NIH Public Access
Author Manuscript
J Adolesc. Author manuscript; available in PMC 2014 August 01.
NIH-PA Author Manuscript
Published in final edited form as:
J Adolesc. 2013 August ; 36(4): 767–776. doi:10.1016/j.adolescence.2013.03.015.
Prevalence and Correlates of Truancy in the US: Results from a
National Sample
Michael G. Vaughn, Ph.D.a,*, Brandy Maynard, Ph.D.b, Christopher Salas-Wright, Ph.D.c,
Brian E. Perron, Ph.D.d, and Arnelyn Abdon, M.A.e
aSchool of Social Work and School of Public Health, Saint Louis University, St. Louis, MO United
States
bMeadows
Center for Preventing Educational Risk,’ College of Education, University of Texas at
Austin, Austin, Texas, United States
cGraduate
School of Social Work, Boston College, Boston, MA United States
NIH-PA Author Manuscript
dSchool
of Social Work, University of Michigan, Ann Arbor, MI United States
eSchool
of Economics, University of the Philippines, Manila, Philippines
Abstract
Truancy has been a persistent problem in the United States for more than 100 years. Although
truancy is commonly reported as a risk factor for substance use, delinquency, dropout, and a host
of other negative outcomes for youth, there has been surprisingly little empirical investigation into
understanding the causes and correlates of truancy using large, nationally representative samples.
Using the adolescent sample (N = 17,482) of the 2009 National Survey on Drug Use and Health
(NSDUH), this study presents the prevalence of truancy and examines individual, school
engagement, parental, and behavioral correlates of truancy. Overall, 11% of adolescents between
the ages of 12–17 reported skipping school in the past 30 days. Results from multinomial logistic
regression models indicate skipping school was robustly associated with an increased probability
of reporting externalizing behaviors, less parental involvement, and engagement and lower grades
in school. Implications for theory, prevention, and policy are discussed.
Keywords
NIH-PA Author Manuscript
truancy; externalizing behaviors; substance use; school dropout
© 2013 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.
*
Corresponding author, Michael G. Vaughn, Tegeler Hall, 3550 Lindell Boulevard, St. Louis, MO 63103 mvaughn9@slu.edu,
314-977-2718, Fax: 314-977-2731.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our
customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of
the resulting proof before it is published in its final citable form. Please note that during the production process errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of Interest Statement
The authors report no conflicts of interest.
Vaughn et al.
Page 2
Prevalence and Correlates of Truancy in the U.S.: Results from a National
Sample
NIH-PA Author Manuscript
Truancy, a persistent problem in the United States for more than 100 years, is associated
with a host of life-course problems (Attwood & Croll, 2006; Garry, 1996). Compared to
most developed nations, the United States fares poorly with respect to tolerating a relatively
high level of truancy and school dropout rate (Willms, 2003). Despite significant efforts and
millions of dollars spent by schools, communities, states, and the federal government to
reduce truancy over the past 20 years, there is little evidence that any positive impact has
been made on school attendance (Attwood & Croll, 2006;Davies & Lee, 2006; National
Center for Education Statistics, 2006; Sheppard, 2007; Stahl, 2008).
