and credit cards. Imagine, for example, a student
with a credit card at an interest rate of 19.4% and a line of credit at an interest rate of
4.5% who wants to purchase $1,000 of goods.

8
AFE Repayment http://www.afe.gouv.qc.ca/en/apresEtudes/remboursement.asp [consulted May
31, 2011]
9
Dugas, Sylvie. La marge de credit personelle: est-ce une bonne affaire? servicevie.com.
http://www.afe.gouv.qc.ca/en/apresEtudes/remboursement.asp (consulted July 11, 2011)
10
Office of Consumer Affairs. http://www.ic.gc.ca/eic/site/oca-bc.nsf/eng/ca01812.html (consulted
July 15, 2011)
11
If the student uses the credit card to finance the purchase and makes only the minimum
payment of $40, he or she will pay $291.47 in one year in interest fees on the credit card
and $52 on the line of credit. The difference is significant.
2.3.2. Main Features of Regular Student Loans from Financial Institutions
We will now study two financial products: credit cards and lines of credit. We will note
that all financial institutions have a section on their website that is easy for students to
access. Some even have a sub-section designed specifically for them, offering various
financial planning tools as well as the institutions different financial products. It is to
be noted that the references used for this section are all found at the end of section 2.3.2.
Credit cards
Table 2-5 illustrates the main features of student credit cards. Credit cards are by far the
most common financial product for young adults. They are offered by six of the eight
financial institutions surveyed: As for the other two institutions, one offers a similar
product, targeted generally to youth, whereas the other one advertises student cards,
but they do not have any special features.
The products are all similar: they have high interest rates and no annual fees. Generally,
however, they offer options meant to cultivate students customer loyalty: five out of
thirteen cards offered have a point system and three offer cash back programs.
Averaging out the annual fees, two institutions offer a reduced interest rate.
It appears that the main objective of student credit cards is to gain customer loyalty,
offering a seemingly free entrance into the world of credit. It is worthwhile to note that
the credit limits are not provided on the websites: none of the financial institutions
mention them, as they vary depending on the borrowers credit rating. Financial
institutions have become known for their propensity to offer consumers excessive credit
limits, leading to more severe legislative framework by the Qubec government, as we
will see later.
Table 2-5: Main features of student credit cards
Product offered Features
Student Credit Card Six out of eight financial institutions offer them.
One offers a similar product for youth
Interest Rate Between 19.4% and 19.99%; two institutions have a
rate of 21.99% for cash advances
Reduced Interest Rate Two institutions offer this option
Annual Fees None of the cards have fees; some options may
incur annual fees
Points Program Five out of thirteen cards have a points program
Cash Back Program Three out of thirteen cards have a cash back
program of between 0.5% and 1% per dollar spent

12
A frequently mentioned argument for student credit cards is building a good credit
rating:
Financial institution 3: Nothing could be simpler! Its a credit card adapted to the realities
of student life: it has flexible admissibility criteria and gives you the opportunity to start
building credit now a real advantage when the time comes to make important first
purchases, such as a car.
11

Other institutions stress specific features, including the points system, which is
intended to increase use of the credit card. In every case, they emphasize the idea that
using a credit card is a solution specially adapted to the needs of students:
Financial institution 6: Finally, a card logically thought out for students. The card offers an
accelerated Moneyback reward program and a host of other great benefits. It's the ideal choice
for students who spend wisely and want to earn valuable cash rewards - that can really add
up! And best of all - there's no annual fee!
12

Financial institution 5: A premium no annual fee credit card. Earn 1 point for each $2 spent.
Redeem points for travel, merchandise, gift certificates/cards & more.
13

The credit card is presented as a complementary means of financing that lets students
enter the world of consumption easily, profitably, and without anxiety. Student lines of
credit will prove to be somewhat different in their presentation.
Lines of credit
We have seen that lines of credit are distinguished from credit cards in several respects.
All financial institutions surveyed here offer student lines of credit. The amounts at play
are very high: the majority of institutions offer a maximum annual line of $10,000 with a
maximum debt level of $40,000 for four years of study. Repayment is made after the
completion of studies, similar to a student loan; however, interest is charged during the
study period. The interest rate is variable; currently, it is 4.5% at financial institution 1,
but varies depending on several criteria. Table 2-7 breaks down the main features of the
surveyed student lines of credit.
Table 2-7: Main features of student lines of credit
Product Offered Features
Student line of credit Offered by all financial institutions

11
Laurentian Bank. Student VISA Black.
https://www.banquelaurentienne.ca/en/personal_banking_services/my_ideas/ideas_student_visa_blac
k.html
12
Scotiabank. Learn VISA card.
http://www.scotiabank.com/cda/content/0,,CID13398_LIDen,00.html (consulted July 11, 2011)
13
RBC Royal Bank. RBC Rewards Visa Gold. http://www.rbcroyalbank.com/credit-cards/student-
credit-cards/index.html (consulted July 11, 2011)
13
Maximum annual
amount for a full-time
undergraduate student
$5,000: 1/8 institutions
Specific features Financial institution 2 offers a $15,000 line of credit
for the first year and $10,000 for subsequent years
Financial institution 4 is the only one to describe its
line of credit as working capital
Maximum borrowing
period
Usually four years, with a maximum debt load of
$20,000 to $45,000 depending on the line of credits
annual maximum
Repayment terms None of the lines of credit must be paid back
during studies, apart from interest.

One out of eight institutions allows a six-month
grace period after the end of studies before
repayment; the others allow one year. One reduces
the grace period to six months for students who
leave school without a diploma.

The length of repayment varies between 7 and 20
years.
Interest rate Variable depending on the financial institution and
individual credit ratings.
While credit cards are presented as a means of paying for some common expenses,
student lines of credit are presented in many different ways. Some financial institutions
advertise them as a complementary means of financing university studies,
compensating for gaps in the Loans and Bursaries Program:
Financial institution 1: If they do not qualify for government financial aid or if it is not enough
to cover their educational costs or if you want o see your financial needs met to study in
peace, sheltered from financial worries. This line of credit is reserved to students, without
age limits.
Financial institution 3: Studying or returning to school made easy! Get a load off your
shoulders with the student line of credit. This line of credit is especially tailored to your
student needs so you can concentrate on your academic success while enjoying a little much
appreciated financial help.
Others suggest financing the entirety of studies on a line of credit:
Financial institution 2: In your first year of college or university, everything is new including
the cost of your education. A Student Line of Credit helps you plan for the expenses you
expect and the ones you dont.
Financial institution 4: What if you dont have all the funds you need to finance your
education? Should that prevent you from going ahead with your plans? No, especially when
you consider that your education is one of the most !important investments youll make in
your life.
14
Moreover, financial institution 4 seems to adopt a discussible approach to student
credit, by describing its line of credit as student working capital, name that has very
little to do with a personal loan. It is also notable that among the financial institutions
surveyed, only four mention the existence of student loans subsidized by the Quebec
government, while financial institution 8refers to an aid program for pan-Canadian
studies, which does not exist in Quebec. The other four institutions, all of which offer
government-guaranteed loans, do not mention this in their documentation for students.
We have reached the end of the discussion on general products offered by financial
institutions. They are mainly of two types: credit cards and lines of credit. The former
are mostly presented as a means of building a credit history; the latter, as a
complementary method of financing, or to finance studies in their entirety.
References (classified by alphabetic al order):
Banque Laurentienne, Marge de crdit tudiante .
https://www.banquelaurentienne.ca/fr/services_particuliers/mes_idees/idees_marge_de_credit_etudiante.html
[Consulte le 11 juillet 2011]
Banque Laurentienne. VISA noire tudiante
https://www.banquelaurentienne.ca/fr/services_particuliers/mes_idees/idees_visa_noire_etudiante.html
Banque de Montral. Marge-crdit Aux tudiants . http://www.bmo.com/accueil/particuliers/services-
bancaires/prets-et-prets-hypothecaires/prets-et-marges-credit/marges-credit/etudiants/etudiants [Consulte le 11
juillet 2011]
Banque Nationale. Fonds de roulement tudiant . http://bnc.ca/bnc/cda/productfamily/0,2664,divId-2_langId-
2_navCode-10020,00.html [Consulte le 11 juillet 2011]
Banque Scotia. VISA Savoir Scotia pour tudiants .
http://www.scotiabank.com/cda/content/0,,CID13507_LIDfr,00.html [Consulte le 11 juillet 2011]
Desjardins, Marge de crdit Avantage tudiant
http://www.desjardins.com/fr/particuliers/produits_services/financement/marge_credit/mc_avantage.jsp
[Consulte le 11 juillet 2011]
RBC Banque Royale. Visa Or RBC rcompense . http://www.rbcbanqueroyale.com/cartes/student-
cards/index.html [Consulte le 11 juillet 2011]
2.3.3. A Targeted Offer: Financial Products Offered by Field of Study
Several institutions offer financial products directed specifically to particular students
registered in programs judged to be more profitable. These products are divided, once
again, into credit cards and lines of credit. It is to be noted that the references used for
this section are all found at the end of section 2.3.3.
Credit cards
Only three institutions offer credit cards targeted by field of study. The benefits are
largely similar to regular credit cards; however, it can be presumed that the credit limits
offered are higher. To note: financial institution 4 is the only institution that offers an
amnesty on annual fees for student credit cards for only two years, while giving
students in specific fields access to the most prestigious credit cards, just as do financial
institution 5 and financial institution 6. This practice encourages the idea that students
belong to the social group corresponding to their future profession, while still having
the financial means of a student. In the section 3.2.2., we will discuss feelings of
attachment to a social group that does not correspond, economically speaking, to
students actual situations, and how this may contribute to a false sense of consumer
spending power, prompting students to exceed their credit capacity.
15
Table 2-8: Features of credit cards offered for targeted fields
Credit card for targeted
fields
Offered by three out of eight institutions
Fields targeted Healthcare: three institutions
Law: two institutions
Administration: one institution
Engineering: one institution
Interest rates Between 19.4% and 19.99%; one institution has a
rate of 21.99% for cash advances
Reduced interest rates One institution offers an optional reduced
interest rate
Annual fees Three institutions out of four offer the product
without annual fees; one offers two years of
exemption from the annual fees on the card
Points program Offered by two institutions
Cash back Offered optionally by one institution

Lines of credit
Seven out of eight financial institutions surveyed offer lines of credit with
higher limits for some targeted fields of study: only financial institution 3
does not. Healthcare professions are the most frequently targeted, with six
institutions, followed by administration and law with five institutions, MBAs
with four institutions and engineering with two lending institutions. The
lines of credit are much more generous. They range from $25,000 to $200,000
for the highest lines. The interest rates are often better. The interest rate is
often more advantageous. The only interest rate surveyed is for financial
institution 4, which offers rates from 3% to 4.5%.
Table 2-9: Main features of student lines of credit by field of study
Student line of credit by
field of study
Offered by seven out of eight financial institutions
Targeted fields of study Healthcare: Six institutions
Administration: Five institutions (mainly for
accounting)
Law: Five institutions
MBA: Four institutions
16
Engineering: Two institutions
Specific features Students must be in the program to receive benefits
of the line. If they change programs, the institution
may require repayment.
Maximum period of
borrowing
Usually four years, with a maximum debt load of
$25,000 to $200,000 according to the line of credits
annual maximum
Repayment Terms None of the lines of credit must be repaid during
study excepting interest payments.

One institution out of seven allows a grace period
of six months after the end of studies before
beginning repayment; the others allow one year.
One reduces the grace period to six months for
students who leave school without receiving a
diploma.

The length of the repayment period varies between
7 and 20 years.
Interest Rates Variable depending on the institution and
individual credit ratings.

Interest rates vary depending on the field of study.
It varies from 3% to 4.75% in financial institution 4.

Lines of credit, as we have seen, are often presented as either a
complementary or principal method of payment for university studies. Table
2-10 illustrates the distribution of debt accumulated at a financial institution,
which will be discussed in the Error! The requested resource was not found
section. Nevertheless, we can note that the vast majority of students hold less
than $5,000 in private debt. However, a small but worrisome amount of
students accumulate more than $20,000 worth of private debt 11% of
students who borrow.
Table 2-10: Distribution of debt accumulated at financial institutions
Accumulateddebt Percentage
oftotal
numberof
students
Percentage
ofstudents
withdebt
0 62%

15,000 21% 55%
500110,000 8% 20%
17
10,00115,000 4% 10%
15,00120,000 2% 5%
20,00125,000 1% 4%
25,00130,000 1% 3%
30,00135,000 0% 1%
35,00140,000 0% 1%
40,00145,000 0% 1%
45,001+ 0% 1%
Total 100%

Totalindebted
students

100%

Sales policies for these products are even more seductive than for regular products.
Financial Institution 4 describes its student line of credit as follows:
The Bank Student Line of Credit is a convenient and flexible solution for: Financing all your
annual tuition and education-related expenses. Become your only transactional source, since
it functions as a Chequing account.
Financial institution 1 emphasizes the high cost of a university education:
Who should get this loan? University students who want to meet substantial financial needs
related to their program of study.
These institutions are selling the possibility of living without financial worries and
without needing to work. And yet, as we will see later, high debt is a source of
significant stress, which students do not necessarily realize, not yet having a credit
history or models among their friends and family. The idea of paying back an entire
years expenses is also mentioned.
Financial institution 2: Pursuing a professional designation demands focus. Now,
Professional or Medical Student Lines of Credit let you stay focused on whats important
your studies.
Financial institution 7: Borrow up to $205,000 (depending on your field of study) to help
cover tuition, rent, books, school supplies, living and residency costs
1
for your program.
14

Costs associated with loans are only rarely mentioned or explained. The consequences

14
CIBC. Professional Edge Student Program https://www.cibc.com/ca/loans/prof-edg-st-pers-ln-
credit.html (consulted July 11, 2011)
18
of dropping out or changing programs are only rarely stated. The products are
marketed in a reassuring way, but incur high costs. Some protection does, however,
exist.
References (classified in alphabetical order) :
Banque de Montral. Marges-crdit aux tudiants (professions librales et mdecine) .
http://www.bmo.com/accueil/particuliers/services-bancaires/prets-et-prets- hypothecaires/prets-et-marges-
credit/marges-credit/etudiants/professions [Consulte le 11 juillet 2011]
Banque Nationale. Fonds de roulement tudiant http://bnc.ca/bnc/cda/content/0,2662,divId-2_langId-
2_navCode-17203,00.html?stab=2 [Consulte le 11 juillet 2011]
CIBC. Programme tudiants service classe professionnels CIBC . https://www.cibc.com/ca/loans/prof-edg-st-
pers-ln-credit-fr.html [consulte le 11 juillet 2011]
Desjardins. Marge de crdit STRATGIQUE tudiant .
http://www.desjardins.com/fr/particuliers/produits_services/financement/etudiant/puissance_d.jsp [Consulte
le 11 juillet 2011]
1.1.1. Protection Offered
Consumer protection laws offer minimal protection. Concerning credit cards, the
interest rate is fixed and cannot be changed without the consent of both parties; also,
the credit limit cannot be raised independently.
15

Bill 24, introduced in June 2011 (Assemble nationale, 2011), proposes closer control of
consumer credit. Some proposals may affect student credit, including the following:
Prohibiting false representation that suggests credit can improve an
individuals financial situation
Prohibiting the granting of a higher credit limit than requested by the
consumer
Imposing minimum payments on loans that will reach 5% of the full term.
We will suggest additions and modifications in the Recommendations section. It is, of
course, possible to declare bankruptcy on a private loan, contrary to a government-
funded student loan. As we have mentioned earlier, this is a painful rocess.
1.4. Public and Private Loans: Advantages and Disadvantages
Which is more advantageous, a public loan or a private loan? Public loans have some
notable advantages. They are subsidized, which means that students will pay more for
the same loan obtained privately than publicly. Table 2-12 gives an example. It is based
on the following information:
A student borrows $2,440 every September for three years, the same amount
as a public loan for a full-time university student registered for 30 credits over eight
months.

15
Office de la protection du consommateur. Credit card
http://www.opc.gouv.qc.ca/webforms/SujetsConsommation/FinancesAssurances/ContratsCredit/Cart
eCredit_en.aspx (consulted July 11)
19
The student has a line of credit with an interest rate of 4.5% and only pays
interest while studying.
Figures have not been adjusted for inflation.
Calculations were made with Desjardins simulation calculator.
16

We have calculated that this loan, which is not subsidized by the government, would
cost a total of $658.80 while the student was in school. A government-funded loan
would cost nothing in interest. Clearly, the amount of private loans can be higher or
distributed differently.

Table 2-12: Interest costs for a loan with a line of credit
Volume of
debt
Annual
repayment
Year 1 $2,440 $109.80
Year 2 $4,880 $219.60
Year 3 $7,320 $329.40
TOTAL $7,320 $658.80
Government-funded loans have other advantages, such as an interest rate that is fixed
by law, a tax credit, the deferred payment program and the debt forgiveness program.
In addition, they are accompanied by assistance in the form of substantial bursaries.

Private loans are much less advantageous. They do present two advantages, however.
The borrower can choose the amount to borrow, while it is fixed in the case of public
loans. It is also possible to declare bankruptcy. However, the interest rates are higher
and legal protection is minimal. Additionally, students with more at risk profiles will
receive less advantageous terms, which goes against a policy of social mobility. The
features of private loans explain why the government of Qubec must subsidize and
regulate student loans!

16
http://www.desjardins.com/fr/simulateurs/marge_avantage/
20
3. Impacts of student indebtedness: what the scientific literature has to
say
The FEUQ has studied the question of student indebtedness and its multifaceted
impacts on the student population many times. We must proceed with a few
preliminary warnings. First, it is difficult to distinguish the specific impact of loans.
Some studies have, but most do not. Studies rarely distinguish forms of indebtedness
and restrict themselves to a study of public student debt. Therefore, there is little
literature on student private debt due to a lack of data. The current study aims to make
up for this loss.
We will start by presenting a few methodological and theoretical elements. Then, we
will discuss indebtedness generally by painting a portrait of the situation of young
Quebeckers. Then we will present the impacts of student indebtedness based on the
university path before, during and after, and synthesize them.
3.1. Methodology
This section aims to provide a certain methodological foundation that will be useful for
what follows. We will present in sequence the notion of methodological individualism
and how it is different from the rational choice theories (RCT), the limits of the review of
literature and an introduction to foreign university systems.
3.1.1. Methodological individualism and rational choice theory
We can perceive the individual in several different ways: the behavioural analysis will
vary depending on whether the analysis is based on the balance of power in a society,
the relationship between social classes, the cognitive processes at work in the decision,
etc. This study is mainly based on the theory of methodological individualism. It is
based on two ideas: a rational and understandable decision process and postulates
limited by the actors behaviour.
We can choose to analyze the individuals behaviour in several different ways. The
postulates adopted will tinge the result: these postulates are all prescriptive, more or
less. The TCR has this fault that methodological individualism (MI) does not. Therefore,
the three postulates regarding the actors behaviour are as follows (Boudon, 2002).
1. Individualism. Any social phenomenon is a product of the individual. Therefore,
the individual actor has absolute theoretical decision-making freedom.
2. Understanding. Any human action is understandable. MI rejects the idea that
human actions are motivated by irrational and evasive motives.
3. Rationality. Any action has a meaning for the individual. The rationality at work,
however, is multifaceted: for example, an individual does not only research a
profit, but his actions may be motivated by the pursuit of happiness, the will to
start a family, etc. Thus, the rationality is much more complex than strictly
economic rationality.
Rationality needs a rational decision process. An example of such a process is the
following (Mercier, 2004, p. 149):
1. Identify values and goals to be met
21
2. Study all possible alternatives to meet these goals;
3. Research information on efficiency/the efficiency of various alternatives;
4. Compare alternatives and their consequences;
5. Choose the alternative that maximizes the values and goals;
6. Implementation;
7. Feedback.
Presenting broader postulates, MI excludes more normative RCT postulates that are
consequentialism, selfishness, and the cost-benefit calculation. It also excludes
Nietzschean analyses from its conceptual framework, which add the will to power, or
Marxist analyses, which add class interests. This also allows for an analysis that re-
examines the critiques of the classic notion of rationality (including bounded rationality
by Herbert Simon
17
). However, at times, the student will be conceptualized in a MI
context with the addition of cost-benefit calculations when necessary.
The typical application of RCT in the case of teaching gives the preconceived notion
according to which new students should operate a simple cost-benefit calculation with
the intention of evaluating the economic profitability of a given university education.
Going beyond simplistic and normative RCT postulates, the scientific literature
presented here suggests behaviour of another nature. It also raises existential cases
which tend to demonstrate the negative effects of student indebtedness on the various
levels of students paths. Therefore, it is clear that like any human being, the student is
not only a cold and calculating being that seeks to optimize his private benefits: he is a
being of flesh and blood, fuelled by various passions.
3.1.2. Difficulties encountered
The review of literature also presents some difficulties of another nature. First, it is often
sometimes difficult to separate the specific affect of student indebtedness from other
factors that play on the participation and access to university studies. The reader
familiar with the questions of access to university studies may notice certain similarities
between affects noticed here and those which we can attribute to the tuition bill, for
example. It is also important to remember a precaution that must always be kept in
mind:
[] social sciences are not exact sciences and may not lead to a unique solution or to
perfectly laid out public policies like mathematicians who would solve equations. In social
matters, problems are rarely solved once and for all. (FCBEM, 2009, p. 238)
This remark is particularly true in the case of student financial assistance systems, which
are complex machines. We cannot evaluate them by the yardstick of a single given
criterion (whether it be indebtedness, accessibility to the assistance system,
perseverance, success, etc.) without the risk of truncating reality. This study presents a
different ambition, by wanting to paint the widest portrait possible of the student
indebtedness situation in Quebec.

17
Bounded rationality affirms that the quality of a decision is limited by the quality of available
information, the deciders cognative limits and the time dedicated to making the decision.
22
3.1.3. Foreign university systems: an introduction
Canadian scientific literature, and moreover, Quebec scientific literature, is relatively
lacking in studies on student indebtedness. Even among the Canadian group, Quebec is
an exception with its own system of loans and bursaries and relatively low tuition fees.
However, when we study the foreign literature (which is much lengthier, mainly in
Great Britain and the United States), it is important to take different social contexts into
account, mainly with regard to access to education policies. Therefore, we will re-
examine the OCDEs categorization (OCDE, 2010, indicator B5), which identifies four
major types of public policies related to access educations:
Low tuition fees and generous assistance: Scandinavian countries
High tuition fees, developed public assistance: Australia, Canada, United
States, New Zealand, The Netherlands, United Kingdom
High tuition fees, little assistance: Korea, Japan
Low tuition fees and little assistance: Austria, Belgium, Spain, France, Ireland,
Italy, Portugal, Czech Republic
With tuition fees on the order of $2,500 in 2009, Quebec qualifies as a jurisdiction with
high tuition fees (more than $1,500 US) and developed public assistance. However, it is
important to note that tuition fees there are substantially lower than in other Canadian
provinces and in the United States, and that assistance is most often given as bursaries.
Access to education in Quebec is also more equal than in these jurisdictions, albeit a
recent tradition. It is important to consider when comparing Quebecs situation to that
of the United States, Great Britain or even other Canadian provinces.
3.2. Indebtedness: a general discussion
Student indebtedness is a subgroup of a larger phenomenon, that is consumer leverage.
However, it is a subgroup with very particular rules. It is not lacking in interest to
present the contours of consumer leverage according to two aspects: the relationship of
students towards credit and explicative factors of indebtedness.
3.2.1. Credit among Quebec students
Lachance, Beaudoin et Robitaille (2005) surveyed 980 young Quebec adults from 18 to 29
years of age on their credit use and consumer leverage. We will use some of these data
to paint a portrait of the indebtedness situation.
First of all, it is important to note that the use of various financial products has
skyrocketed from 1994 to 2004, as shown in Error! Reference source not found.. Credit
ard use has increased 32%. The number of students with a line of credit doubled and the
number of students with a personal loan increased significantly
18
.

18
The data from Young, 1995, do not allow us to know precisely if student loans are included in
the personal loans perimeter. Therefore, we included both types of data from Lachance,
Beaudoin and Robitaille.
23
Figure 3-1: Use of various credit products among students 18 to 29 years old, progression from 1994 to 2004

Source: Young, 1995 and Lachance, Beaudoin and Robitaille, 2005
However, this increase in the use of credit is not accompanied by enough knowledge.
Students with at least one debt had an average success rate of 4.7/9, 52.2%, on the nine-
question questionnaire on credit. The authors noticed that the main sources of
information about credit are the family (37.8%) and personal experiences (25.4%). School
takes third place with 13.1% of respondents. Attitudes towards credit are mixed with
positive and negative opinions: according to the authors, young adults who enter the
labour market, who must use credit, were thought to have a more positive perception
than young adults who live with their parents or who are students.
As for knowledge of the mechanics of various forms of indebtedness, secondary school
graduates have little knowledge. The FCBEM (2006) revealed that secondary school
students knew more about credit cards (46% reported having minimal knowledge about
this topic) than government loans (40%).
Credit is increasingly common among students, which brings about vulnerabilities with
regard to financial knowledge. In addition, financial products are increasingly common
among students as well.
3.2.2. Explanatory factors of indebtedness
Duhaime (2001) studied the situation of over-indebted households in Quebec. Following
a content analysis of 49 interviews with over-indebted Quebec households, it presents a
modeling of what he calls the cycle of over-indebtedness. The first phase, the initiation
phase, is what interests us the most. Unfortunately, the author does not study students
specifically. However, in many cases, consumption habits seem to start at the university
level: the author mentions the case of a doctor whose over-indebtedness started during
his premed.
First they note that of the households studied, 23.5% of university graduates expect to
finish their undergraduate studies with a student debt beyond $15,000, which
corresponds to the average consumer debt mentioned by Duhaime for over-indebted
0,00%!
10,00%!
20,00%!
30,00%!
40,00%!
50,00%!
60,00%!
70,00%!
80,00%!
Carte!de!crdit! Marge!de!
crdit!
Prt!personnel!
sans!prt!
tudi ant!
Prt!personnel!
avec!prt!
tudi ant!
1994!
2004!
24
households. An important number of students have a debt that could lead them into the
over-indebtedness cycle. High debt is a necessary but not sufficient condition for entry
into the over-indebtedness cycle.
The credit initiation phase starts with the beginning of independent life, outside of the
family home. The credit desensitization process begins, given the few competing
obligations with regard to indebtedness. The emergence of unexpected or inappropriate
purchases puts the subject on the track to over-indebtedness. Social pressures from the
immediate environment also contribute to making the people concerned go into debt.
Overall, three major explanatory factors may contribute to the entry into the over-
indebtedness cycle:
1. Traumatic events. Cameron and Golby (1990) validated this exogenous shock
hypothesis on two factors, namely the bankruptcy of his business and the quick
lost of income. This factor is applied difficultly to our study: students are
generally relatively young and, therefore, have less to loose. However, marked
instability in the labour market could cause financial problems, especially since
student jobs are frequently affected by economic cycles, and young workers are
often laid off during economic downturns.
2. Insufficient income. Cameron and Golby did not validate this hypothesis:
however, students are a particular group, who must necessarily go into debt in
many cases given the make-up and insufficiency of sources and methods of
financing university studies, combined with a sometimes deficient financial
support.
3. Compulsive buying, which can be expanded to psychological factors. On the
whole, students are a population with an aversion to lower indebtedness than the
general population, already indebted.
Lea, Webley and Walker identified the factors that influence the level of debt and the
propensity to go into debt (1995(, which can be grouped into social variables and
individual variables. They used this model to study the case of British consumers
based on attitudinal scales built in order to have a faithful portrait of the relationship
with indebtedness. These attitudes and situations are interpreted based on various
levels of debt: no debt, average debt and heavy debt. Overall, three factors predict
indebtedness: economic factors, financial management style and economic socialization.
The others dont have an independent impact on indebtedness.
Social variables
Economic and demographic factors. The fact of being a woman, working part
time, being a housewife or being unemployed all foretells high indebtedness.
Weak incomes, being a renter and the number of children as well. The level of
poverty foretells more than half of the cases being assigned to one
indebtedness group or another.
Social support for indebtedness. Translate public attitudes towards
indebtedness. Indebtedness is more tolerated than it was at the beginning of
the twentieth century. In addition, credit is clearly more wide spread, as we
mentioned earlier. Indebted consumers believed that their circle would be
more understanding overall.
25
Economic socialization. Variable which translates family attitudes towards
indebtedness: people from more well off families seem to have less aversion
towards indebtedness. This is the third major factor.
Social comparisons. Identifying the consumer in a social group that is not his:
this means that he might live beyond his means. The survey confirms it.
Individual variables
Financial management styles. Seriously indebted people are often unable to
suitably manage their finances. After economic factors, this is the most
important factor.
Consumer behaviour. Consumers who classify unnecessary expenses as
necessary expenses are more at risk of going into debt. In the study, they say
that Christmas gifts, cigarettes, the car and the telephone are priorities.
Time horizon. Consumers who live on short time horizon risk not being able
to delay acquiring property.
Attitude towards debt. A positive attitude towards indebtedness is usually
correlated with higher indebtedness. However, the study does not confirm
this.
Control centre. People who feel that they have their financial situation under
control risk a lower instance of indebtedness. However, the study does not
confirm this.
Therefore, we see that the situation of indebtedness among students is not necessarily
easy. With little knowledge about sound credit use, they can easily fall into the over-
indebtedness trap. Financial products designed for them, in which the principal does
not have to be repaid during their education, and which tends to encourage a social
comparison that leads to over-indebtedness, creates unnecessary risks and problematic
indebtedness situations.
3.3. Preventing the start of university education
The first category of effects that we will study is the affects that are noticeable before
education begins. Student indebtedness delays students starting university.
Multifaceted risk is perceived differentially depending on social origins. Families play a
crucial role in reducing the aversion to indebtedness, and the costs tend to guide student
choices irrespective of more academic volitions.
3.3.1. Delaying or preventing students from starting university because of financial
constraints
Julie Dubois, of Human Resources and Skills Development Canada (HRSDC), studied
the trends of indebtedness of students in the Canadian Student Grants Program (CSGP)
in 1990, 1995 and 2000 (Dubois, 2006). A government loan increases the probability of a
student delaying education 1.5 times among men and 1.8 times among women. A
parental contribution or employment income do not have the same impact. It is also
interesting to note that the effect is much lower in college (1.2 times among men, 1.1
26
times among women, where the amount of debt is generally lower and the path is
shorter.
19
Such a delay may be explained in several ways, among others, by the different
profile of students who take out a government loan (students from less well off families,
for example). Combined with the debt aversion phenomenon, which we will see later
we can presume that a considerable portion of students who wind up taking out a loan
delay beginning their education in order to create a personal savings to limit the need
for other forms of indebtedness during their education.
A delay in beginning a university education decreases both a diplomas individual and
collective profitability, by decreasing the portion of the new graduates active life, in
which he contributes fully to the governments finances by paying higher taxes, a
generally recognized result of a salary associated with a university education.
Other data are troubling. Several surveys from Statistics Canada ask secondary school
graduates to give their option about the reasons that prevent them from attending
university (FCBEM, 2004, p. 103):
Financial obstacles represented the main reason for not attending university
among 20% of respondents (according to the survey of graduating students
and the study on postsecondary study participation);
Financial obstacles are cited by 36% of secondary school graduates who are
not enrolled in postsecondary studies as one of the factors that prevents them
from currently pursuing a university education, or from pursuing one in the
future (data from the JET).
Financial reasons, including, first and foremost, the likelihood of taking out a major debt
without the assurance of being able to repay it, seems to play a major role. More
psychological factors are also involved.
3.3.2. A multifaceted risk: debt aversion and sticker price
Archer and Hutchings (2000) suggest that the perception of risk varies based on social
class. Therefore, British working class students have a greater aversion to risk than their
counterparts from other social classes. The authors postulate that building the notion of
value, that is questions of risk, cost and benefit, varies based on social class. In focus
groups led with 109 British working class students, the others observed that the
discourse on diploma profitability was firmly established and, in every respect, seemed
to be the prevailing discourse. However, access to university was often qualified as
being a risky business, on the financial level (high opportunity cost) and academic level
(risk of failure). Later on, we will see that student indebtedness is correlated with
dropping out.
Two other mechanisms are at work. First, students from economically disadvantaged
environments present, in their conception of the cost of studies, a propensity to over
estimate the cost of education and under-estimate its profitability.

19
In 1995, average debt was $9,500 at the college level and between $12,500 and $14,000 at the
university level, depending on the level of studies (Finnie, 2001 in Dubois, 2006).
27
Table 3-1: Canadians perceptions of costs and benefits of university education in 2003

Source: FCBEM, 2004, p. 109
The rationality would, therefore, be modulated based on socioeconomic origins
(Vossensteyn, 2005, in FEUQ, 2010). Finally, according to Finnie and Laporte, the
aversion to indebtedness might dissuade 2% of Canadian youth from pursuing a
postsecondary education. We note, however, that other financial reasons may be at
work: the proportion of students completely put off by the possibility of going into debt
is significant, by considering the attractiveness of university education.
Callender and Jackson (2005) studied the aversion to indebtedness and the perception of
university diploma profitability based on social class among British students. Like
several authors before them, they conclude that the aversion to indebtedness is greater
among the more disadvantaged, even if the perception of costs and benefits of
university is similar based on social class.
The question of general effectiveness of loans for education was explored in 2007 by the
FEUQ as part of a feasibility study on the conversion of loans offered by SFA into
education bursaries. Citing St-Jean (St-John, 1990 in FEUQ, 2007), the author affirms that
the increase in assistance awarded, as loans did not help access to education for more
disadvantaged students, the ones who are Student Financial Assistances main
customers.
3.3.3. The role of the family
Family plays an important role in access to education. We placed it at the beginning of
university studies: however, we will see later that family plays a role in every step of the
university path. In this respect, changes in student debt policies may have varied
impacts. Christie and Munro studied this question in England. The relation to student
indebtedness is modelled as follows (Christie et Munro, 2003):
Table 3-2: Individual attitudes towards the financial costs and benefits of university education
Aversion to
indebtedness
None Average Strong
Perception of
economic
profitability
None B A
Average E D C
Strong
These are ideal types that are only rarely found among the population. We find
economic profitability on one side and aversion to indebtedness on the other: this is a
modified form of the cost-benefit calculation, in which costs are mainly indebtedness
and benefits, the perceived economic profitability.
28
The authors studied the British case, questioning 49 students from two universities.
They identify three types of students, mainly based on their relation to indebtedness:
those who avoid debt (type C), those who are neutral (type D) and those who choose to
go into debt (type E). Types A and B were not in the authors samples, and we can
assume that they do not attend university, given that they do not perceive any economic
benefits of doing so
20
. We also note that the British loan system is qualified as income
contingent, contrary to Quebecs system, and often presented as a cure-all to lift the
negative affects of student indebtedness.
The increase in the cost of education has varied impacts on students and their place on
the aversion to indebtedness scale is made based on family characteristics, according to
two criteria: their financial resources and their cultural capital
21
. These two resources
condition entry into university. According to the authors, the strong presence of these
resources tends to change the students aversion to indebtedness:
Students who consciously choose to go into debt simultaneously benefit from
higher cultural capital and financial resources. Their parents value education
and help their children financially. Among them, access to university
education is considered the natural next step and students have access to a
plethora of information about university education. These results are
consistent with those of Callender and Jackson (2005), who present perceived
profitability and family encouragement as major attendance factors. Both
family capital factors are fully at work here and encourage the pursuit of
university education.
Those who are neutral towards indebtedness have more limited family
capital. Information on the actual costs of attending university is fewer, which
makes the amount of accumulated debt during university more of a shock.
Here, debt is not a choice but an obligation that we must try to limit.
Those who are impervious to indebtedness generally receive a major parental
contribution, which allows debt to be avoided. Others have had difficult
experiences with their circle with respect to indebtedness, which makes them
impervious. Their studies are not necessarily economically profitable. To this
category, we can add students from low-income families, who are more likely
to have an aversion to indebtedness, which will determine access to university
education or not, according to Callender and Jackson (2005).
The authors conclude that we must proceed with caution in changing student debt
policies. It is imperative to consider parents financial resources and the often under-
estimated costs associated with entering university. Merani et al (2010) demonstrated so