NIH-PA Author Manuscript
Although truancy is one of the major issues facing schools and the education of youth in the
United States (Heaviside, Rowand, Williams, & Farris, 1998), estimating the prevalence of
truancy has been fraught with challenges. Despite federal requirements for states to report
truancy, definitions of truancy and the reporting standards are not uniform across states. Due
to this lack of uniformity, calculating a national rate of truancy by aggregating state level
data is, at best, problematic (National Center for School Engagement, 2006). Several large
inner-city schools systems report thousands of unexcused absences each day while some
estimate hundreds of thousands of youth being absent from school on a regular basis (Baker,
Sigmon, & Nugent, 2001). Henry (2007), who examined the prevalence and correlates of
skipping school among 8th and 10th grade youth using data from the 2003 wave of the
Monitoring the Future study, found that nearly 11% of 8th graders and 16% of 10th graders
reported recent truancy. Data from other non-peer reviewed sources indicate a wide range of
truancy prevalence. For example, the National Comorbidity Survey (Adolescent
Supplement) interviewed 9,244 students across the country and asked students questions on
truant-related behaviors. From this self-report data, 27.04% of adolescents reported that they
have ever played hookey or skipped a whole day of school, with adolescents skipping on
average 3.78 days of school during the month that they skipped school the most (Kessler,
2001–2004). Another national survey, the School Crime Supplement to the National Crime
Victimization Survey, also asked students in the sample about skipping. From this survey,
the prevalence of skipping in the four weeks prior to the survey was 5.5% for students
between the ages of 12 and 18 (United States Department of Justice, 2007).
NIH-PA Author Manuscript
While current prevalence estimates lack the accuracy needed to determine the specific
magnitude of the problem, there is substantial evidence that truancy is linked to serious
consequences. Studies have found that students who are chronically absent from school are
more likely to drop out of school and less likely to be employed 6 months after the end of
compulsory schooling, which in turn negatively impacts their earning potential over their
lifetimes (Attwood & Croll, 2006; Garry, 1996). Truancy has also been associated with a
variety of risk behaviors that can negatively impact the development and wellbeing of truant
youth. Prior studies have linked truancy to negative outcomes such as the use of tobacco,
alcohol, and other drugs; delinquency and crime; poor academic performance; and school
expulsion (Best, Manning, Gossop, Gross, & Strang, 2006; Dynarski & Gleason, 1999;
Henry, 2010; Henry & Huizinga, 2007; Lochner & Moretti, 2004; Loeber & Farrington,
2000; Perez, Ariza, Sanchez-Martinez, & Nebot, 2010). The associations between truancy
and delinquency and substance use suggest that truancy can best be conceptualized as part of
the externalizing spectrum (e.g., Krueger et al., 2002; Krueger, Markon, Patrick, & Iacono,
2005; Vaughn et al., 2011).
Notwithstanding the extant research, there has been little attention given to the examination
of truancy as a focal problem using nationally representative samples. Much of what is
known about truancy is derived from studies examining the consequences and costs of
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 3
NIH-PA Author Manuscript
truancy or studies examining other problematic behaviors. Studies of truancy are plagued by
small and/or non-probability convenience samples often comprised of students from urban
and disadvantaged areas, or the studies have employed qualitative designs. Few studies have
identified truancy rates and correlates using large, nationally representative samples.
Conceptual Underpinnings
NIH-PA Author Manuscript
The problem of truancy is increasingly recognized as a developmentally complex and
heterogeneous problem that can be influenced by a number of factors in multiple domains
including school, family and individual domains (Kearney, 2008; Kim & Streeter, 2006). As
such, this study is guided by a developmental-ecological framework that views truancy as an
outcome influenced by dispositional and contextual factors across multiple domains that
adolescents traverse. Within this overarching framework, truancy is theorized in two major
ways: as an externalizing behavior closely corresponding to delinquency and as an indicator
of low school engagement (i.e., disengagement). Although more recent research is pointing
to a reciprocal relationship between engagement and delinquency (Hirschfield & Gasper,
2011), it is unclear whether truancy is better theorized as low school engagement, or if
truancy is indeed more aptly conceived within the externalizing continuum, in which truancy
is just one of several other problem behaviors comprising a syndrome of externalizing
problem behavior in adolescence that often persists into adulthood (Donovan, Jessor &
Costa, 1998; Jessor, 1991; Krueger et al., 2002, 2007). Therefore, this study aims to examine
truancy from a dual largely intertwined framework that considers truancy within an
overlapping engagement perspective and externalizing spectrum in adolescence.
Present Study Purpose
Understanding the correlates of truancy is important to the development of prevention and
intervention strategies. Although numerous prevention and intervention efforts are in
operation across the United States, they have done little to impact truancy (Maynard,
McRea, Pigott, & Kelly, 2012). This study improves upon and expands the current
knowledge base on truancy by examining correlates of truancy in multiple domains from an
engagement and problem behavior theory/externalizing behavior framework while
controlling for key confounding variables; exploring differences between students who
report no skipping, some skipping, and high rates of skipping; and utilizing a large,
nationally representative sample to provide a broader, more comprehensive and
generalizable view of truancy in the United States.
NIH-PA Author Manuscript
Specifically, this study considers five research questions: 1) What is the prevalence of
truancy? 2) What are the sociodemographic and mental health correlates of truancy? 2)
What associations does school engagement have on skipping school? 3) To what extent are
youth who skip school less likely to have a parent involved in their lives and in what
aspects? and 4) To what extent does the externalizing spectrum of behaviors increase the
likelihood of skipping school? We also explore the relative associations among youth who
reported higher rates of truancy (4 or more days in the prior 30) compared to moderate rates
of truancy (1–3 days in the prior 30). Our overarching hypothesis is that truancy is part of
the externalizing spectrum of behavior and, as such, correlates with other externalizing
behaviors will have the strongest effects even after controlling for the confounding effects of
age, gender, race/ethnicity, family income, and internalizing behavior (lifetime anxiety and
depression).
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 4
Methods
Sample and Procedures
NIH-PA Author Manuscript
This study is based on data from the 2009 National Survey on Drug Use and Health
(NSDUH) (SAMHSA, 2009). NSDUH is designed to provide population estimates of
substance use and health-related behaviors in the U.S. general population. It utilizes
multistage area probability sampling methods to select a representative sample of the U.S.
civilian, non-institutionalized population aged 12 years or older for participation in the
study. Study participants include household residents, residents of shelters, rooming houses,
and group homes, residents of Alaska and Hawaii, and civilians residing on military bases.
NIH-PA Author Manuscript
NSDUH study participants were interviewed in private at their places of residence. Potential
participants were assured that their names would not be recorded and that their responses
would be kept strictly confidential. Participants were paid 30 dollars for their participation.
All field interviewers signed a confidentiality agreement, and the procedures and protections
were carefully explained to potential participants in the informed consent protocol. The
NSDUH interview utilized a computer-assisted interviewing (CAI) methodology to increase
the likelihood of valid respondent reports of illicit drug use behaviors (SAMHSA, 2009).
The CAI methodology includes a combination of computer-assisted personal interviewing
(CAPI) and audio computer-assisted self-interviewing (ACASI) methodologies. ACASI is
designed to provide the respondent with a highly private and confidential means of
responding to questions and is used for questions of a sensitive nature.
A total of 68,736 respondents aged 12 years or older completed the 2009 survey. The study
analytic sample was confined to youth age 12–17 (N = 18,819). Weighted response rates
were 89% for household screening and 74.4% for interviewing (SAMHSA, 2009). Each
independent, cross-sectional NSDUH sample was considered representative of the U.S.
general population. NSDUH design and data collection procedures have been reported in
detail elsewhere (SAMHSA, 2009).
The mean age of the study sample is 14.6 years old (SD = 1.7). The respondents are evenly
distributed between males (51.0%) and females (49.0%) but are unevenly distributed in
terms of race/ethnicity. More than half of the respondents are White (58.7%), 17.7% are
Hispanic, and 14.0% are African American. The annual family income of 15.8% of the
sample is less than $20,000; 32.6% have income between $20,000 and $49,999; 19.4% have
income between $50,000 and $74,999; and 32.2% have more than $75,000 annual family
income.
NIH-PA Author Manuscript
Measurement
Skipping school—The NSDUH survey queried youth on how many days they missed
school from skipping in the past 30 days. We subsequently coded this variable into nonschool skipping (0 days), moderate skipping (1–3 days), and high levels of school skipping
(4+ days).