20
Other models might come to different conclusions: this one is particularly simple, even
simplist, because it only considers one form of rationality. The choice of studies, as we will see
later, is a complex multifaceted dynamic.
21
Concept of Bourdian sociology. Pierre Bourdieu analyzes contemporary society based on the
field theory, in which the accumulation of capital given is the condition of entry and progression
in a given field, which is its own microcosm of the social world that is governed by certain laws.
In this case, the entry and prosperity in the university field is conditioned by the acquisition of
cultural capital (promotion of education, effort) related to the university field.
29
for medicine. Comparing sociodemographic characteristics of Quebec and non-Quebec
medical students, the other points out the imposition of a tuition fee system and very
high student debt in the other Canadian provinces had profound, regressive impacts on
programs social make-up: it was 1.22 times more likely for a Quebec student to come
from a low income environment.
Combined with the conclusions from Archer and Hutchings and Callender and Jackson
(2003), we can infer a varied influence on the impact of student debt, depending on the
typology in table 3.1.
Students who favour fields with little economic profitability risk quitting their
studies if the assumed economic benefits are lower than the costs (type A).
Students from more disadvantaged families tend to under-estimate the
benefits and to over-estimate the costs; as loans increase more, the more they
risk not attending university (type B)
Students from more advantaged families see access to education as natural
and will go into debt without concern (type E) or have the required financial
resources to avoid student indebtedness (type C)
Students from more modest families, but who pursue education, are
somewhat sensitive to student indebtedness, and will try to limit it as
possible, while demonstrating a current but limited capacity of indebtedness.
The student risks under-estimating the actual impact of the debt from their
education. (type D)
Others are straight out allergic to debt for various reasons and will avoid it at
all costs (type C). Such is the case with students from low-income families,
who have a strong aversion to indebtedness (Callender et Jackson, 2005).
Students with a negative attitude towards indebtedness (types A, C and D) are more at
risk. In a British study, students who had more positive attitudes towards indebtedness
were 1.25 times more likely to study at university (Callender, 2003 in Pennell et West,
2005). And yet, these students usually present a more advantaged family profile.
Therefore, a high indebtedness policy tends to stratify education based on social classes,
even before entry.
3.3.4. Guiding the choice of field or area of study
If we go back to the typology of Error! Reference source not found., certain students
present greater vulnerability than others. Their family capital is relatively low, and their
aversion to risk is high. Type C and D students, those who present an average to high
aversion to risk, are particularly vulnerable. In a classic analysis based on RCT, we
should theoretically expect students to choose fields of study with high private
profitability, instead of less profitable fields, which are more research based (such as
social sciences see Government of Quebec, 2008 in FEUQ, 2010a, p. 69 for rates of
private profitability). Since these fields are perceived as less profitable, it is possible that
the students academic path, which presents a greater aversion to risk, is changed for
reasons that go beyond the students deep motivations. Others will simply not pursue a
university education due to a lack of economic motivation to pursue them in the field of
their choice.
Such an explanation allows us to understand why certain others have not made a
30
Such an explanation allows us to understand why certain others have not made a
connection between indebtedness and the duration of doctoral studies in the United
States (including Kim and Otts, 2010): the selection was made upstream and students
who reach doctoral either have a low aversion to indebtedness because of their family,
or others, capital, or because they have sufficient resources to fulfill their project.
However, such results contradict those observed by the CNCS-FEUQ (2001) in a study
on the actual duration of graduate studies. Indebtedness is not always directly related to
duration of studies: it is also the propensity to work during university, brought about by
paid work, which causes studies to be extended.
In this respect, medicine offers an interesting case. A highly specialized and
academically demanding profession, it is often accompanied with colossal bills and
debt. It is also a very strategically important field for the states. However, several
studies report that the very high levels of student debt incurred by medical students
tend to influence career choices. For example, Rosenblatt and Andrilla (2005) reported a
certain, but modest, influence on student debt with two aspects. The influence is
negative with regard to the likelihood of practicing family medicine, usually less valued
and definitely less profitable. However, the influence is neutral with regard to the
likelihood of going to study in underserved regions, which would be explained in
various ways, including sociodemographic characteristics of respondents as well as the
existence of incentive programs to practice in underserved rural regions.
However, the results are partially contradicted by Callender and Jackson (2008), who see
an impact of the aversion to indebtedness on the choice of teaching institution for the
most disadvantaged and middle class, and on employment possibilities during studies
for the most disadvantaged. With universities located far from urban centres and
offering a more limited range of programs, we may, therefore, assume that a financial
constraint exists with regard to the choice of field of study (we will come back to this
later). We must also mention that the others note that the perception of the costs and
benefits of education have an influence in choosing more profitable fields of study. The
economic effect on students is different: rather than choosing a more profitable subject,
at-risk students prefer to stay at home longer or to choose a less expensive university. In
extreme cases, this could create a two-speed university system, for example in the
context of a deregulated university system or differentiated by institution or by field of
study. However, we note that the study examines the entry to university education; it
does not give indicators on abandoning studies based on aversion to indebtedness. Debt
tends to create dropouts, as we will see later. It is possibly at this stage of the path that
the greatest effect is felt by students who choose fields based on the economic
profitability of a given diploma.
The differentiation of tuition fees, mentioned earlier, has had detrimental impacts on
social mobility, reducing the poorest students participation at Guelph University by
40% (FCEE 2001 in FEUQ, 2010a). The participation of students from families making
less than $60,000 dropped from 35.7% to 14.9% between 1997 and 2000 (ACPPU, 2001 in
FEUQ, 2010a). We can expect a similar impact if indebtedness profiles differ
significantly between fields of study. We already see it (see section 1.4.4) at the graduate
level, which increases the burden of debt in many cases. Fortunately, Quebec does not
have any differentiation of tuition fees or burden of debt.
The choice of institution is an interesting variable to study. Frenette (2007 in FEUQ,
2009a) studied the question of studying far from home. Costs associated with a move for
31
university are high, which increases the risk taken on by the student. It is here, with
much more acuteness, that we notice the influence on the choice of postsecondary
studies that will be pursued. A typical student in a far away region may have six typical
choices:
1. Not to pursue postsecondary studies
2. Complete a diploma of collegial studies and enter the labour force directly
3. Complete a distance program.
4. Complete a program in a delocalized institution (see FEUQ, 2008) however, the
program offering is usually limited
5. Complete a program in a regional city (Saguenay in the Saguenay-Lac-Saint-Jean
for example), and incur either high transportation costs (car maintenance), or
moving expenses for the entire duration of the program, for training that remains
somewhat limited (FEUQ, 2009 b).
6. Complete a program in a major urban area (Montreal, Quebec City or Sherbooke
in some cases), which offers a wider range of programs.
A student in a major area is usually only faced with choices 1, 2, 3 or 6, with very
different costs. The choice of attending university is made easier by lower costs, a direct
consequence of the slightest requirement to relocate.
3.4. During university: a break
It is usually believed that student indebtedness hinders academic success. It tends to
create dropouts, especially among the most disadvantaged, both directly and indirectly,
by generating a need to provide for oneself by working rather than indebtedness. The
psychological impacts may be great, and a major debt harms access to graduate studies.
3.4.1. A source of dropping out for at-risk students
Several studies have observed that student loans have negative effects on academic
success. Don bin Kim (2007) studied the impact of total indebtedness on program
completion. To do so, she studies the first-year loan amount and the likelihood of
academic success. The author drew troubling conclusions:
Each $1,000 annual loan increase has a greatly varying impact based on
income. We observe an increase in dropouts of 1.6% for each $1,000 loan
increase for students from the most disadvantaged families (Kim, 2007, p. 85).
The relationship exists, but is much less marked among the middle class,
where as high income students are positively influenced by a loan increase.
Increasing student loans tends to be a regressive policy.
The graduation rate decreases as loans are increased among students from
minority groups (Asians, Hispanics and African Americans); the opposite
effect is seen among white students, usually more advantaged. African
Americans are the most disadvantaged.
We also highlight that previous studies show a negative link between loans and success
based on ethnic origin (Finke, Porter et Dub rock, 2000; Thomas, 1998 in Kim, 2007) or
family income (Johnson, Mont Marquette et Deckle, 2003).
32
From a more general point of view, student financial assistance usually has a positive
impact on success. However, research tends to demonstrate that loans are a less effective
form of assistance than bursaries, which are non-refundable. (FCBEM, 2009, p. 78)
Therefore, we can conclude that relying on a high debt system tends to stratify
education, by discriminating against students from more disadvantaged families or who
have an unfavourable attitude towards postsecondary education.
3.4.2. Avoiding indebtedness by paid work
First, remember that paid work time has increased steadily since 1976, as shown in a
recent FCBEM study:
Table 3-3: Evolution of the rate of employment for Canadian students enrolled in full time studies between 1976
and 2008, September to April

Source: Mote et Schwartz, 2009, p. 3
The authors give the hypothesis that the major increase of tuition fees in Canada is
probably related to this increase in the frequency of paid work. We may operate a
similar reasoning with student indebtedness, and scientific literature tends to say we are
right.
First, we mentioned earlier (section 1.3.3) that certain students, especially those with
relatively low family capital, fear student indebtedness and tend to limit it. Various
strategies may be implemented. Later on, we will look at the characteristics of university
students in various regards. When we analyze the sources of student financing,
however, we are led to believe that they are limited (see FEUQ, 2010b) :
Public assistance (SFA);
Paid work;
Family contributions;
Excellence bursaries;
Private debt.
33
The student who wishes to avoid going into debt has little recourse, unless he increases
his number of paid work hours. The phenomenon may be observed in practical terms.
Pennel and West (2005 : p. 133 in FEUQ, 2007) highlighted the existence of a link
between indebtedness and paid work: students work to avoid going into debt. This paid
work tends to have two impacts: one on academic success, similar to the one observed in
Canada by Motte and Schwartz (2005), the other on the duration of studies.
3.4.3. Often deep psychological impacts
Student indebtedness has negative effects on psychological health. Leonard (1995)
studied the behaviours of Sociology students at Queens University in Belfast. We learn
that 25% of students lived with anxiety caused by indebtedness which created negative
stress during education, lowering academic performance.
This anxiety tends to increase based on debt. Merani et al., (2010) also studied the notion
of financial stress. Quebec medical students, who have a median debt three times lower,
had financial stress that was 0.43 times that of their counterparts in other provinces in
2007. This proportion was 0.60 in 2011, although the relationship between Quebec
median debt and median debt outside Quebec was similar, but the median debt outside
Quebec was much lower ($50,000 compared to $90,000 in 2007).
Psychological impacts are varied (Scott). However, several observed an increase in the
likelihood of depression for highly indebted graduates, and an increase in the likelihood
of depression when confronted with a high level of financial difficulties during the
course of a students education (Andrews et Wilding, 2004);
Effective financial assistance, which minimizes recourse to loans, improves the quality
of education in another way: it allows for a better commitment to education both in
academic and social dimensions (Hossler, 2008 in FCBEM, 2009). An increase in student
commitment tends to improve academic success.
3.4.4. Abandoning graduate studies
Pursuing graduate studies is a crucial issue. It would seem that students who pursue a
Masters are less indebted than those who join the labour force directly an average
difference of $3,200 (PRA inc., p. 17). Therefore, there seems to be a relationship.
Millett (2003) studied the impact of debt acquired during undergraduate studies on
admission to a graduate program (Masters or Doctorate) in 1993/1994. The results are
eloquent. Graduates without debt applied to a graduate studies program at a rate of
58.7% compared to 46.1% for the most indebted ($15,000 and more). Those with debt
between $100 and $14,999 applied at a proportion from 46.6% to 50.6%. Statistical
analyses conducted confirm the effect: students with a debt between $5,000 and $10,000
had 1.6 less of a chance of applying to graduate studies: the ratio is 1.4 times for debts
between $10,000 and $14,999. The interest of this study, beyond its unequivocal results,
is that it refutes several preceding studies with contradictory results with a much wider
sample as well as robust statistical methods.
3.5. After university
Student debts also had an impact when the student graduates. Repayment and the
default rate are a highly studied dimension, and common debts have a major impact on
34
ex-students finances, who must devote a major portion of their revenues to pay them
off.
3.5.1. Impacts on the government and ex-students: repayment of student debts
Constantine Kapsalis studied the factors influencing the repayment of student loans
(Kapsalis, 2006). Although 24% of Bachelors students who applied to the Canadian
Student Grants Program (CSGP) had trouble repaying their debt. Personal revenue
appeared to be the main factor for defaults of payment. The researcher concluded that,
in fact, two factors influenced default.
First, future income is directly correlated with the rate of default: the higher the revenue
is, the lower the rate of default. She stated that every $1,000s of income decreased the
rate of default by 1.2% (Kapsalis, 2006, p. 10).
Second, high debts cause a considerable jump in defaults of payment. Whereas the
default remains somewhat stable for debts between $0 and $20,000, the rate of default
increased by a spectacular 20 points for debts owed to the CSGP of $20,000 and more.
However, it is important to note the limits of the research: first, its perimeter of student
indebtedness remained very restricted, essentially limited to the federal student grants
program (CSGP). Therefore, Quebec is not included, just like provincial student loan
programs. We also note that, in fact, private loans are excluded. The data are also very
old, with the most recent dating from 1997 for consolidations made in 1994-1995.
Therefore, the more a debt is high, the more the program costs will also be high. We
estimated that a student loan awarded in 2009 cost about 17 cents per dollar loaned by
the provincial government in terms of administrative fees, interest fees and loans that
were simply lost
22
. In return, civil servants from Human Resources and Skills
Development Canada estimated the same management cost at 30 to 40% of the loan
value (FCBEM, 2009, p.168). Hence, the more the size f the students debt increase, the
more we can expect an increase in defaults of payment on loans, which leads to
considerable fees
23
for the government, as well as considerable impacts on the new
graduates finances.
Other studies examined defaults of payment. Such is the case of Laure Greene Knapp
and Terry G. Seaks, who studied the situation in the United States (Greene Knapp et
Seaks, 1992). They determined four significant variables in the study of loan defaults:
family situation, graduation, parental income and race.
The family situation had an impact. A united family decreased the default rate
by 2.7%. Family income has a significant, but low impact: each $10,000 of
income decreased the rate of default by 2%. Thus, students from
disadvantaged environments are empirically in a more difficult financial
situation after their education, which may be explained in several ways. The
American university system remains largely unequal and it is not easy to
make comparisons with the situation in Quebec. However, we can expect

22
FEUQs calculations confirmed by SFA civil servants.
23
See section 2.2.3 for various repayment mechanisms in Quebec
35
similar behaviours in Quebec: parents may act as financial support for a new
graduate starting their career.
Receiving a diploma improves the default rate by 10%. Dropouts obviously do
not have access to employment so doors are opened by university diplomas.
The fact of being or not being African American negatively impacts the rate of
default by 10%. African Americans have particularly serious economic
situations.
This being said, the authors reject the idea that institutions of higher education cause
situations of default that are worse than others. Rather, they prefer to note the
importance of student characteristics.
Wilms, Moore and Bolus highlight that in California, personal characteristics like
dropping out, ethnicity, citizenship status or family income tend to influence the rate of
default.
Flint (1997) presents a different interpretation in which the fact of being older, male or
black all tend to increase the rate of default, along with available income. Other
characteristics, he says, do not have an impact. Hence, he suggests exploring borrowers
psychological motivations in a default of payment situation. It is also important to note
that this study concludes that loan payment is first and foremost a question of money.
However, the amount of debt does not seem, to have an impact on the existence, or lack
thereof, of a default of payment in this case.
Other studies are more adapted to the Canadian situation. Questions from the National
Graduates Survey (NGS) evaluate if students had had difficulties repaying their student
loans three years after graduating. At the university level, 12% of Bachelors graduates
had less debt (less than $10,000), 22% of respondents with an average debt ($10,000 to
$24,999) and 43% of respondents with high debt ($25,000 or more) had difficulty
repaying (FCBEM, 2009, p. 219), contradicting the previous results.
We can draw a general conclusion of the analysis of literature on defaults of payment.
Although elements like adequate financial information or the improvement of certain
administrative procedures generally seem interesting to public decision makers (see,
among others, the Committee of experts on the ways of repaying student debt, 1997),
the question of the default rate is mainly based on the graduates income and the
amount of the debt. It is, therefore, upon final analysis, a question of money.
Default of payment, however, is an extreme situation. Evaluating the impacts of student
debt requires going further, by studying the finances of new graduates.
3.5.2. Impacts on new graduates finances
May Luong, in a recent article published by Statistics Canada, explores some of the
impacts of student debt on new Canadian graduates (Luong, 2010). The presence of a
student debt has several impacts, most of them negative; we do not see impacts on
income, which is the only variable not influenced. First, the savings rate is greatly
affected: 47% of non-borrowers had a savings compared to 39% of borrowers. Property
is also negatively affected, with the likelihood of owning a home being 53% compared to
60% for non-borrowers.
36
The most probing impact is on the wealth of student borrowers and non-borrowers. The
amount of total debt remains relatively stable between borrowers and non-borrowers.
However, the nature of the debt varies, which implies that the average asset of non-
borrowers [$106,300] is clearly much higher than that of borrowers [$60,700].
What can we infer from these results? First, a considerable impact on economic growth.
If total household debt remains stable, the more a student debt increases in frequency
and in size, the ability to borrow decreases. There is also a major iniquity in
postsecondary education results, which does not translate in as direct of a manner as we
would believe. The social characteristics of borrowers and non-borrowers, and first and
foremost, their social origin, will prove to be very different. Student debt harshly affects
social mobility in postsecondary studies, by greatly reducing the strictly economic
profitability of a specific category of students: those who need a student loan to attend
university. The more student indebtedness increases [whether it be by loan cap
increases or a decrease in the level of assistance, which forces students to turn to private
loans], the more we create two categories of students: the more privileged one and the
other that repays its student debts.
It is also important to note that the history of credit almost always plays a role in
awarding loans. A student confronted with payment difficulties will have his credit
affected, lowering his possibility of obtaining a loan for other purchases.
3.6. Modelling of the impacts of student indebtedness
After examining the impacts of student indebtedness on various facets of student, it is
appropriate to compile the major impacts, which we will do in Error! Reference source
not found.. Remember the main determining factors of debt: in order they are financial
situation, financial management abilities and economic socialization. Students depend
on them more and more, despite their often-lacking financial training.
Tableau 3-4 : Impacts of student indebtedness
37
Tableau 3-4 : Impacts of student indebtedness
Before During After
Students with a loan have 1.5 to
1.8 more likelihood of starting
university late [Canada];
Financial reasons are the main
reason for not attending
university [Canada]
Aversion to debt: variable based
on social class.
Future students are sensitive to
prices as shown
The family [financial resources
and willingness to help the
student] play a major role
Guide the choice of field or
region of study based on the cost
and not the talent
Kim: +$1,000 in annual loan
causes the dropout rate among
the most disadvantaged to
increase by 1.6%, with a greater
impact among cultural
minorities. [United States]
Loans are less effective than
bursaries in terms of access
Students avoid debt through paid
work, which causes them to drop
out.
Increase in stress and
psychological distress based on
the amount of debt.
Abandoning graduate studies

The default of payment is based
on income on the labour market
High debts increase the rate of
default.
The more the rate increases, the
more the costs of managing loan
programs increases.
In Canada: 12% to 43% of
Bachelors graduates have had
payment problems based on the
size of their debt.
Student debt significantly
reduces graduates financial
flexibility.
Some theoretical conclusions must be made. First the student, like any human being, is not
purely rational. Although a rational actor, his decision to attend university or not is
influenced by numerous factors including family environment, available financial
support and aversion to indebtedness. The cost-benefit calculation for postsecondary
education generally favours participation, but the benefits are difficult to estimate. The
cognitive processes during university education play a role, and indebtedness may act
as a catalyst in a decision to drop out of university.
We also note that new graduates are confronted with multiple challenges. Beyond
joining the labour force, those who are in debt must repay their loans with a
considerable possibility of defaulting on the payment and having a less profitable
diploma than expected.
Finally, student indebtedness, particularly public debt, raises an important question of
social fairness. Is a social policy that puts the most disadvantaged students into debt and
the most at risk first fair? We will re-examine this in the conclusion.

39
4. Methodology
This section aims to present the methodological parameters of the quantitative study
presented in sections 5 to 8. For more detailed statistical analysis, the reader is invited to
consult appendix II, which includes the gross tables.
4.1. Statistical methods
All the data have been analyzed in the following manner:
Descriptive statistics
Presentation of raw data. The continuous variables were converted into
ordinal variables for the presentation of data.
Analysis of the indebtedness rate
Presentation of the raw data by crossed table. The continuous variables were
converted into ordinal variables, and certain ordinal variables were
manipulated for the presentation of the tables only (groups of values
distanced from the median).
Analysis of the statistical correlation. In the case of dichotomous variables, we
have used a phi correlation (). We should note that the coefficient of this
correlation is a bit different from the others: its highest limit is determined by
the distribution between the two possible positions of the dichotomous
variable.
In the case of continuous variables, we have used the Pearson correlation
(noted r
pb
), which efficiently approximates the point bi-serial correlations,
which are the most appropriate, but the SPSS do not calculate directly. In one
case, we have used the -c of Kendall.
Analysis of the moment of the debt
Presentation of the raw data by comparison of averages. The continuous
variables have been converted into variable ordinals for the presentation of
tables.
Analysis of the statistical correlation. We have used the Pearson correlation
(r), which can be raised to the square to illustrate the amount of the variance
explained by the correlation.
However, the tables adequately present the fact that the relations are not
always linear, or distributions that are not necessarily normal. In such cases,
we will note it in the analysis. Furthermore, we have used the rho of
Spearman (), which is non-parametric. His interpretation differs somewhat
from the Pearson correlation, but it consists of the same type of test.
For the purposes of the analysis of correlation, we have retained three levels of
significance, namely p<,001 (***), p<,01 (**) and p<,05 (*). The most interesting data are
presented in the following sections. Unless there is a contrary indication, all the
significance tests are bi-directional tests. The correlation rates include two significant
data: the sign, which indicates the sense of the relation, and the value, which indicates
40
the force of the relation. The more we approach 1 or -1, the stronger the relation.
However, it is difficult to establish a universal rule for the force of correlations. As you
will see, most of the effects are significant, but relatively weak. It is important to note
that a significant statistical effect is not necessarily interesting: it only essentially signals
the fact that the difference is not due chance. This rule is particularly true for large-size
samplings: this explains why we find more statistically significant correlations for
indebtedness due to SFA for indebtedness than for indebtedness to the family. x
4.2. Sampling and representativeness
AS the methodological report of the survey mentioned, the data gathered trace a
representative portrait of the Quebec student population. However, it presents a certain
number of general and specific limitations that we have to take into account. We first
have to note that certain educational institutions refused to participate in the survey:
this consists of McGill University and HEC, for the population pool that concerns us.
However, the final data remain closed to the observed reality. There is also a specific
limit to the use of data of the survey. In fact, the questionnaire was distributed to
students during studies. Thus, the data does not entirely account for the real situation of
students, but rather their projection of their financial situation when they leave with
their diploma in hand. There may thus be a certain bias in the data. The data on total
indebtedness as well as indebtedness to SFA are both reliable, corresponding to official
data. We can thus proceed with confidence.
We worked on a subsample: we only analyzed data for students enrolled full-time
(n=9006). When we work on the entire sample, the margin of error is 1%, 19 times out of
20. Of course, the more we descend into subsamples, the higher the margin of error
becomes. In the case of very small samples, warnings are noted. The high rate of
response allows us to affirm that the study is representative of the Quebec university
population.
We also have to note that the methodology calculation differs from the one used in the
survey. In fact, we are using the SPSS program and there exist slight differences in the
weighting of the data, which can explain some marginal differences in the average and
median values. Other differences can emerge from missing values, which can vary the
average amounts of the debt in the crossed tables according to the control variables.
Other differences can emerge from the missing values, which can make vary the average
amounts of debts in the crossed tables according to the control variables.
4.3. Treatment of the data
All the data were treated according to three methods. First, we examined the variation
of the rate of indebtedness. The variables were derived from the projected level of
indebtedness at the end of undergraduate studies: this is a dichotomous variable. The
rate of indebtedness is defined as the percentage of students that expect to leave their
studies with a debt from one of three identified sources, or simply that expect to leave
with at least one debt for the total rate of indebtedness.
Subsequently, we have analyzed the variation of the amount. To be counted in the
calculations, a student must present at least one dollar of projected debt at the end of
studies: this method of calculation allows us to determine the variation of the amount of
debt for the population that is already indebted.
41
Finally, at the end of each section, we present a table summarizing the statistical tests.
Table 4-1 presents an example of a table.
Table 4-1: Example of table summarizing statistical tests
SFA Private Family/friends Total
Duration
Rate;r
pb
,05*** ,06*** ,01 ,04***
Amount;r ,04** ,09*** ,07** ,06***
Amount; ,07*** ,12*** ,01 ,10***
The first column includes the variables studies in bold: here, it is the duration of studies.
Next, the line Rate; r
pb
indicates that the line presents the variation in the rate of
indebtedness by using the point bi-serial correlation. Amount; r indicates that we are
analyzing the variation of the amount of indebtedness. Finally, Amount; indicates that
we are analyzing the variation of the amount according to of Spearman.
The other columns present the coefficients of the correlation according to three forms of
student indebtedness retained, and total indebtedness.
Various tables were simplified to facilitate comprehension. Statistical appendices in
Excel format are available from the FEUQ for the reader who wishes to deepen his
understanding.
42
5. Student indebtedness: main characteristics
What are the main characteristics of student indebtedness in Quebec? To answer this
question, we will exploit the data of the survey on the living conditions of
undergraduate students. As mentioned in the introduction, we will only focus on full-
time undergraduate students. We will mainly evaluate the percentage of students that
indebt themselves as well as the volume of this debt.
5.1. Characteristics of the rate of indebtedness
A high percentage of students must indebt themselves to successfully complete their
study project. Error! Reference source not found. illustrates the rate of indebtedness
based on the source. As we can easily see, it is the Quebec student financial assistance
program that is the most frequent source of indebtedness, with 47%. However, 35% of
students expect to leave their studies indebted to a financial institution, and 17% expect
they will owe money to their family or their friends. In total, 65% of students expect to
leave their studies with at least one debt.
Figure 5-1: Rate of indebtedness according to source Figure 5-1: Rate of indebtedness according to source

5.1.1. The number of sources of debts
A debt often comes alone. 36% of students only present one source of debts. However,
22% cumulate two, while 7% cumulate three distinct sources of indebtedness that we
have identified.
43
Figure 5-2: Number of sources of debts

The fact of cumulating various debts from several different sources influences in an
important manner the amount of the debt. On the one hand, the total debt grows with
the number of debts (r=0,368 to p <0,001), which is logical and normal. The different
forms of debt behave in a more or less erratic manner. The debt to SFA grows with the
number of debts, but its growth is weak (r=0,063 to p<0,001). The private debt remains
stable regardless the number of debts, not showing any significant statistical relation.
The family debt presents a more interesting motive: its amount weakens according to
the number of debts (r=-0,172 to p<0,001), indicating that it consists of a debt of last
recourse, which does not obey the normal financial rules. Once again, we can observe
that indebtedness leads to indebtedness.
Figure 5-3: Variation of the level of debt according to the number of sources of debts

!"!!!!#!!
!$!%%%!!#!!
!&%!%%%!!#!!
!&$!%%%!!#!!
!'%!%%%!!#!!
!'$!%%%!!#!!
()*! +,-./! )01-223401-5! 67802!
&!93:3!
'!93:35!
;!93:35!
44
5.1.2. Characteristics of the volume of debt
The amounts of debts contracted are often quite high, as is shown by Error! Reference
source not found..
Table 5-1: Characteristics of debts by source Table 5-1: Characteristics of debts by source
Average 12 785 $ 8 043 $ 5 234 $ 13 967 $
1
st
quartile 6 000 $ 1 500 $ 1 000 $ 5 000 $
Median 10 000 $ 5 000 $ 3 000 $ 11 000 $
3rd quartile 18 000 $ 10 001 $ 6 000 $ 20 000 $
Obviously, it is in student financial assistance that we find the highest level of debts,
with an average debt of $12,785 upon completion of studies. This average debt is higher
than the median debt, at $10,000, indicating that a distribution with a low concentration
of relatively small debts. This is explained by the functioning of the loans and bursaries
program, which grants loans based on the maximum loan limit. Furthermore, a student
that benefits from loans and bursaries will generally benefit from this throughout his
studies path, within the limits imposed by the program. We must also note the existence
of a relatively high indebtedness limit, which limits the loan maximum. A detailed
study of the sources of indebtedness turns out to be more troubling, by revealing the
existence of an important minority of strongly indebted students.
Error! Reference source not found. illustrates the distribution of debt amounts due to
SFA. 47% of the recipients will owe between $5,000 to $15,000 to the program at the end
of their undergraduate degree. It is nonetheless worrying to see that 29% of students,
nearly one-third, expect to owe more than $15,000 at the end of their undergraduate
studies, and 10%, more than $25,000.
Figure 5-4: Distribution of the amounts of debt due to the SFA Figure 5-4: Distribution of the amounts of debt due to the SFA

The second source of debt in importance comes from financial institutions, with an
average debt of $8,043 for the 35% of students that indebt themselves from this source.
!"#$
!%#$
!&#$
'(#$
)#$
'&#$
'*&&$+$,&&&*&&$
,&&'*&&$+$'&&&&*&&$
'&&&'*&&$+$',&&&*&&$
',&&'*&&$+$!&&&&*&&$
!&&&'*&&$+$!,&&&*&&$
!,&&&-$
45
We notice, however, as in the case of family debts, that the first quartile, at $1,500,
reveals a high level of small and relatively modest loans.
Error! Reference source not found. explores the distribution of debt amounts coming
from financial institutions. 55% of these are less than $5,000, as revealed by the median.
However, this important concentration of debt amount must not hide its corollary logic:
a significant portion of students contract very high debts in the private sector. A debt of
$15,000 and more is the lot of 16% of students that have a debt of this type.
We must note that that a debt due to a private institution is often less advantageous than We must note that that a debt due to a private institution is often less advantageous than
a public loan, due to the less advantageous repayment conditions, a higher interest rate
and no exoneration of the payment of interest during studies.
Figure 5-5: Distribution of debt amounts from financial institutions

The last source of debts in importance is family debt, of which the average debt is
situated at $5,234 and the median debt at $3,000. Error! Reference source not found.
allows us to observe that family debts are generally of a lesser amount than the other
two sources. Once again, we find an important portion of students that have these types
of loans and relatively high debts, of the order of $5,000 and more.
Figure 5-6: Distribution of amounts of family debts
!"#$
%&#$
'&#$
'(#$
')&&$*$!&&&)&&$
!&&')&&$*$'&&&&)&&$
'&&&')&&$*$'!&&&)&&$
'!&&')&&+$
46
Figure 5-6: Distribution of amounts of family debts

The amount of the total debt, at $13,967, is quite high. Furthermore, the 3
rd
quartile, at
$20, reveals the presence of an important percentage of students that cumulate quite
voluminous debts.
Figure 5-7: Distribution of total amounts of debt

5.2. The impact of already contracted indebtedness
We could believe that the forms of indebtedness replace each other: in this case, the
presence of a form of indebtedness would reduce the frequency of the other forms of
!"#$
%&#$
%'#$
()**$+$%&**)**$
%&*()**$+$&***)**$
&**()**,$
47
indebtedness. Now, it is rather the inverse relation we can observe: in all forms of debt,
there is a statistically significant correlation with the probability of indebtedness and the
average amounts of debts. The only exception to the rule is familial indebtedness, whose
level of debt diminishes as the number of sources of debt increases.
We will study in sequence the three forms of indebtedness that we have retained, based
on the three relations: the evolution of the rate of indebtedness, the variation of the rate
of indebtedness according to the amount of debt accumulated and the relations between
the amounts of debt. The first series of indicator allows us to know if having a debt to a
given source has an influence on the fact of subscribing to another source of debts, the
second if the fact of being indebted to a first and second source has an influence on the
level of debt of the first source, and the third if the increase in the amounts of debts for
someone who cumulates the two selected sources influence each other.
5.2.1. Relations between the rates of indebtedness
The fact of being enrolled in SFA tends to increase the probability of leaving studies
with a private debt: we can explain 9% of the variance of the rates of indebtedness with
this relation (=0,299 for p<0,001). A quarter of students have both a debt with SFA and
a private debt, while 22% have a debt with SFA, but no private debt. Only 13% have a
private debt without a debt at SFA. We must note that more than half the respondents
with a debt at SFA also expected to leave their studies with a private debt. Nearly two-
thirds of students that have a private debt also had a debt with SFA.
Table 5-2: Relation between the rate of indebtedness to SFA and the rate of private indebtedness
Absenceofa
privatedebt
Presenceof
aprivate
debt
Total
AbsenceofadebtatSFA
%line 76,50% 23,50% 100,00%
%column 65,40% 34,50% 54,00%
%oftotal 41,30% 12,70% 54,00%
PresenceofadebtatSFA
%line 47,60% 52,40% 100,00%
%column 34,60% 65,50% 46,00%
%oftotal 21,90% 24,10% 46,00%
Total
%line 63,20% 36,80% 100,00%
%column 100,00% 100,00% 100,00%
%oftotal 63,20% 36,80% 100,00%
The relation between indebtedness at SFA and family debt is similar, but not as strong:
it explains 1% of the variance (=0,112 p < 0,001). 11% of respondents cumulate a
family debt and a debt at SFA: this nonetheless consists of more than half the
respondents declaring a family debt, and nearly a quarter of students with a debt from
SFA.
48
Table 5-3: Relation between the rate of indebtedness at SFA and the rate of family indebtedness
Absenceofa
familydebt
Presenceof
afamily
debt
Total
AbsenceofanSFAdebt
%line
85,40% 14,60% 100,00%
%column
57,60% 43,30% 54,90%
%oftotal
46,90% 8,00% 54,90%
PresenceofadebtatSFA
%line
76,60% 23,40% 100,00%
%column
42,40% 56,70% 45,10%
%oftotal
34,50% 10,50% 45,10%
Total
%line
81,40% 18,60% 100,00%
%column
100,00% 100,00% 100,00%
%oftotal
81,40% 18,60% 100,00%
We also find a relation of average strength between private indebtedness and family
indebtedness: this explains 5% of the variance. Only 11% of respondents cumulated a
private debt with a family debt: however, 58% of students with a family debt also had a
private debt.
Table 5-4: Relation between the rate of private indebtedness and the rate of family indebtedness
Absenceofa
familydebt
Presenceof
afamily
debt
Total
Absenceofaprivatedebt
%line 87,90% 12,10% 100,00%
%column 69,50% 41,90% 64,30%
%oftotal 56,50% 7,80% 64,30%
Presenceofprivatedebt
%line 69,70% 30,30% 100,00%
%column 30,50% 58,10% 35,70%
%oftotal 24,80% 10,80% 35,70%
Total
%line 81,40% 18,60% 100,00%
%column 100,00% 100,00% 100,00%
%oftotal 81,40% 18,60% 100,00%
As we will se further on, the same type of motive repeats itself on the variation of the
indebtedness rate and on the amounts of the debt.
5.2.2. Student financial assistance
The fact of contracting a debt from student financial assistance should normally, procure
a sufficient standard of life to be able to prevent having to indebt oneself more in the
course of studies. It is at least one of the objectives of the program.
49
The amount of the debt due to SFA has a certain influence on the debts contracted from
other sources, as is shown by Error! Reference source not found.. The relation is
stronger in the case of private loans: the amount of the debt due to SFA explains 8.9% of
the variance in the private rate of indebtedness. The relation with the presence of a
family debt is lower.
Table 5-5: Relation between the amount of the SFA debt and the presence of other sources of debt Table 5-5: Relation between the amount of the SFA debt and the presence of other sources of debt
Private Family/friends
1,00-5000,00 49,90% 27,40%
5001,0010000,00 49,50% 18,00%
10001,0015000,00 50,80% 22,70%
15001,0020000,00 56,40% 23,70%
20001,0025000,00 61,60% 23,30%
25001,0030000,00 55,80% 25,10%
30001,0035000,00 63,00% 38,50%
35001,0040000,00 59,60% 42,60%
40001,0045000,00 80,00% 40,00%
45001,00+ 76,00% 19,00%
Total 52,50% 23,40%
The relation also exists in relation to the average amount of debt, but is lower. The
amounts of debt to student financial assistance explains 2% of the variance in the
amounts of private and family loans. We must nevertheless note that the average
amounts of private loans reach a maximum limit at about $30,000 to $35,000 of debt due
to SFA, suggesting a more parabolic relation than linear. We must also note that the
population remains very small at such levels of debt.
We can issue two hypotheses regarding this relation. On the one hand, it is largely
recognized that the assistance granted to the loans and bursaries program is insufficient,
which could push certain beneficiaries of the program to turn toward private loans. On
the other hand, some students stop being eligible for the program because they have
exceeded the time limit allowed for the program. Private loans can then be an
interesting exit door for certain students.
We also have to note that the size of the debt due to student financial assistance is the
main explanatory factor of the size of the total debt: it explains 64% of the variance.
50
Table 5-6: Variation of the average amounts of debt according to the amount of debts due to student financial
assistance
Private Family/friends Total
1,00-5000,00 6002$ 3439$ 5972$
5001,0010000,00 8275$ 3852$ 12427$
10001,0015000,00 8155$ 4006$ 17431$
15001,0020000,00 9278$ 4497$ 23557$
20001,0025000,00 10976$ 5479$ 29068$
25001,0030000,00 10617$ 6116$ 33351$
30001,0035000,00 12513$ 6396$ 38325$
35001,0040000,00 10520$ 4606$ 41506$
40001,0045000,00 9090$ 7189$ 45798$
45001,00+ 8886$ 9139$ 49606$
Total 8355$ 4231$ 16353$
We can conclude that students begin to indebt themselves during their studies do not
stop in the course of the route. Two factors can explain this. On the one hand, there is a
financial need that is not filled. An alternative explanation rests on the notion of
aversion to risk: a student that takes out a loan is partially desensitized to the negative
impacts of indebtedness, permitting him to take out higher loans. We will see further on
that the first explanation is often more convincing.
5.2.3. Loans from financial institutions
The second source of loans in importance, loans coming from financial institutions, are
strongly related to other forms of indebtedness. The presence of a loan from the private
source predicts 9% of variance in the amounts of the debt of the SFA and 5% of the
variance of the family loan, two effects of average force (R=0,299 and r=0,224).
Table 5-7: Variation of the indebtedness rate by source based according to the size of the private loan
SFA Family/friends
1,00-5000,00 62,80% 31,30%
5001,00-10000,00 67,30% 28,00%
10001,00-15000,00 73,70% 29,00%
15001,00-20000,00 64,80% 29,70%
20001,00-25000,00 67,00% 34,30%
25001,00-30000,00 61,40% 25,00%
30001,00+ 75,40% 33,30%
65,50% 30,30%

Here again, the amounts of private loans are positively correlated with the amounts of
other sources of loans and respectively explain 2%, 5% and 43% of the variance of loans
of SFA, family loans and the total debt.
51
Table 5-8: Variation of the average amounts of debts per source according to the size of the private loan
SFA Family/Friends Total
1,00-5000,00 11913$ 3537$ 9920$
5001,00-10000,00 14580$ 4690$ 17441$
10001,00-15000,00 16176$ 4507$ 24138$
15001,00-20000,00 15692$ 5154$ 27464$
20001,00-25000,00 14123$ 4400$ 31536$
25001,00-30000,00 14880$ 7708$ 35996$
30001,00+ 16109$ 11425$ 42092$
Total 13482$ 4382$ 15706$
5.2.4. Family loans
The family loans are those that present the greatest stability relative to other sources of
debts. Their presence predicts 1% of the variance of debt amounts of the SFA and 5% of
private loans.
Table 5-9: Variation of the rate of indebtedness by source based on the size of the family loan
SFA Private
1,00-2500,00 64,90% 67,00%
2501,00-5000,00 52,80% 50,00%
5001,00-7500,00 59,50% 62,70%
7501,00-10000,00 47,60% 48,30%
10001,00-12500,00 50,00% 45,50%
12501,00-15000,00 45,30% 56,00%
15001,00+ 37,20% 42,70%
Total 57,30% 58,50%

Similar relations (2% and 5%) exist in relation to the amounts of family loans. Fact to
note: the size of the family loan only explains 17% of the variance of the total debt.
Table 5-10: Variation of the average amount of debts by source based on the size of the family loan
SFA Private Total
1,00-2500,00 11223$ 5986$ 11336$
2501,005000,00 14602$ 4137$ 14278$
5001,00-7500,00 18653$ 10276$ 17830$
7501,0010000,00 16517$ 8714$ 18205$
10001,0012500,00 13970$ 6582$ 18271$
12501,0015000,00 15382$ 14408$ 23135$
15001,00+ 15637$ 13963$ 30739$
Total 13186$ 7387$ 14522$
This is all explained quite simply: the amounts of family loans are generally much lower
than the two other forms of loans for studies, which makes them often fill a lesser need.
The low variation of the presence of a family loan is based on the size of the SFA loan.
52
5.3. A spiral of indebtedness
The data presented earlier suggests the existence of a sort of spiral of indebtedness. On
the one hand, the fact of being a source is correlated positively with that of being
indebted to other sources. Thus, the SFA program shows again one of its deficiencies:
the fact of being indebted to SFA predicts in many cases a loan to at least another source
of indebtedness. The number of sources of indebtedness has a striking influence on total
indebtedness, but negatively influences family debt. In the other two cases, it is when
we attain the useful limit of the program, where we tend to turn to another form of
indebtedness. However, the analysis of the amounts in relation in relation with the rates
denotes an interesting motive: the amount of family indebtedness is negatively
correlated with all other forms of indebtedness. Thus, the student with very high debts
will have a little less of a tendency to borrow from other sources. We can believe that
students that borrow a lot from their family are no longer eligible for SFA and private
loans.
Table 5-11: Summary of statistical tests on indebtedness
Numberofsourcesof
debt

Amount;r ,06*** ,00 -,17*** ,37***
SFA
Raterate; 1 ,30*** ,11***
Amountrate;rb 1 ,09*** ,18
Amountamount;r 1 ,15*** ,15*** ,80***
Private
Rate-rate; ,30*** 1*** ,22***
Amountrate;rb ,39* 1*** -,05***
Amountamount;r ,15*** 1*** ,23*** ,65***
Family/friends
Raterate; ,11*** ,22*** 1***
Amountrate;rb -,13*** -,12*** 1***
Amountamount;r ,15*** ,22*** 1*** ,41***
Furthermore, the amounts of debts are all positively correlated between themselves,
thus indicating an increase of the amount of a debt tends to come with debts from other
more voluminous sources. The size of a debt also influences the propensity of a student
to contract debt from other sources. It thus does not seem abusive to talk about a spiral
of indebtedness. Is it a question of financial vulnerability or a reduction of the aversion
to risk? This is what we will see in the following sections.