Sociodemographic and mental health covariates—The following demographic
variables were used: Age, gender, race/ethnicity (non-Hispanic white, non-Hispanic black,
Hispanic, and other [American Indian or Alaska Native, Asian, other Pacific Islander or
Native Hawaiian, and persons reporting more than one race]), education level, father in the
home (0 = no, 1 = yes), incarcerated in the past year (0 = no, 1 = yes), worked in the past
year (0 = no, 1 = yes), and total annual family income (less than $20,000; $20,000 to
$49,999; $50,000 to $74,999; and $75,000 or more). Family income was ascertained by
asking respondents, “Of these income groups, which category best represents your total
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 5
NIH-PA Author Manuscript
combined family income during the previous calendar year?” Because adolescents are often
unable to provide accurate estimates about family household income, responses from an
adult or other household member were provided. Additionally, we also examined lifetime
history of depression and anxiety. This was based on whether respondents were told by a
doctor or medical professional that they had either of these disorders.
School engagement—Seven variables were used to assess various aspects of school
engagement. Two items included past semester grades (A or B, C, D or lower) and the
number of school based activities attended (none, 1 or 2, 3 or more). Five items queried
youth about their feelings toward school. These included such questions as “How interesting
are courses at school?”; “How often you felt school work was meaningful?”; and “How
often did a teacher let you know you were doing a good job?” These five questions
originally had a response format of always, sometimes, seldom, and never. These were
subsequently dichotomized into always/sometimes and seldom/never to enhance
interpretability.
NIH-PA Author Manuscript
Parental involvement—Five items were used to assess various forms of parental
involvement. Like the school engagement response format, this original response format of
always, sometimes, seldom, and never was also dichotomized into always/sometimes and
seldom/never to enhance interpretability. Sample items included “During the past 12
months, how often did your parents provide help with your homework when you needed
it?”; “During the past 12 months, how often did your parents limit the amount of time you
went out with friends on school nights?”; and “During the past 12 months, how often did
your parents provided positive reinforcement, such as telling you they were proud of you for
something you had done?”
Externalizing behavior—A host of deviant behavior and risk-related variables, including
delinquent behaviors and substance use were used. Delinquent variables were self-reported
past-year selling of illegal drugs, stealing an item worth 50 dollars or more, attacking
someone with the intent to injure, serious fighting at school or work, and carrying a
handgun. These were measured dichotomously (i.e., yes or no), with the exception of serious
fighting at school or work for which four categories of frequencies were utilized. Substance
use variables assessed were self-reported past-year use of alcohol, marijuana, and illicit
drugs (hallucinogens, cocaine/crack, ecstasy, and heroin). These were also measured
dichotomously as use and non-use. Two dichotomously coded items were used to assess risk
propensity: “How often do you get a real kick out of doing things that are a little
dangerous?” and “How often do you like to test yourself by doing something a little risky?”
NIH-PA Author Manuscript
Data analysis—Weighted prevalence estimates and standard errors were computed using
Stata 11SE (StataCorp, 2009). This system implements a Taylor series linearization to adjust
standard errors of estimates for complex survey sampling design effects including those
found in clustered data. A series of multinomial logistic regression analyses were conducted
to assess the associations between categories of skipping school (moderate and high levels)
and demographic, school engagement, delinquency and risky behavior, and parental
involvement variables. Youth who did not skip school served as the reference group for all
unadjusted and adjusted odds ratios. We chose to use individual variables from the school
engagement, externalizing behaviors, and parental involvement domains. We could have
indexed these items as a form of data reduction but given that we had adequate statistical
power this approach would provide less information about the differential associations
across these domains. Final adjusted models controlled for the influences of age, gender,
race/ethnicity, family income, and mental health (lifetime anxiety and depression). Adjusted
odds ratios (AORs) and 95% confidence intervals (CIs) are presented to reflect association
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 6
strength. AORs were considered statistically significant only if associated confidence
intervals did not include the value 1.0.