53
6. The impact of student characteristics
In this section we will study the impact of two broad categories of student
characteristics. We will begin with the socioeconomic characteristics: gender, age, family
income, place of residence, the study area, changing study domains, and the presence of
dependent children and status of first-generation students. Subsequently, we will look
at school characteristics, such as debt already accumulated, academic progress, the
duration of the degree, prolonging previous studies and the field of study itself.

6.1. The impact of socioeconomic characteristics
For the purposes of this study we retained a select number of socioeconomic
characteristics for analysis: age, family income, place of residence, the area of study,
changing study domains, the presence of a dependent child student and status of first
generation students.
6.1.1. Gender
The student population is predominantly female at 58%. As shown in Figure 1-1, gender
has no influence on the debt ratio, regardless of the source.
Figure 1-1: Change in debt ratio by source by gender

Gender does however, have an influence on the increase of debt: being a male
significantly but weakly increases in terms of total and private debt acquired (rpb =-,
rpb =- 06 and, at 08 p <, 001). This is probably due in part to the preferred fields of study
of men, such as engineering, where studies are longer and therefore see higher debt
rates (see below section 1.2.5). Other influencing factors, could however, be taken into
account.

Figure 6-1 : Change in average amounts of debt by source by gender

!"#$ %&'()$ "*+',,-.*+'/$ 012*,$
31++-$ 45657$ 89657$ :;6<7$ 946<7$
"-++-$ 456=7$ 8>6?7$ :;687$ 9?6:7$
<6<7$
:<6<7$
=<6<7$
8<6<7$
4<6<7$
?<6<7$
9<6<7$
5<6<7$
54

Overall, gender has little influence on the level of debt, but studies show that males
have a slightly higher amount of debt than females. For a more detailed study of the
dynamics between men and women in higher education, please consult (FEUQ, 2011c),
where we see, among other things, that women still face many barriers when pursuing
higher education.

6.1.2. Age
The study population is relatively young, as shown in Table 1-1. However, it is
noteworthy to remember that 22% of students are over the age of 25, and thus
correspond to a more atypical profile: non-traditional students (FEUQ, 2010).
Table 6-1: Description of students age

Age
Average 23
1st quarter 21
2
nd
quarter 22
3rd quarter 24

Figure 6-2 : Table 6-2: Description of students age

!"#$ %&'()$ "*+',,-.*+'/$ 012*,$
31++-$ 45$467$8$$ 9$:77$8$$ ;$<9<$8$$ 4<$=99$8$$
"-++-$ 46$;59$8$$ =$<4=$8$$ ;$>74$8$$ 45$599$8$$
>$8$$
6$>>>$8$$
<$>>>$8$$
7$>>>$8$$
9$>>>$8$$
4>$>>>$8$$
46$>>>$8$$
4<$>>>$8$$
47$>>>$8$$
55

Age has a significant influence on the amount of accumulated debt at the time of
graduation. There is a wide range of forms of debt as well as significant statistical
variations in their level of debt with medium effects on the debt to the FAE, private debt
and total debt, as well as a weak effect on family debt. While 48% of students 19 and
younger expect to leave school in debt, 86% of students 26 years and over are in the
same situation. Consequently, it appears that non-traditional students have to resort to
higher proportions of debt to finance their education.
Figure 6-3: Change in debt ratio by source according to age

Similar trends in the amount of debt incurred are not reflected from a statistical point of
view. Significant correlations can be found, but they are weaker (for example, family
loans have no significant correlation and remain fairly stable), however, changing
averages allow us to observe significant changes (see Figure 1.5). Age alone only
explains 5% of the variance in total student debt.
!"#
$%"#
$!"#
$&"#
$'"#
!"#
%"#
$$"#
&"#
$(#)*+#,-#
./0*+#
12# 1$# 11# 1'# 13# 1%# 1&#4#'2#
)*+#
'$#)*+#,-#
567+#
!"#
$!"#
%!"#
&!"#
'!"#
(!"#
)!"#
*!"#
+!"#
,!"#
$!!"#
$,#-./#
01#
234./#
%!# %$# %%# %&# %'# %(# %)#5#&!#
-./#
&$#-./#
01#678/#
9:;# <=4>?# :-24770@-24/# A31-7#
56

With the exception of family debts which are relatively stable, we also note a continued
growth in the average amount of student debt accumulated by the time of graduation,
starting at the age of 21.
Figure 6-4: Change in average amounts of debt by source according to age

This trend can be partly explained through a better understanding of students personal
financial situation: the more a student is nearing the end of their studies, the better the
likelihood that they will know the precise amount of debt that they will have
accumulated by the time they graduate. However, it seems that this explanation is
insufficient: literature reviews indicate that age and debt were closely associated.
One factor that is embedded in the actual financial situation can be added to this
question: older students are probably part of a different debt profile, as they are less
reliant on significant parental support, a factor which is important in reducing student
debt (see below). As we shall see, the duration of studies has a significant on student
debt: which is probably an important explanatory factor. We must also remember that
older students are more frequently non-residents, a feature that is highly correlated with
student debt in all its forms as student autonomy is generally combined with the use of
a lack of resources to finance the furthering of their education.
6.1.3. Family Income
One might think that private student debt as well as family debt would be more
common in certain income brackets, especially among those in the upper middle classes.
These are families where family capital is often higher and the propensity to borrow for
education would be increased accordingly, given a better understanding of the
economic benefits of university education. However, this is not what the data shows.
Lets first bear in mind the characteristics of the family income of students, presented in
Table 1-2. We divided the distribution into quartiles to facilitate the statistical analysis of
the data, which gives a good overview of the variations that occur depending on the
gross family income.
!"!!!!#!!
!$!%%%!!#!!
!&%!%%%!!#!!
!&$!%%%!!#!!
!'%!%%%!!#!!
!'$!%%%!!#!!
&(!)*+!
,-!
./0*+!
'%! '&! ''! '1! '2! '$! '3!4!1%!
)*+!
1&!)*+!
,-!567+!
89:! ;<0=>! 9).066,?).0+! @/-)6!
57
Table 6-3: Description of gross family income
NetFamily
Income
Average 79128$
1stquarter 35000$
2
nd
quarter 65000$
3rdquarter 112500$

Figure 1-6 shows the importance of family income in determining student debt. We
have divided the gross family income according to income quartiles which are
presented in our distribution of family income. It shows that the rate of student debt is
significantly negatively correlated with family income, regardless of the source of the
loan. The correlation is strong in terms of student financial aid (which is mainly due to
program rules), but also for private loans, family loans and total debt (rpb =-, 16, rpb = -,
10, rpb =-, 34 at p <.001).
Figure 6-5 : Change in debt ratio by source based on family income

The changes are less clear on the average amount of debt. The average debt, as is the
case for the SFA debt, is negatively correlated with gross family income (r =- 17 and r =-,
18, 3% of variance explained). There is no statistically significant relationship associated
to the amount of private and family loans. There is however a slight increase in average
private loans: one can assume that there are less relatively small debts (typically credit
card debt). This relative stability of the amounts of debt, however, should not excuse the
most striking finding, that is to say, the ratio of debt.
!"#$%&'(&)*&
***+&
,-&---+&.&/*&
***+&
0-&---+&.&
12,&***+&
123&---+&(4&
567%&
89:& 00;--<& 3=;2-<& )2;/-<& 13;*-<&
>?#@A& ,0;=-<& ,);--<& )/;2-<& 23;=-<&
9BC#66(DBC#%& 2,;=-<& 2-;--<& 1*;)-<& 12;0-<&
E"4B6& =/;--<& 02;0-<& 30;0-<& )*;)-<&
-;--<&
1-;--<&
2-;--<&
)-;--<&
,-;--<&
3-;--<&
/-;--<&
0-;--<&
=-;--<&
*-;--<&
1--;--<&
58
Table 6-4: Change in average amounts of debt by source based on gross family income

What can be deduced is that student debt is first and foremost about money and
financial need. The idea that students from wealthier backgrounds take on more debt in
the private sector seems invalid. We will indirectly see this again with the issue of
parental contributions toward education, which has similar characteristics in several
respects.
6.1.4. Place of Residence
Most students leave home during their studies, or they already left before they started
them. More than two-thirds of students do not live with their parents, which has a
disproportionately large impact on student finances: it involves, among other things,
spending significantly more (FEUQ, 2010). These expenditures also have a significant
impact on student debt. First, not living with ones parents leads to much higher debt
amounts: the relationship is statistically significant for all forms of debt, and stronger for
the student financial aid and 'total debt. Three quarters of students who do not live with
their parents go into debt, versus 43% of those who remained at home while pursuing
studies. A student who does not reside with his parents has a probability of being in
debt which is 1.7 times greater than a student who remains at home.
!"#$%&'(&
)*&***+&
,-&---+&.&
/*&***+&
0-&---+&.&
12,&***+&
123&---+&
(4&567%&
89:& 1,&0-*;0)&&+&& 11&<0<;/2&&+&& 1-&)-,;0<&&+&& 1-&,/*;,<&&+&&
=>#?@& 0&)**;*/&&+&& <&))-;-,&&+&& <&/*1;,)&&+&& <&**/;3)&&+&&
9AB#66(CAB#%& 3&2//;0,&&+&& ,&,/2;**&&+&& 3&-02;<2&&+&& /&-0-;2*&&+&&
D"4A6& 1/&/2-;*-&&+&& 1,&1-/;10&&+&& 11&<-/;*-&&+&& 1-&<,);*-&&+&&
&E&&&&+&&
2&---;--&&+&&
,&---;--&&+&&
/&---;--&&+&&
<&---;--&&+&&
1-&---;--&&+&&
12&---;--&&+&&
1,&---;--&&+&&
1/&---;--&&+&&
1<&---;--&&+&&
59
Figure 6-6: Change in debt ratio by source by place of residence

The amount of debt is also higher for non-Quebec resident students. The average total
debt is always higher, $ 15 277 for non-Quebec residents versus $ 8,780 for Quebec
residents (r =-, 23 at p <.001). We can observe significant statistical relationships here.
Figure 1 8 shows these results.
Figure 6-7: Change in average amounts of debt by source based on the basis of place of residence

It is also important to remember that students from lower income backgrounds leave
the family more frequently, probably because they are more often tenants. (rpb =-, 126 at
p <.001). It is not necessarily possible for the student to stay with their parents during
adulthood.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3*4'2-$56-7$/-/$8*&-92/$ :;<=>?$ :@<A>?$ =;<:>?$ A;<>>?$
BC6*4'2-$8*/$56-7$/-/$
8*&-92/$
@D<>>?$ A;<A>?$ :=<E>?$ FA<:>?$
><>>?$
=><>>?$
:><>>?$
;><>>?$
A><>>?$
@><>>?$
G><>>?$
F><>>?$
D><>>?$
AlL rlve
lamllle/
amls
1oLal
PablLe chez ses parenLs 9 073 $ 3 633 $ 4 028 $ 8 781 $
n'hablLe pas chez ses
parenLs
13 446 $ 8 631 $ 3 316 $ 13 278 $
- $
2 000 $
4 000 $
6 000 $
8 000 $
10 000 $
12 000 $
14 000 $
16 000 $
18 000 $
60
6.1.5. Area of Study
The area of study, and distance from place of residence, are two factors that may have a
significant influence on debt. To facilitate data analysis, we grouped students according
to the region in which students reside: Montreal, Quebec, the central regions (Outaouais,
Estrie and Mauricie) and remote areas (Saguenay Lac-St-Jean, Abitibi-Tmiscamingue ,
Bas-St-Laurent). The majority of students reside Montreal, as illustrated in Table 6-4.
Table 6-5 : Description of the area of origin of students
Proportion
(%)
Montreal 60
Quebec 18
Central Regions 16
Remote Areas 6
Distance 0,2
Total 100

Figure 1-9 shows the debt ratio by source and region. Other than percentages of family
debt, all debt ratios are higher outside the Greater Montreal region.
Figure 6-8: Change in debt ratio by source according to the study area

The relationship is somewhat different in respect to the average debt accumulated
during studies. Indebtedness to the SFA remains almost stable, while private debt is
slightly higher in Quebec and in the central regions than in Montreal or in remote areas.
However, there are some very interesting conclusions which can be drawn out from
this.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3142&)*,$ 56789:$ 6;769:$ <=789:$ >?7<9:$
@A)B-C$ 5=7=9:$ 6D769:$ <>759:$ >D769:$
E)F'14/$C-42&*,-/$ D?7;9:$ 6=759:$ <=769:$ >8789:$
E)F'14/$),1'F4)-/$ ><7D9:$ 5D7D9:$ ?<7?9:$ ;>7;9:$
9799:$
<9799:$
?9799:$
69799:$
59799:$
D9799:$
>9799:$
;9799:$
89799:$
=9799:$
61
Table 6-6 : Average debt according to the study area

The relatively small sample size for remote areas however, requires us to use some
caution in interpreting the data. We note, however, a somewhat more difficult financial
situation for students studying in remote areas. The debt rate is significantly higher ,
demanding that attention be paid to students located in these regions. The well
documented economic situation which is often more difficult for them, may be an
explaining factor, because of a less dynamic labor market and, therefore, a lower
diversity of sources and methods of financing university education.
6.1.6. Moving Away From Home
One-third of students must leave their home towns to complete their final educational
projects. These students, who must move, face many challenges: the financial challenge
of adjusting to full autonomy, as well as integration challenge to what is truly a
microcosm of new home.
However, the change in the region seems to weaken student financial situations. The
debt rate is higher, especially in terms of student financial assistance ( =- 14 to p <.001).
A student who is required to leave their home town is 1.2 times more likely to go into
debt during their studies. Figure 1-10 illustrates the variables: there is a statistically
significant relationship for all forms of debt, it is often low.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3142&)*,$ 56$777$$8$$ 9$:;;$$8$$ <$5:=$$8$$ 5=$97;$$8$$
>?)@-A$ 56$;<:$$8$$ 7$9;9$$8$$ :$7=5$$8$$ 5:$B::$$8$$
C)D'14/$A-42&*,-/$ 56$5=5$$8$$ ;$57E$$8$$ <$67;$$8$$ 5:$B76$$8$$
C)D'14/$),1'D4)-/$ 5=$<66$$8$$ 9$6<9$$8$$ E$:B:$$8$$ 5<$5B5$$8$$
$F$$$$8$$
6$BBB$$8$$
:$BBB$$8$$
E$BBB$$8$$
7$BBB$$8$$
5B$BBB$$8$$
56$BBB$$8$$
5:$BBB$$8$$
5E$BBB$$8$$
62
Figure 6-9: Change in debt ratio by source based on moving away

In regards to the study of the amounts of debt, two forms of debt are statistically
significantly influenced by leaving ones home town: private debt and debt faced by
loved ones (rpb =-, 11 and rpb =-, 11 at p <010). This variation seems to explain why the
total debt is higher (rpb =-, 07 at p <010). We can see that the average debt is $ 9,299 for
those who have private loans private and are required to leave home, versus $ 7,241 for
those in the opposite situation.
Figure 6-10: Change in average amounts of debt by source based on the change of region

!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
34*56-+-52$7-$&)6'15$ 89:;<=$ >?:@<=$ A<:B<=$ ;>:;<=$
C-/2)$7*5/$/*$&)6'15$
7D1&'6'5-$
>@:E<=$ @E:B<=$ ?;:A<=$ E?:<<=$
<:<<=$
?<:<<=$
A<:<<=$
@<:<<=$
><:<<=$
8<:<<=$
E<:<<=$
;<:<<=$
9<:<<=$
AlL rlve
lamllle/
amls
1oLal
ChangemenL de reglon 12 603 $ 9 299 $ 3 394 $ 14 770 $
8esLe dans sa reglon
d'orlglne
12 429 $ 7 241 $ 4 236 $ 13 096 $
- $
2 000 $
4 000 $
6 000 $
8 000 $
10 000 $
12 000 $
14 000 $
16 000 $
63
Students who are required to leave their home towns are therefore more vulnerable
profile. The presence of larger private loans amongst students who leave their home
towns demonstrates, once again, significant gaps in our student financial assistance
system which inadequately covers students financial needs, including those who must
leave their home towns.
6.1.7. Presence of a dependant child
A student of twenty must balance school and family life. This average rises to 20.3% for
students over the age of twenty-four (FEUQ, 2010a: 22). The FEUQ has frequently
emphasized the precarious financial situation of students who must juggle studies, work
and family (see among others: FEUQ, 2010a; FEUQ, 2010b; FEUQ, 2009a; CNCS-FEUQ,
2011) and the data which we will present in relation to student debt supports this view.
It is much more likely for a student with dependent children to be in debt, regardless of
the source. All these changes are statistically significant. Students with dependent
children have a loan rate 1.4 times higher than those without dependent children.
Figure 6-11 : Change in debt ratio by source according to the presence of dependent children

Three types of debts which see increases when there is presence of a dependent child, as
shown in Figure 1 13: they are debt against the AFE, private and total debt. There is an
average of about $4,000 higher debt and the average debt for students with dependent
children goes above $20,000. In the case of the AFE, we believe that the increase in debt
is caused by a prolongation of studies.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3*4/$-45*42$6$78*&9-$ :;<;=>$ ?@<A=>$ BA<C=>$ @?<:=>$
#45*42$6$78*&9-$ AD<@=>$ ;@<:=>$ DE<==>$ AC<A=>$
=<==>$
B=<==>$
D=<==>$
?=<==>$
:=<==>$
;=<==>$
@=<==>$
C=<==>$
A=<==>$
E=<==>$
B==<==>$
64
Figure 6-12: Change in average amounts of debt by source according to the presence of dependent children

Inadequate financial support for student parents, combined with the general
shortcomings of the program of loans and bursaries, place these students in an
extremely precarious financial situation. This can seriously compromise their ability to
continue their academic studies.
6.1.8. Status of First-Generation Students (FGS)
The university system developed largely in the periods when the government expanded
access to university studies. The result is that 44% of Quebec students are first-
generation students. Remember that parents with a university education generally have
higher incomes. Conversely, the FGS are generally more vulnerable, which is explained
in part by the lack of university educations amongst families. The literature suggests a
greater aversion to debt due to a lower cultural capital, which is mainly, manifested in
university studies admissions. It is generally recognized that these students also need
special attention to avoid prolonging their university studies (Auclair et al, 2008).
However, it is easy to see that the status of FGS has a considerable influence on student
debt. There is a statistically significant correlation between debt ratio for FGS ( = - 20),
private debt ( = -, 08) and total debt ( = - 14). Figure 1-14 shows these key variations:
there maybe variations of 7 to 14 percentage points in all cases except that of family
debt.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3*4/$-45*42$6$78*&9-$ :;$<=<$$>$$ ?$@;A$$>$$ A$;B=$$>$$ :C$<<A$$>$$
#45*42$6$78*&9-$ :B$A?C$$>$$ @$A;@$$>$$ <$@::$$>$$ ;=$@D:$$>$$
$E$$$$>$$
$A$===$$>$$
$:=$===$$>$$
$:A$===$$>$$
$;=$===$$>$$
$;A$===$$>$$
65
Figure 6-13: Change in debt ratio by source based on the generation of the student

The situation is somewhat less clear with respect to amounts of debt incurred. Two
types of debt emerge: on the one hand, debt faced by family or friends (rpb = .07 at p
<010) is weakly correlated with the students generation as parents with a university
degree better understand the reality of university requirements and are therefore more
willing to lend money to their children if necessary. However, this does not eliminate
overall debt from all sources tolled, to be higher (rpb =-, 08) for first-generation
students; the idea that they are most vulnerable is therefore valid.
Figure 6-14: Change in average amounts of debt by source based on the generation of the student

6.1.9. The socio-economic characteristics determine a significant portion of the accumulated debt
What can be learned from this analysis of socioeconomic characteristics? Table 6-6
summarizes the statistical tests performed in this section.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3245'*62$5-$7&-+'8&-$
9)6)&*:16$
;<=>?@$ AB=<?@$ B>=C?@$ CB=A?@$
3245'*62$5-$5-4D'8+-$
9)6)&*:16$14$7,4/$
E<=F?@$ EA=B?@$ BF=E?@$ ;C=<?@$
?=??@$
B?=??@$
G?=??@$
E?=??@$
A?=??@$
;?=??@$
<?=??@$
C?=??@$
>?=??@$
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3245'*62$5-$7&-+'8&-$
9)6)&*:16$
;<$=;;$$>$$ ?$;@=$$>$$ A$?=B$$>$$ ;A$@;<$$>$$
3245'*62$5-$5-4C'8+-$
9)6)&*:16$14$7,4/$
;D$D<;$$>$$ ?$=;D$$>$$ B$?BA$$>$$ ;D$?DE$$>$$
$F$$$$>$$
D$===$$>$$
A$===$$>$$
G$===$$>$$
?$===$$>$$
;=$===$$>$$
;D$===$$>$$
;A$===$$>$$
;G$===$$>$$
66
Table 6-7 : Summary of statistical tests on socioeconomic characteristics
AFE Priv Famille/amis Total
Gender
Rate; ,00 ,02 ,00 ,01
Amount;r
pb
-,03 -,08*** -,03 -,06***
Age
Rate;r
pb
,25*** ,17*** ,07*** ,19***
Amount;r ,16*** ,11*** ,02 ,22***
FamilyIncome
Rate;r
pb
-,42*** -,16*** -,10*** -,34***
Amount;r -,17*** ,03 ,02 -,16***
Amount; -,22*** ,04 ,05* -,24***
Residence
Rate; ,32*** ,17*** ,10*** ,30***
Amount;r
pb
,16*** ,13*** ,09** ,23***
Movingforstudies
Rate; -,14*** -,04*** -,05*** -,14***
Amount;r
pb
-,01 -,11*** -,11*** -,07**
DependentChildren
Rate; ,17*** ,09*** ,06*** ,11***
Amount;r
pb
,13*** -,05* -,01 ,17***
FirstGeneration
Students

Rate; -,20*** -,08*** ,01 -,14***
Amount;r
pb
-,04* -,01 ,07** -,08***

It is clear that some students are more vulnerable than others. As for the average debt,
the inability to live with ones parents, having children or being a first generation
student all increase ones chances of being in debt. The same associations are observed
in the amounts of debt. Aging tends to lead to greater and higher debt which is likely
related to having children or not being dependent on ones parents. Age, however, has
no influence on family debt.
Students in remote areas often have a more difficult financial situation as their debt is
more frequent and higher. Having to leave ones home town in pursuit of a higher
education also places students in a situation where debt becomes more frequent and
larger, especially in the private sector.
The biggest factor in student debt is parental income: the more one comes from a
wealthy family, the less chances of getting into debt. However, the amounts associated
to private debt and household debt is stable, contrary to the amounts borrowed from the
SFA.
Finally, gender predicts the propensity to borrow wrong: there is a statistically
significant but weak from being a man and debt levels. Forms of debt have similar
profiles. The SFA is influenced by all the socioeconomic factors except sex. The private
debt is slightly more stable factors as: the amount is not influenced by parental income.
Family debt is the most stable of all, with variations low, but significant regardless of the
characteristics. Turning to the analysis of school characteristics.
6.2. School Characteristics
The questions used in this survey posed very few questions in regards to school
characteristics, and many of those posed are not of interest for the purposes of this
study. However, we are able to extract some information about academic progress, the
67
length of time for completing the degree, extending the length of time required to
complete ones studies and the field of study.
6.2.1. Previously Existing Debt
Before going further, it is important to look at the official statistics of the student
financial aid program. Indeed, one of the weaknesses of the study is that it includes only
students enrolled at the undergraduate level. However, an important determinant of
student debt is academic progress. Table 1 7 illustrates this reality.
Tableau 6-8 : Profile of borrowers based on academic progress
Undergraduate
studieswheredebt
isincurred
Lenders Proportion AverageDebt
Amounts
Undergraduate 11324 63% 11295$

CEGEP 6565 37% 16001$
Total 17889 100% 13022$

Source : AFE, 2010, p. 73
It can be seen quite easily that being in debt in college tends to increase overall student
debt: the average loan is 1.4 times higher for students who are in debt in college and at
the undergraduate level than for students who only seek loans at the undergraduate
level. Loan limits are lower in college: however, study periods are longer than at the
university, which leads to a similar annual debt.
6.2.2. Academic Progress
We can also estimate the increase in the debt average under consideration in other ways:
by isolating the students according to their year of study. This gives a very rough
estimate of a longitudinal study, which is by far the most appropriate tool for such an
exercise. In the absence of such a tool, the year of study allows a satisfactory estimate.

To do this we convert the year of school entry to estimate the students current year of
study and current debt variables (not debt estimated at the time of graduation). The year
of study is calculated based on the year of entry into the school and the current school
year. Thus, a student who registers in August 2009 will be considered in their first year
of study: however, a student who enrolls in school for time in January 2009 would be
considered to be in their second year of studies.
To begin with, Figure 1-16 shows the division of students according to their status in
their current program of study. As can be seen, 17% of students are enrolled for more
than three years which is beyond the normal time required for completing a
bachelors degree. This may partly be explained by factors such as needing to prolong
ones studies as well as the initial duration of the program itself.
68
Figure 6-15 : Distribution of students according to their progress in their program

The debt ratio is not stable from year to year. As for the SFA, the debt ratio increases
from 2 to 8% per year, with similar variations in all forms of debt. This reflects the fact
that the original financial plan put in place for studies does not always last in its
entirety. Students who do not necessarily want have debt at the beginning of their
studies will probably be pushed to do so as they are confronted with the financial
realities of the university system.
Figure 6-16 : Change in debt ratio from source depending on year of study

The average debt follows a similar slope. Students who have loans complete their first
year, with an average of about $7,918 in debt. As noted earlier, we also find that many
students are on loans beginning as early as upon entering the CEGEP system. The
annual variances are however, less significant than that of the proportion of students
with debt: of course, every year, new students begin to go into debt of some kind;
!"#$
"%#$
"&#$
'"#$
(#$
)*+,-.*+$/001+$ 2+34-.,+$
/001+$
5*6-7-.,+$
/001+$
83/9*-.,+$
/001+$
:-0;3-.,+$
/001+$+9$<=37$
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
-./012./$
3445/$
6/78120/$
3445/$
9.:1;120/$
3445/$
<73=.120/$
3445/$
>14?7120/$
3445/$/=$@A7;$
BCD$ -.1E5$ C301AA/F301;$ 9:=3A$
69
nevertheless, some patterns emerge from this practice.
The SFA debt increases at an average of $ 1,500 per year, which is lower than the loan
cap. Many students probably apply for the program, which reduces the growth of debt.
Private debt on the other hand increases steadily each year at a rate of approximately $
1,000 per year. Meanwhile, family debt is the most stable, remaining around $4,000 on
average.
Figure 6-17 : Change in average amounts of debt by source according to year of study

Progress in ones program of study also increases the number of sources of debt. One
quarter of students had multiple two or three sources of debt in their first year of study
while the rate stands at 36% of fourth-years, which corresponds to the last year of study
for many including those who may have prolonged their studies. Thus, the more
students advance in their studies, the more the vulnerable populations will accumulate
more and more sources of debt.
We can also predict the impact of debt on the prolonging of studies (see below):
Students who are in debt usually take slightly longer to complete their studies than
those who are not, thus increasing their representation in enrolment of studies for four
years or more.
!"!!!!#!!
$!%%%!!#!!
&!%%%!!#!!
'!%%%!!#!!
(!%%%!!#!!
)%!%%%!!#!!
)$!%%%!!#!!
)&!%%%!!#!!
)'!%%%!!#!!
)(!%%%!!#!!
*+,-./+,!
0112,!
3,45./-,!
0112,!
6+7.8./-,!
0112,!
940:+./-,!
0112,!
;.1<4./-,!
0112,!,:!=>48!
!?@A!! !*+.B2!! !@0-.>>,C0-.8!! !67:0>!!
70
Figure 6-18 : Year of study and number of sources of debt

6.2.3. Duration of the Degree
For this component we asked students to identify the year and month they began and
would finish their degree. Therefore, the data presented here does not adequately reflect
the duration of the academic progress as it does not take into account changes in
programs.
Not surprisingly, most students (46%) expect to complete their degree in three years; a
minority (12%) are enrolled in shorter programs; while 32% of students who expect to
complete their degree in four years: a requirement for some programs (education,
engineering), but a sign of prolonging studies for others. Bear in mind that the average
length of a degree is 6.7 semesters of full-time studies (MELS, 2010: 83) which is the
equivalent of slightly less than three and a half years.
Figure 6-19 : Description of the expected length of the degree

!"#$
!%#$
!&#$
!'#$
'%#$
!(#$
!(#$
!%#$
!'#$
!!#$
')#$
''#$
'!#$
'(#$
'"#$
%#$
(#$
(#$
*#$
+!#$
)#$ +)#$ ')#$ !)#$ ,)#$ %)#$ &)#$ ()#$ ")#$ *)#$ +))#$
-./012./$3445/$
6/78120/$3445/$
9.:1;120/$3445/$
<73=.120/$3445/$
>14?7120/$3445/$/=$@A7;$
B7C74/$D/E/$ F4/$;:7.C/$ 6/78$;:7.C/;$ 9.:1;$;:7.C/;$
71

A longer degree is usually a guarantee for increased debt. Students who enroll in short
programs, however, have a higher tendency to take out a loan, as shown in Figure 1 21.
However, the distribution is surprising: students who expect to complete their degree in
one or two years (a rare phenomenon among students enrolled full-time) have a greater
tendency to take out a loan for their studies.
We observed a statistically significant relationship for all forms of debt except for loans
from family members, which is relatively low, which is reflected the more parabolic
appearance of the distribution line.

Figure 6-20 : Change in debt ratio by source according to the duration of the diploma

!"#
$"#
%&"#
!'"#
("#
!"#
)#*+# '#*+,# !#*+,# %#*+,# -#*+,# &#*+,#./#01/,#
72

Again, the distributions of the average debt proved a little more sporadic. However, a
clear upward trend in average in a linear way in comparison to the duration of studies is
noted. There are linear relationships still here, but a small effect on the duration of the
degree of average debt can be noted. On average the high debt of respondents who
expect to complete their degree in one or two years is probably due to accumulated debt
over the course of a previous degree (here we are referring primarily to the college level,
where the potential of being in debt, while lower than that of university studies, remains
high). Due to the current length of time required to complete a degree, a student
completing a bachelor's degree certificate would also have a high accumulation of debt.

Figure 6-21 : Change in average amounts of debt by source according to the duration of the diploma

The next section illustrates the relationship between debt and longer studies in a more
shocking way.
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
-!"!!#$
%$./$ &$./0$ '$./0$ ($./0$ )$./0$ *$./0$12$
3420$
567$
89:;<$
6.=:44>?.=:0$
@1A.4$
!"!!!!#!!
$!%%%!!#!!
&!%%%!!#!!
'!%%%!!#!!
(!%%%!!#!!
)%!%%%!!#!!
)$!%%%!!#!!
)&!%%%!!#!!
)'!%%%!!#!!
)(!%%%!!#!!
$%!%%%!!#!!
)!*+! $!*+,! -!*+,! &!*+,! .!*+,! '!*+,!
/0!120,!
345!
6789:!
4*;822<=*;8,!
>/?*2!
73
6.2.4. Already Increased the Time to Complete Studies
Close to 35% of students said they have been required to extend their studies. However,
this extension is associated with higher debt rates, as shown in Figure 1-23. These
differences are statistically significant, the largest being in the rate of private debt. It can
be assumed that extending ones studies may lead some students to entering debt.
However, literature suggests that to go into debt causes students to prolong their
studies by encouraging the student to avoid debt by maximizing their main source of
independent funding: paid work during their studies. In fact, students who have
extended their studies have a total debt rate which is 1.2 times higher than those who
have never extended their university studies.
Figure 6-22 : Change in debt rates based on extensions of past university studies

The effect found on the debt ratio earlier is not reflected in the average debt. The only
significant statistical differences are on the SFA and total overall debt. This is normal as
a one-year extension adds about $ 2 440 to the debt of a student who subscribes to the
SFA. The surprising stability of other sources of debt confirms the idea that students
tend to adopt behaviours that limit their personal debt. A student who chooses to
extend his studies, for example by signing up for twelve credits per term rather than
fifteen in order to dedicate some of their time to work while there are pursuing their
studies which minimizes their personal debt.

Figure 6-23: Change in debt amounts based on extensions of past university studies

!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
!$3&)4-5-++-62$&*,,167)$
/-/$)285-/$
9:;<=>$ ?9;@=>$ A?;<=>$ BA;==>$
CD*$E*+*'/$-8$F$&*,,167-&$
/-/$)285-/$
??;?=>$ ::;?=>$ <G;9=>$ G=;H=>$
=;==>$
<=;==>$
A=;==>$
:=;==>$
?=;==>$
9=;==>$
G=;==>$
B=;==>$
H=;==>$
74

6.2.5. Areas of Study
We grouped the study areas based on their profitability. To do this, we categorized the
fields of study presented in the survey into two areas: high and low profitability
(Government of Quebec, in 2008 FEUQ, 2010a, p. 69). These categories are presented in
Table 1-8.
Table 6-9 : Classification of the curriculum according to their profitability

High profitability Low Profitability
Arts
Letters
Education
Humanities
Law
Applied Sciences
Business Administration
Health Sciences
Pure and Applied Sciences

Students in these two categories often possess different characteristics. In the fall of
2009, students in areas with higher economic efficiency had an employment rate which
was smaller, were required to spend more on school supplies, came from easier
upbringings, were less frequently first generation students, were often parents and have
a somewhat higher spending limits. Overall, they have, more favorable socioeconomic
characteristics but higher expenses for studies and lower course loads. Table 1 9
presents these correlation coefficients.

AlL rlve
lamllle/
amls
1oLal
A precedemmenL rallonge
ses eLudes
14 014,73 $ 8 170,37 $ 3 396,89 $ 13 266,01 $
n'a [amals eu a rallonger
ses eLudes
12 006,68 $ 7 931,06 $ 3 073,29 $ 13 114,11 $
- $
2 000,00 $
4 000,00 $
6 000,00 $
8 000,00 $
10 000,00 $
12 000,00 $
14 000,00 $
16 000,00 $
18 000,00 $
75
Table 6-10 : Correlation coefficients between field of study and certain characteristics

Coefficient
correlation
AutonomousEmployment
Rate2009
r
pb
=-,13***
SchoolSuppyRates r
pb
=,17***
FirstGenerationStudents =,06**
NetFamilyIncome r
pb
=,08**
Presenceofadependent
children
=-,05***

As shown in Figure 1-25, there is no indication that significant differences exist in the
debt ratio between economically viable low and high programs. The only statistically
significant difference (p <, 050) is that of student financial aid which remains marginally
lower.

Figure 6-24 : Change in debt ratio from source depending on the degree of economic profitability

Debt owed to the SFA does not vary depending on profitability of the degree chosen;
however, other forms of debt do have somewhat higher averages. This is quite normal,
as fortunately the loans and bursaries program does not fluctuate depending on
program of study. The differences are however, much greater in private and family
loans. The amounts loaned are much larger: about 1.5 times higher on average in both
cases. Financial needs are likely higher, pursuing studies in fields such as medicine and
engineering are very academically demanding, and often take longer to complete.
Another aspect that must be taken into consideration is the cost of specialized
equipment required in these programs (computer software, working tools), which are
often much higher than in other areas which the AFE does not cover.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
"*'3,-$&-42*3','2)$ 567689$ :67589$ ;<7:89$ =>7=89$
"1&2-$&-42*3','2)$ 5=7=89$ :?7589$ ;67<89$ =57@89$
87889$
;87889$
@87889$
:87889$
587889$
>87889$
=87889$
?87889$
76
In addition, we saw this in the previous section, financial institutions directly target
students enrolled in certain fields of study when giving credit loans. High debt rates are
likely the result of a low concentration of private loans which are abnormally high. In
fact, the total debt of students in private funding is only 1.1 times higher than the debt of
those registered in the areas of low profitability.
Figure 6-25 : Change in average amounts of debt by source based on the profitability of the diploma

6.2.6. At Risk Student Profiles
Some students have characteristics which identify them as having high-risk profiles.
Table 6-9 presents the results of statistical tests presented in the previous section.
Table 6-11 : Summary of statistical tests on school characteristics
AFE Private Family/Friends Total
Length
Rate;r
pb
,05*** ,06*** ,01 ,04***
Amount;r ,08*** ,14*** ,09** ,11***
Amount; ,07*** ,12*** ,01 ,10***
PastExtensions
Rate; -,08*** -,12*** -,09*** -,11***
Amount;r
pb
-,10*** -,01 -,02 -,09***
Profitability
Rate; -,02* -,01 -,01 -,02
Amount;r
pb
,00 ,18*** ,16*** ,07***
The length of time taken to complete ones studies partly explains the rising amount of
debt, particularly in the private sector, however, the relationship between the two is
weak. In addition, the more one continues in their academic career, the more they go
into debt and as a result, the amount of debt increases, regardless of its source. If we
start going into long lasting debt it is usually only once we have completed our studies
that we are able to repay the loan.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
"*'3,-$&-42*3','2)$ 56$7859:;$$<$$ =$8779;5$$<$$ >$:>89?>$$<$$ 58$5@>9;5$$<$$
"1&2-$&-42*3','2)$ 56$7:59:;$$<$$ ?$75?96@$$<$$ =$67;9=7$$<$$ 5>$75>9>@$$<$$
$A$$$$<$$
6$:::9::$$<$$
>$:::9::$$<$$
=$:::9::$$<$$
7$:::9::$$<$$
5:$:::9::$$<$$
56$:::9::$$<$$
5>$:::9::$$<$$
5=$:::9::$$<$$
77
The field of study has no impact on the debt ratio: however, students in areas of high
profitability are often more in debt than their colleagues in other disciplines. This can be
explained by higher material costs, greater difficulty balancing work and study and
aggressive financial offers for these students.
Lastly, extending ones studies has an influence on the interest rates as well as the
amount of debt one acquires throughout their academic career.
The analysis of socio-economic characteristics singled out students who are more
vulnerable and more prone to debt: the older students from lower socio-economic
backgrounds, who must leave home, those with children, and those who have extended
their studies or study in economically viable programs. Being first-generation students
also puts the student in a risky situation. In short, debt is more common in people who
need support: often those where financial support is low.
78
7. The Impact of Sources and Methods of Funding
In this section, we present the impact of total funding on student debt, as well as
the effect of paid work and parental contribution, which comprise two major sources of
funding. For a more detailed analysis on this subject, the reader is invited to consult the
survey, Sources and Methods of Funding for undergraduate students, published by the
Quebec Federation of University Students (FEUQ).
7.1. Total Funding
It is necessary to recall a few facts regarding total funding of undergraduate
studies. On one hand, we are talking about relatively low funding, with the average
falling at $13,819 and the median sitting at $12,500. This includes loans and bursaries,
merit grants, paid work, various family contributions and alimony.
Figure 7-1: Description of Total Funding

It is also interesting for us to focus on the number of available funding sources.
We have narrowed it down to four major sources of funding: paid work, family
contribution (which includes contributions from both parents and spouses), student
loans and various bursaries granted out of merit or for special projects. Figure 1-2
describes the results. The majority of students use two different sources of funding in
order to successfully carry out their studies. It is rare to be without any sources of
funding (a result mainly attributed to non-responses), or to have access to all four
sources of funding.
!"#
$%"#
&$"#
&'"#
$("#
)"#
*"#
*"#
+# $,++#-#
*+++,++#
*++$,++#-#
$++++,++#
$+++$,++#-#
$*+++,++#
$*++$,++#-#
&++++,++#
&+++$,++#-#
&*+++,++#
&*++$,++#-#
!++++,++#
!+++$,++.#
79
Figure 7-2: Description of Number of Funding Sources

With the exception of family debt, all forms of debt increase with an increase of
total funding. However, as Figure 1-3 illustrates, the effect is closer to resembling a
parabolic relationship for loans from student financial aid (SFA). It is natural that this
percentage would vary in such a way: the program allocates substantial financial
assistance (with the potential of acquiring close to $10,000 for the average university
year of thirty credits), and thereby raises the average amount of funding for which it is
one of the major constituents. The variation in private debt can be explained by a rise in
the tendency towards debt or by entirely private funding of university studies. We must
nonetheless note that in contrast to AFE loans, personal loans are not taken into account
when calculating the average funding of a university year.
Figure 7-3 : Variation in debt percentage by sources as a function of total funding

The average debt amount presents a more unforeseeable pattern. The only
statistically significant variation lies in the area of total debt. Debt owed to the AFE
!"#
!!"#
$%"#
!&"#
!"#
&# '# !# (# %#
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
!$ %"!!$-$
)!!!"!!$
)!!%"!!$-$
%!!!!"!!$
%!!!%"!!$-$
%)!!!"!!$
%)!!%"!!$-$
&!!!!"!!$
&!!!%"!!$-$
&)!!!"!!$
&)!!%"!!$-$
'!!!!"!!$
'!!!%"!!.$
/01$
23456$
078499:;784<$
=>=/?$
80
presents a barely linear pattern, better reflected by non-parametric tests (=0.09 to
p<0.001). However, the average debt is higher for students with funding ranging from
$20 000 to $30 000 per year; we can suppose that these are the students who reached
their debt limit with the AFE, or who, by working more, wish to avoid accumulating
supplementary debt for the sake of finishing their studies.
The average personal debt is a bit higher for students with total funding ranging
from $1 to $10 000; for some students, personal debt can replace AFE as a source of
funding, though the pattern is not widespread like we saw in section Error! Reference
source not found.*** As a general rule, the increase in funding negatively correlates
with the the level of personal debt (r=-0.03 to p<0.050), a sign that students are
attempting to avoid debt by working.
As for the variation in amounts of family loans, it is not statistically significant, As for the variation in amounts of family loans, it is not statistically significant,
despite an increase for students with high funding.
Finally, the total debt tends to grow as a function of total funding; here, there is
probably an effect of including AFE loans in the interpretation of total funding.
Figure 7-4: Variation of average amounts of debt by source as a function of total funding

Total funding varies in a sometimes unpredictable way with student debt;
however, we notice an increase in the average debt in relation to total funding, up to a
peak which takes place at the income bracket of $20 000 to $25 000. We are probably
finding a variation that is parallel to SFA student loans, since the study of other kinds of
debt is either inconclusive (family loans) or indicating a negative relationship (private
loans).
7.2. Paid work
Paid work represents 55% of the total funding of undergraduate studies. In most
cases, this is the only source of independent funding for students. However, without
studying the phenomenon in detail, we must remember that working throughout ones
studies has an overall negative impact on academic success and perseverance. Also,
literature uncovers a complex relationship between the two phenomena: many students
!"!!!!#!!
$!%%%!!#!!
&!%%%!!#!!
'!%%%!!#!!
(!%%%!!#!!
)%!%%%!!#!!
)$!%%%!!#!!
)&!%%%!!#!!
)'!%%%!!#!!
)(!%%%!!#!!
%! )*%%!"!
+%%%*%%!
+%%)*%%!"!
)%%%%*%%!
)%%%)*%%!"!
)+%%%*%%!
)+%%)*%%!"!
$%%%%*%%!
$%%%)*%%!"!
$+%%%*%%!
$+%%)*%%!"!
,%%%%*%%!
,%%%)*%%-!
./0!
12345!
/673889:673;!
<=<.>!
81
try to avoid debt by working. In the next section, by looking at the employment rate in
autumn 2009, at the numbers of hours worked in autumn 2009, and at the gross income
from paid work, we will see whether or not this phenomenon can be quantified. We
took autumn 2009 since it is the period for which we have the most valid data. That said,
the study of the winter 2009 period may reveal interesting changes in the employment
rate, but this falls outside of this studys objectives.
7.2.1. Presence or Absence of Paid Employment in Autumn 2009
During autumn 2009, 63% of students held at least one paying job. Balancing
work with study is therefore more a reality than ever. The rate of employment has no
discernible effect on total debt, which is surprising. However, there is a clearer effect on
the sources of debt. Statistically, the presence of an SFA loan has a lower effect on the
rate of employment in autumn 2009 (= -0.06). Thus, the presence of financial aid
seems to allow many students to devote themselves fully to a term of studies, which is
one of the programs goals. Yet conversely, the percentage of private debt is positively
correlated with the employment rate (=0.08). We can suppose that private loans
complete the funding obtained by paid work, particularly for students whose public
funding is insufficient via the student financial aid program (SFA).
Figure 7-5 : Variation in debt percentage by souce as a function of the presence of a job in autumn 2009

The study of average amounts of debt does nevertheless validate the hypothesis, as is
demonstrated by Figure 7.6. When students cannot hold a job, either because they are
registered for the entire year, including the summer (some are in a cooperative
program), or due to problems on the job market, it seems like debt is what compensates
for lack of funds. In fact, all the relationships are statistically significant and correlate
holding a job with having a lower debt. We have two plausible explanations for this:
1. On one hand, paid work is used as a strategy to limit accumulated debt, even for
students who are in debt. Hence, a substantial concern prevails regarding the
amount of accumulated debt, which propels students to work. The debt amounts
would then also be the determining factor in a students likelihood of working.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3*4/5-+6,1'$ 789:;<$ =:9>;<$ :;9?;<$ >798;<$
@AAB6*'2$*B$+1'4/$B4$
-+6,1'$
?79:;<$ ?;9C;<$ 8D9?;<$ >?97;<$
;9;;<$
8;9;;<$
:;9;;<$
=;9;;<$
?;9;;<$
7;9;;<$
>;9;;<$
C;9;;<$
82
2. On the other hand, we might conclude that students without jobs who finance
their studies by going into debt do so because on a practical level, it is impossible
for them to work during their studies. Earlier in this study, we labelled debt in
terms of fields of study. Often fields of higher economic profitability are more
difficult to balance with paid work and the costs are the highest. In these cases,
the lack of a job is mainly compensated for by going into debt.
When students cannot hold a job, either because they are registered for the entire year,
including the summer, or due to problems on the job market, it seems like debt is what
compensates for the lack of funds.
Figure 7-6 : Variation of average debt by source as a function of the employment rate

As we will see further on, the absence of parental contribution also plays a role in
student debt. Managing the costs of study without a job or the benefit of parental
contribution makes for a very arduous task and could easily be associated with high
debt.
7.2.2. Average number of hours worked per job in autumn 2009
Quebecois students work a lot. In order to obtain the following data, we made an
average number of hours worked per job in autumn 2009. This handling of data is not
ideal. Unfortunately, the data does not allow us to know the precise working periods.
Adding up the data would give markedly elevated results and would overrepresent
students who held several jobs concurrently during the given period. We must also
note that the median and the mean are very close to one another; we have therefore
favoured the mean. This method seems to underestimate the number of hours worked
over the course of study; SFAs study on students living conditions found an average
number of around 17 hours of paid work per week (MELS, 2009:58).
Of the 63% who combined work and full-time study in autumn 2009, 22%, that is, nearly
a quarter, worked 21 hours or more per week. The majority, 55%, worked between 11
and 25 hours per week. The average number of worked hours is around 16. The
breakdown is presented in Figure 1-8.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3*4/5-+6,1'$ 78$99:;:8$$<$$ =$9>8;?7$$<$$ >$@>:;A9$$<$$ 79$@B8;@7$$<$$
CDDE6*'2$*E$+1'4/$E4$
-+6,1'$
7A$A=7;>8$$<$$ :$88B;:=$$<$$ @$@=B;8=$$<$$ 78$7A7;?:$$<$$
$5$$$$<$$
A$BBB;BB$$<$$
@$BBB;BB$$<$$
>$BBB;BB$$<$$
?$BBB;BB$$<$$
7B$BBB;BB$$<$$
7A$BBB;BB$$<$$
7@$BBB;BB$$<$$
7>$BBB;BB$$<$$
7?$BBB;BB$$<$$
83
Figure 7-7 : Numbers of hours worked per week, autumn 2009

The gathered statistical testing show significant statistically differences for private loans
(rpb= 0.10 to p< 0.001). Students who take private loans seem to work more hours,
likely so theyll succeed in paying the interest associated with their loans. A correlation
also exists with the total debt (r
pb
=0.05 to p<0.010).
Error! Reference source not found. shows that the lowest percentage of total debt
occurs in students who work less than 20 hours a week; it is higher for those who do not
work, and for those who work more than 20 hours. Once again, the idea of a debt
avoidance strategy seems to be confirmed. We can assume that when choosing between
taking a loan from the AFE and working, many students are choosing to combine work
with studies. In many cases, however, we should not call this a choice, but rather an
obligation. SFAs criteria are notorious for their inability to acknowledge the reality
faced by students. An example among others is their policy on parental contribution.
We can therefore add a circumstantial obligation to work to the debt-avoidance strategy.
The data on private debt tends to confirm such a scenario. For example, the percentage The data on private debt tends to confirm such a scenario. For example, the percentage
of personal debt increases in a marked and statistically significant manner with the
number of hours worked during the course of study in autumn 2009. Resorting to
personal loans for living expenses seems to coincide with an intense combination of
work and study.
!!"#
$$"#
$%"#
!&"#
!'"#
("#
)"#
("#
!"#
!#*#+#,# %#*#!'#,# !!#*#!+#,#!%#*#$'#,#$!#*#$+#,#$%#*#)'#,#)!#*#)+#,#)%#*#('#,# (!,#-#
84
Figure 7-8 : Variation of the level of debt by source as a function of the number of hours worked per week per job in
autumn 2009

The model does not reveal any statistically significant relationships on the link between
the amount of accumulated debt and the number of hours worked. The level of debt
accumulated with the AFE seems pretty stable, with total variations being less than
$1000, apart from extreme values. To a great extent, this can be explained by the cap on
loans, which consolidates the size of loans. We saw that the main impact is on the debt
percentage and not on the debt amount.
In terms of private debt, however, it seems that students who work less than 20 hours a
week run into more debt than those who work more than 20 hours a week. Family
loans do not present a clear statistical relationship with the level of debt. However, they
are higher for students who worked few hours of paid work, a sign that these loans
respond more directly to an expressed financial need, a bit like private loans.

!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
-./$ 01234$ .5627789562:$ ;<=57$
%$>$)$?$
*$>$%!$?$
%%$>$%)$?$
%*$>$&!$?$
&%$>$&)$?$
&*$>$'!$?$
'%$>$')$?$
'*$>$(!$?$
(%?$@$
85
Figure 7-9: Variation of average amounts of debt by source as a function of the number of hours worked in 2009

Overall, students who work little, either by choice or obligation, go deeply into debt, just
like those who work a lot. The first case can be explained by a choice; the second, by
financial obligation. In the middle, we find students who work moderately, probably to
avoid accumulating supplementary debt, or else to fill the holes in their loans and
bursary program through whatever way they can.
7.2.3. The presence or absence of a job during the year
Its no surprise that the annual rate of employment operates in about the same way as
the employment rate for autumn 2009. Students without jobs are a bit more at risk of
going into debt and a bit less at risk of acquiring private debt. The other relationships
are not statistically significant.
Figure 7-10: Variation in debt percentage by source as a function of job status

!"!!!!#!!
!$!%%%!!#!!
!&!%%%!!#!!
!'!%%%!!#!!
!(!%%%!!#!!
!)%!%%%!!#!!
!)$!%%%!!#!!
!)&!%%%!!#!!
!)'!%%%!!#!!
!)(!%%%!!#!!
*+,! -./01! +23/445623/7! 89:24!
)!"!;!<!
'!"!)%!<!
))!"!);!<!
)'!"!$%!<!
$)!"!$;!<!
$'!"!=%!<!
=)!"!=;!<!
='!"!&%!<!
&)<!>!
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3*4/5-+6,1'$ 7897:;$ <=9>:;$ =?9=:;$ >797:;$
@AAB6*'2$*B$+1'4/$B4$
-+6,1'$
7>98:;$ <C98:;$ ?C9D:;$ >79D:;$
:9::;$
?:9::;$
=:9::;$
<:9::;$
7:9::;$
E:9::;$
>:9::;$
D:9::;$
86
Globally, students who dont have a job have a higher level of debt. All the statistical
testing is significant. The amount of family debt is again much higher for students
without jobs.
Figure 7-11: Variation of average debt amount by source as a function of job status

7.2.4. Gross Annual Income from a Job
Final indicator taken into account: the gross annual income from paid work. Contrary
to the two indicators taken into account earlier, this one gives an idea of the students
annual financial capacity. However, we must note that the data act in very similar
ways. The average annual income from work is around $8929, and the median is
around $7500. If we compare the annual income to the actually incurred spending (infra
section 8.1), we can thus easily see that the majority of students cannot subsist on paid
work. It is also important to remember that the period of 2009 was marked by a
significant rise in unemployment for young people during the summer season.
Figure 7-12 : Description of gross annual income from a job
Gross Income
Average $8 929
1st quartile $1 500
Median $7 500
3rd quartile $13 000
Note: for the statistical testing, we reported that people without a working income as
missing values. Hence, we are only targeting students who worked throughout the
year.
There are statistically significant relationships for all kinds of debt, but not for total debt.
We observe a decrease in total SFA debt as a function of total gross revenue income, but
the level of decline is low.
Once again, the percentage of private debt grows with the gross annual income (r= .07).
This result seems counter-intuitive; its not impossible that we find students who have
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
3*4/5-+6,1'$ 78$9:9$$;$$ <$=>>$$;$$ 9$:?>$$;$$ 7>$>:>$$;$$
@AAB6*'2$*B$+1'4/$B4$
-+6,1'$
7=$>C:$$;$$ 9$D77$$;$$ C$:88$$;$$ 78$:?9$$;$$
$5$$$$;$$
=$???$$;$$
C$???$$;$$
:$???$$;$$
D$???$$;$$
7?$???$$;$$
7=$???$$;$$
7C$???$$;$$
7:$???$$;$$
7D$???$$;$$
87
gone from full-time employee status to being full-time students. Hence, during their
studies, they find themselves with financial obligations which surpass their financial
capacities. However, a high gross annual income makes someone more eligible for
certain personal loans. It is also conceivable that these students have very high living
expenses which demand a great deal of paid work to simply make ends meet.
As for the percentage of family debt, it decreases slightly for those who work a lot. In
the end, the total percentage of debt remains quite stable, regardless of the level of gross
annual income from working. We can conclude that paid work replaces some funding
sources, but does not replace debt on its own. That said, it does not have the same effect
on the level of debt.
Figure 7-13 : Variation on debt percentage by source as a function of gross annual income

The total gross annual income has a marked impact on the average debt amount, and
this occurs regardless of the source. We find statistically significant correlations for all
forms of debt, with the most notable variations being with government financial aid, or
loans from family or friends. The weakest correlation, which is with private loans, can
explain the slightly elevated debts of students with a higher gross annual income (in
and around more than $20 000).
We must also note a major variation in the amount of money borrowed from parents or
family when students dont find a job. We can probably call this last resort support,
occurring when the student is unable to find a job and requires funding that a private
institution will not provide. Recall that families who lend the greatest amounts are
those in the first and fourth quartile of the gross parental income; we can assume that
the former have little money and cannot make a parental contribution, while the latter
are probably less sensitive to the issue of student debt.
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
-./$ 01234$ .5627789562:$ ;<=57$
!$ %"!!$>$)!!!"!!$ )!!%"!!$>$%!!!!"!!$
%!!!%"!!$>$%)!!!"!!$ %)!!%"!!$>$&!!!!"!!$ &!!!%"!!$>$&)!!!"!!$
'!!!%"!!?$
88
Figure 7-14 : Variation in average amounts of debt by source in function of gross annual income

It is here that we really see the role played by paid work. We can conclude without
hesitation that a high working income can, to a great extent, help limit debt accumulated
over the course of studies. Those who benefit from a high income during their studies,
particularly throughout the summer season, can establish a financial cushion that will
help them throughout the year and thus allow them to avoid accumulating too much
debt. Others do go into debt.
7.3. Parental Contribution
Parental contribution is the second greatest source of funding for registered full-time
undergraduate students. As the FEUQ restated (2011b), parental contribution is a legal
obligation written in the Civil Code of Quebec. The loans and bursaries program also
calculates a parental contribution in function of a contribution table. However, this
table is very far from actual reality. The effect of this table is that SFA contribution
correlates directly with family income; the more a family makes in a year, the less the
chance a student will have access to the program. The distribution of parental
contribution is proportional to family revenue. The richer a family is, the more likely a
student will receive a parental contribution. Table 1-1 presents the results of statistical
testing. As we can see, the presence of a parental contribution is strongly influenced by
gross annual income, and the amount of the contribution grows in function of income.
Table 7-1 : Correlation between family income and family contribution
Presence of parental contribution Amount of Parental contribution
Gross Annual Family Income r
pb
= -0.33*** r = 0.13***
The percentage of parental contribution remains low. For students who are presumed
to receive a contribution and are eligible for the loans and bursaries program (those
from a family with a gross income of between $30 000 and $60 000), only 52% actually
receive a parental contribution. The median contribution for these families is around
$2000.
!"!!!!#!!
!$!%%%!!#!!
!&%!%%%!!#!!
!&$!%%%!!#!!
!'%!%%%!!#!!
()*! +,-./! )01-223401-5! 67802!
%! &9%%!"!$%%%9%%! $%%&9%%!"!&%%%%9%%!
&%%%&9%%!"!&$%%%9%%! &$%%&9%%!"!'%%%%9%%! '%%%&9%%!"!'$%%%9%%!
:%%%&9%%;!
89
As a result, the presence and amount of the parental contribution should vary with
family income. However, it is possible in certain cases that a student does not receive a
parental contribution. This is the case for families whose relationships have
deteriorated, or for adult students who are returning to school and are generally
financially independent.
7.3.1. The presence of a parental contribution
According to our study, 61% of students receive a parental contribution. Receiving a
parental contribution greatly decreases the rate of borrowing from SFA (r
pb
= 0.37) and
from private institutions (r
pb
= 0.17). The difference is lower for loans coming from close
friends (r
pb
= - 0.01

to p< 0.010). The impact of not receiving family assistance is very
significant; parental contribution makes up 22% of the total funding for studies.
Without it, it seems to be very difficult for students to make it financially. The debt
percentage for students who dont receive parental assistance is 80%, 1.5 times higher
than for those who do. Recall that for the most part, these students come from poorer
families, or else they are going back to school. Thus, they make up vulnerable
populations.
Figure 7-15: Variation of debt percentage by source as a function of the presence of a parental contribution

A similar, if more limited pattern can be observed in the average amounts of
accumulated debt. The difference is not significant except for the debt owed to SFA and
for the total debt. Given the cap on loans, the $3800 difference on the average debt
owed to SFA, is troubling. We can conclude that many students without a declared
parental contribution began to go into debt from cegep, thus increasing the average
amount of accumulated debt. Others may have had to stretch out their studies beyond
the bursary eligibility period (recall that debt has a detrimental effect on the length of
study), and thus can only receive loans. Not receiving financial aid from parents
multiplies total average debt by 1.5.
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
%&)/-34-$5673-$
4132&'87913$:*&-32*,-$
;<=;>?$ ;@=<>?$ @A=B>?$ BC=D>?$
!8/-34-$5673-$
4132&'87913$:*&-32*,-$
D>=<>?$ CE=C>?$ @E=F>?$ E>=>>?$
>=>>?$
@>=>>?$
<>=>>?$
;>=>>?$
C>=>>?$
B>=>>?$
F>=>>?$
D>=>>?$
E>=>>?$
A>=>>?$
90
Figure 7-16: Variation of average debt amounts by source as a function of the presence of a parental contribution

7.3.2. Amount of Parental Contribution
We will now study the amount of received parental contribution. First of all, note that it
is important to remember the characteristics of families who provide parental
contributions: generally, the families are more comfortable financially, and often have a
university background. The majority of students who receive a parental contribution,
63%, receive a contribution of less than $3000. Thus we are talking about a contribution
that, at most, covers the cost of an average university year. 12% receive between $3000
and $4000, which covers tuition fees and some expenses, mainly the cost of
transportation. Finally, one quarter receive more than $4200. For the distribution set,
the average is from $3689, against a median contribution of in and around $2600.

AlL rlve
lamllle/
amls
1oLal
resence d'une
conLrlbuuon parenLale
10 601 $ 7 408 $ 3 941 $ 11 492 $
Absence d'une
conLrlbuuon parenLale
14 412 $ 8 728 $ 4 163 $ 16 727 $
- $
2 000 $
4 000 $
6 000 $
8 000 $
10 000 $
12 000 $
14 000 $
16 000 $
18 000 $
91
Figure 7-17: Description of Levels of Parental Contribution

For students receiving a parental contribution, is there an influence on student debt? It
seems like there definitely is. On one hand, the debt percentage from student financial
aid and the percentage of total debt diminish quite significantly. For students receiving
more than $4000 in assistance from their parents, we are talking about a debt percentage
that oscillates around 25%. Such a contribution, according to SFA calculations, would
be required from a gross annual income of $47 000, though a family earning $37 444
represents two parents working full time for $9.00 an hour, the minimum wage that was
in effect May 1
st
, 2009. As for the percentage of private debt, it diminishes slightly.
However, the percentage of family debt increases slightly with parental contributions
(r= 0.03 to p< 0.050). We can conclude that some parental contributions are accompanied
with loans, or that some contributions were considered to be loans.
Figure 7-18 : Variation in debt percentage by source as a function of the level of parental contribution

!"#$
%&#$
%%#$
!%#$
'#$
(#$
!&#$
!)**$+$
!***)**$
!**!)**$+$
%***)**$
%**!)**$+$
&***)**$
&**!)**$+$
,***)**$
,**!)**$+$
(***)**$
(**!)**$+$
-***)**$
-**!)**.$
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
%"!!$-$
%!!!"!!$
%!!%"!!$-$
&!!!"!!$
&!!%"!!$-$
'!!!"!!$
'!!%"!!$-$
(!!!"!!$
(!!%"!!$-$
)!!!"!!$
)!!%"!!$-$
*!!!"!!$
*!!%"!!.$
/01$ 23456$ 078499:;784<$ =>?79$
92
Many interesting tendencies emerge by studying Figure 7-20. On one hand, we can see
that students who receive assistance from $2000 to $3000 have the lowest average debt.
Recall that this is a group of students who receive assistance that corresponds
approximately with tuition fees. Those who receive more assistance also have higher
levels of average debt.
We also see a dramatic leap in family loans: 15% of the variation in the level of family
debt is explained by the level of parental assistance granted. Strangely, the more people
rely on their parents to pay for their studies, the more their family debt increases. The
other relationships are not statistically significant, though we do notice higher amounts
of private loans for the group with a parental contribution ranging from $3000 to $5000.
Figure 7-19: Variation of average debt amounts as a function of the level of parental contribution

7.4. Merit grants
In 2009, nearly 15% of students reported receiving a merit grant, or a grant for
internships. Bursaries are of an average value of $2616, but the median is much lower,
at $1721. They also make up a fairly marginal source of funding; they represent just 2%
of funding for full-time study in 2009 (FEUQ, 2010 : 42). We must also note that in this
specific case, since such a low percentage of students receive this funding, the data is
somewhat weak. Thus, the descriptive analyses are shown for your interest only. They
dont constitute as official data, since only statistical testing can truly be considered
significant.
!"!!!!#!!
!$!%%%!!#!!
!&!%%%!!#!!
!'!%%%!!#!!
!(!%%%!!#!!
!)%!%%%!!#!!
!)$!%%%!!#!!
!)&!%%%!!#!!
)*%%!"!
)%%%*%%!
)%%)*%%!"!
$%%%*%%!
$%%)*%%!"!
+%%%*%%!
+%%)*%%!"!
&%%%*%%!
&%%)*%%!"!
,%%%*%%!
,%%)*%%!"!
'%%%*%%!
'%%)*%%-!
./0! 12345! /673889:673;! <=>68!
93
Figure 7-20: Description of amounts for merit grants and grants for internships

Increasing the volume of institutionally awarded grants is a solution that is frequently
called-upon in order to reduce student debt. It is notable that schools in other provinces
have a much more substantial budget for financial aid than we do in Quebec. Does the
Quebecois experience allow us to conclude that these grants help reduce student debt?
7.4.1. The presence or absence of a merit grant
Figure 7-22 reveals a few slight differences in the profiles of students who receive
bursaries versus those who do not. Their debt percentage with the SFA is somewhat
higher (up to p< 0.050) and their private debt a little lower (up to p<0.010). However,
these figures do not represent very significant statistical implications.
Figure 7-21: Variation of debt percentage by source as a function of the presence or absence of a merit bursary

!"#$
%%#$
&&#$
"#$
'#$
&%#$
&())$*$
&)))())$
&))&())$*$
%)))())$
%))&())$*$
!)))())$
!))&())$*$
+)))())$
+))&())$*$
,)))())$
,))&())-$
!"#$ %&'()$
"*+',,-.
*+'/$
012*,$
!3/-45-$6784-$318&/-$6-$
+)&'2-$18$*82&-$
9:;<=>$ ?<;@=>$ AB;9=>$ :9;9=>$
%&)/-45-$6784-$318&/-$6-$
+)&'2-$18$*82&-$
@=;:=>$ ??;@=>$ AC;C=>$ :9;<=>$
=;==>$
A=;==>$
D=;==>$
?=;==>$
9=;==>$
@=;==>$
:=;==>$
C=;==>$
94
No linear statistical correlation can be observed in the average level of debt, as
illustrated in Figure 1-23
Figure 7-22: Variation of average debt amount by source as a function of the presence or absence of merit bursaries

It is likely that since institutional bursaries are generally only granted for a period of a
year, they dont have a longstanding effect on student finances. Also, some bursaries
are only granted for specific projects, such as an internship or the completion of a
project. Hence, bursaries only represent a partial remuneration and mostly follow the
correlations between work income and debt.
7.4.2. Level of Bursary
As we saw above, the size of institutionally granted bursaries at the
undergraduate level is often small. Does the size of a bursary have an observable impact
on debt? We already observed that the absence or presence of a bursary has little
impact. Figure 1-24 illustrates the variation in the average debt amount by source as a
function of the size of institutionally granted bursaries. No statistical test provides
conclusive evidence.
AlL rlve
lamllle/
amls
1oLal
Absence d'une bourse de
merlLe ou auLre
12 839 $ 7 948 $ 3 272 $ 13 993 $
resence d'une bourse de
merlLe ou auLre
12 664 $ 8 703 $ 3 077 $ 13 862 $
- $
2 000 $
4 000 $
6 000 $
8 000 $
10 000 $
12 000 $
14 000 $
16 000 $
95
Figure 7-23: Variation of debt percentage by source in function of the size of institutional granted bursary

Figure 1-25 illustrates the variation in debt amount as a function of the size of the
institutional bursary. The only statistically significant relationship is with the amount of
private debt ( = 0.12 to p < 0.010). However, as we can see, this does not constitute a
linear relationship. There are two foreseeable explanations. One is that the presence of
private debt is parallel to large bursaries granted for projects that require students to
travel. A second possibility is that large bursaries are granted to students who present
significant financial difficulties that may put the completion of their education at peril.
Figure 7-24: Variation of average debt amount by source as a function of the size of institutional bursaries

We are far from affirming the uselessness of student bursaries outside of those offered
by the loans and bursary program. To the contrary, such bursaries provide financial
respite to students who need them, and allow for the realization of projects that could
not be achieved otherwise. That said, this method of funding is much more useful for
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
-!"!!#$
%"!!$.$
%!!!"!!$
%!!%"!!$.$
&!!!"!!$
&!!%"!!$.$
'!!!"!!$
'!!%"!!$.$
(!!!"!!$
(!!%"!!$.$
)!!!"!!$
)!!%"!!/$
012$ 34567$ 1895::;<895=$ >?@8:$
!"!!!!#!!
$!%%%!!#!!
&!%%%!!#!!
'!%%%!!#!!
(!%%%!!#!!
)%!%%%!!#!!
)$!%%%!!#!!
)&!%%%!!#!!
)'!%%%!!#!!
)(!%%%!!#!!
)*%%!"!
)%%%*%%!
)%%)*%%!"!
$%%%*%%!
$%%)*%%!"!
+%%%*%%!
+%%)*%%!"!
&%%%*%%!
&%%)*%%!"!
,%%%*%%!
,%%)*%%-!
./0! 12345! /673889:673;! <=>68!
96
students enrolled in graduate studies, and it does not seem to be a particularly effective
method in this very specific case of student debt. Nevertheless, the increase in the
amount of private debt amongst students who receive large institutional bursaries
remains difficult to explain.
7.5. Establishing the complex and multifaceted links between funding and debt
We must dissect the funding in order to properly understand the relationship existing
between funding and debt. Analysis of total funding has become more complex by the
presence of SFA loans. The number of funding sources influences all types of debt,
except for family loans. The more students must combine sources of funding, the more
they are at risk of going into debt. Table 7-2 presents results from statistical testing.
97
Table 7-2 : Summary of statistical testing between sources of funding and debt
SFA Private Family/Friends Total
Total Funding
Rate; r
pb
0.09*** 0.10*** -0.02 0.07***
Amount; r 0.03 0.01 0.04 0.08***
Amount; 0.09*** 0.03* 0.00** 0.12***
Number of Funding
Sources

Percentage; -c 0.30*** 0.05*** 0.01 0.19***
Presence of
independent work
2009

Percentage; -0.06*** 0.08*** -0.02* -0.01
Amount; r
pb
-0.07*** -0.11*** -0.13*** -0.10***
Average hours
worked, autumn 2009

Percentage; r
pb
0.00 0.10*** 0.01 0.05**
Amount; r -0.01 0.00 -0.03 -0.02
Amount; 0.00 0.00 -0.04 0.00
Presence of Job
Percentage; -0.02*** 0.05*** -0.02* 0.00
Amount; r
pb
-.05** -.06** -.17*** -.07***
Total employment
income

Percentage; r
pb
-0.04** 0.07*** -0.04** -0.01
Amount; r -0.07*** -0.01 -0.02 -0.04
Presence of parental
contribution

Percentage; 0.37*** 0.17*** -0.01*** 0.26***
Amount; r
pb
.20*** .07*** -.12*** .23***
Amount of parental
contribution

Percentage; r
pb
-0.15*** -0.07*** 0.03* -0.11***
Amount; r -0.04 0.05 0.38*** 0.06**
Amount; -0.05* 0.06** 0.30*** -0.02

Presence of another
bursary

Percentage; 0.03* -0.04** -0.02 0.01
Amount; r
pb
-0.01 0.03 -0.01 -0.00
Amount of Bursary
Percentage; r
pb
-0.02 -0.02 -0.04 -0.04
Amount; r -0.03 0.10 0.10 0.05
Amount; 0.00 0.12** 0.11 0.04
The presence of a parental contribution, which is correlated with the gross family
income, predicts the level of student debt owed to the SFA and in private loans. It is the
greatest observable influence. The fact of receiving parental assistance significantly
diminishes the SFA and private debt percentages. The presence of a parental
contribution does not however have any impact on the level of private debt, and it
increases the amount of family debt.
Paid work is the main source of funding for studies. Not working in autumn 2009
tended to increase debt percentage with the SFA, though only slightly. The presence of
a job also slightly reduced the amount of accumulated debt. However, the number of
98
hours worked in autumn 2009 did not have a notable influence: the hypothesis that
paid work helps to avoid debt is still validated by the study on employment rate. The
rate of employment over the course of the year showed a similar result.
The gross annual income from work has but one influence on the amount of total debt:
the more students work a paying job, the less theyll be in debt.
Finally, student bursaries of various types, with the exception of SFA bursaries, do not
have a significant influence on debt. At this point, they are not useful tools in the
struggle against student debt, at least not for undergraduate students.