NIH-PA Author Manuscript
Results
What is the prevalence and the sociodemographic and mental health correlates of
skipping school?
The prevalence of past month skipping school was 11%, with 9% reporting having skipped
1–3 days and 2% reporting having skipped 4 or more days of school. Table 1 shows the
sociodemographic and mental health characteristics of youth who reported moderate and
high levels of skipping school in the past 30 days. Following adjustments for confounding,
youth reporting a moderate and high level of skipping school were significantly more likely
to be older (31 and 35% respectively) than youth not reporting skipping school. Only high
level school truants were less likely to report an annual family income of $75,000 or more.
With respect to not having a father in the household only the moderate group was
significant. For mental health variables, high level school skippers were nearly 3 times more
likely to report depression than non-school skippers and approximately two and one-half
times more likely to report anxiety, whereas moderate truants were just significantly more
likely to report anxiety.
NIH-PA Author Manuscript
What associations does school engagement have on skipping school?
Table 2 compares the correlations of various indicators of school engagement of moderate
and high level school skipping groups. Consistent effects were found with both groups being
significantly less engaged in school and more likely to receive lower grades. The odds ratios
were comparable for both the moderate and high levels groups with respect to engagement
indicators except participating in 3 or more school activities. However, high level truants
were far more likely to receive C grades and D grades or lower. Overall, both groups were
significantly less likely to like or kind of like school.
To what extent are youth who skip school less likely to have a parent involved in their
lives and in what aspects?
Table 3 compares the associations between various forms of parental involvement to
moderate and high levels of skipping school. Results of AORs indicate a uniform pattern
such that youth who reported skipping school were significantly less likely to have a parent
involved. AORs ranged from 0.40 to 0.75. Although high level school skippers generally
had less parental involvement than moderate school skippers, the differences were
negligible.
NIH-PA Author Manuscript
To what extent does the externalizing spectrum of behaviors increase the likelihood of
skipping school?
The models depicted in Table 4 examine externalizing behaviors. Several variables were
found to have large AORs. However, effects were uniformly larger for high level truants
compared to moderate truants including serious fighting at school or work 6 or more times,
carrying a handgun, selling illegal drugs, frequent stealing, attacking someone, testing
oneself by doing risky things, doing dangerous things, using alcohol, using marijuana, and
any other illicit drug use.
Sensitivity Analysis
In order to assess the sensitivity of our findings across the developmental period of 12 to 17
years of age we reanalyzed our data for 12–14 year olds and 15–17 year olds separately.
Examining the associations between moderate and severe truancy and its correlates among
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 7
NIH-PA Author Manuscript
adolescents between the ages of 12–14 and 15–17 year olds suggested no significant
differences across these two developmental time periods in as much as all confidence
intervals for the two developmental subgroups were overlapping and directionality was the
same.
Discussion
This study examines the prevalence and correlates of truancy using a large, nationally
representative sample and provides a more accurate and generalizable picture of truancy
than was previously available. The findings of the present study clearly demonstrate that
truancy is a significant problem in the United States. Of the 17,482 youth between the ages
of 12–17 years who completed the NSDUH survey, 11% reported skipping school within the
30 days prior to completing the survey. Henry (2007) also found a prevalence of 11% of
truancy using data from the 2003 Monitoring the Future Study with a similarly constructed
question. With an estimated 17.2 million students enrolled in grades 9–12 (U.S. Census
Bureau, 2011), approximately 2 million students skip school at least once in a given month.