99
8. The impact of the levels of expenses
We saw that the sources of financing have a limited impact, but important. As a general
rule, the higher the sources of funding, the lower the rate of indebtedness. Cumulating
at least two sources of funding also seems to have an impact. What about expenses?
We will focus on the levels of expenses of students. We will proceed to an analysis in
four sequences: first, the level of total expenses, to then analyze university expenses,
living expenses and other expenses.
8.1. Total expenses
It costs more and more to study. Total expenses include the entirety of the expenses
incurred by students, such as tuition fees, school material, rent, food, transportation,
expenses for dependent children, paid alimony, recreational expenses and other
expenses.
One-third of students spend from $10,000 to $15,000 in a year; one-quarter, from
$15,000 to $20,000. The level of average expense is $16,434, compared to a median of
$14,640, indicating an asymmetrical distribution, as illustrated by Error! Reference
source not found.. We must note that 23% of students annually spend more than
$20,000 in tuition fees and living expenses.
We should note that the level of expenses is influenced, among other things, by the We should note that the level of expenses is influenced, among other things, by the
degree of autonomy of the student (place of residence, age, presence of dependent
children). This is often determined by the income level of parents: it is more probable for
a student from a less well-off family to leave the family home. As we will see, the level
of expenses has a direct influence on the level of indebtedness.
Figure 8-1: Distribution of total expenses

The level of expenses is directly correlated with the level of indebtedness, but in an often
more surprising manner. First, we should note that all relations are statistically
significant, the highest with total debt and private indebtedness.
!"#
$%"#
&'"#
!'"#
$$"#
%"#
("#
$)**#+#
,***)**#
,**$)**#+#
$****)**#
$***$)**#+#
$,***)**#
$,**$)**#+#
!****)**#
!***$)**#+#
!,***)**#
!,**$)**#+#
&****)**#
&***$)**-#
100
In regard to debt due to SFA, the most critical bracket is found in the $10,000 to $25,000
of annual expenses, namely 79% of students. This is where we find the highest rate of
indebtedness (from 50% to 57%). It subsequently drops. Expenses and income are
intimately related, and we can infer that students that spend more than $25,000 per year
have considerable income this is probably a return to studies or studies supported by
accumulated personal savings on the job market. Conversely, low levels of expenses
(less than $10,000) are probably related to living with ones parents.
The rate of private indebtedness grows rapidly, but manifests itself in full expansion
beginning at the level of expense of $15,001 and more, where it stabilizes around 45%.
The rate of family indebtedness grows in a slower manner, but also increases according
to the level of expenses in a quite important manner. Its maximum limit is 25%, for
students earning $25,000 and more. It is the only form of indebtedness that does not
decrease at the level of high expenses, which can be a sign of considerable financial
precariousness (which renders one ineligible for private loans) combined with a level of
high expenses, such as the one that is engendered by the presence of a dependent child,
and the absence of adapted student financial assistance. The close family then becomes
the last recourse.
Finally, the rate of total indebtedness follows a similar slope to that of the rate of
indebtedness in student financial assistance. Consequently, the more we spend, the
more we risk indebting ourselves as the sources and modes of funding run dry: this
seems particularly true for students presenting a level of expense between $10,000 and
$25,000 per year, which is far from being excessive.
Figure 8-2: Variation of the indebtedness rate by source based on the level of expenses

The relations between the level of expenses and the average level of indebtedness are
much more linear: a rapid glance at Error! Reference source not found. immediately
reveals it. All the relations are statistically significant.
The link between the level of expenses and indebtedness is surprising in student The link between the level of expenses and indebtedness is surprising in student
financial assistance. Students that present a level of relatively low expenses are probably
those that stay with their parents, which explains a much lower average indebtedness
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
%"!!$-$
)!!!"!!$
)!!%"!!$-$
%!!!!"!!$
%!!!%"!!$-$
%)!!!"!!$
%)!!%"!!$-$
&!!!!"!!$
&!!!%"!!$-$
&)!!!"!!$
&)!!%"!!$-$
'!!!!"!!$
'!!!%"!!.$
/01$ 23456$ 078499:;784<$ =>?79$
101
($8,000 to $9,000 accumulated, namely a little more than six semesters according to
regular maximum loan limit of SFA) than their colleagues with higher expenses, whose
average indebtedness upon leaving is between $12,000 and $15,000.
The amount of private indebtedness is distributed in a somewhat particular manner:
however, the relation is globally linear and the more expenses increase, the more the
average amount of the debt increases. The same motive can be observed in family loans.
Figure 8-3: Variation in the average amounts of indebtedness according to the level of expenses

The more we spend, the more we tend to indebt ourselves, and the more the debt
becomes heavier. There are nevertheless grounds to ask which expenses have significant
impacts on indebtedness?
8.2. University expenses
University expenses are of two kinds: tuition fees, which are all mandatory fees that
must be paid to enroll in a university course, and school material expenses.
8.2.1. Tuition fees
The link between tuition fees and indebtedness is polemical. By presenting its policy of
tuition fee hikes, the government of Quebec pretends that there will not be an increase
in student indebtedness. Now, as we will see, in the current situation, where there exist
variations between institutions in regard to mandatory institutional fees, the variation of
mandatory fees for studies has a significant impact on student indebtedness.
Firstly, the amount of tuition fees is, let us remember, fixed by the government of
Quebec. However, we also include ancillary fees, which vary according to the
institution, even the faculty. These amounts vary according to the number of university
credits, twelve being generally the minimum for being considered enrolled full-time in a
given term. In view of the simplified statistical treatment, we are using for this section
only the tuition fees of students who are of Quebec origins and permanent residents, the
other students have tuition fees that are much higher, as illustrated in Error! Reference
source not found.. For Quebec students, the average is $2,618 for a year studies.
Table 8-1: Description of the amounts of tuition fees according to citizenship status
!"!!!!#!!
$!%%%!!#!!
&!%%%!!#!!
'!%%%!!#!!
(!%%%!!#!!
)%!%%%!!#!!
)$!%%%!!#!!
)&!%%%!!#!!
)'!%%%!!#!!
)(!%%%!!#!!
$%!%%%!!#!!
)*%%!"!
+%%%*%%!
+%%)*%%!"!
)%%%%*%%!
)%%%)*%%!"!
)+%%%*%%!
)+%%)*%%!"!
$%%%%*%%!
$%%%)*%%!"!
$+%%%*%%!
$+%%)*%%!"!
,%%%%*%%!
,%%%)*%%-!
./0! 12345! /673889:673;! <=>68!
102
Table 8-1: Description of the amounts of tuition fees according to citizenship status
Average Median N
CanadianresidentinQuebec 2618$ 2550$ 8013
CanadianresidentoutsideQuebec 5031$ 4809$ 171
Permanentresident 2833$ 2700$ 348
Non-residentforeignerwithstudiespermit 7315$ 4900$ 321
Non-residentforeignerwithoff-campusworkpermit 8202$ 4572$ 38
Total 2865$ 2600$ 8938
Error! Reference source not found. describes the amounts of tuition fees paid in 2009 by
the students retained. Obviously, the majority of students paid between $1,000 and
$3,000 in tuition fees. However, 17% paid between $3,001 and $4,000, which can
correspond to two terms of study full-time and one part-time during the summer.
Figure 8-4: Description of amounts of tuition fees paid in 2009 by permanent residents and students of Quebec Figure 8-4: Description of amounts of tuition fees paid in 2009 by permanent residents and students of Quebec
origins

The rate of indebtedness is influenced by the amount of tuition fees, The most
significant relations are on private debt and family indebtedness (r
pb
=,04 et r
pb
=,05
p<,001), though relatively low. Error! Reference source not found. shows the variations:
influence is low, but present. The correlation with SFA is much lower and less
significant, which emanates from the very structure of the program.

!"#
$"#
%&"#
'("#
()"#
*"#
+"#
!# (,!!#-#
(!!!,!!#
(!!(,!!#-#
%!!!,!!#
%!!(,!!#-#
$!!!,!!#
$!!(,!!#-#
'!!!,!!#
'!!(,!!#-#
*!!!,!!#
*!!(,!!.#
103
Figure 8-5: Variation of rates of indebtedness by source according to the level of tuition fees paid

The amount of tuition fees explains between 1% and 2% of the variance of the level of
indebtedness. In brief, the higher the tuition fees in a given year, the higher the
indebtedness. Error! Reference source not found. shows it well (note that students who
do not pay tuition fees are a very small population, who present particularities that are
unique to them). We see a constant growth in the amount of indebtedness based on the
level of tuition fees paid.
Figure 8-6: Variation of the average amounts of indebtedness by source according to the level of tuition fees paid Figure 8-6: Variation of the average amounts of indebtedness by source according to the level of tuition fees paid

These results confirm on a small scale what should be in evidence: if tuition fees
increase, student indebtedness increases!
!"#
$!"#
%!"#
&!"#
'!"#
(!"#
)!"#
*!"#
+!"#
!# $,!!#-#
$!!!,!!#
$!!$,!!#-#
%!!!,!!#
%!!$,!!#-#
&!!!,!!#
&!!$,!!#-#
'!!!,!!#
'!!$,!!#-#
(!!!,!!#
(!!$,!!.#
/01# 23456# 078499:;784<# =>?79#
!"!!!!#!!
!$!%%%!!#!!
!&%!%%%!!#!!
!&$!%%%!!#!!
!'%!%%%!!#!!
!'$!%%%!!#!!
!(%!%%%!!#!!
%! &)%%!"!
&%%%)%%!
&%%&)%%!"!
'%%%)%%!
'%%&)%%!"!
(%%%)%%!
(%%&)%%!"!
*%%%)%%!
*%%&)%%!"!
$%%%)%%!
$%%&)%%+!
,-.! /0123! -45166784519! :;<46!
104
8.2.2. School material
Mandatory school material is the other form of expense contracted during the course of
studies. This consists of an annual average expense of $607, although an important
portion of students, namely 15.8%, spent more than $1,000 in mandatory school
material. The detail is presented in Error! Reference source not found..
Figure 8-7: Description of the amounts paid in school material Figure 8-7: Description of the amounts paid in school material

High expenses in school material also lead to indebtedness. Error! Reference source not
found. illustrates the variation. In addition to the students who declared no expense,
namely 0.4% of respondents, we see a continuous growth of indebtedness with school
material expenses. The strongest effect (r
pb
=,10 to p<,001) is on private indebtedness,
which is explained by the imperfect coverage of the loans and bursaries program for
this type of expenses (we grant a lump sum and not an amount linked to the expenses
really incurred). The impact is obviously low, which explains the relatively minimal
amounts in play, namely 4% of total expenses.
Figure 8-8: Variation of indebtedness rates according to the level of expenses on school material
!"#$%
&'"#$%
(#"#$%
&)"($%
&("*$%
+")$%
#"&$%
+",$%
!% &"!!%-%
.+!"!!%
.+&"!!%-%
+!!"!!%
+!&"!!%-%
*+!"!!%
*+&"!!%-%
&!!!"!!%
&!!&"!!%-%
&.+!"!!%
&.+&"!!%-%
&+!!"!!%
&+!&"!!/%
105
Figure 8-8: Variation of indebtedness rates according to the level of expenses on school material

The same type of relation is observable for the level of indebtedness. Here, the impact is
stronger for private indebtedness and family indebtedness (r=,12 and r=,13 to p<,001).
However, all the forms of indebtedness are influenced by higher expenses in school
material, as we can see in Error! Reference source not found..
Figure 8-9: Variation of the average amounts of indebtedness by source according to the level of expenses in Figure 8-9: Variation of the average amounts of indebtedness by source according to the level of expenses in
school material

8.3. Living expenses
Paying for the university bill does not exempt students from spending on their living
expenses. We will study three types of expenses, namely rent, food and transportation
expenses.
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
-!"!!#$
!$ %"!!$.$
&)!"!!$
&)%"!!$.$
)!!"!!$
)!%"!!$.$
+)!"!!$
+)%"!!$.$
%!!!"!!$
%!!%"!!$.$
%&)!"!!$
%&)%"!!$.$
%)!!"!!$
%)!%"!!/$
012$ 34567$ 1895::;<895=$ >?@8:$
!"!!!!#!!
$!%%%!!#!!
&!%%%!!#!!
'!%%%!!#!!
(!%%%!!#!!
)%!%%%!!#!!
)$!%%%!!#!!
)&!%%%!!#!!
)'!%%%!!#!!
)(!%%%!!#!!
$%!%%%!!#!!
%! )*%%!"!
$+%*%%!
$+)*%%!"!
+%%*%%!
+%)*%%!"!
,+%*%%!
,+)*%%!"!
)%%%*%%!
)%%)*%%!"!
)$+%*%%!
)$+)*%%!"!
)+%%*%%!
)+%)*%%-!
./0! 12345! /673889:673;! <=>68!
106
8.3.1. Rent
For the series of tests that follow, we have preferred to analyze the rent expense of
students who do not live with their parents. We have already seen, in any case, the fact
that being a resident or not had a very important influence on the level of indebtedness
as on the rate of indebtedness (supra section Error! Reference source not found.).
A student spends on average $6,899 for his rent, which includes the cost of rent, A student spends on average $6,899 for his rent, which includes the cost of rent,
electricity, telephone, heating and furniture. The median expense is $6,000 annually, a
sign of the existence of very high rent expenses in the distribution. Most students, 65 %,
spend between $2,501 and $7,500 per year for their housing, which corresponds to a
monthly expense ranging from $208 per month to $625 per month. Various factors
influence the cost of rent, such as the number of housemates, the place of residence, and
the duration of the rental. A student that returns to his parents in the summer will save
substantial amounts. Various students have high rent expenses: we can think this
consists of student parents or students returning to studies. Given that we have
excluded non-residents, students that have rent expenses of $2,500 and less per year is
very low (5%): we can believe that rent is taken into account by a partner, or (more
probably) students that were only tenants for autumn 2009, or only for study terms.
Figure 8-10: Description of rent expenses

The more we spend on rent, the more we become indebted: the relation is true for all
sources of indebtedness, as is illustrated by Error! Reference source not found.. It is
statistically significant for all forms of indebtedness, with the strongest relation for
private indebtedness (r=,11 to p<,001).
Figure 8-11: Variation of rates of indebtedness by source according to the level of expenses on rent
!"#
$%"#
$&"#
&!"#
'"#
("#
%"#
&)**#+#
(!**)**#
(!*&)**#+#
!***)**#
!**&)**#+#
,!**)**#
,!*&)**#+#
&****)**#
&***&)**#+#
&(!**)**#
&(!*&)**#+#
&!***)**#
&!**&)**-#
107
Figure 8-11: Variation of rates of indebtedness by source according to the level of expenses on rent

The same type of relation is at work on the amount of indebtedness, except that it is the
amount of student financial assistance that presents the strongest correlation (r=,11 to
p<,001), while the link is very weak (r=,06 p<,010) for private indebtedness.
Figure 8-12: Variation of the average amounts of indebtedness by source according to the level of expenses on rent

Since the expenses associated with rent are generally the most important among all
expenses, it is normal that they are the most frequently correlated with student
indebtedness. Now, this is an expense that is rarely compressible, and rather dependent
on the local market. Regions with rates of very low vacancy, such as Abitibi-
Tmiscamingue, do not leave much choice to students that must find a place to live. In
other cases, it is the presence of dependent children that increases in a significant
manner the costs associated with rent. The only solution then becomes co-tenancy,
which is a widespread phenomenon: 28,9% of recipients of SFA shared a dwelling in
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
-!"!!#$
%"!!$.$
&)!!"!!$
&)!%"!!$.$
)!!!"!!$
)!!%"!!$.$
+)!!"!!$
+)!%"!!$.$
%!!!!"!!$
%!!!%"!!$.$
%&)!!"!!$
%&)!%"!!$.$
%)!!!"!!$
%)!!%/$
012$ 34567$ 1895::;<895=$ >?@8:$
!"!!!!#!!
!$!%%%!!#!!
!&%!%%%!!#!!
!&$!%%%!!#!!
!'%!%%%!!#!!
!'$!%%%!!#!!
&(%%!"!
'$%%(%%!
'$%&(%%!"!
$%%%(%%!
$%%&(%%!"!
)$%%(%%!
)$%&(%%!"!
&%%%%(%%!
&%%%&(%%!"!
&'$%%(%%!
&'$%&(%%!"!
&$%%%(%%!
&$%%&*!
+,-! ./012! ,34055673408! 9:;35!
108
2007, and 23.8% lived with a spouse, against 17.1% and 18.1% for non-recipients (MELS,
2009, p. 47). Allowable expenses in SFA are notoriously low, and we can well believe
that it is students who rely on it that are the most adversely affected, thus having to take
out a private debt that they necessarily wouldnt have had to contract otherwise.
8.3.2. Food
Food expenses are the second most important item of expenses. The average annual
expense is $3,294, with a median expense of $2,400. This consists of an expenses item
that is very stable based on the socioeconomic characteristics of students, but that seems
to be influenced by age (expenses increase with age).
We should note that only half of students succeed in feeding themselves with $7 per day
at least, the criteria of SFA (FEUQ, 2010: 55). Error! Reference source not found.
describes the characteristics of food expenses.
Figure 8-13: Description of food expenses Figure 8-13: Description of food expenses

The rate of indebtedness presents variations that are difficult to define when we put in
relation with food expenses. However, we find significant statistical variations of the
indebtedness rate for all sources when food expenses increase. The presence of
dependent children just like age are determinants of the level of expenses on food, and
this can probably explain a part of the relation between expenses on food and
indebtedness.
We should note this is a form of expense that is difficult to compress without putting in
peril ones own health, which is not desirable for reasons that do not need explanation.
Thus, the rate of indebtedness, just as the variations in the amounts of the debt, cannot
be contained other than an increase in available funding, preferably in the form of
bursaries.
It is particularly worrisome to see an important increase in the rate of private
indebtedness in students, who for various reasons, present significant expenses on food.
It very probably consists of credit card expenses, which incur very high fees.
!"#
$%"#
&$"#
%'"#
%%"#
("#
)"#
%*++#,#
%+++*++#
%++%*++#,#
$+++*++#
$++%*++#,#
&+++*++#
&++%*++#,#
'+++*++#
'++%*++#,#
-+++*++#
-++%*++#,#
(+++*++#
(++%*++.#
109
Figure 8-14: Variation of the rate of indebtedness by source according to the level of expenses on food

All the relations between expenses on food and average indebtedness are statistically
significant (for family loans, it is significant to p < ,050). The different forms of debt
stabilize in amounts of expenses on food of more than $3,000, except for private loans,
which drop for the bracket ranging from $5,001 to $6,000 of annual expenses. Food
expenses, however, only explain a low part of total indebtedness: 1.3% of the variance. It
is, of course, an expense that is difficult to compress.
Figure 8-15: Variation of the average amounts of indebtedness by source according to the level of expenses on food

From the moment we indebt ourselves, it seems we have to limit our expenses in regard
to food: doing the contrary creates a significant problem of indebtedness. This augurs
badly in the current period of food inflation that we are experiencing. This is
particularly true for student parents, who have to assume high food expenses.
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
-!"!!#$
%!!"!!#$
%"!!$.$
%!!!"!!$
%!!%"!!$.$
&!!!"!!$
&!!%"!!$.$
'!!!"!!$
'!!%"!!$.$
(!!!"!!$
(!!%"!!$.$
)!!!"!!$
)!!%"!!$.$
*!!!"!!$
*!!%"!!/$
012$ 34567$ 1895::;<895=$ >?>0@$
!"!!!!#!!
$!%%%&%%!!#!!
'!%%%&%%!!#!!
(!%%%&%%!!#!!
)!%%%&%%!!#!!
*%!%%%&%%!!#!!
*$!%%%&%%!!#!!
*'!%%%&%%!!#!!
*(!%%%&%%!!#!!
*)!%%%&%%!!#!!
*&%%!"!
*%%%&%%!
*%%*&%%!"!
$%%%&%%!
$%%*&%%!"!
+%%%&%%!
+%%*&%%!"!
'%%%&%%!
'%%*&%%!"!
,%%%&%%!
,%%*&%%!"!
(%%%&%%!
(%%*&%%-!
./0! 12345! /673889:673;! <=<.>!
110
8.3.3. Transportation
Transportation expenses are an important part, but difficult to define, of student
expenses, given the variety of modes of transportation and varied distances. The
average annual expense is $851 per year, and the median is $549 per year, which
corresponds to a CAM at a reduced rate. The description of transportation expenses
presented in Error! Reference source not found. is nonetheless richer. 16% of students
do not have regular transportation expenses (a sign they favor an active transportation
or benefit from a universal collective transportation program, such as at the Universit
de Sherbrooke), and the majority spends less than $1,000 per year on transportation.
Figure 8-16: Description of expenses on transportation Figure 8-16: Description of expenses on transportation

Error! Reference source not found. describes the most regularly used mode of
transportation in autumn 2009. Obviously, it is the personal vehicle that is the most
expensive, with an average annual expense of $1,548. We should note that automobiles
loans are not included in private indebtedness: we can thus believe that most students
that have to use a car use an already paid used vehicle, or borrow that of their parents or
a relation. Furthermore, this is only 19% of students. However, more than three-quarters
of students use a collective or active mode of transportation to attend university, which
are much less expensive.
Table 8-2: Description of the mode of transportation used most regularly in autumn 2009 and average annual
!"#$
%!#$
&'#$
!(#$
"#$
)#$
*$ !+**$,$(**+**$ (*!+**$,$
!***+**$
!**!+**$,$
!(**+**$
!(*!+**$,$
&***+**$
&**!+**-$
111
Table 8-2: Description of the mode of transportation used most regularly in autumn 2009 and average annual
expense associated
Percentage Average
annual
expenses
Publictransit 56% 837$
Bicycle 4% 347$
Walking 17% 208$
Car-pooling 3% 1190$
Personalvehicle 19% 1548$
The offer of a public transit service is an important determinant of the mode of
transportation used. It is 88% of students that study in Montreal who choose active or
collective travelling, compared to 36 % of students in remote regions, who nonetheless
very often favor walking. Error! Reference source not found. illustrates the modes of
transportation used. We can conclude that the quality of the collective transportation
offered is an important determinant of the costs of transportation: however, when it is
absent or deficient, students nonetheless search the most economical mode of
transportation, and look toward active transportation and car-pooling to reduce the
costs associated with transportation.
Table 8-3: Description of the mode of transportation used most regularly in autumn 2009 based on the region of Table 8-3: Description of the mode of transportation used most regularly in autumn 2009 based on the region of
studies
Montreal Quebec Central
regions
Remote
regions
Publictransit 73% 40% 31% 6%
Bicycle 4% 5% 2% 1%
Walking 11% 23% 27% 29%
Car-pooling 2% 4% 5% 7%
Personalvehicle 9% 27% 32% 55%

We clearly note that those who spend a lot on transportation have higher rates of
indebtedness, as revealed in Error! Reference source not found.. We can also see a
difference of nearly five percentage points between the total rate of indebtedness of
students that spend less than $500 and those that spend between $501 to $1,000 per year.
This roughly consists of the difference between a CAM bus pass at a reduced price and
another at a regular price. Students with levels of expenses ranging from $1,001 to
$2,000 correspond to more onerous collective transportation services, generally
associated with the Montreal suburbs. The lower effect of indebtedness can be explained
by the fact of their living with their parents.
The model detects significant statistical correlations for SFA and private indebtedness: The model detects significant statistical correlations for SFA and private indebtedness:
these are nonetheless often low (r
pb
=-,03 p<,010 et r
pb
=,08 to p<,001). The distribution is
not very linear given the significant jumps between expenses for students travelling on
foot or bicycle, on public transit and personal vehicle.
112
Figure 8-17: Variation of the rates of indebtedness by source according to the level of expenses on transportation

The motive remains very similar for Error! Reference source not found.: however, we
do not find significant statistical correlations. The descriptive analysis does not allow us
to ascertain particularly interesting motives.
Figure 8-18: Variation in the average amounts of indebtedness by source according to expenses on transportation Figure 8-18: Variation in the average amounts of indebtedness by source according to expenses on transportation

We can thus conclude that contrary to a received idea, students generally seem to travel
according to their means. Transportation expenses are a quite bad indicator of student
indebtedness, and a surprising portion of these have no expense on transportation, a
sign that they choose active transportation (walk, bicycle), that they live close to their
place of studies or that they benefit from a universal pass, such as at the Universit de
Sherbrooke.
!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
!$ %"!!$-$
)!!"!!$
)!%"!!$-$
%!!!"!!$
%!!%"!!$-$
%)!!"!!$
%)!%"!!$-$
&!!!"!!$
&!!%"!!.$
/01$ 23456$ 078499:;784<$ =>?79$
!"!!!!#!!
$!%%%&%%!!#!!
'!%%%&%%!!#!!
(!%%%&%%!!#!!
)!%%%&%%!!#!!
*%!%%%&%%!!#!!
*$!%%%&%%!!#!!
*'!%%%&%%!!#!!
*(!%%%&%%!!#!!
%! *&%%!"!
+%%&%%!
+%*&%%!"!
*%%%&%%!
*%%*&%%!"!
*+%%&%%!
*+%*&%%!"!
$%%%&%%!
$%%*&%%,!
-./! 01234! .5627789562:! ;<=57!
113
8.4. Other expenses
Two other types of expenses are difficult to classify. This consists of expenses for
dependent children, recreational expenses and other expenses.
8.4.1. Expenses for dependent children
We saw earlier that the fact of having a child or not leads to very important expenses
and increased indebtedness. Now, the level of expense for dependent children also has
an impact.
24
Annually, expenses for children represent for respondents that are parents,
an average expense of $5,587 and a median expense of $4,347. Error! Reference source
ot found. describes the situation. Most parents have expenses for dependent children
that oscillate between 0 and $6,000, with most spending from $2,000 to $4,000. The
scarcity of daycare services in a university setting (CNCS-FEUQ, 2011) probably leads to
its lot of additional expenses in many cases. Furthermore, this here consists of expenses
assumed by one of the two parents, and not expenses incurred by the total family unit.
Figure 8-19: Description of expenses for dependent children

Error! Reference source not found. illustrates the variation of the rate of indebtedness
by sources according to expenses for dependent children. We do not really see a very
interesting motive, and only one relation is statistically significant: private indebtedness
increases with the level of expenses for dependent children (r
pb
=,11 p <,050). Private
credit is thus more widespread in student parents who have high expenses for their
children, whether it be through the presence of various dependent children or because
of the age of the child.
Figure 8-20: Variation of the rates of indebtedness by source according to the level of expenses for dependent

24
This is a very small sub-sample (408 respondents). We thus have to proceed with caution!
!"#
$%"#
!&"#
&'"#
("# ("#
$)"#
)# $*))#+#
&)))*))#
&))$*))#+#
%)))*))#
%))$*))#+#
,)))*))#
,))$*))#+#
-)))*))#
-))$*))#+#
$))))*))#
$)))$*)).#
114
Figure 8-20: Variation of the rates of indebtedness by source according to the level of expenses for dependent
children

The correlations are difficult to make on the level of indebtedness. Furthermore, we
often find ourselves with very small samples. There are no statistically significant linear
relations, as we can see in Error! Reference source not found.. However, the level of
indebtedness in SFA and the private are correlated with the level of expenses for
dependent children ( = -,11 and = ,13 to p<,050). The level of indebtedness in SFA
diminishes with expenses for dependent children, whereas the level of total
indebtedness increases. We should note that the samples are miniscule and should be
taken with caution.
Figure 8-21: Variation of the average amounts of indebtedness by source according to the level of expenses for Figure 8-21: Variation of the average amounts of indebtedness by source according to the level of expenses for
dependent children

!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,!"!!#$
-!"!!#$
%!!"!!#$
!$ %"!!$.$
&!!!"!!$
&!!%"!!$.$
(!!!"!!$
(!!%"!!$.$
*!!!"!!$
*!!%"!!$.$
,!!!"!!$
,!!%"!!$.$
%!!!!"!!$
%!!!%"!!/$
012$ 34567$ 1895::;<895=$ >?@8:$
!"!!!!#!!
!$!%%%!!#!!
!&%!%%%!!#!!
!&$!%%%!!#!!
!'%!%%%!!#!!
!'$!%%%!!#!!
%! &(%%!"!
'%%%(%%!
'%%&(%%!"!
)%%%(%%!
)%%&(%%!"!
*%%%(%%!
*%%&(%%!"!
+%%%(%%!
+%%&(%%!"!
&%%%%(%%!
&%%%&(%%,!
-./! 01234! .5627789562:! ;<=57!
115
The level of expenses for dependent children has few significant statistical relations with
indebtedness. Such a situation is contrary to what prevails for the difference between
students with and without dependent children, we must note.
8.4.2. Recreational expenses and other expenses
Often blamed in the popular media to explain the financial problems of students,
recreational expenses and other expenses remain a relatively low part of student
expenses. The average expense is $2,339, but the median, at $1,300, is much lower. This
indicates a weak concentration of very high expenses, as revealed by Error! Reference
source not found.. A low percentage of students (3%) do not declare any recreational
expenses; conversely, 9% declare annual expenses of more than $5,000. The vast
majority spend less than $3,000 annually. We must also point out a small concentration
of students with very high annual funding compared to average funding, which distorts
the picture somewhat.
Figure 8-22: Description of recreational and other expenses Figure 8-22: Description of recreational and other expenses

How do these expenses compare with the expenses of Quebec households? To evaluate
this difference, we have used the data of the Survey of Household Spending, edition
2008 (the last available in the democratization initiative of the data of Statistics Canada).
To obtain a similar portrait to that of sources and modes, we aggregated various data:
recreation, reading material, expenses on tobacco and alcoholic drinks as well as other
expenses. It emerges that student expenses are lower than those of Quebec households:
!"#
$%"#
!&"#
&'"#
("#
)"#
%"#
*# &+**#,#
&***+**#
&**&+**#,#
$***+**#
$**&+**#,#
!***+**#
!**&+**#,#
)***+**#
)**&+**#,#
-***+**#
-**&+**.#
116
Table 8-4 : Comparison of measures of the central trend of recreational expenses for Quebec households and
students
Typeofhousehold Average Median
Oneperson 3410$ 2295$
Coupleonly 5628$ 4433$
Couplewithchildren,singlesonly 7997$ 6621$
Couplewithotherrelatedpersonornotrelated 8006$ 4984$
Singleparentwithoutadditionalperson 4711$ 3775$
Otherhouseholdwithrelatedperson(s) 5147$ 2492$
Otherhouseholdwithnon-relatedperson(s) 4860$ 3774$
Total 5477$ 4085$
Students 2339$ 1300$
The most comparable category, namely that of a household composed of only one
person, presents much higher average and median expenses than students: this trend is
true for all types of households. However, these data are more difficult to compare
given that they include more than one individual.
How are these expenses funded? On the one hand, it is important to note that students
that have a debt to SFA spend a little less on recreation than colleagues without financial
assistance. The same motive (significant to p < ,010) can be observed on the total debt. A
very low positive effect can be noticed for those who indebt themselves to the private
(significant to p <,050), while no trend is ascertained for family loans. Error! Reference
source not found. visually illustrates this in a very striking manner.
We should note that very high expenses on recreation can be associated with a We should note that very high expenses on recreation can be associated with a
particular project (a trip, for example), for which the student was able to economize
during a certain number of years, or can self-finance through a job. The negative
correlation between the rate of indebtedness and expenses on recreation seem to show
this thesis.
Figure 8-23: Variation of expenses on recreation according to the rate of indebtedness by source

!"!!#$
%!"!!#$
&!"!!#$
'!"!!#$
(!"!!#$
)!"!!#$
*!"!!#$
+!"!!#$
,-.$ /0123$ -45166784519$ :;<46$
!$ %"!!$=$%!!!"!!$ %!!%"!!$=$&!!!"!!$
&!!%"!!$=$'!!!"!!$ '!!%"!!$=$(!!!"!!$ (!!%"!!$=$)!!!"!!$
)!!%"!!>$
117
The study of the average amounts of debt do not allow us to ascertain significant
relations, except on the level of total debt, that varies negatively, but in a very weak
manner, with the level of expenses (r=-,03 to p<,050). We notice, however, that the level
of indebtedness to SFA diminishes up until expenses of $2,000 to $4,000, to then rise
again. The level of private indebtedness also seems a bit higher, but not in a significant
manner.
Figure 8-24: Variation in recreational expenses and other expenses according to the average amounts of
indebtedness by source

Recreational expenses do not seem to be a good explanatory factor of student
indebtedness. We even notice that students compress their expenses in this item when
they already have current indebtedness. Non-parametric tests reveal two negative
statistical relations, on the level of indebtedness to SFA ( = -,05 to p<,010) and on the
total level of indebtedness (=-,05 to p<,001). The higher expenses on recreation are
probably only the effect of a minority that benefits from relatively lower living expenses.
8.5. The level of living expenses increases the level of indebtedness
The increase in the level of expenses has a general impact on student indebtedness. In
fact, the higher the expenses, the more we tend to indebt ourselves, and the higher
indebtedness will be. The level of total expenses alone explains 3% of variance in total
indebtedness. However, certain expenses have more influence than others.
Mandatory expenses (tuition fees and school material) have a direct impact on
indebtedness: the higher the cost, the more we will have recourse to indebtedness.
Living expenses, such as food and rent, have more or less the same behavior.
Other expenses behave in a more surprising manner. The relation between
transportation expenses and SFA is negative. We can then think that students who tend
to indebt themselves will reduce their transportation expenses, by favoring collective or
active transportation, or by living near the university.
Expenses for dependent children do not have a significant impact on the level of
indebtedness. It always consists of a very small sub-sample where indebtedness is
!"!!!!#!!
$!%%%!!#!!
&!%%%!!#!!
'!%%%!!#!!
(!%%%!!#!!
)%!%%%!!#!!
)$!%%%!!#!!
)&!%%%!!#!!
)'!%%%!!#!!
*+,! -./01! +23/445623/7! 89:24!
%! );%%!"!)%%%;%%! )%%);%%!"!$%%%;%%!
$%%);%%!"!<%%%;%%! <%%);%%!"!&%%%;%%! &%%);%%!"!=%%%;%%!
=%%);%%>!
118
already very high. Students do not tend to decrease the level of expenses devoted to
dependent children, which is reassuring.
Finally, recreational expenses and other expenses are not a source of indebtedness. In
fact, students who use SFA spend a bit less than their colleagues who do not use it.
However, there is a slight positive relation in regard to private indebtedness.
Table 8-5: Statistical analysis of the links between the sources of expenses and indebtedness
SFA Private Family Total
Totalexpenses
Rate;r
pb
,08*** ,14*** ,11*** ,10***
Amount;r ,15*** ,12*** ,15*** ,18***
Tuitionfees
Rate;r
pb
,02* ,04*** ,05*** ,03*
Amount;r ,15*** ,12*** ,13*** ,10***
Schoolmaterial
Rate;r
pb
,06*** ,10*** ,07*** ,08***
Amount;r ,08*** ,12*** ,13*** ,11***
Rent
Rate;r
pb
,06*** ,11*** ,07*** ,05***
Amount;r ,11*** ,06** ,08** ,14***
Amount; ,13*** ,07*** ,09*** ,15***
Food
Rate;r
pb
,05*** ,09*** ,08*** ,06***
Amount;r ,08*** ,07*** ,06* ,12***
Transportation
Rate;r
pb
-,03** ,08*** ,01 ,00
Amount;r -,02 ,01 -,05*** -,01
Dependent
children

Rate;r
pb
-,06 ,11* ,03 -,06
Amount;r -,06 ,09 -,03 -,00
Amount; -,11* ,13* -,07 -,04
Recreational
expenses

Rate;r
pb
-,06*** ,03** ,00 -,03**
Amount;r -,03 ,01 ,02 -,03*
Amount; -,05** ,03 -,01 -,05***

119
9. The situation in graduate studies
9.1. Methodology and limits
The data of graduate students are drawn from the survey Sources and modes of funding
graduate students, published in 2007 by the CNCS-FEUQ. However, it is more limited
than the survey on undergraduate students.
The sample is of 1638 respondents: for the purposes of study, we have only retained
full-time students and those at the writing stage, which gives us 1,370 respondents. This
gives us a margin of error of 2.6%, 19 times out of 20.
Students enrolled in the masters without writing status could present vey different
characteristics, mainly because they do not have a right to the scholarships of research
granting agencies. Unfortunately, the survey had a too little number of respondents of
this type to be bale to analyze this population.
The questionnaire does not include questions on the levels of expenses or on
indebtedness. The only question asked was on the level of accumulated indebtedness
during studies, which greatly limits the analysis that could be made. Furthermore, we
cannot distinguish respondents that have a low level of debts from those that have none
at all, which limits the possibilities of analysis.
9.2. Description of levels of indebtedness
9.2.1. The debt due to SFA on completion
Student financial assistance is an important source of funding for graduate studies. In
2006, 23,349 students were enrolled at the masters level full-time and 11,786 at the
doctorate (MELS, 2008). 51% of masters students and 21% of doctoral students enrolled
full-time benefited from the student financial assistance program (AFE, 2008, p. 40,
calculation of the author)
As we saw earlier, students enrolled in graduate studies are subject to a higher
maximum loan limit than their undergraduate colleagues. Furthermore, those who
indebt themselves to SFA have substantial debts: $16,304 at the masters and $23,405 on
average at the doctoral level.
Table 9-1: Number of borrowers that had to assume at the end of their studies the repayment of their loans
obtained according to the amount of the student debt, 2008-2009
masters Doctorate
1-5000$ 578 56
5001-10000$ 1407 125
10001-15000$ 1467 136
15001$or
more
3171 685
Total 6623 1002
Average 16304$ 23405$
Source : AFE, 2010, p. 65
120
9.2.2. Description of the accumulated debt
Students enrolled in graduate studies often have very significant levels of debt.
Although 47% of students had accumulated in 2006 less than $6,000 of debt, many were
dragging very high debts, as indicated by Figure 9-1. One student in ten had
accumulated more than $30,000 of debts. The average accumulated debt was situated at
$11,970, and the median debt at $10,500.
Figure 9-1: Description of the level of indebtedness of graduate students

It is methodologically difficult to use this data to characterize indebtedness in graduate
studies. We will, however, examine the characteristics of students according to another
criteria: the fact of benefiting or not from the student financial assistance program,
according to certain socioeconomic characteristics.
9.3. Role of sources and modes of funding
Certain academic characteristics seem to have an influence on the propensity to indebt
ourselves or not to SFA. On the other hand, the fact of belonging to a research group is
associated to the fact of not needing student financial assistance, as is shown by Table
9-2. There is a significant statistical difference, but weak (=-,08 p<,010).
Table 9-2: Belonging to a research group and participation in SFA
Masters Doctorate Presenceof
SFA
Absenceof
SFA
Total
Researchgroup 37,00% 64,30% 37,70% 46,10% 43,30%
Noresearchgroup 63,00% 35,70% 62,30% 53,90% 56,70%
100,00% 100,00% 100,00% 100,00% 100,00%
Merit grants play a major role in the funding of graduate studies, thus contrasting with
undergraduate studies. 44.1% of the funding of full-time students is made up of merit
grants or conferences and internships (CNCS-FEUQ, 2007, p. 42). Now, the fact of
receiving one or various grants/bursaries reduces the propensity of indebting ourselves
during studies to SFA, (=-,11 p<,001), as we can see in Table 9-3.
!"#$
%&#$ %&#$
%%#$
'#$
%(#$
)*+,-$./$
'(((0$
'((%0$1$
%2(((0$
%2((%0$1$
%3(((0$
%3((%0$1$
2!(((0$
2!((%0$1$
&((((0$
456-$./$
&((((0$
121
Table 9-3 : Presence or absence of merit grants
SFA
Absenceof
SFA
Total
Ameritgrantormeritaward 17,80% 35,40% 17,50% 24,20% 22,00%
of2grants 6,90% 12,70% 6,90% 8,90% 8,20%
Morethan2grants 2,50% 9,40% 2,30% 5,10% 4,20%
Nogrant 72,80% 42,50% 73,30% 61,90% 65,60%
100,00% 100,00% 100,00% 100,00% 100,00%
The presence of at least one grant for conferences or internships has the same effect, but
a little less strong ( =-,06 p<,050).
Table 9-4 : Presence or absence of a grant for conference or internship
SFA
Absenceof
SFA
Total
Grantforinternship,
conference
12,40% 34,50% 14,30% 19,20% 17,60%
No 87,60% 65,50% 85,70% 80,80% 82,40%
100,00% 100,00% 100,00% 100,00% 100,00%
The family contribution is a more marginal mode of funding for graduate studies: it
only represents 6.1% of the total funding of full-time studies (CNCS-FEUQ, 2007, p. 42).
Once again, the presence of a family contribution (from the parent or spouse) makes a
bit less probable enrollment in loans and bursaries ( =-,07 p<,010).
Table 9-5: Presence or absence of a family contribution
SFA
Absenceof
SFA
Total
Familycontribution 36,70% 29,10% 30,00% 37,20% 34,80%
Nofamilycontribution 63,30% 70,90% 70,00% 62,80% 65,20%
100,00% 100,00% 100,00% 100,00% 100,00%
Remunerated work in graduate studies is analyzed according to two criteria: the
internal or external character to the university and the type of job. Now, as illustrated in
the two next tables, there does not seem to be an impact of the presence of these jobs on
indebtedness to student financial assistance, contrary to the situation prevailing at the
undergraduate level. We can believe that this consists of the more marginal sources of
funding in relation to grants, which are the main determinant of funding graduate
studies.
122
Table 9-6: Presence or absence of an internal job at the university
SFA
Absenceof
SFA
Total
Teachingassistantship 17,70% 20,90% 20,20% 17,60% 18,50%
Researchassistantship 9,00% 15,00% 13,10% 9,10% 10,40%
Lecturer 1,50% 7,90% 1,50% 3,70% 2,90%
Otherjob 9,40% 6,50% 8,30% 8,90% 8,70%
Nojob 62,40% 49,70% 56,90% 60,80% 59,50%
100,00% 100,00% 100,00% 100,00% 100,00%
Table 9-7: Presence or absence of an external job to the university
SFA
Absenceof
SFA
Total
Externaljob 51,90% 26,50% 53,50% 42,30% 46,00%
Noexternaljob 48,10% 73,50% 46,50% 57,70% 54,00%
100,00% 100,00% 100,00% 100,00% 100,00%
This brief analysis of the situation prevailing in graduate studies allows us to affirm that
contrary to the situation at the undergraduate level, the best way to reduce student
indebtedness would be to increase the number and amount of merit grants offered so
that they respond to demand. We will return to this in the section on recommendations.
123
10. Analysis and recommendations
Student debt load in Quebec is difficult to analyze. As we have seen, there are various
angles of approach. Ours is to review the literature as well as a major quantitative study.
Both the review and the study seem to make the same point: student debt load is first
and foremost a question of money. We will analyze this situation in two stages. First, we
will look at debt loads, starting from the choice to get into debt and the level of
indebtedness, to the impact of debt in relation to six key factors: family background,
individual characteristics, credit history, academic choices, external factors, and choices
following the first degree or diploma. We will discuss the major differences between the
three forms of student debt. Finally, we will present a model of the effects discussed in
this section and a series of recommendations to limit student debt load.
10.1. Student debt: Key factors and impact
10.1.1. Family background
The choice to get into debt can be related to family background. Four factors play
important roles: the view of credit, the family income, the parental residence, and the
level of schooling attained.
The first factor is sociological. Lea, Webley, and Walker (supra section 3.3.2) related
greater social support to indebtedness or at least a more tolerant attitude towards debt.
Data from Quebec seems to support this theory, as seen in Figure 3-1. Christie and
Munroe go further; stating that aversion to debt is lower among students from families
with greater cultural and financial capital (supra section 3.3.3).
A parental contribution or lack thereof plays an even more important role. Not receiving
a parental contribution multiplies by 1.5 both the rate of indebtedness and the amount
of debt. The only exception is with loans, as the opposite occurs in relation to the
amount owed (supra section 7.3). The greater the contribution received, the lower the
rate of indebtedness even while the amount owed stays stable. Once again, family loans
are an exception. Of course, family financial capacity does have its limits, thus very high
parental contributions are often accompanied by some loans.
Financial revenue has a decisive influence on student debt loads (supra section 7.1). It is
responsible for a 3% variance in debt load. Our results demonstrate that regardless of
the source of debt, coming from a low-income family indicates a greater likelihood of
indebtedness. It is not, however, a decisive factor in the amount owed, except to student
financial assistance. Family revenue determines, to a great extent, the presence of a
parental contribution and its amount. Thus, students coming from more privileged
backgrounds have greater chances of leaving university without financial problems.
Where parents live is also important. Distance plays a role, as noted by Frenette (supra
section 3.3.3). In addition to uprooting students from their hometown, the further the
university is from the family home, the greater the cost to the student. Moving in order
to study multiplies the rate of indebtedness by 1.22 and the average total amount owed
by 1.12 (supra section Error! Reference source not found.). Thus, moving to study at
university is an important factor in student debt loads.
First generation university students have a riskier academic profile and require special First generation university students have a riskier academic profile and require special
guidance. They are more often in debt than second generation students (supra section
124
6.1.8). In addition, students who come from families without a university education are
usually averse to higher levels of debt since they have little concrete proof of the benefits
of university education (supra section 3.3.2).
Being born in the wrong family greatly determines the rate of indebtedness and the
level of student debt. The most vulnerable students, those whom we try to help with our
financial assistance system, are also those who must take on more debt in order to
succeed in university. Any increase in student debt loads hits them hardest. Next, we
shall look at individual characteristics.
10.1.2. A students individual characteristics
Many individual characteristics are involved when we look at student debt loads, in
respect to choices and results. They include age, place of residence, whether one is a
parent or not, paid work, and recreational expenses.
Age plays a decisive role in student debt loads (supra section 6.1.2) as it is responsible for
a 5% variance in accumulated debt. The older the student is when starting university,
the greater role debt plays in financing education. Their debt load, like their rate of
indebtedness, is twice as high. This is particularly true for students who are 25 or older.
They correspond generally with the profile of a student returning to school. Being older
is also associated with various life choices, such as having children or no longer living
with ones parents. These characteristics correlate with student debt. It must also be
noted that student debt delays university education and is a major reason for not
attending, as Dubois has asserted (supra section 3.3.1.).
We have already seen that the place of residence is a decisive factor. Leaving the family
home involves major housing and food expenses, which, previously, were probably not
borne by the student. We have seen that total expenditures have a decisive influence on
indebtedness (supra section 8.1) . Not living with ones parents entails a major increase
in the rate of indebtedness as well as in the amount owed. It is 1.7 times more likely that
a student who has left the family home will leave university in debt. That same
students average debt will be 1.7 times higher than a student who did not leave home.
The reverse is probably also true. Choosing to stay with ones parents is probably
influenced by the high cost of independence, including the substantial debt load that
accompanies it.
Being a parent creates other major expenses. The presence of a dependent child (supra
section 6.1.7) multiplies by 1.4 the probability of indebtedness and doubles the
probability of taking out a loan from student financial assistance. The amount owed
also increases by 1.5 times. One in twenty students are parents and being a parent
entails an average annual expenditure of about $5587. Such a drastic increase in debt can
only have serious effects on the ability of already vulnerable students to continue their
studies.
The literature generally confirms that students do paid work to avoid debt (supra section
3.4.2). However, beyond a certain limit, paid work becomes detrimental to success that
may lead students to take longer to graduate. Not working during school increases the
average student debt load. The impact is greatest for students who did not work during
the Fall 2009 semester. In addition, students who do not work and who work a lot (more
than thirty hours a week during the Fall 2009 semester) have something in common.
They have a higher rate of indebtedness and more debt than those who took on a
125
moderate workload. The hypothesis that paid work is employed as a strategy to avoid
debt is confirmed.
Recreational expenses and other expenses determine a small portion of student debt
loads (supra section 8.4.2), as they are nonessential. Students in debt tend to cut these
expenses. They are already lower than such expenses for the general population in
Quebec. The only exception is with private debt. We might say that certain students
decide to consciously go into debt for a reason related to recreation. This connection is
however very weak (r
pb
=,03 to p<,010).
Finally, mental heath can be affected by debt load (supra section 3.4.3). A high debt load
causes a lot of stress, increases the probability of depression, and gives rise to anxiety.
When financial institutions state that their lines of credit allow students to pursue their
education free of financial worry, they are misleading them. Individual characteristics
play an important role in the decision of whether to get into debt. Being older, a parent,
or leaving the family home tends to increase debt loads. Students do adopt strategies to
limit their indebtedness by working during school, limiting their nonessential expenses,
and delaying life choices such as having children or leaving home. Debt loads also have
an impact on students, as higher debt loads seem to entail a greater likelihood of
working and greater psychological effects.
10.1.3. Credit history: the debt spiral
Previous experience with debt greatly influences student behaviour in relation to debt.
We mentioned in the section 5.3 that there exists a debt spiral phenomenon. Taking on
debt from one source is more often than not accompanied by indebtedness to other
sources. Increasing the amount of debt also increases the rate of debt from other sources
except for family debt, which is generally taken on as a last resort. The amount of debt
accumulated from one source tends to determine the amount accumulated from
another. This is not insignificant. While our system of student financial assistance
should limit student debt, its outdated parameters lead to the accumulation of debt from
various sources. Certain characteristics of private debt, such as only having to pay the
interest on most student debt, encourage spiraling debt, which we will discuss further
on.
In addition, students often have weak financial skills. They find themselves in the initial
stage of what Duhaime has called, the cycle of debt overload (supra section 3.3.2). They
are particularly vulnerable because they have low incomes and aspire to a high level of
social mobility. Their financial literacy is also very weak. Among 18 to 29 year-olds,
knowledge is limited and the main sources of information are their family and personal
experience (supra section 3.2.1). It bears repeating that previous family experience can
influence attitudes towards student debt (supra section 3.3.3).
Thus, university students have many characteristics that make them more vulnerable to
debt. They are less conscious of the associated risks and take on debt that often goes
beyond their financial means. Many are at risk of finding themselves overloaded with
debt.
10.1.4. Academic choices
Academic choices and debt loads are interlinked. Indebtedness tends to lead to certain
choices, just as some academic choices are influenced by debt load.
126
We saw earlier that indebtedness reduced access to university education (supra section
3.3.1). Financial reasons are the most frequently cited in cases where students do not go
to university. Students who do go into debt tend to go to university later than those who
do not take on debt. In an American study, Dongbin Kim came to the conclusion that
debt negatively influences perseverance (supra section 3.4.1). He found that an increase
of $1000 in loans causes a 1.6% increase in drop out rates for students from less
fortunate backgrounds. Thus, loans are generally a less effective tool than bursaries
(FCBEM, 2009, p. 78).
Academic career plans also play a role. The amount of debt varies depending on when
debt is first taken on (supra section 6.2.1). The further on in a program, the more likely a
student is to take on debt (supra section 6.2.2). The rate of SFA debt increases from 2% to
8% a year, depending on the how long the student has been studying. The longer it
takes to finish the degree, the greater the debt load (supra section 6.2.3). This is
confirmed when we look at students who take longer to graduate. Their rate of
indebtedness is 11% higher (supra section 6.2.4). The relationship between debt and
work and debt load itself increases the time it takes to graduate. The impact is
interlinked: debt loads increase the probability of taking more time before graduating
and this, in turn, increases debt loads.
The same type of relationship is at work when we look at fields of study. Students who
are more sensitive to risk when going to university will tend to favour subjects that are
practical and economically viable rather than those disciplines that are more
traditionally associated with university (supra section 3.3.4). A policy of high debt loads
puts disciplines that are less economically practical at a disadvantage. Specializing a
given field may influence debt loads, specifically private student debt (supra section
6.2.5.). Certain factors are at work. The most economically practical areas of study are
often also the most demanding, reducing the time available for paid work. In addition,
there is a sizable market for financial products aimed at students in certain disciplines
(supra section 2.3.3.). The supply of financial products combined with a lack of financial
knowledge seems to lead to an increase in student debt loads in order to meet needs
unmet by student financial aid. Such practices can also encourage compulsive buying,
one of the causes of debt overload, by giving students the impression that they have the
social status of someone much richer than themselves, thereby increasing their
expenditures and debt load (supra section 3.2.2).
Academic success has an impact on debt loads, especially at the graduate level. Merit
scholarships, awarded by provincial and federal granting agencies seem to effectively
address student debt loads (supra section 9.3). Since debt is a reason for dropping out,
we can say that being in debt has a negative impact on academic success. Thus we risk
creating a vicious circle where students who receive merit scholarships are mainly
students who come from more privileged backgrounds. The scholarships offered by
universities at the undergraduate level do not have the same effect (supra section 7.4),
probably because the amounts are much smaller.
Debt loads and academic choices are interlinked. The prospect of debt reduces access to
education and debt lengthens the time it takes to get a degree, which in turn increases
debt loads. The ability to finish a course of study is also negatively affected. Debt loads
affect the subject studied and this can, in turn, have an influence on debt loads. A policy
of high debt loads probably accentuates these characteristics.
127
10.1.5. External factors
External factors are those over which the student has no control. They influence the rate
of indebtedness and the amount owed. They can be under the governments control or
open to the whim of market forces. The four major external factors are the cost of
studying, the mix of loans and bursaries in a student financial assistance program, the
availability of private credit and the cost of living.
From time to time, the Quebec government sets tuition fees. However, ancillary fees are
under the jurisdiction of universities. Even though differences between institutions are
currently relatively minor, the fees do influence student debt load (supra section 8.2.1.).
This is the relatively weak but still important link: as the mandatory costs of tuition go
up, so do both the tendency to get into debt and the amount owed. This is counter-
intuitive because students who attend more affordable institutions (the Universit du
Qubec network) usually come from lower socioeconomic backgrounds (a greater
proportion of first generation students and students studying part time, for example)
than those who study at more expensive schools. Even if the government does not
change the loan limit, it is clear that the next increase in tuition fees will increase student
debt load since student financial needs will increase while their resources will not. As a
general rule, higher tuition leads to higher student debt load.
The Quebec government has established a loans and bursaries program, described in
detail in section 2.2. It is the government that determines a sizable part of student debt
load by setting a loan limit as well as the repayment terms. Many students thus do not
choose to go into debt but are obliged to do so. In order to have access to student
financial aid, a student must accept to take on an, often considerable, amount of debt.
This indebtedness results from a choice made by the government.
The supply of credit from financial institutions probably increases the level of student
debt. We took note of this in section 2.3. There are many products that target students
and they often provide very high loan limits. This is particularly true for students who
are enrolled in programs considered to be economically practical. The supply of credit
targeting them is very impressive, with lines of credit that can go up to $200 000. The
prospect of incurring debt without having to pay the capital while in school can be quite
attractive. However, the profitability of a course of study is never assured and this can
place new graduates in an unfortunate financial position. We will come back to this in
the next section. The Quebec government should better regulate financial products by
adopting provisions that specifically target financial products aimed at students.
A final external factor is the cost of life. The section 8.3 presents the relationship between
living expenses and debt load. Generally, higher rents and food costs push up
borrowing and debt loads. These two categories are each responsible for a 2% variation
in the amount owed. These are usually essential expenditures and thus students often
find themselves at the mercy of rental markets and food prices. They do cut nonessential
expenditures as a way of reducing debt loads. Students who receive SFA spend a bit less
on transport and rely on methods of transport that are free or less expensive.
Many factors that students do not control directly influence debt levels. Only
government action can have a real influence on student debt, whether it in relation to
the cost of tuition, the level of indebtedness, the supply of credit, or living costs.
128
10.1.6. Choices after the first degree
Before and during university, student debt has a multifaceted impact, influencing some
choices and being affected by other choices. We tend, wrongly, to believe that student
debt load has no impact and that graduates pay off their loans quickly. However, after
graduation, debt loads have major effects on the pursuit of graduate studies, loan
reimbursement, and a new graduates finances.
A high debt load often entails abandoning plans to undertake graduate studies.
Students who go to graduate school carry less debt than those who leave school to enter
the workforce (supra section 3.4.4). All else being equal, the higher the debt, the lower
the probability of enrolling in graduate school. It is clear that increasing student debt
load in a context where graduate level specialization is becoming a necessity is a
counterproductive policy.
The reimbursement of debt after school is often difficult (supra section 3.5.1). It is
determined mainly by future revenue although the size of the debt also plays a role. The
size of the debt seems to have an impact on the cost of financial aid programs. A loan in
Quebec costs about 17 per dollar, a Canadian loan costs the State between 30 to 40
per dollar. In Canada, 22% of graduates owing $10 000 to $24 999 had difficulty making
payments, while 43% of those owing more than $25 000 had the same difficulty.
Student debt loads slow down economic growth by reducing the financial capacity of
new graduates (supra section 3.5.2). Young university graduates generally have the
same borrowing power. However, if they have student loans, it becomes more difficult
for them to borrow money for a house or a car. Student borrowers have assets that are
an average of $40 000 lower than those who did not take out a loan.
10.2. Characteristics of different loans
The three major categories of loans (public loans, private loans, and loans from family or
friends) that we have identified are very similar. However, they do have a few unique
characteristics that we shall now explore.
Public loans have many advantages that make them more attractive: they are
subsidized, their interest rate is lower, they only have to be reimbursed once schooling
is complete, and their reimbursement is facilitated by certain SFA programs (supra
section 2.2). In addition, they are often accompanied by bursaries, which are a very
effective form of student aid. They do have certain limitations. They decline in relation
to increases in family income, which means that many students who have financial need
are not covered by the program. In addition, SFA is not well adapted to nontraditional
academic career paths (FEUQ, 2010, p. 122). The program creates a heavy debt load and
does not cover the cost of all needs. As such, SFA debt is a predictor of debt from other
sources. They really are subsistence loans, since students receiving SFA must tighten
their belts and, often, borrow elsewhere.
Private loans are the second most important source of debt. Their terms are less
favorable than SFA (supra section 2.3) since their interest rates are higher and interest
must be paid during university. Contrary to what we might think, the rate of private
debt declines in relation to parental revenue. The amounts borrowed do not follow this
pattern. The supply of credit and unmet needs seem to explain why students in
disciplines that are considered economically practical take out more of these loans, since
avoidance strategies are more difficult to maintain. Generally speaking, this is also
129
subsistence debt but one that is not about to replace SFA. This is the debt taken on by
students who work more during their education and whose gross income levels are
higher. This is a worrying trend that encourages drop out (infra section 7.2), particularly
because it targets consumer debt loads. This is a form of credit that is not sensitive to
increasing the time it takes to graduate. The amount of debt stays stable regardless of
whether more time has been spent to finish the degree. In addition, financial institutions
are about to be more strictly regulated by the Quebec government. In the next section,
we will propose regulations that specifically address student debt loads.
Loans from family or friends resemble private loans. The one difference is that they are
much larger when they are the only source of debt. We believe that this reveals an
important characteristic of these debts: they are the last resort for students who do not
want to or cannot borrow from other sources but who still have financial needs to be
met.
10.3. Student debt load: A model
The following figures and tables summarize the general model presented in sections
10.1 to 10.6. Generally, we can see that student debt load has various undesirable effects.
It encourages selecting students based on their ability to pay rather than on their talents
and it acts regressively towards the least privileged students. In addition, it has a
negative impact on academic success. A policy of high student loans can provide the
capital necessary to ensure wide access to university education but it does not guarantee
the successful completion of a course of study. Even after university, student debt loads
considerably limit the financial leeway of new graduates. In summary, putting students
into debt is a game that is not worth the candle. The next section presents a series of
recommendations that aim to concretely reduce the impact of student debt load on
Quebec students.
130
Figure 10-1: Modelling of the six key factors of student debt.

Figure 10-2: Classification of key factors in relation to the choice to take on debt and the impact of debt
1.Family
backgroud
Social
relationshipwith
credit
Familyincome
Parentalhome
Previous
educationlevel
2.Individual
characteristics
Age
Placeof
residence
Beingaparent
ornot
Paidworkand
debtload
Recreational
expenses
Psychological
impact
3.Credit
history
Levelof
accumulated
debt
Pastexperiences
Financial
management
4.Academic
choices
enrollment,
durationof
studies,and
perseverance
Fieldofstudy
Merit
scholarships
5.External
factors
Costofstudying
Mixofloans/
bursaries
Supplyofcredit
Livingexpenses
6.Choices
afterthegirst
degree
Pursuitof
graduatestudies
Loan
reimbursement
andbankruptcy
Anewgraduate's
Ninances
131

Thechoicetotakeondebt
andtbelevelof
indebtedness
1.Familybackground
2.Individualcharacteristics
Age,residence,beingaparentor
not,paidwork.
3.Credithistory
4.Academicchoices
5.Externalfactors
Impactofdebtload
2.Individualcharacteristics
Age,residence,beingaparentor
not,paidwork,recreational
expenses,psychologicalimpact.
4.Academicchoices
6.ChoicesaftertheNirstdegree
132

Table 10-1: Description of the relation between debt load, the six key factors, and studies on academic career paths
Keyfactor Element Impactonaccess Impactonthetotalrate
ofindebtedness
Impactonthetotal
amountowed
Impactduringandafter
university
1.Family
background
Socialrelationship
tocredit

Attitudesthataremoretolerantdiminishdebtaversion
Familyincome Decreaseswhenincome
increases
Responsiblefor3%of
variance
Absenceof
parental
contribution
1.5timeshigher 1.5timeshigher Possibleimpacton
defaults
Parentalhome Theneedtomovecan
discourageordelay
academicplans
1.2timeshigherwhen
thereisaneedtomoveto
anotherregion
1.12timeshigherwhen
thereisaneedtomoveto
anotherregion
Previous
educationlevel
1.4timeshigheriffirst
generation
1.15timeshigher
2.Individual
characteristics
Age Theneedtogointodebt
increasestheprobability
ofdelayingenrollmentby
1.5to1.8times
Therateincreaseswith
age
Responsiblefor5%of
variance
Placeofresidence 1.7timeshigherforthose
whodontlivewiththeir
parents
1.7timeshigherforthose
whodontlivewiththeir
parents
Beingaparentor
not
1.4timeshigherfor
student-parents
1.5timeshigherfor
student-parents
Paidworkand
debtload
Paidworkisusedasastrategyto
avoiddebt
Recreational
expensesand
otherexpenses
Expensescutinorderto
limitdebtload

Psychological
impact
Debtloadsincrease
financialstress
133
ofindebtedness
Impactonthetotal
amountowed
university
3.Credithistory Levelof
accumulateddebt
Beingindebtleadsto
moredebt:apossible
studentdebtspiral

Pastexperiences Rendersdebtadverse Littleexperienceandpossiblemisunderstandingofdebt.
However,itamajorsourceoffinancialinformationfor
students
Financial
management
Studentfinancialmanagementisoftenlackingduetoa
lackoffinancialtraining,whichincreasestheir
vulnerability
4.Academic
choices
Enrollment,
durationof
studies,and
perseverance
$1000moreinloans=
1.6%fewerstudentsfrom
lowersocioeconomic
backgrounds
Takingmoretimeto
completeadegree
increasesby1.2timesthe
probabilityofhavingtogo
intodebt.
Takingmoretimeto
completeadegree
increasesthelevelofdebt
by1.2times.
InCanada,financial
considerationsarethe
mainreasonfordropping
outofschool
Fieldofstudy Choiceofdisciplinebased
oneconomicprofitability
Studentsinprofitable
programsaremorelikely
totakeoutprivateloans

Academicsuccess Meritscholarshipsreducedebtloadsatthegraduate
level
5.Externalfactors Costsofstudying Hightuitionfeeslower

accesstouniversity
Asthecostofstudying
goesup,sododebtloads

Mixof
loans/bursaries
Highdebtloadsblock
access
Thegovernmentsetsthe
levelofpublicstudent
debt

Supplyofcredit Supplyofstudentcreditisabundant,attimesdeceiving,
andinsufficientlyregulated,whichcanincreasedebt
loads
Livingexpenses Foodandrentareeach
responsiblefora2%
varianceintheamountof
debt

134
ofindebtedness
Impactonthetotal
amountowed
university
6.Choicesafter
thefirstdegree
Pursuitof
graduatestudies
Highdebtloads
discouragemanystudents
frompursuinggraduate
studies(1.4to1.6times
dependingonthesizeof
thedebt)
Loan
reimbursement
andbankruptcy
Asstudentdebtloads
increasesodoesthecost
ofstudentfinancialaid
programsanddefault
rates
Anewgraduates
finances
Graduateswhoborrowed
moneyhaveassetsworth
$40000less.

135
10.4. Recommendations
Various measures can be put into place to concretely reduce student debt loads. They
can be filed under four general categories: policy regarding university tuition fees, the
student financial aid program, the financing of granting agencies, and the regulation of
financial institutions and financial information.
We believe that all these recommendations should be integrated into a complete and
coherent strategy against student debt. A quick fix cannot properly address student debt
load since this issue is complex and interlinked. The SFA, OPC [consumer protection
bureau], ministre de la justice, financial institutions, and universities all need to agree
on efforts to limit student debt load.
1st Recommendation
That the government of Quebec develop and implement a strategy to fight against
student indebtedness that puts in place: strategy to address student debt load that
includes:
A tuition fee freeze as of 2012 accompanied by a better regulation of ancillary fees;
Improvements to loans and bursaries and merit grants from granting agencies;
A better regulation of financial institutions;
Communication mechanisms to improve the financial skills of youth.
10.4.1. Freeze Tuition Rates
In its last budget, the Quebec government decided to raise university tuition fees by
$325 per year over five years, thus bringing the cost of one years tuition to $4700 by
2017.
Figure-3: Student tuition fee rates from 1994 to 2017 (projected)

Currently, tuition fees are the same among institutions and between disciplines and thus
increased student costs are synonymous with greater student debt loads (infra section
!"!!#$##
%!!"!!#$##
&#!!!"!!#$##
&#%!!"!!#$##
'#!!!"!!#$##
'#%!!"!!#$##
(#!!!"!!#$##
(#%!!"!!#$##
)#!!!"!!#$##
)#%!!"!!#$##
%#!!!"!!#$##
&
*
*
)
#
&
*
*
%
#
&
*
*
+
#
&
*
*
,
#
&
*
*
-
#
&
*
*
*
#
'
!
!
!
#
'
!
!
&
#
'
!
!
'
#
'
!
!
(
#
'
!
!
)
#
'
!
!
%
#
'
!
!
+
#
'
!
!
,
#
'
!
!
-
#
'
!
!
*
#
'
!
&
!
#
'
!
&
&
#
'
!
&
'
#
'
!
&
(
#
'
!
&
)
#
'
!
&
%
#
'
!
&
+
#

136
8.2). Without a doubt, this new policy will create higher student debt loads despite
maintaining the maximum loan limit (this will be discussed further on). The FEUQ has
previously investigated the impact of higher tuition fees on various aspects of student
life. It is clear that students do not have the ability to pay. The Quebec government
should reconsider its decision, cancel the tuition fee increase, and prioritize a tuition rate
freeze.
2
nd
Recommendation
That the Quebec government renounce tuition fee increases announced in the 2011-2012
Budget.
The main reason for variations in student fees among educational institutions relates to
mandatory institutional fees. As we can see in the following graph, anglophone
institutions and the HEC charge the highest mandatory institutional fees.
Figure -4: Mandatory undergraduate institutional fees by university (2010-2011)

Source : FEUQ, 2011
A new tuition fee freeze must include a strict policy regulating mandatory institutional
fees. University administrations have shown time and again that MIO levels should be
supervised by the Quebec government. Regulations currently exist, but universities can
bypass them in various ways and have done so frequently in the past. A new umbrella
act would be the best way to limit MIO increases.
3rd Recommendation
That the National Assembly of Quebec adopt a law to regulate the mandatory
institutional fees required by university institutions (as well as their components) and
stipulate that such fees cannot be imposed unless the nature, amount and modalities of
these fees are the object of an agreement between the institution and the recognized
student association as representatives of the students concerned. (CAU-643).
0 200 400 600 800 1000 1200
UQTR
ETS
UQAC
UQAR
UQAT
UQO
UQAM
Sherbrooke
Polytechnique
Montral
Laval
Concordia
Bishop's
HEC
McGill

137

10.4.2. Loans and bursaries adapted to student needs.
Addressing student debt loads requires an effective and generous loans and bursaries
scheme. The Quebec government has various tools within reach to intervene here. We
note five of them: the loan limit, the level of allowable expenses, parental contribution,
student financial aid, and bankruptcy.
In order to ensure efficiency, we should rely mainly on bursaries, the most effective
form of aid. Thus, the loan limit must be maintained at its current level. This would be
the first and most direct tool that the government can and should use to act.
4th Recommendation
That the maximum loan limit of student financial assistance not be increased.
Quebec claims that its policy will not increase student debt loads. However, the
government continues to provide a special allowance to students who only receive
loans, and who, thus, increase their debt load. Currently, more than a third of recipients
see their student debt load increase each year. This is unacceptable. If our aim is to limit
student debt, we should simply convert the special allowance into a bursary.
5th Recommendation
That the special allowance covering the increase of tuition fees be paid out in the form of
a bursary to all recipients without exception, and that it not lead to any increase in
indebtedness.
The amount of aid provided to students is insufficient. For example, as students spend
more on living expenses, they must also borrow more (supra section 8). This is probably
the reason that students who borrow from SFA often must also borrow from other
sources in order to finish their year (supra section 5). These students, who already have
greater debt loads and various financial difficulties, must also make do with a program
that is becoming more and more obsolete in terms of the standard of living that it can
ensure.
By not indexing allowable expenses, the Quebec government is widening the gap
between the loans and bursaries program and reality. This will probably increase the
need for private loans among program recipients. The situation needs to be addressed
by increasing and indexing the level of allowable expenses.
6th Recommendation
That Student Financial Assistance increase the amount of allowable expenses for
recipients of student financial assistance, notably through an improvement of the
amounts for living expenses and transportation expenses for students that do not have
access to public transit. Transportation expenses for non-residents and Internet expenses
must also be included in allowable expenses.
7th Recommendation
That the Ministry of Education, Leisure and Sports introduce an automatic annual
indexation mechanism for all allowable expenses in the calculation of Student Financial

138
Assistance. That this indexation be equivalent to the Consumer Price Index (CPI) for the
year concerned.

The student financial aid program is not sufficiently accessible. Currently, a number of
students who should be able to benefit, do not, even though receiving a parental
contribution is an important indicator of student debt load (supra section 7.3).
The main reason is the current method used to calculate the parental contribution. The
government has proposed a modest increase in the parental contribution threshold. It
must go much further and bring it to a minimum of $45 000 and index it annually. Such
a threshold would be equivalent to a family in which both parents work full-time at
minimum wage.
8th Recommendation
That the exemption amounts for the maintenance of the family unit be established at
$45,000 and subsequently indexed.
We have seen that certain students experience particularly difficult financial situations,
especially, older students (supra section 6.1.2) and student-parents (supra section 6.1.7).
Programs are not well adapted to these two groups. Loans and bursaries does not take
into account non-traditional pathways to education. As a result, balance between family
and school is not easily found. With regard to continuing education, work needs to be
done in order to increase the access these groups have to a university education.
9th Recommendation
That the government of Quebec adapt the loans and bursaries program with a view to
recognizing the diversity of academic paths and studies-work-family integration.
Restricting the ability to declare bankruptcy on a student loan for seven years stems
from a desire to balance government budgets and not from a consideration of the needs
of new graduates. Limiting student debt loads would reduce the number of students
who would consider bankruptcy. As we noted earlier, bankruptcy is a last resort and a
painful process that should not be encouraged, but is sometimes necessary.
10th Recommendation
That the Government of Canada abolish the discriminatory provisions toward students
in the Bankruptcy and Insolvency Act.
10.4.3. More excellence awards
Graduate students rely heavily on excellence awards provided by granting agencies.
These awards are a major source of financing for students who receive them but they do
not meet current demand. Today, too many deserving students are unable to find
funding for their studies. We believe that addressing student debt loads must involve
better financial support for the next generation of scientific research in Quebec.
11th Recommendation
That governments increase the level of financing for the different granting agencies.
(CNCS-426 [2.3.])

139
10.4.4. Better regulation of financial institutions
The Quebec government announced its intention to table Bill 24, which targets the
excessive consumer debt loads, in the National Assembly. This bill includes numerous
measures, three of which are of particular interest:
False representation to consumers that credit may improve their financial
situation is to be prohibited.
Granting of a higher credit limit than that requested by the consumer is to be
prohibited.
Minimum payments on outstanding credit card balances are to be equal to a
certain percentage of that balance. This percentage will progressively increase
until it reaches 5%.
We think these measures are valid and we believe that the FEUQ should support them
while proposing other guidelines. Private student loans are a lot less advantageous to
students than public loans. The Quebec government considerably subsidizes financial
institutions in order to ensure the operation of loans and bursaries. The government
takes on responsibility of defaulted loans and, in some cases, subsidizes interest. We
think that, taking this into consideration, the Quebec government should ask more of
financial institutions.
12th Recommendation
That the FEUQ support the provisions of bill 24 aiming at limiting the over-
indebtedness of consumers.
In many cases, the promotional practices of financial institutions veer rather close to
false representation. Peace of mind and the innocuousness of credit are two major
selling points for student lines of credit. This idyllic vision is in direct conflict with the
major financial difficulties students may suffer due to credit.
13th Recommendation
That the government of Quebec forbid false representation in regard to student
indebtedness according to which credit can allow one to go through university studies
sheltered from financial troubles and on the future profitability of studies.
Many elements of student lines of credit are unclear. This is particularly true for those
that target certain areas of study. It is difficult to determine what happens when a
program of study is dropped or changed. Those provisions should be clarified.
14th Recommendation
That the government of Quebec impose on financial institutions the duty of clarifying
financial products offered directly to students, among others, by stating clearly the
consequences of changing program or dropping out on the repayment of the debt.
Loan limits offered by financial institutions are much higher than that which is offered
by student financial aid. Given the risk inherent in undertaking university education,
we believe that it is in the interest of institutions and students that the maximum loan
amounts offered by financial institutions be lowered. Such a measure would contribute
to limiting new graduate debt loads and bankruptcies.

140
15th Recommendation
That financial institutions downwardly assess the maximum loan limits for students.
As we have seen, one of the causes of consumer debt overload is the impression of
belonging to a social group that is not your own. We believe that financial institutions
offering rebates on prestigious credit cards to students in certain specializations are one
source of this problem. Such practices should be prohibited. When students receive
rebates or special offers, they should be identified as such. Desjardins and the
Laurentian Bank already operate in this manner. To do otherwise is to encourage
detrimental social comparisons.
16th Recommendation
That the government of Quebec forbid financial institutions from proposing to students
discounts on financial products that are not clearly or specifically destined for them.
10.4.5. Ensure the distribution of financial information
Financial institutions seem to exploit a lack of financial literacy among students. In
many cases, they do not publicize the loans and bursaries program and, in most cases,
the advantages of loans and bursaries are not clearly stated. We believe that financial
institutions who take part in the loans and bursaries program, on very favourable terms,
should be forced to show the terms of the loans offered by loans and bursaries as well as
the advantages of those loans over conventional ones.
17th Recommendation
That the government of Quebec impose on financial institutions the obligation of
presenting on their Internet sites and informational folders presenting their student
products the student financial assistance program and its modalities.
Young Quebeckers, and thus students, lack financial literacy. We believe that is
necessary to put and end to this situation. A good way to proceed would be to mandate
that the Office de la protection du consommateur [consumer protection bureau] produce
student targeted informational material on the reality of private student loans.
18th Recommendation
That the Office de la protection du consommateur, in collaboration with student financial
assistance and university institutions, develop and distribute information material on
student indebtedness, with an emphasis on private indebtedness and credit targeted
toward students.

141
11. General Conclusion
This study aimed to quantify the situation of student debt and evaluate its key factors
and impact in order to issue recommendations. We undertook this study by defining
student debt load as all debt accrued for professional or postsecondary education,
regardless of the issuer, that allows a student to pay educational and living expenses,
excluding mortgages and car purchases. The three major loan issuers are the
government, financial institutions, and family. We then examined the characteristics of
two major student loan providers. Government loans have a subsidized interest rate and
various programs are offered to facilitate loan reimbursement. These loans often come
with substantial bursaries and the maximum loan limit is regulated. However the loan
limit can be bypassed by a special allowance that increases the debt load for those who
only receive loans, about a third of recipients. In addition, the assistance offered does
not sufficiently cover student needs, thus pushing more and more students towards
private loans.
Private loans are less advantageous to students since their interest rates are higher and
the interest is charged immediately. This generates higher costs and higher potential
debt loads. Many of the practices put in place by financial institutions ought to be
challenged. This is why we have recommended a series of measures to tighten
legislative control of financial institutions.
Next, we examined the scientific literature on student debt load. Access to private credit
is increasing among youth and students. This increase is combined with insufficient
financial literacy. Too many young people rely on their previous experiences to dictate
their behaviour towards credit. In addition, student debt load has a major impact on all
aspects of a students academic career path. It delays or blocks access to education for
many students, is a reason for dropping out, and is a considerable obstacle to success.
Finally, according to the research, debt loads lead to financial difficulties and reduce
student financial capacity.
We added to this review of the literature, an exhaustive examination of the key factors
and impact of student debt load. We came away with six major categories.
Family background is a main factor in student debt load. Family income and the family
capacity to contribute financially to tuition play an important role for many in the rate
and level of indebtedness. In addition, certain sociological aspects play a role, such as
the attitude towards credit and the level of education achieved by the students parents.
The place of residence also has an effect. A student who must leave the parental home
faces higher expenses than one who is able to stay at home.
Individual characteristics can go both ways with regard to debt. Being older, leaving the
family residence, and being a parent are predictors of higher debt loads. Conversely,
there are various debt avoidance strategies including doing more paid work during
school and reducing nonessential expenses.
Credit history tends to determine the level of student indebtedness. Accumulated debt
loads are a bad influence that can lead to a debt spiral. The more a student borrows, the
more likely that student is to borrow from different sources, thereby accumulating large
debts from several sources. In the end, the same student is left with a level of debt that
he or she is unable to reimburse. This situation is made worse by low student financial

142
literacy and financial institutions whose supply of credit encourage students to take on
more debt.
Academic choices both influence and are influenced by levels of indebtedness.
Enrollment, the duration of studies, and perseverance are all negatively affected by
student debt or the prospect of it. The field of study may also be affected by debt load,
since disciplines considered to be more financially practical may be favoured over those
that carry greater risk. However, being enrolled in a more practical discipline tends to
increase debt from private and familial sources. This may be due to being less able to
work during school, the higher cost of required materials, and an aggressively
marketed, and often attractive, supply of credit. Finally, a strategy unique to graduate
students is to receive merit scholarships. This can only occur if a student is judged
sufficiently worthy by granting agencies.
Many external factors influence student debt load. The cost of studying has a direct
impact on debt load. Our study shows that increasing the cost of tuition also increases
the rate and level of indebtedness. Governmental choices relating to financial assistance
and the relationship between loans and bursaries also play decisive roles. The generally
under-regulated supply of financial products directly marketed to students is often
deceptive. Credit is presented as an easy way to get an education without financial
problems, while the reality is very different. Stricter regulation is necessary. A final
factor outside of student control is the cost of living. Rent and food costs are difficult to
cut. The increase in food costs due to the global food crisis and the constant increase in
rents risk pushing students into more debt. Strategies to cut rent costs such as rooming
with other students or living with ones parents are already frequently used.
Debt loads continue to make their effects felt after the completion of a first degree. Being
greatly in debt can jeopardize the pursuit of graduate studies. A new graduates
finances are usually in worse shape if that graduate had to borrow for university. In
addition, bankruptcy is always a possibility even though federal law prohibits declaring
bankruptcy on public loans for seven years after graduation.
This analysis of debt load leads us to present five main sets of recommendations to
stabilize the price of tuition, increase student financial assistance and merit scholarships,
better regulate the student loan market, and ensure better dissemination of financial
information.
In the end, we believe that a policy of access to university that is based on high tuition
rates and increasing private debt is contrary to its own principles. Students from lower
socioeconomic backgrounds are hit hardest by debt, which affects their ability to pursue
their studies free from financial worry. Thankfully, the current situation in Quebec is
better than that of other Canadian provinces. Unfortunately, it is deteriorating each and
every year, as tuition fee increases accumulate and student financial assistance lags
behind. Current government complacency is jeopardizing Quebecs future. If major
remedies are not brought forth soon, Quebecs future as well as its capacity to continue
to develop with an aging population will be at risk. The Quebec government must put
in place a real strategy addressing student debt load
Scientifically, there is still much left to explore. Student debt load in Quebec is a
phenomenon that has not been properly investigated. This first study uses bivariate
analysis, a relatively simple statistical model. Further studies using more complex

143
statistical models would be able to more precisely isolate certain factors. A longitudinal
study could look at the evolution of student debt over the course of studying for a
degree.