NIH-PA Author Manuscript
The present study uniquely extends the knowledge base on truant youth by examining
correlates of truancy in multiple domains; exploring differences between youth who report
moderate and high rates of truancy; and providing a broader, more comprehensive and
generalizable understanding of truancy and truant youth in the United States. The findings
suggest that truant youth are multi-problem youth who exhibit risk factors in multiple
domains. Truant youth are more likely to be older and report substance use, externalizing
and internalizing behavior problems, lower school engagement, less parental involvement,
and lower grades than their non-truant counterparts. Although we anticipated that AfricanAmerican and Hispanic youth would have a higher likelihood of reporting truant behavior
relative to White youth based on prior research and strong correlations between truancy,
poverty status and single-parenthood, the results of this study found that while Hispanic
youth were more likely to report truancy, African-American youth were not. We speculate
that African-American adolescents, who are at the highest risk for truancy and dropping out
of school, may also be more likely to be absent from school for longer periods of time and
therefore were not surveyed.
NIH-PA Author Manuscript
The results of this study support our hypothesis that youth who skip school more frequently
are at higher risk and exhibit more serious externalizing behaviors and substance use than
youth who skip school less frequently. Our exploratory analysis revealed that youth who
reported higher rates of truancy were 1.5 to 2 times more likely than moderately frequent
truants to report alcohol and drug use, serious fighting at school, carrying a handgun, selling
illegal drugs, stealing/trying to steal 3 or more times, and attacking with intent to seriously
harm. Moreover, our findings suggest a consistent pattern of differences between the types
and severity of externalizing behaviors. It is apparent from an examination of Table 5 that
high frequency truants were much more likely to report engaging in more severe
externalizing behaviors than truants reporting less skipping. As these results indicate, the
two groups of truants varied in important ways across the substance use and externalizing
behavior spectrum. However, we should also point out that the largest effect for highly
truant youth were poor academic grades. Although poor academic grades can be viewed as a
sign of disengaging from school, it also can be seen as intertwined with externalizing
behavior as prior studies indicate that delinquent adolescents also are likely to receive poor
grades (Hinshaw, 1992).
Despite stronger correlates found for highly truant youth compared to moderately truant
youth, this does not mean that intervention resources should solely be directed for this
severe group; it seems probable that moderately truant youth may become more severe in
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 8
NIH-PA Author Manuscript
their frequency of skipping school over time. It could be argued that prevention resources
may be best utilized with moderate skippers given the possibility that highly truant youth are
a smaller proportion of the population and may be more difficult to reach.
Study findings suggest that frequency of skipping school has implications for youth both in
terms of the likelihood of engaging in problematic risk behaviors, as well as the pattern of
behaviors in which they engage. Because most prior studies do not examine the effects of
frequency of skipping on outcomes, differences between youth who report different rates of
truancy has not been adequately examined in prior research. This analysis extends the
literature on truant youth by demonstrating that frequency of truancy has important
implications. Future research must more closely examine frequency of skipping as a
moderator of outcomes for truant youth to better guide policy and practice in this area. In
addition, future longitudinal studies should also attempt to disentangle the relationship
between externalizing behaviors and the processes of disengaging from school.
NIH-PA Author Manuscript
In addition to variation between moderate and high frequency skippers on substance use and
externalizing behaviors, we also observed variation in reports of lifetime anxiety and
depression between the moderate and high skippers. Interestingly, youth who reported
higher rates of skipping were 3 times more likely to report depression and 2.5 times more
likely to report anxiety than moderate truants, whereas moderate skippers were just
significantly more likely to report anxiety. These findings point to differential mental health
risks and needs of youth who exhibit more chronic patterns of truancy.