145
Bibliography
Official documents
AFE (2009). Enqute sur les conditions de vie des tudiants de la formation professionnelle, du
collgial et de luniversit. Qubec : Aide financire aux tudes. 294 pages.
AFE (2010a). Statistiques, rapport 2008-2009. Qubec : Aide financire aux tudes. 85
pages.
AFE (2011a). Prts et bourses : des rponses vos questions (2011-2012). Qubec : Aide
financire aux tudes. 24 pages.
AFE (2011 b). Une aide votre porte 2011-2012.Qubec : Aide financire aux tudes. 51
pages.
Assemble nationale (2011). Projet de loi n. 24 : Loi visant principalement lutter contre le
surendettement et moderniser les rgles relatives au crdit la consommation. Qubec :
Assemble nationale. 36 pages.
MELS. (2009). Enqute sur les conditions de vie des tudiants de la formation professionnelle, du
collgial et de luniversit. Qubec : MELS. 294 pages.
MELS. (2010). Indicateurs de lducation dition 2010. MELS : Direction gnrale des
politiques, de la recherche et de la planification stratgique. 134 pages.
OCDE. (2010). Regards sur lducation 2010 : Les indicateurs de lOCDE. Paris : Les ditions
de lOCDE. 506 p.
Statistique Canada. Guide de lutilisateur : fichier de microdonnes grande diffusion. Enqute
sur les dpenses des mnages, 2008. Division de la statistique du revenu, Statistique
Canada. 34 pages.
Scholarly articles
Allen, Mary et Chantal Vaillancourt. (2004). Promotion de 2000 : profil des diplms du
postsecondaire et endettement des tudiants. Ottawa : Ministre de lindustrie. 52 pages.
Andrews B et JM. Wilding. (2004) The relation of depression and anxiety to life-stress
and achievement in students . British journal of psychology.
Archer, L., & Hutchings, M. (2000). Bettering yourself? Discourses of risk, cost and
benefit in ethnically diverse, young working-class non-participants constructions of
higher education . British Journal of Sociology of Education, 21(4), 555574
Auclair, Raymond et al. (2008). Note 2 Les tudiants de premire gnration : un
concept prometteur ? . Montral : FCBEM. 67 pages.
Boudon, Raymond. (2002). Thorie du choix rationnel ou individualisme
mthodologique . Sociologie et Socits. Vol. XXXIV, no1, p. 9-34
Callender, Claire et Jonathan Jackson. (2008) Does the fear of debt deter students from
higher education ? . Journal of Social policy. Vol 34, p. 509-540.

146
Callender, Claire et Jonathan Jackson. (2008) Does the fear of debt constrain choice of
university and subject of study? . Studies in Higher Education. Vol 33 : issue 4. 2008,
p. 405-429.
Cameron, S. et D. Golby. (1990). An economic analysis of personal debt . Bulletin of
economic research. 42 :3. p. 241-247.
Davies, Emma et Stephen E.G. Lea. (1995). Student attitudes to student debt . Journal
of economic psychology. Vol. 16. P. 663-679.
Flint, Thomas A. (1997). Predicting student loan defaults . The journal of higher
education. Vol. 68, no. 3. p. 322-354
Greene, Laura L. (1989) An economic analysis of student loan default . Educational
evaluation and policy analysis. Vol. 11. No. 1. p 61-68
Kim, Dongbin. (2007). The effect of loans on students degree attainment : differences
by student and institutional characteristics . Harvard educational review. Vol. 77. No. 1,
p. 64-127
Lea, Stephen E.G., Paul Webley et Catherine M. Walker. (1995). Psychological factors
in consumer debt : money management, economic socialization, and credit use .
Journal of economic psychology. Vol 16, p. 681-701.
Merani, S., Abdulla, S., Kwong, J. C., Rosella, L., Streiner, D. L., Johnson, I. L. and
Dhalla, I. A. (2010), Increasing tuition fees in a country with two different models of
medical education . Medical Education, 44: 577586. doi : 10.1111/j.1365-
2923.2010.03630.x
Millett, Catherine M. (2003). How Undergraduate Loan Debt Affects Application and
Enrollment in Graduate or First Professional School . The Journal of Higher Education.
Vol. 74, No. 4 (Jul. - Aug., 2003) (pp. 386-427)
PRA inc. (2007). Rapport sur lendettement des tudiants : Sondage du Consortium canadien de
recherche sur les tudiants du niveau collgial et Sondage du Consortium canadien de
recherche sur les tudiants du premier cycle universitaire. Montral : Fondation
canadienne des bourses dtudes du millnaire. 30 p.
Scott, Adrian J. Student debt in the UK : an economic psychological perspective .
University of Bath.
Smith, Margaret (2002). Le traitement des prts aux tudiants selon la Loi sur la faillite
et linsolvabilit . http://dsp-psd.pwgsc.gc.ca/Collection-R/LoPBdP/BP/prb0126-
f.htm#LE%20PROGRAMME%20CANADIEN%20DE%20PR%C3%8ATS%20AUX%20
%C3%89TUDIANTS [Consulte le 31 mai 2011]
Wilms, Welfword W., Richard W. Moore et Roger E. Bolus. (1987) Whose Fault Is
Default? A Study of the Impact of Student Characteristics and Institutional Practices
on Guaranteed Student Loan Default Rates in California . Educational evaluation and
policy analysis. Vol. 9, no. 1. p. 41-54

147
Monographs
CNCS-FEUQ (2009). Avis sur les besoins financiers des organismes subventionnaires.
Montral : Conseil national des cycles suprieurs de la Fdration tudiante
universitaire du Qubec.
CNCS-FEUQ (2011). Note dinformation sur les services de garde en milieu universitaire.
Montral : Conseil national des cycles suprieurs de la Fdration tudiante
FCBEM (2006). Accrotre laccs : linformation importe-t-elle ? . Montral : FCBEM.
11 pages.
FEUQ (2008). Dlocalisation des campus universitaires. Montral : Fdration tudiante
FEUQ (2009a). Laccessibilit aux tudes universitaires : encore aujourdhui, une ncessit!
Montral : Fdration tudiante universitaire du Qubec.
FEUQ (2009 b). Vers une plus grande diversit doffre de formation universitaire en rgion?
Montral : Fdration tudiante universitaire du Qubec.
FEUQ (2010a). Enqute : conditions de vie des tudiants de premier cycle. Montral :
Fdration tudiante universitaire du Qubec.
FEUQ (2010 b). Trousse sur laide financire aux tudes. Montral : Fdration tudiante
FEUQ (2010c). Les frais de scolarit et luniversit. Montral : Fdration tudiante
FEUQ (2011a). Mesures fiscales et ducation postsecondaire. Montral : FEUQ. 91 pages.
FEUQ (2011 b). Hausse des frais de scolarit et contribution parentale. Montral : FEUQ. 3
pages.
FEUQ (2011c). Lducation universitaire : un outil pour passer de lgalit de droit lgalit de
fait. Montral : FEUQ. 37 pages.
Internet sites
Dugas, Sylvie. La marge de crdit personnelle : est-ce une bonne affaire?
servicevie.com. http://www.servicevie.com/depenses/la-marge-de-credit-
personnelle-est-ce-une-bonne-affaire/a/1652 [Consulte le 11 juillet 2011]
Raymond, Chabot inc. La faillite personnelle
http://www.avocat.qc.ca/public/iifailliteperson.htm (consulte le 1er juin 2011)
Statistical references
Christensen, Howard B. (1986). La statistique : dmarche pdagogique programme.
Montral : Gatan Morin diteur. 656 pages.
Field, Andy. (2009). Discovering statistics using SPSS, third edition. Londres : SAGE
publications ltd. 822 pages.

148
Howell, David C. Mthodes statistiques en sciences humaines. Bruxelles : Groupe De Boeck.
762 pages.
Yaffee, Robert A. (2003). Common correlation and reliability analysis with SPSS for
Windows . http://www.nyu.edu/its/statistics/docs/correlate.html. [Consulte en
juin 2011].

149
Annexe I - Caractristiques des prts privs
Tableau I-1 : Offre de cartes de crdits rgulires par institution financire
Nom Intrt Intrtrduit Fraisannuels? Autres
Institution
financire1
VisaDesjardins()
pourtudiants
19,4% 11,8%;enoption Non
Servicederservationde
billets;protection-cellulaire;
assurancevoyagede3jours
Institution
financire2
MasterCard()SPC
AIRMILES
19,50% 12,90%
35$/anpourintrt
rduit
Programmeairmiles;rabais
tudiantsSPC
MasterCardBMO
SPCRemises
19,50% 12,90%
35$/anpourintrt
rduit
Remiseenespcesde0,5%
Institution
financire3
Visanoire 19,99% ND Non
VisanoireMa
rcompense
19,99% ND 3,50$/mois
Programmedepointset
d'assurance
Institution
financire4
Nondisponible
Institution
financire5
VisaOr()
Rcompenses
19,99%
ND Non
Programmedepoints
CarteVisaRemiseen
argent
19,99%
ND Non
1%deremiseenespces
Signature()
RcompensesVisa
19,99%
ND Non
Programmedepointset
rcompensesRBC
Institution
financire6
Visascne
19,99%;
21,99%
pourles
avancesde
fonds
ND Non
Programmedepoints
changeablescontredela
musique,desfilms,etc.
Visasavoir 1%deremiseenespces
Institution
financire7 CarteClassique()
Visapourtudiants
19,99%;
21,99%
pourles
avancesde
fonds
ND Non
Assuranceaccidentpour
transporteurspublics
(100000$)
Institution
financire8
Nondisponible


150
Tableau I-2 : Marges de crdit rgulires
Nom Montant Remboursement Tauxd'intrt
Institution
financire1
Margedecrdit
Avantagetudiant()
10000$/anpouruniversittemps
plein;Maximum40000$1
er
cycle
ou60000$cyclessuprieurs
5000$tempspartiel
Intrtsseulementdurantles
tudes;6moisdedlaiaprsla
diplomation;10anspour
rembourser(universit)
Variabledurantles
tudes;fixeinfrieur
auprtpersonnel
aprs
Institution
financire2
Marge-crditAux
tudiants
15000$premireanne;10000$
pourlesannessubsquentes
jusqu'45000$
tudes;Reportducapitaldurant
12mois
Institution
financire3
Margedecrdit
tudiante
5500$/an,20000$/4ansou
3000$/5ans(cyclessuprieurs)
rembourser
Nondisponible
Institution
financire4
Fondsderoulement
tudiant
Non-spcifi Intrtsseulementdurantles
rembourser
Nondisponible
Institution
financire5
MargedeCrdit()
pourtudiants
5000$/an1
er
cycle;10000$/an
cyclessuprieurs
tudes;Reportducapitaldurant
12mois
Prfrentiel+1%
Institution
financire6
LignedecrditScotia
pourtudiants
Jusqu'10000$/an;maximum
40000$au1ercycle;maximum
20000$cyclessuprieurs;la
moititempspartiel
Tauxdebase+2%
(enfonctiondu
dossier)
Institution
financire7
Margedecrdit
tudes
Jusqu'15000$/an,40000$pour
leprogramme,pasdelimite
annuelleau3emecycle
diplomation(6moispourune
sortiesansdiplme);jusqu'20
anspourrembourser
Variableselonletaux
prfrentiel[]
Margedecrdit
ressource-totCIBC
Rductiondutaux
d'intrtenmettant
engarantiesa
propritoucellede
sesparents
Institution
financire8
Lignedecrdit
tudiant
10000$/an,maximum40000$
diplomation
Nondisponible
Lignedecrdit
tudiant()pour
professionnelset
diplms
8000$/anpendantdeuxans:
maximum48000$avecladettedu
1ercycle(pourcyclessuprieurs)


151
Tableau I-3 : Offre de carte de crdit cible par institution financire
Domaines Intrt
Intrt
rduit
Fraisannuels Autres
Institution
financire1
Nondisponible
Institution
financire2
Nondisponible
Institution
financire3
Nondisponible
Institution
financire4
Santsaufmdecine
19,50% ND
125$/anaprs2
ans
CartePlatine;Programmedepoints
Droitetnotariat 75$/anaprs2ans
Carteor;Programmedercompenses
encourageantl'utilisationdelacarte
Gnie
125$/anaprs2
ans
CartePlatine;Programmedercompenses
Infirmier
125$/anaprs2
ans
CartePlatine;Programmedercompenses
HECMontral 75$/anaprs2ans
Comptabilit 75$/anaprs2ans
Institution
financire5
Mdecineet
optomtrie
19,99% ND Non
CarteVISA*Or()Rcompenses;Programme
depointsRBCrcompense
Institution
financire6
Professionslibrales
19,99%;
21,99%
pourles
avancesde
fonds
ND 39$/an
VisaMomentum(;)Remisesenespces(2%
pourachatsdesubsistance,sinon1%)
ND Non
VISAor()sansfraisannuels;Bonislisaux
voyages
Institution
financire7
Nondisponible
Institution
financire8
Nondisponible


152
Tableau I-4 : Marges de crdit offertes par institution financire selon le domaine d'tude
Domaine Montant Remboursement Tauxd'intrt
Institution
financire1
mdecine,mdecinedentaire,
optomtrieetpharmacie;
Maximumnon-spcifi,
"lev"
Intrtsseulementdurant
lestudes;reportdu
capitaljusqu'6mois
aprslafindestudes
"Tauxavantageux"
durantlestudes;fixe
ouvariableaprsles
tudesendessousdu
prtpersonnel
actuariat,droit,gnieet
pharmacologie;
chiropratique,mdecinevtrinaire,
podiatrieetsciencescomptables.
Institution
financire2
Mdecine 200000$
maximum/50000$par
an
lestudes;12moisde
dlaiaprsladiplomation
Nondisponible
Dentisterie 200000$
maximum/50000$par
an
Chiropractie 80000$
maximum/20000$par
an
Optomtrie
Droit
Comptabilit
Pharmacie
Mdecinevterinaire
MBA 50000$
maximum/10000$par
an
Institution
financire3
Nondisponible
Institution
financire4
Sant 120000$saufmdecine
etmdecinedentaire
(200000$)
lestudes;12moisde
Base(2,50%)
Droitetnotariat 7000$/anpourquatre
ans+7000$pourdes
tudesau2ecycle.
Maximum35000$
Rduitde0,50%
(4,75%annuelau1er
mars2011)
Gnie Non-spcifi Base+0,50%(3%au
1ermars2010)
Infirmier 5500$/anpour4ans+
6000$pour2ecycle:
25000$terme
Base+0,75%(3,25%
au1ermars2010)
HECMontral 7000$/anpourquatre
ans+7000$pourdes
tudesau2ecycle.
Maximum35000$
Rduitde0,50%
(4,75%annuelau1er
mars2011)
Comptabilit
Institution
financire5
optomtrie,mdecinevtrinaire,
pharmacie,podiatrie,sciences
infirmires,chiropractie,droit,gnie,
ergothrapie,physiothrapie,
MBA/EMBAetcomptabilit
De40000$150000$
maximumselonle
dossier
lestudes;12moisde
Tauxprfrentiel+1%
Mdecine,dentisterie 200000$maximum,
sansrestrictionsurle
montantannuel
lestudes;12moisde
Tauxprfrentiel+1%
Institution
financire6
MBA 40000$maximum Intrtsseulementdurant
lestudes;12moisde
Nondisponible
Institution
financire7
Mdecine/dentisterie 200000$maximum Intrtsseulementdurant
lestudes;12moisde
(6moispourunesortie
sansdiplme);jusqu'20
anspourrembourser
Variableselonletaux
prfrentiel() Mdecinenaturopathique 140000$maximum
Mdecinevtrinaire 125000$maximum
Optomtrie,pharmacie,,
comptabilitouchiropratique
80000$maximum
MBAouMBApourcadres 70000$maximum
Sciencesinfirmiresoupodiatrie 55000$maximum
Institution
financire8
Chiropratique 62000$maximum Intrtsseulementdurant
lestudes;12moisde
Nondisponible
Dentisterie 20000$maximum;
62000$lapremire
anne

153
Droit 80000$pour3ans
MBA 80000$maximum
Mdecine 20000$maximum;
62000$lapremire
anne
Optomtrie,Pharmacie,Mdecine
vtrinaire
80000$maximum


154
Appendice I - Code SPSS premier cycle
Cration de nouvelles variables
*************************************************************
*CRATEUR:Louis-PhilippeSavoielpsavoie@gmail.com
*Datedederniremisejour:21juillet2011
*
*Cefichiercomprendl'entiretdesvariablescresdansle
*cadreduprojetderecherchesurl'endettementtudiant
*Normalement,excutertoutlefichiersuffitrecrer
*l'entiretdesvariablesutilises.
*
*Lefichierrenommelesvariablespourlesrendrelisibles.
*************************************************************
*Crationdevariables:residence,recodagedek75_m1(rsidenceprincipale).
*1=chezsesparents;2=autonome;sinonmanquante.
recodek75_m1
(7=1)
(1thru6=2)
(8=8)
(9=9)
(sysmis=sysmis)
intoresidence.
VARIABLELABELSresidence
'Lieudersidencedel''tudiant.Recodagedek75_m1'.
VALUELABELSresidence
1'Habitechezsesparents'
2'N''habitepaschezsesparents'
8'Autre'
9'Jeprfrenepasrpondre'.
MISSINGVALUESresidence(8,9).
EXECUTE.
*Divisiondesges.24etmoins=1;25etplus=2;sinonmanquante.
recodeq76
(lothru24=1)
(25thru98=2)
(99=99)
INTOageMediane.
VARIABLELABELSageMediane'gedivisselonlamdiane.Recodagedeq76'.
VALUELABELSageMediane
1'24ansetmoins'
2'25ansetplus'
execute.
*Recodagedelavariablegeq76pourregrouperlesvaleursextrmes.

155
RECODEq76
(0thru19=1)
(20=2)
(21=3)
(22=4)
(23=5)
(24=6)
(25=7)
(26thru30=8)
(31thru98=9)
(99=99)
INTOAgeStrat.
VARIABLELABELSAgeStrat'q76(ge)recodepourfaciliterletraitement'.
VALUELABELSAgeStrat
1'19ansetmoins'
2'20'
3'21'
4'22'
5'23'
6'24'
7'25'
8'2630ans'
9'31ansetplus'
execute.
*Prsenceouabsenced'aidefinancireauxtudes.
*1=sansAFE.
*2=AFE,prtsseulement.
*3=AFE,boursesseulement.
*Sinon,valeurmanquante.
computestatutAFE=$sysmis.
ifq12=2statutAFE=1.
ifq12=1&q13>0statutAFE=2.
ifq12=1&q14>0statutAFE=3.
ifq12=9statutAFE=9.
ifq12=$sysmisstatutAFE=$sysmis.
execute.
VARIABLELABELSstatutAFE'Statutl''aidefinancireauxtudes.Recodagedeq12et
q13'.
VALUELABELSstatutAFE
1'Sansprtsetbourses'
2'Bnficiaredeprtsseulement'
3'Bnficiairedeprtsetdebourses'
execute.
*Sparationdelavariablerevenuparentalselonlamdiane.
*1=Moinsde65000$paran.

156
*2=65000$etplus.
computerevParentsMediane=99.
ifq78<=9revParentsMediane=1.
ifq78>9&q78<98revParentsMediane=2.
ifq78=98revParentsMediane=8.
ifq78=99revParentsMediane=9.
EXECUTE.
VARIABLELABELSrevParentsMediane'Revenuparentaldivisselonlamdiane.Recodage
deq78.'.
VALUELABELSrevParentsMediane
1'Moinsde65000$'
2'65000$etplus.'
8'Nesaispas'
EXECUTE.
*Recodagedelavariableq78enquartilesderevenuparental.
RECODEq78(1thru6=1)(7thru9=2)(10thru13=3)(14thru17=4)(98=8)(99
=9)intoq78quartile.
VARIABLELABELSq78quartileRevenuparentaldivisenquartiles.
VALUELABELSq78quartile1'Moinsde39999$'2'40000$69999$'3'70000$124
999$'4'125000$etplus'8'Nesaispas'9'Jeprfrenepasrpondre'.
MISSINGVALUESq78quartile(8,9).
EXECUTE.
*Crationdelavariabletudiantdepremiregnration.
*Onconsidrecommetudiantdepremiregnration(EPG)untudiantdontaucundes
deuxparentsn'aunescolarituniversitaire.
*1=EPG.
*2=non-EPG.
*3=nesaitpas.
computeEPG=99.
if(q79a>=1&q79a<=3)|(q79b>=1&q79b<=3)EPG=1.
if(q79a>=4&q79a<=5)|(q79b>=4&q79b<=5)EPG=2.
if(q79a=8)|(q79b=8)EPG=3.
if(q79a=9)|(q79b=9)EPG=9.
if(q79a=$Sysmis)|(q79b=$Sysmis)EPG=$sysmis.
execute.
VARIABLELABELSEPG'tudiantdepremiregnration,drivedeq79aetq79b'.
VALUE LABELS EPG 1 'tudiant de premire gnration' 2 'tudiant de deuxime
gnrationouplus'3'Nesaitpas'9'Jeprfrenepasrpondre'.
MISSINGVALUESEPG(39).
execute.
*CrationdelavariabledureeBaccetrecodagedecelle-ci.
COMPUTEdureeBacc=q7B-q8b.
RECODEdureeBacc(20thruHI=sysmis)(lothru0=sysmis).

157
EXECUTE.
*RecodagededureeBaccpourarrter6ans
RECODE dureeBacc (1 = 1 ) (2=2) (3=3) (4=4) (5=5) (6 thru HI = 6) into

dureeBaccTronquee.
VARIABLELABELSdureeBaccTronquee'Duredudiplme,tronque6ans'.
VALUELABELSdureeBaccTronquee
6'6ansouplus'.
EXECUTE.
*Crationd'unenouvellevariablerentabiliteDiplomequiestunrecodagedudomaine
d'tudes(q9).
RECODEq9
(1,3,6,8=1)
(2,4,5,7,9=2)
(10=3)
(11=4)
(96=8)
(99=9)
intorentabiliteDiplome.
VARIABLELABELSrentabiliteDiplome
'Recodagedeq9endomainesrentabilitconomiqueforteetfaible'.
VALUELABELSrentabiliteDiplome
1'Faiblerentabilit'
2'Forterentabilit'
3'tudesplurisectorielles'
4'Noninscrit'
8'Nesaispas/tudesplurisectorielles'
MISSINGVALUESrentabiliteDiplome(3thru9).
EXECUTE.
********************************************************************************
*MANIPULATIONSDESDONNESSURL'ENDETTEMENTTUDIANT*******************
********************************************************************************
*
*Conversiondesvariablesq43aq43denformatbinaireo0=nonet1=ouiet
*crationdeq43bc.
*.
*AFE.
RECODEq43a
(1thruhi=1)
(0=0)
(sysmis=sysmis)
intoq43aTaux.
VARIABLELABELSq43aTaux
'Prsenceouabsenced''unedettel''AFE:projectiond''icilafindestudesde
premiercycle'.


158
VALUELABELSq43aTaux
0'Non'
1'Oui'.
EXECUTE.
*Endettementpriv.
*Crationdeq43bcquiestlasommedeq43betq43c.
COMPUTEq43bc=sum(q43b,q43c).
VARIABLELABELSq43bc'Prsenced''endettementpriv,avec0.Sommedeq43betq43c'.
EXECUTE.
*Recodagedeq43bc.
RECODEq43bc
(1thruhi=1)
(0=0)
(sysmis=sysmis)
intoq43bcTaux.
VARIABLELABELSq43bcTaux
'Prsenceouabsenced''unedetteuneinstitutionfinancire:projectiond''icila
findestudesdepremiercycle'.
VALUELABELSq43bcTaux
0'Non'
1'Oui'.
EXECUTE.
*Prtfamilial.
RECODEq43d
(1thruhi=1)
(0=0)
(sysmis=sysmis)
intoq43dTaux.
VARIABLELABELSq43dTaux
'Prsenceouabsenced''unedetteauprsdelafamilleoud''amis:projectiond''ici
lafindestudesdepremiercycle'.
VALUELABELSq43dTaux
0'Non'
1'Oui'.
EXECUTE.
*Prsenceouabsencededette.
COMPUTEq43=sum(q43a,q43b,q43c,q43d).
VARIABLELABELSq43'Dettetotale.Sommedeq43a,q43b,q43cetq43d'.
MISSINGVALUESq43(0).
EXECUTE.

159
RECODEq43
(1thruhi=1)
(0=0)
(sysmis=sysmis)
intoq43Taux.
VARIABLELABELSq43Taux
'Prsenceouabsenced''unedette:projectiond''icilafindestudesdepremier
cycle'.
VALUELABELSq43Taux
0'Non'
1'Oui'.
EXECUTE.
********************************************************************
*Mmemangequeplustt,maisenutilisantl'endettementactuel.*
********************************************************************
RECODEq42a
(1thruhi=1)
(0=0)
(sysmis=sysmis)
intoq42aTaux.
VARIABLELABELSq42aTaux
'Prsenceouabsenced''unedettel''AFE:projectiond''icilafindestudesde
premiercycle'.
VALUELABELSq42aTaux
0'Non'
1'Oui'.
EXECUTE.
*Crationdeq43bcquiestlasommedeq43betq43c.
COMPUTEq42bc=sum(q42b,q42c).
VARIABLELABELSq42bc'Prsenced''endettementpriv,avec0.Sommedeq42betq42c'.
EXECUTE.
RECODEq42bc
(1thruhi=1)
(0=0)
(sysmis=sysmis)
intoq42bcTaux.
VARIABLELABELSq42bcTaux
'Prsenceouabsenced''unedetteuneinstitutionfinancire:projectiond''icila
findestudesdepremiercycle'.
VALUELABELSq42bcTaux

160
0'Non'
1'Oui'.
EXECUTE.
*Prtfamilial.
RECODEq42d
(1thruhi=1)
(0=0)
(sysmis=sysmis)
intoq42dTaux.
VARIABLELABELSq42dTaux
'Prsenceouabsenced''unedetteauprsdelafamilleoud''amis:projectiond''ici
lafindestudesdepremiercycle'.
VALUELABELSq42dTaux
0'Non'
1'Oui'.
EXECUTE.
COMPUTEq42aBIS=q42a.
MISSINGVALUESq42aBIS(0,50001thru99999999).
VARIABLELABELSq42aBIS'Detteactuellel''AFE'.
EXECUTE.
COMPUTEq42bcBIS=sum(q42b,q42c).
MISSINGVALUESq42bcBIS(0,50001thru99999999).
VARIABLELABELSq42bcBIS'Detteactuelleuneinstitutionfinancire'.
EXECUTE.
COMPUTEq42dBIS=q42d.
MISSINGVALUESq42dBIS(0,50001thru99999999).
VARIABLELABELSq42dBIS'Detteactuellelafamilleoulesamis'.
EXECUTE.
COMPUTEq42BIS=sum(q42a,q42b,q42c,q42d).
MISSINGVALUESq42BIS(0,50001thru99999999).
VARIABLELABELSq42BIS'Dettetotaleactuelle'.
EXECUTE.
*Prsenceouabsencededette.
COMPUTEq42=sum(q42a,q42b,q42c,q42d).
VARIABLELABELSq42'Dettetotale.Sommedeq42a,q42b,q42cetq42d'.
EXECUTE.
RECODEq42
(1thruhi=1)
(0=0)
(sysmis=sysmis)
intoq42Taux.

161
VARIABLELABELSq42Taux
'Prsenceouabsenced''unedette:projectiond''icilafindestudesdepremier
cycle'.
VALUELABELSq42Taux
0'Non'
1'Oui'.
EXECUTE.
*nbSrcDettes:crationd'unenouvellevariable.
COMPUTEnbSrcDettes=sum(q43aTaux,q43bcTaux,q43dTaux).
VARIABLE LABELS nbSrcDettes 'Nombre de sources de dettes diffrentes. Somme de
q43aTaux,q43bcTauxetq43dTaux'.
EXECUTE.
**nbSrcFinancement:crationd'unenouvellevariable
*presenceCF,surlaprsenced'unecontributionfamiliale(parentaleouduconjoint.
Nedel'agglomrationdeq23etq26.
computepresenceCF=$sysmis.
ifq23=1|q26=1presenceCF=1.
ifq23=2&q26=2presenceCF=0.
ifq23=9|q26=9presenceCF=$sysmis.
ifq23=$sysmis|q26=$sysmispresenceCF=$sysmis.
VARIABLE LABELS presenceCF 'Prsence ou absence d''une contribution familiale.

Agglomrationdeq23etq26.'.
VALUELABELSpresenceCF1'Prsenced''unecontributionfamiliale'0'Absenced''une
contributionfamiliale'9'Jeprfrenepasrpondre'.
MISSINGVALUESpresenceCF(9).
execute.
*presenceBourses,quiestdrivedeq15etq19.
computepresenceBourses=1.
if((q15=3)AND(q19=3))presenceBourses=0.
if(q15=9ORq19=9)presenceBourses=$sysmis.
EXECUTE.
VARIABLE LABELS presenceBourses 'Prsence ou absence d''une bourse de mrite ou

autre.Agglomrationdeq15etq19.'.
VALUELABELSpresenceBourses1'Prsenced''uneboursedemriteouautre'0'Absence
d''uneboursedemriteouautre'9'Jeprfrenepasrpondre'.
MISSINGVALUESpresenceBourses(9).
execute.
*presenceTravail,quel'ondrivedeq3237.
recodeq3237
(sysmis=sysmis)
(0=0)
(1thru100000,000=1)

162
(100001,000thruhi=SYSMIS)
intopresenceTravail.
VARIABLE LABELS presenceTravail 'Prsence ou absence d''un emploi rmunr pour la

priodel''tude,drivedeq3237'.
VALUELABELSpresenceTravail1'Prsenced''aumoinsunemploi'0'Sansemploi'.
EXECUTE.
*Modificationdeq12(prsenceouabsenced'AFE)pourdclarer9(nesaispas)en
manquante.
RECODEq12
(1=1)
(2=0)
(9=sysmis)
intopresenceAFE.
VARIABLELABELSpresenceAFE'RecodagedeQ12pourletraitementdunombredesources
definancement'.
VALUELABELSpresenceAFE0'SansAFE'1'AvecAFE'9'Jeprfrenepasrpondre'.
MISSINGVALUESpresenceAFE(9).
EXECUTE.
* nbSrcFinancement : Compilation du nombre de sources de financement. Addition de

presenceCF, q12, presenceBourses, presenceTravail avec des + pour qu'une variable
manquantefasse"choker"lesystme.
COMPUTEnbSrcFinancement=presenceAFE+presenceCF+presenceBourses+presenceTravail.
VARIABLELABELSnbSrcFinancement'Nombredesourcesdefinancementd''untudiant'.
EXECUTE.
* Travail rmunr en cours d'tude : nombre d'heures travailles pour la session

Automne2009.
*Additiondek32c3etk37c3
COMPUTEk3237c3=mean
(a32c3,b32c3,c32c3,d32c3,e32c3,f32c3,a37c3,b37c3,c37c3,d37c3,e37c3,f37c3).
VARIABLELABELSk3237c3'Nombred''heurestravaillesparsemaine,automne2009'.
MISSINGVALUESk3237c3(0,71thru9999).
EXECUTE.
*Valeurpourdiscriminertouslestudiantsquitravaillentplusde15heurespar
semaineenmmetempsqueleurstudes.
recodek3237c3
(sysmis=sysmis)
(0=0)
(1thru14=1)
(15thru70=2)
(71thru9999=sysmis)
intotravail15h.
VARIABLE LABELS travail15h 'Proportion des tudiants qui travaillent plus ou moins
que15heures'.

163
VALUELABELStravail15h
1'Travaille14heuresetmoinsparsemaine'
2'Travaille15heuresetplusparsemaine'.
MISSINGVALUEStravail15h(0).
EXECUTE.
*Variablepourdiscriminertouslestudiantsquitravaillentplusde20heurespar
semainel'automne2009.
recodek3237c3
(sysmis=sysmis)
(0=0)
(1thru19=1)
(20thru70=2)
(71thru9999=sysmis)
intotravail20h.
VARIABLE LABELS travail20h 'Proportion des tudiants qui travaillent plus ou moins
quevingtheures'.
VALUELABELStravail20h
1'Travaille19heuresetmoinsparsemaine'
2'Travaille20heuresetplusparsemaine'.
MISSINGVALUEStravail20h(0).
EXECUTE.
* Variable qui dtermine si l'tudiant travaillait ou non l'automne 2009. Une

petiteerreurstatistiquepeuts'insrerduaumodedecalculquiutiliselessysmis
dek3237c.
*L'approximationestsuffisantepourlesbesoinsdelacause.
COMPUTEtravailA09=$sysmis.
RECODEk3237c3
(sysmis=0)
(0=0)
(1thru70=1)
(71thru9999=sysmis)
INTOtravailA09.
VARIABLELABELStravailA09'Tauxd''emploi,automne2009'.
VALUELABELStravailA09
0'Sans-emploi'
1'Occupaitaumoinsunemploi'.
EXECUTE.
*Variablequidterminesil'tudianttravaillaitounonen2009.Unepetiteerreur
statistiquepeuts'insrerduaumodedecalculquiutiliselessysmisdek3237.
*L'approximationestsuffisantepourlesbesoinsdelacause.
COMPUTEpresenceTravail=$sysmis.
RECODEq3237

164
(sysmis=0)
(0=0)
(1thru100000,000=1)
(100001,000thruHI=sysmis)
INTOpresenceTravail.
VARIABLELABELSpresenceTravail'Tauxd''emploi'.
VALUELABELSpresenceTravail
0'Sans-emploi'
1'Occupaitaumoinsunemploi'.
EXECUTE.
*k16d20bestlasommedesbourses(mriteetstages),soitk16d+k20b.
COMPUTEk16b20d=sum(k16d,k20b).
VARIABLE LABELS k16b20d 'k16d20b est la somme des bourses (mrite et stages), soit
k16d+k20b.'.
MISSINGVALUESk16b20d(30001toHI).
EXECUTE.
*lessriesq43*BISmodifientlgrementq43a-denajoutantunevaleurmanquante
0.
COMPUTEq43aBIS=q43a.
MISSINGVALUESq43aBIS(0,50001thru99999999).
VARIABLE LABELS q43aBIS 'Dette estime la fin des tudes de premier cycle
l''AFE'.
EXECUTE.
COMPUTEq43bcBIS=sum(q43b,q43c).
MISSINGVALUESq43bcBIS(0,50001thru99999999).
VARIABLELABELSq43bcBIS'Detteestimelafindestudesdepremiercycleune
institutionfinancire'.
EXECUTE.
COMPUTEq43dBIS=q43d.
MISSINGVALUESq43dBIS(0,50001thru99999999).
VARIABLE LABELS q43dBIS 'Dette estime la fin des tudes de premier cycle la
familleoulesamis'.
EXECUTE.
COMPUTEq43BIS=sum(q43a,q43b,q43c,q43d).
MISSINGVALUESq43BIS(0,50001thru99999999).
VARIABLELABELSq43BIS'Detteestimelafindestudesdepremiercycle'.
EXECUTE.
*Prsenceounond'enfantcharge.RecodagedeQ85B1(gedel'enfant).
RECODEq85b1
(sysmis=0)
(1thru8=1)
(9=9)

165
intopresenceEnfant.
VARIABLELABELSpresenceEnfant'Prsenceouabsenced''unenfantcharge.Recodage
deq85b1'.
VALUELABELSpresenceEnfant
1'Enfantcharge'
0'Sansenfantcharge'
EXECUTE.
* Dpenses selon le primtre de sources et modes. On compte donc : frais de

scolarit,fraisaffrents,matrielscolaire,loyer,vtements,nourriture.
COMPUTEdepensesESMFPC=sum(k49x,k50x,k53,k54).
VARIABLE LABELS depensesESMFPC 'Dpenses totales en matire de frais scolaires,
logement,nourritureetvtements(k49x,k50x,k53,k54'.
EXECUTE.
*Insertiondesdpensesdetransportetdeloisir.
COMPUTEdepensesTotales=sum(k29b,k49x,k50x,k53,k54,k56ax,k56bx,k58x).
VARIABLELABELSdepensesTotales'Dpensestotales,incluantletransportetleloisir
[k29b,k49x,k50x,k53,k54,k56ax,k56bx,k58x]'.
EXECUTE.
*Lefinancementtotalcomprendtouteslessourcesdefinancementdestudiants.
COMPUTEfinTotal=sum(q13,q14,k16d,k20b,k24,k27,k28b,q3237).
VARIABLELABELSfinTotal'Financementtotal:sommedesprtsetbourses,desbourses
de mrite, des contributions familiales et du travail
(q13,q14,k16d,k20b,k24,k27,k28b,q3237)'.
EXECUTE.
* Pour des raisons que je ne comprends pas, magTotal est trs loin des rsultats
prsentsdansESMFPC.OnvacrerdeuxMAG:magESMFPCetmagTotal.
* Il va falloir revrifier le calcul. Pour l'instant, je vais crer un nouvel
indicateur,mag1000.
COMPUTEmagESMFPC=finTotal-depensesESMFPC.
VARIABLE LABELS magESMFPC 'manque gagner selon le primtre de subsistance de
sourcesetmodespouruneanne'.
COMPUTEmagTotal=finTotal-depensesTotales.
VARIABLE LABELS magTotal 'manque gagner total pour une anne. finTotal-
depensesTotales'.
COMPUTEmagAFE=sum(finTotal,-q13,-depensesESMFPC).
VARIABLE LABELS magAFE 'Manque gagner total sans prts de l''AFE. finTotal-q13-
depensesESMFPC'.
*CrationdemagPourcent,utilisepourconnatrelenombred'tudiantsensituation
demanquegagner.
RECODEmagTotal

166
(lothru0=1)
(1thruhi=0)
(sysmis=sysmis)
intomagPourcent.
VARIABLE LABELS magPourcent 'Recodage de magTotal pour obtenir la prsence ou
l''absencedemanquegagner'.
VALUELABELSmagPourcent
0'Surplus'
1'Manquegagner'.
EXECUTE.
*Crationdemag1000,unevariabledemanquegagnerquicreunecatgorieautour
du0quiliminetouslesmanquegagner
* ou les surplus de moins de 500$, qui sont probablement des accumulations
d'erreurs.
RECODEmagTotal
(lothru-1000=2)
(-999thru1000=1)
(1001thruHI=0)
(sysmis=sysmis)
INTOmag1000.
VARIABLELABELSmag1000'RecodagedemagTotaloonenlvelestudiantsprsentant
unmanquegagnerouunsurplusde1000$etmoins'.
VALUELABELSmag1000
0'Surplus'
1'Entre-999$et1000$'
2'Manquegagner'.
EXECUTE.
* Cration de annee, qui est un recodage de q8b et estime l'anne d'tudes de

l'tudiant.
*Lavariableestcomptabiliseenfonctiondel'annescolaire:ainsi,untudiant
quicommenceenjanvier2008vatrerendu,l'automne2009,satroisimeanne.
COMPUTEannee=$sysmis.
IFq8b=2009andq8a>=8annee=1.
IFq8b=2009andq8a<8annee=2.
IFq8b<2006annee=5.
IFq8b>2010annee=9.
IFq8a>12annee=9.
EXECUTE.
VARIABLELABELSannee'Anned''tudedel''tudiant,drivedeq8aetq8b.'.
VALUE LABELS annee 1 'Premire anne' 2 'Deuxime anne' 3 'Troisime anne' 4
'Quatrime anne' 5 'Cinquime anne et plus' 9 'Ne sais pas/Je prfre ne pas
rpondre'.

167
MISSINGVALUESannee(09).
*Regnrationdescasiers.UtilesiSPSSbouffetout!.
*Regroupementvisuel.
*k24.
RECODEk24(MISSING=COPY)(LOTHRU0.0=1)(LOTHRU1000.0=2)(LOTHRU2000.0=3)(LO
THRU 3000.0=4) (LO THRU 4000.0=5) (LO THRU 5000.0=6) (LO THRU 6000.0=7) (LO THRU
HI=8)(ELSE=SYSMIS)INTOk24Casiers.
IF(VALUE(k24)EQ0.0)k24Casiers=0.0.
VARIABLELABELSk24Casiers'Q24*:Q24A+Q24B+Q24C(Regroupparcasiers)'.
FORMATSk24Casiers(F5.0).
VALUE LABELS k24Casiers 1 '<= ,00' 2 '1,00 - 1000,00' 3 '1001,00 - 2000,00' 4
'2001,00-3000,00'5'3001,00-4000,00'6'4001,00-5000,00'7'5001,00-6000,00'
8'6001,00+'.
MISSINGVALUESk24Casiers(0.0,50001.0THRU9.9999999E7).
VARIABLELEVELk24Casiers(ORDINAL).
EXECUTE.
*k3237c3.
RECODEk3237c3(MISSING=COPY)(LOTHRU0.0=1)(LOTHRU5.0=2)(LOTHRU10.0=3)(LO
THRU15.0=4)(LOTHRU20.0=5)(LOTHRU25.0=6)(LOTHRU30.0=7)(LOTHRU35.0=8)(LO
THRU40.0=9)(LOTHRUHI=10)(ELSE=SYSMIS)INTOk3237c3Casiers.
IF(VALUE(k3237c3)EQ0.0)k3237c3Casiers=0.0.
VARIABLE LABELS k3237c3Casiers "Nombre d'heures travailles par semaine, automne
2009(Regroupparcasiers)".
FORMATSk3237c3Casiers(F5.0).
VALUELABELSk3237c3Casiers1'0'2'1,00-5,00'3'6,00-10,00'4'11,00-15,00'
5'16,00-20,00'6'21,00-25,00'7'26,00-30,00'8'31,00-35,00'9'36,00-
40,00'10'41,00+'.
MISSINGVALUESk3237c3Casiers(0.0,71.0THRU9999.0).
VARIABLELEVELk3237c3Casiers(ORDINAL).
EXECUTE.
*q3237.
RECODEq3237(MISSING=COPY)(LOTHRU0.0=1)(LOTHRU5000.0=2)(LOTHRU10000.0=3)
(LO THRU 15000.0=4) (LO THRU 20000.0=5) (LO THRU 25000.0=6) (LO THRU HI=7)
(ELSE=SYSMIS)INTOq3237Casiers.
VARIABLE LABELS q3237Casiers "Q3237B* : Revenu d'emploi total (Regroup par
casiers)".
FORMATSq3237Casiers(F5.0).
VALUE LABELS q3237Casiers 1 '0' 2 '1,00 - 5000,00' 3 '5001,00 - 10000,00' 4
'10001,00-15000,00'5'15001,00-20000,00'6'20001,00-25000,00'7'25001,00+'.
MISSINGVALUESq3237Casiers(100001.0THRU9.99999999E8).
VARIABLELEVELq3237Casiers(ORDINAL).
EXECUTE.
*k16b20d.
RECODEk16b20d(MISSING=COPY)(LOTHRU0.0=1)(LOTHRU1000.0=2)(LOTHRU2000.0=3)
(LOTHRU3000.0=4)(LOTHRU4000.0=5)(LOTHRU5000.0=6)(LOTHRUHI=7)(ELSE=SYSMIS)
INTOk16b20dCasiers.
VARIABLELABELSk16b20dCasiers'k16d20bestlasommedesbourses(mriteetstages),
soitk16d+k20b.(Regroupparcasiers)'.
FORMATSk16b20dCasiers(F5.0).

168
VALUE LABELS k16b20dCasiers 1 '0' 2 '1,00 - 1000,00' 3 '1001,00 - 2000,00' 4
'2001,00-3000,00'5'3001,00-4000,00'6'4001,00-5000,00'7'5001,00+'.
MISSINGVALUESk16b20dCasiers(30001.0THRUHI).
VARIABLELEVELk16b20dCasiers(ORDINAL).
EXECUTE.
*finTotal.
RECODE finTotal (MISSING=COPY) (LO THRU 0.0=1) (LO THRU 5000.0=2) (LO THRU
10000.0=3) (LO THRU 15000.0=4) (LO THRU 20000.0=5) (LO THRU 25000.0=6) (LO THRU
30000.0=7)(LOTHRUHI=8)(ELSE=SYSMIS)INTOfinTotalCasiers.
VARIABLELABELSfinTotalCasiers'Financementtotal:sommedesprtsetbourses,des
bourses de mrite, des contributions familiales et du travail
(q13,q14,k16d,k20b,k24,k27,k28b,q3237)(Regroupparcasiers)'.
FORMATSfinTotalCasiers(F5.0).
VALUE LABELS finTotalCasiers 1 '0' 2 '1,00 - 5000,00' 3 '5001,00 - 10000,00' 4
'10001,00-15000,00'5'15001,00-20000,00'6'20001,00-25000,00'7'25001,00-
30000,00'8'30001,00+'.
VARIABLELEVELfinTotalCasiers(ORDINAL).
EXECUTE.
*depensesTotales.
RECODE depensesTotales (MISSING=COPY) (LO THRU 0.0=1) (LO THRU 5000.0=2) (LO THRU
30000.0=7)(LOTHRUHI=8)(ELSE=SYSMIS)INTOdepensesTotalesCasiers.
VARIABLE LABELS depensesTotalesCasiers 'Dpenses totales, incluant le transport et
le loisir [k29b, k49x, k50x, k53, k54, k56ax, k56bx,k57x, k58x] (Regroup par
casiers)'.
FORMATSdepensesTotalesCasiers(F5.0).
VALUELABELSdepensesTotalesCasiers1'0'2'1,00-5000,00'3'5001,00-10000,00'
4'10001,00-15000,00'5'15001,00-20000,00'6'20001,00-25000,00'7'25001,00-
30000,00'8'30001,00+'.
VARIABLELEVELdepensesTotalesCasiers(ORDINAL).
EXECUTE.
*k54.
THRU750.0=4)(LOTHRU1000.0=5)(LOTHRU1250.0=6)(LOTHRU1500.0=7)(LOTHRUHI=8)
(ELSE=SYSMIS)INTOk54Casiers.
VARIABLELABELSk54Casiers'Q54*:TOTALQ54A,Q54BQ54C(Regroupparcasiers)'.
VALUELABELSk54Casiers1'<=,00'2'1,00-250,00'3'251,00-500,00'4'501,00-
750,00' 5 '751,00 - 1000,00' 6 '1001,00 - 1250,00' 7 '1251,00 - 1500,00' 8
'1501,00+'.
MISSINGVALUESk54Casiers(7501.0THRU9.9999999E7).
EXECUTE.
*k49x.
RECODEk49x(MISSING=COPY)(LOTHRU0.0=1)(LOTHRU2500.0=2)(LOTHRU5000.0=3)(LO
THRU7500.0=4)(LOTHRU10000.0=5)(LOTHRU12500.0=6)(LOTHRU15000.0=7)(LOTHRU
HI=8)(ELSE=SYSMIS)INTOk49xCasiers.
VARIABLELABELSk49xCasiers'Q49*A:Moyenneannuelle(Regroupparcasiers)'.
FORMATSk49xCasiers(F5.0).

169
VALUELABELSk49xCasiers1'0'2'1,00-2500,00'3'2501,00-5000,00'4'5001,00-
7500,00' 5 '7501,00 - 10000,00' 6 '10001,00 - 12500,00' 7 '12501,00 - 15000,00' 8
'15001,00+'.
MISSINGVALUESk49xCasiers(36001.0THRU9.99999999E8).
VARIABLELEVELk49xCasiers(ORDINAL).
EXECUTE.
*k50x.
THRU 3000.0=4) (LO THRU 4000.0=5) (LO THRU 5000.0=6) (LO THRU 6000.0=7) (LO THRU
HI=8)(ELSE=SYSMIS)INTOk50xCasiers.
3000,00' 5 '3001,00 - 4000,00' 6 '4001,00 - 5000,00' 7 '5001,00 - 6000,00' 8
'6001,00+'.
EXECUTE.
*k56ax.
RECODEk56ax(MISSING=COPY)(LOTHRU0.0=1)(LOTHRU500.0=2)(LOTHRU1000.0=3)(LO
THRU1500.0=4)(LOTHRU2000.0=5)(LOTHRUHI=6)(ELSE=SYSMIS)INTOk56axCasiers.
VARIABLELABELSk56axCasiers'Q56A*A:Moyenneannuelle(Regroupparcasiers)'.
FORMATSk56axCasiers(F5.0).
VALUELABELSk56axCasiers1'0'2'1,00-500,00'3'501,00-1000,00'4'1001,00-
1500,00'5'1501,00-2000,00'6'2001,00+'.
MISSINGVALUESk56axCasiers(6001.0THRU9.9999999E7).
VARIABLELEVELk56axCasiers(ORDINAL).
EXECUTE.
*k57x.
THRU 6000.0=4) (LO THRU 8000.0=5) (LO THRU 10000.0=6) (LO THRU HI=7) (ELSE=SYSMIS)
INTOk57xCasiers.
6000,00'5'6001,00-8000,00'6'8001,00-10000,00'7'10001,00+'.
EXECUTE.
*k58x.
INTOk58xCasiers.
3000,00'5'3001,00-4000,00'6'4001,00-5000,00'7'5001,00+'.
EXECUTE.

170
*q43aBIS.
RECODEq43aBIS(MISSING=COPY)(LOTHRU0.0=1)(LOTHRU5000.0=2)(LOTHRU10000.0=3)
(LOTHRU15000.0=4)(LOTHRU20000.0=5)(LOTHRU25000.0=6)(LOTHRU30000.0=7)(LO
THRU35000.0=8)(LOTHRU40000.0=9)(LOTHRU45000.0=10)(LOTHRUHI=11)
(ELSE=SYSMIS)INTOq43aCasiers.
IF(VALUE(q43aBIS)EQ0.0)q43aCasiers=0.0.
VARIABLELABELSq43aCasiers"Detteestimelafindestudesdepremiercycle
l'AFE(Regroupparcasiers)".
FORMATSq43aCasiers(F5.0).
VALUE LABELS q43aCasiers 1 '<= ,00' 2 '1,00 - 5000,00' 3 '5001,00 - 10000,00' 4
'10001,00-15000,00'5'15001,00-20000,00'6'20001,00-25000,00'7'25001,00-
30000,00'8'30001,00-35000,00'9'35001,00-40000,00'10'40001,00-45000,00'11
'45001,00+'0.0'0'.
MISSINGVALUESq43aCasiers(0.0,50001.0THRU9.9999999E7).
VARIABLELEVELq43aCasiers(ORDINAL).
EXECUTE.
*q43bcBIS.
RECODE q43bcBIS (MISSING=COPY) (LO THRU 0.0=1) (LO THRU 5000.0=2) (LO THRU
30000.0=7)(LOTHRUHI=8)(ELSE=SYSMIS)INTOq43bcCasiers.
IF(VALUE(q43bcBIS)EQ0.0)q43bcCasiers=0.0.
VARIABLELABELSq43bcCasiers'Detteestimelafindestudesdepremiercycle
uneinstitutionfinancire(Regroupparcasiers)'.
FORMATSq43bcCasiers(F5.0).
VALUE LABELS q43bcCasiers 1 '0' 2 '1,00 - 5000,00' 3 '5001,00 - 10000,00' 4
'10001,00-15000,00'5'15001,00-20000,00'6'20001,00-25000,00'7'25001,00-
30000,00'8'30001,00+'.
MISSINGVALUESq43bcCasiers(0.0,50001.0THRU9.9999999E7).
VARIABLELEVELq43bcCasiers(ORDINAL).
EXECUTE.
*q43dBIS.
RECODEq43dBIS(MISSING=COPY)(LOTHRU0.0=1)(LOTHRU2500.0=2)(LOTHRU5000.0=3)
(LO THRU 7500.0=4) (LO THRU 10000.0=5) (LO THRU 12500.0=6) (LO THRU 15000.0=7) (LO
THRUHI=8)(ELSE=SYSMIS)INTOq43dCasiers.
IF(VALUE(q43dBIS)EQ0.0)q43dCasiers=0.0.
VARIABLELABELSq43dCasiers'Detteestimelafindestudesdepremiercyclela
familleoulesamis(Regroupparcasiers)'.
FORMATSq43dCasiers(F5.0).
VALUELABELSq43dCasiers1'0'2'1,00-2500,00'3'2501,00-5000,00'4'5001,00-
7500,00' 5 '7501,00 - 10000,00' 6 '10001,00 - 12500,00' 7 '12501,00 - 15000,00' 8
'15001,00+'.
MISSINGVALUESq43dCasiers(0.0,50001.0THRU9.9999999E7).
VARIABLELEVELq43dCasiers(ORDINAL).
EXECUTE.
*q43BIS.
RECODEq43BIS(MISSING=COPY)(LOTHRU0.0=1)(LOTHRU5000.0=2)(LOTHRU10000.0=3)
(LOTHRU15000.0=4)(LOTHRU20000.0=5)(LOTHRU25000.0=6)(LOTHRU30000.0=7)(LO
THRU 35000.0=8) (LO THRU 40000.0=9) (LO THRU 45000.0=10) (LO THRU HI=11)
(ELSE=SYSMIS)
INTOq43Casiers.

171
IF(VALUE(q43BIS)EQ0.0)q43Casiers=0.0.
VARIABLE LABELS q43Casiers 'Dette estime la fin des tudes de premier cycle
(Regroupparcasiers)'.
FORMATSq43Casiers(F5.0).
VALUE LABELS q43Casiers 1 '<= ,00' 2 '1,00 - 5000,00' 3 '5001,00 - 10000,00' 4
'10001,00-15000,00'5'15001,00-20000,00'6'20001,00-25000,00'7'25001,00-
30000,00'8'30001,00-35000,00'9'35001,00-40000,00'10'40001,00-45000,00'11
'45001,00+'0.0'0'.
MISSINGVALUESq43Casiers(0.0,50001.0THRU9.9999999E7).
VARIABLELEVELq43Casiers(ORDINAL).
EXECUTE.
*k53.
INTOk53Casiers.
VARIABLELABELSk53Casiers'Q53*:TOTALQ53A,Q53BQ53C(Regroupparcasiers)'.
VALUELABELSk53Casiers1'0'2'1,00-1000,00'3'1001,00-2000,00'4'2001,00-
3000,00'5'3001,00-4000,00'6'4001,00-5000,00'7'5001,00+'.
MISSINGVALUESk53Casiers(30001.0THRU9999999.0).
EXECUTE.
* Prparation de donnes pour les tests : dclarations de nouvelles valeurs

manquantes, gnralement, il s'agit des non-rponses, pour permettre des tests
statistiqueslogiques.
*Profilsocioconomique.
MISSINGVALUESageStrat(99).
MISSINGVALUESq78quartile(8,9).
MISSINGVALUESresidence(8,9).
MISSINGVALUESregion2(6,99).
MISSINGVALUESq64a(9).
MISSINGVALUESpresenceEnfant(9).
MISSINGVALUESepg(3,9).
MISSINGVALUESrentabiliteDiplome(3thru9).
MISSINGVALUESq3237(0,100001thru999999999).
EXECUTE.
*Rtiquetagedesvariablespourlesrendrelisibles.
VARIABLELABELSq76'q76:gedel''tudiant'.
VARIABLELABELSq78x'q78x:Revenufamilialbrut,convertienvariablecontinue'.
VARIABLE LABELS q78quartile 'q78quartile : Revenu familial brut, dvisi en
quartiles'.
VARIABLELABELSq64a'q64a:Changementdergion'.
VARIABLELABELSq84'q84:Probabilitderussite'.
VARIABLELABELSq82a'q82a:Faitd''avoirallongprcdemmentsestudes'.

172
VARIABLELABELSq23'q23:Prsencedelacontributionparentale'.
VARIABLELABELSk24'k24:Montantdelacontributionparental,moyenneannuelle'.
VARIABLELABELSk3237c3'k3237c3:nombred''heurestravailles,automne2009'.
VARIABLE LABELS travailA09 'travailA09 : Prsence ou absence d''emploi, automne
2009'.
VARIABLELABELSpresenceTravail'presenceTravail:Prsenceouabsenced''emploi'.
VARIABLELABELSq3237'q3237:Revenubrutannuelenprovenancedutravailrmunr'.
VARIABLE LABELS k53 'k53 : Dpenses annuelles en matire de frais de scolarit et

autresfraisobligatoires'.
VARIABLELABELSk54'k54:Dpensesannuellesenmatrielscolaire'.
VARIABLE LABELS k49x 'k49x : Dpenses annuelles pour les frais de logement (loyer,
lectricit,tlphone,chauffage,meubles,fournitures)'.
VARIABLELABELSk50x'k50x:Dpensesannuellesennourriture'.
VARIABLELABELSq52'q52:Dpensesannuellesenvtementsetchaussures'.
VARIABLELABELSk56ax'k56ax:Dpensesannuellesentransport'.
VARIABLELABELSk58x'k58x:Loisirsetdpensesautres'.
Code de gnration des tableaux
*************************************************************
* NOM : ESMFPC - Syntaxe de gnration des tableaux croiss et des comparaison de
moyennes.sps
*
*Cefichiercomprendl'entiretdesanalysestatistiquesdu
*projetderecherchesurl'endettementtudiant.
*
*Lesanalysessontd'aborddescriptives,etonprocdeparlasuite
*auxtestsstatistiques.Deuxcasspciauxsonttraitslafin.
*
*ILFAUTCOMMENCERPAREXCUTER
*ESMFPC-Endettement-Nouvellesvariablesetrecodages.
*
*associeravecunebasededonnesrcentedel'enqute
*sourcesetmodesdefinancementdestudiantsdepremiercycle.
*************************************************************
*Slectiondelavariabledepondration.ElleatconstruiteparLgerMarketing
selonlesadmissionsdel'automne2006.
* La pondration est une pondration sur marge en fonction du sexe, du rgime

d'tudesetdel'tablissementd'enseignement.
WEIGHTbypond.
* Slection de l'chantillon: on ne traite que les tudiants temps plein. La

variablecontrleestq2_m1.
USEALL.
COMPUTEfilter_$=(q2_m1=1).
VARIABLELABELSfilter_$'q2_m1=1(FILTER)'.
VALUELABELSfilter_$0'NotSelected'1'Selected'.

173
FORMATSfilter_$(f1.0).
FILTERBYfilter_$.
EXECUTE.
*************************************************************
*SECTION1:Statistiquesdescriptives.
*
* Onsortlesfrquencespourlesvariables
*discrtesetlamoyenne,lamdiane,ladviationstandardetles
*quartilespourlesvariablescontinues.
*************************************************************
*Caractristiquessocioconomiques
Sexe:sexedel'tudiant
AgeStratetq76:gedel'tudiant-ageStratcoupelesdonnespourfaciliter
laprsentation
q78quartileetq78x:Revenufamilialbrut
residence:lieudersidence
region2:rgionsd'tudes
q64a:allongementprcdentdestudes
presenceEnfant:prsenced'unenfantcharge
epg:statutd'tudiantdepremiregnration.
FREQUENCIESVARIABLES=sexeAgeStratq78quartileresidenceregion2q64apresenceEnfant
epg
/ORDER=ANALYSIS.
FREQUENCIESVARIABLES=q76q78x
/FORMAT=NOTABLE
/NTILES=4
/STATISTICS=STDDEVMEANMEDIAN
/ORDER=ANALYSIS.
FREQUENCIESVARIABLES=anneedureeBaccTronqueerentabiliteDiplomeq82a
/ORDER=ANALYSIS.
*Financement.
*FinTotal:Financementtotal
*q23etk24:contributionparentale(prsenceetmnontant).
* travailA09 et k3237c3 : Travail rmunr l'automne 2009 - prsence et nombre
d'heures.
* presenceTravail et q3237 : Travail rmunr durant l'anne - prsence et revenu
brut.
*presenceBoursesetk16b20dCasiers:boursesd'tudes.
FREQUENCIESVARIABLES=finTotalCasiers
q23k24Casiers
travailA09k3237c3casiers
presenceTravailq3237casiers
presenceBoursesk16b20dCasiers
/ORDER=ANALYSIS.
FREQUENCIESVARIABLES=finTotalk24k3237c3q3237k16b20d
/FORMAT=NOTABLE

174
/NTILES=4
/ORDER=ANALYSIS.
* Dpenses. Deux dpenses font l'objet de traitements particuliers (k53 et k49 -

fraisdescolaritetloyer),
*ilfautsereporterlafindufichier.Touteslesdpensessontpourunan.
*depensesTotales:Dpensestotales.
*k54Casiers:Dpensesdematrielscolaire.
*k49x:Loyer
*k50x:Nourriture
*k56ax:Transport
*k57x:Fraispourenfantcharge
(ici,lespersonnessansenfantn'ontpasdevaleurassocie
nousn'avonspasbesoindelesexclure.
*k58x:Dpensesdeloisirsetautresdpenses.
FREQUENCIES VARIABLES=depensesTotalesCasiers k54Casiers k50xCasiers k56axCasiers

k57xCasiersk58xCasiers
/ORDER=ANALYSIS.
FREQUENCIESVARIABLES=depensesTotalesk54k50xk56axk57xk58x
/FORMAT=NOTABLE
/NTILES=4
/ORDER=ANALYSIS.
*Typesdedettes
*q43a=AFE
*q43bc=Priv
*q43d=Famille/amis
*q43=total
* Les variables 'bis' sont simplement les variables normales, o les '0' ont t
convertisenvaleurmanquante.
FREQUENCIES VARIABLES=q43aTaux q43bcTaux q43dTaux q43Taux q43aCasiers q43bcCasiers

q43dCasiersq43Casiers
/ORDER=ANALYSIS.
FREQUENCIESVARIABLES=q43aBisq43bcBisq43dBisq43Bis
/FORMAT=NOTABLE
/NTILES=4
/ORDER=ANALYSIS.
*************************************************************.
*SECTION2:Comparaisonsdemoyennesettableauxcroiss.
*************************************************************.
*Version"silencieuse".Onsortlestableauxcroissetlescomparaisonsdemoyenne
avec:
*Corrlationsphietpourcentagedecolonnepourles
tableauxcroiss.

175
* Moyenne et mdiane pour les comparaisons de
moyennes.
* Les tests de corrlations phi ne sont utiles que
pourcertainesvariables,quisontnotesencommentaires.
* Pour les autres, on calcule plus loin les
coefficientsdecorrlationdePearson(r)etdeSpearman(rho)
*Comparaisondestauxd'endettementenfonctiondelatailledeladette.
CROSSTABS
/TABLES=q43aCasiers q43bcCasiers q43dCasiers q43Casiers by q43aTaux q43bcTaux
q43dTauxq43Taux
/FORMAT=AVALUETABLES
/STATISTICS=PHI
/CELLS=ROW
/COUNTROUNDCELL.
MEANSTABLES=q43aBISq43bcBISq43dBISq43BISBYq43aCasiersq43bcCasiersq43dCasiers
q43Casiers
/CELLSMEANMEDIAN.
*Comparaisondestauxd'endettement.
CROSSTABS
/TABLES=q43aTauxBYq43bcTauxq43dTauxq43Taux
/STATISTICS=PHI
/CELLS=COUNTROWCOLUMNTOTAL
/COUNTROUNDCELL.
CROSSTABS
/TABLES=q43bcTauxBYq43aTauxq43dTauxq43Taux
/STATISTICS=PHI
/COUNTROUNDCELL.
CROSSTABS
/TABLES=q43dTauxBYq43aTauxq43bcTauxq43Taux
/STATISTICS=PHI
/COUNTROUNDCELL.
CROSSTABS
/TABLES=q43TauxBYq43aTauxq43bcTauxq43dTaux
/STATISTICS=PHI
/COUNTROUNDCELL.
*Tests:profilsocioconomique.
*Phiestutilepourresidence;q64a;presenceEnfant;
epg

176
CROSSTABS
/TABLES=sexe AgeStrat q78quartile residence region2 q64a presenceEnfant epg BY
q43aTauxq43bcTauxq43dTauxq43Taux
/STATISTICS=PHI
/CELLS=ROW
/COUNTROUNDCELL.
MEANS TABLES=q43aBIS q43bcBIS q43dBIS q43BIS BY sexe AgeStrat q78quartile residence

region2q64apresenceEnfantepg
/CELLSMEANMEDIAN.
EXECUTE.
*Sectionsurleprofilscolaire.
*PhiestutilepourrentabiliteDiplome;q82a.
CROSSTABS
/TABLES=dureeBaccTronqueerentabiliteDiplomeq82aBYq43aTauxq43bcTauxq43dTaux
q43Taux
/STATISTICS=PHI
/CELLS=ROW
/COUNTROUNDCELL.
MEANSTABLES=q43aBISq43bcBISq43dBISq43BISBYdureeBaccTronqueerentabiliteDiplome
q82a
/CELLSMEANMEDIAN.
EXECUTE.
*Sectionsurlessourcesetmodesdefinancement,
*PhiestutilepourtravailA09presenceTravailpresenceBourses.
CROSSTABS
/TABLES=finTotalCasiers
q23k24Casiers
BYq43aTauxq43bcTauxq43dTauxq43Taux
/STATISTICS=PHI
/CELLS=ROW
/COUNTROUNDCELL.
MEANSTABLES=q43aBISq43bcBISq43dBISq43BIS
BYfinTotalCasiersq23k24Casiers
/CELLSMEANMEDIAN.

177
EXECUTE.
*Sectionsurlestypesdedpenses.
*Phin'estjamaisutile.
CROSSTABS
/TABLES=depensesTotalesCasiers k54Casiers k50xCasiers k56axCasiers k57xCasiers
k58xCasiersBYq43aTauxq43bcTauxq43dTauxq43Taux
/STATISTICS=PHI
/CELLS=ROW
/COUNTROUNDCELL.
MEANS TABLES=q43aBIS q43bcBIS q43dBIS q43BIS BY depensesTotalesCasiers k54Casiers

k50xCasiersk56axCasiersk57xCasiersk58xCasiers
/CELLSMEANMEDIAN.
EXECUTE.
********************************************************************************.
*SECTION3:Coefficientsdecorrlation.
********************************************************************************.
*Gnrationdescoefficientsdecorrelationpourlestaux.
* On ne calcule les coefficients de corrlation que pour les relations avec des
variablescontinues:pourlesrelationsentrevariablesdichotomiques,phiestdj
gnr.
*Endettement.
CORRELATIONS
/VARIABLES=q43aTauxq43bcTauxq43dTauxq43Tauxq43aBISq43bcBISq43dBISq43BIS
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
EXECUTE.
*Caractristiquessocioconomiques
CORRELATIONS
/VARIABLES=q43aTauxq43bcTauxq43dTauxq43Tauxq76q78x
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
CORRELATIONS
/VARIABLES=q43aBIS q43bcBIS q43dBIS q43BIS sexe q76 q78x residence region2 q64a
presenceEnfantepg
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
EXECUTE.
NONPARCORR
/VARIABLES=q43aBISq43bcBISq43dBISq43BISq76q78x

178
/PRINT=SPEARMANTWOTAILSIG
/MISSING=PAIRWISE.
*Caractristiquesscolaires.
CORRELATIONS
/VARIABLES=q43aTauxq43bcTauxq43dTauxq43TauxdureeBaccrentabiliteDiplome
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
EXECUTE.
CORRELATIONS
/VARIABLES=q43aBISq43bcBISq43dBISq43BISdureeBaccrentabiliteDiplomeq82a
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
EXECUTE.
NONPARCORR
/VARIABLES=q43aBISq43bcBISq43dBISq43BISdureeBacc
/MISSING=PAIRWISE.
*Sourcesetmodesdefinancement.
CORRELATIONS
/VARIABLES=q43aTauxq43bcTauxq43dTauxq43TauxfinTotalk24k3237c3q3237k16b20d
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
CORRELATIONS
/VARIABLES=q43aBIS q43bcBIS q43dBIS q43BIS finTotal q23 k24 travailA09 k3237c3
presenceTravailq3237presenceBoursesk16b20d
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
NONPARCORR
/VARIABLES=q43aBISq43bcBISq43dBISq43BISfinTotalk24k3237c3q3237k16b20d
/MISSING=PAIRWISE.
EXECUTE.
*Dpenses
CORRELATIONS
/VARIABLES=q43aTaux q43bcTaux q43dTaux q43Taux depensesTotales k54 k50x k56ax
k57xk58x
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
EXECUTE.
CORRELATIONS

179
/VARIABLES=q43aBIS q43bcBIS q43dBIS q43BIS depensesTotales k54 k50x k56ax k57x
k58x
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
NONPARCORR
/VARIABLES=q43aBIS q43bcBIS q43dBIS q43BIS depensesTotales k54 k50x k56ax k57x
k58x
/MISSING=PAIRWISE.
EXECUTE.
* Quelques corrlations entre la rentabilit du diplme et des caractristiques

triessurlevolet.
*Lestestsdesignificationsontunidirectionnels.
CORRELATIONS
/VARIABLES=rentabiliteDiplome travailA09 k54 q82a q84 EPG q78x depensesTotales
presenceEnfant
/PRINT=TWOTAILNOSIG
/STATISTICSDESCRIPTIVES
/MISSING=PAIRWISE.
* Tests statistiques spciaux. Pour deux types de dpenses, soit les frais de
scolaritetleloyer,nousexcluonsunepartiedel'chantillon.
USEALL.
COMPUTEfilter_$=(q2_m1=1&(residence=2)).
VARIABLELABELSfilter_$'q2_m1=1&(q10=1|q10=3)(FILTER)'.
FILTERBYfilter_$.
EXECUTE.
FREQUENCIESVARIABLES=k49xCasiers
/ORDER=ANALYSIS.
FREQUENCIESVARIABLES=k49x
/FORMAT=NOTABLE
/NTILES=4
/ORDER=ANALYSIS.
CROSSTABS
/TABLES=k49xCasiersBYq43aTauxq43bcTauxq43dTauxq43Taux
/STATISTICS=CHISQPHICORR
/CELLS=ROW
/COUNTROUNDCELL.
MEANSTABLES=q43aBISq43bcBISq43dBISq43BISBYk49xCasiers
/CELLSMEANMEDIAN

180
/STATISTICSANOVALINEARITY.
CORRELATIONS
/VARIABLES=q43aTauxq43bcTauxq43dTauxq43Taux
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
EXECUTE.
CORRELATIONS
/VARIABLES=q43aBisq43bcBisq43dBisq43bisk49x
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
EXECUTE.
NONPARCORR
/VARIABLES=q43aBISq43bcBISq43dBISq43BISk49x
/MISSING=PAIRWISE.
EXECUTE.
USEALL.
FILTERBYfilter_$.
EXECUTE.
* Exclusion temporaire des tudiants non-Qubcois pour les fins du traitement

statistiquedesfraisdescolarit.
USEALL.
COMPUTEfilter_$=(q2_m1=1&(q10=1|q10=3)).
VARIABLELABELSfilter_$'q2_m1=1&(q10=1|q10=3)(FILTER)'.
FILTERBYfilter_$.
EXECUTE.
CROSSTABS
/TABLES=k53CasiersBYq43aTauxq43bcTauxq43dTauxq43Taux
/STATISTICS=PHI
/CELLS=ROW
/COUNTROUNDCELL.
MEANSTABLES=q43aBISq43bcBISq43dBISq43BISBYk53Casiers
/CELLSMEANMEDIAN.
EXECUTE.
CORRELATIONS

181
/VARIABLES=q43aTaux q43bcTaux q43dTaux q43Taux q43aBIS q43bcBIS q43dBIS q43BIS
k53
/PRINT=TWOTAILNOSIG
/MISSING=PAIRWISE.
EXECUTE.
NONPARCORR
/VARIABLES=q43aTaux q43bcTaux q43dTaux q43Taux q43aBIS q43bcBIS q43dBIS q43BIS
k53
/MISSING=PAIRWISE.
USEALL.
FILTERBYfilter_$.
EXECUTE.


182
Appendice II - Code SPSS cycles suprieurs
*************************************************************
*
*Cefichiercomprendl'entiretdesvariablescreset
*modifiesainsiquelestableauxgnrsdanslecadredu
*projetderecherchesurl'endettementtudiant
*Normalement,excutertoutlefichiersuffitrecrer
*l'entiretdesvariablesutilises.
*
*Lefichierrenommelesvariablespourlesrendrelisibles.
*************************************************************
*Constructiond'unenouvellepondrationpond2fondesurlecycled'tudes.
RECODEq1a
(1,2=1.22158789693945)
(3,4=0.601297833323734)
(96=0)
intopond2.
WEIGHTbypond2.
VARIABLELABELSpond2'Nouvellepondrationsurlecycled''tudes'.
*Recodagedeq1apourdiscriminerdeuximeettroisimecycleuniquement.
RECODEq1a
(1,2=1)
(3,4=2)
(96=9)
intocycleEtudes.
VARIABLELABELScycleEtudes'Cycled''tudes'.
VALUELABELScycleEtudes
1'Deuximecycle'
2'Troisimecycle'
9'Non-rponse'.
MISSINGVALUEScycleEtudes(9).
*Recodagedeq40pourenrduirelenombredeclasses.
RECODEq40
(1,2=1)
(3,4=2)
(5,6=3)
(7,8=4)
(9,10=5)
(11,12,13,14,15,16=6)
(99=9)
intoq40bis.

183
VARIABLELABELSq40bis'Recodagedeq40pourrduirelenombredeclasses'.
VALUELABELSq40bis
1'Moinsde6000$'
2'6001$12000$'
3'12001$18000$'
4'18001$24000$'
5'24001$30000$'
6'Plusde30000$'
9'Jenesaispas'.
EXECUTE.
*Recodagedevariablespourlesrendredichotomiques.
recodeq10
(1thru3=1)
(4=2)
intopresenceBourseMerite.
VARIABLELABELSpresenceBourseMerite'Prsenced''aumoinsuneboursedemrite'.
VALUELABELSpresenceBourseMerite
1'Aumoinsunebourse'
2'Aucunebourse'.
RECODEq20a_m1
(1thru4=1)
(99=2)
intopresenceTravailInterne.
VARIABLE LABELS presenceTravailInterne 'Prsence ou absence d''un emploi

l''universit'.
VALUELABELSpresenceTravailInterne1'Emploiinterne'2'Pasd''emploiinterne'.
*DescriptiondestudiantsdeCSenfonctionducycled'tudes.
USEALL.
COMPUTEfilter_$=(q1b=1&q1a<>96).
VARIABLELABELSfilter_$'q1b=1(FILTER)'.
FILTERBYfilter_$.
EXECUTE.
CROSSTABS
/TABLES=q1dq1jq10q14a_m1q18q20a_m1q28BYcycleEtudesq8a
/CELLS=COLUMN

184
/COUNTROUNDCELL.
CROSSTABS
/TABLES=q1jpresenceBourseMeriteq14a_m1q18presenceTravailInterneq28BYq8a
/STATISTICS=PHI
/CELLS=COLUMN
/COUNTROUNDCELL.

En - Student Indebtedness

Uploaded by

Copyright:

Available Formats

En - Student Indebtedness

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

En - Student Indebtedness

Uploaded by

Copyright:

Available Formats

Fdration tudiante universitaire du Qubec

Undergraduate 11324 63% 11295$

5.Externalfactors Costsofstudying Hightuitionfeeslower

Student indebtedness : Comprehensive study

RECODE dureeBacc (1 = 1 ) (2=2) (3=3) (4=4) (5=5) (6 thru HI = 6) into

Student indebtedness : Comprehensive study

VARIABLE LABELS presenceCF 'Prsence ou absence d''une contribution familiale.

VARIABLE LABELS presenceBourses 'Prsence ou absence d''une bourse de mrite ou

VARIABLE LABELS presenceTravail 'Prsence ou absence d''un emploi rmunr pour la

* nbSrcFinancement : Compilation du nombre de sources de financement. Addition de

* Travail rmunr en cours d'tude : nombre d'heures travailles pour la session

* Variable qui dtermine si l'tudiant travaillait ou non l'automne 2009. Une

* Dpenses selon le primtre de sources et modes. On compte donc : frais de

* Cration de annee, qui est un recodage de q8b et estime l'anne d'tudes de

* Prparation de donnes pour les tests : dclarations de nouvelles valeurs

VARIABLE LABELS k53 'k53 : Dpenses annuelles en matire de frais de scolarit et

* La pondration est une pondration sur marge en fonction du sexe, du rgime

* Slection de l'chantillon: on ne traite que les tudiants temps plein. La

* Dpenses. Deux dpenses font l'objet de traitements particuliers (k53 et k49 -

FREQUENCIES VARIABLES=depensesTotalesCasiers k54Casiers k50xCasiers k56axCasiers

FREQUENCIES VARIABLES=q43aTaux q43bcTaux q43dTaux q43Taux q43aCasiers q43bcCasiers

MEANS TABLES=q43aBIS q43bcBIS q43dBIS q43BIS BY sexe AgeStrat q78quartile residence

MEANS TABLES=q43aBIS q43bcBIS q43dBIS q43BIS BY depensesTotalesCasiers k54Casiers

* Quelques corrlations entre la rentabilit du diplme et des caractristiques

* Exclusion temporaire des tudiants non-Qubcois pour les fins du traitement

VARIABLE LABELS presenceTravailInterne 'Prsence ou absence d''un emploi

You might also like