NIH-PA Author Manuscript
The results of the present analysis demonstrate that while truancy is found to be correlated
with engagement factors, the most salient correlates of truancy are substance use and
externalizing behaviors. It is obvious from our analyses that truant youth are not only
significantly more likely to report engaging in substance use and deviant behavior than nontruant youth but that more frequent skippers exhibited consistent patterns of higher
probability and greater severity across multiple categories of problematic risk behaviors
compared to truant youth who reported moderate frequency of skipping. While there a
several competing theoretical perspectives on truancy, the present investigation suggests that
truancy is part and parcel of an externalizing behavior spectrum. These results are consistent
with prior research that report a positive association between truancy and delinquency, drug
use, aggression, and an overall propensity toward risk (Henry, 2007; Henry, 2010; Hallfors,
Vevea, Iritani, Cho, Khatapoush, & Saxe, 2002; Maynard, Vaughn, Salas-Wright & Peters,
2012). Recent research on the externalizing spectrum indicates a common liability that
manifests early, is partly heritable, and is comorbid with a broad array of not only risk
behaviors but also psychiatric problems over the life-course (Dick et al., 2009; Krueger et
al., 2002, 2007; Markon & Krueger, 2005). These findings also lend support to Jessor’s and
Jessor’s (1977) Problem Behavior Theory in which problematic behaviors (i.e., drinking,
marijuana use, delinquent behavior and sexual intercourse) cluster to constitute a syndrome
of problem behavior in adolescence (Donovan & Jessor, 1985). The present results support
the notion of a syndrome of problem behaviors, with truancy being an additional problem
behavior highly correlated with other delinquent behaviors, indicating an underlying factor
or latent variable across a range of externalizing behaviors, including truancy.
The externalizing behavior perspective contrasts with other views on truancy and school
dropout phenomena, which cast this problem in the context of school disengagement. The
disengagement perspective places greater onus on the school and school attachment. From
this view, truancy and dropout are part of a process of disengaging that begins early as
children are not engaged or fully bonded with the school. Multiple academic, psychological,
cognitive, and behavioral factors are involved, but at the heart of the problem lay the lack of
connection with school. Prior research has often used truancy as a measure or predictor of
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 9
NIH-PA Author Manuscript
school engagement (Archambault, Janosz, Fallu, & Pagani, 2009), while others have found
that skipping school clustered much more strongly with measures of delinquency than
measures of engagement (Hirschfield and Gasper, 2011). Although truant youth in this study
were more likely to report lower school engagement than non-truant youth, effects of the
engagement indicators were weaker compared to substance use and externalizing behavior
indicators. These findings suggest that truancy conceptualizations should include the
externalizing spectrum of behavior rather than viewed merely as a school engagement
problem.
Toward a theory of truancy
NIH-PA Author Manuscript
We suggest that a theory of truancy reflects interplay between externalizing behavior and
school disengagement, each mutually reinforcing one another during childhood and
adolescence. Normative cognitive, behavioral, familial, and ecological challenges that
children face developmentally may give rise to, or are aggravated by, a tendency toward
externalizing. In turn, this amplifies the prospects for disengaging from school resulting in a
harmful spiral where stronger supports and scaffolds are badly needed. Although
externalizing and engagement perspectives on truancy and dropout overlap, we speculate
that the externalizing spectrum may occur prior to disengagement. Because our present
cross-sectional analysis does not permit us to disentangle temporal ordering of variables, this
speculation is based largely on evidence from chronic school skippers who demonstrate a
very high risk for juvenile delinquency, criminal behavior and serious psychiatric problems
reflecting a general externalizing. Prior research studies have indicated that externalizing
behaviors manifest early in childhood (Caspi, Moffitt, Newman, & Silva, 1996; Vaughn,
Beaver, DeLisi, & Wright, 2009; Vaughn, Perron, Beaver, DeLisi, & Wexler, 2010). Thus,
disengagement and dropout may ultimately be a consequence of the externalizing spectrum,
though at various points in time during adolescence one can show that engagement in school
predicts externalizing behaviors (Li et al., 2011). Other longitudinal research suggests that
delinquency and engagement affect each other, although the effects were not consistent
across all engagement domains (i.e., cognitive, emotional, and behavioral; Hirschfield &
Gasper, 2011).
Limitations and conclusions
NIH-PA Author Manuscript
Although the size, scope, and long-term stability of the NSDUH are impressive, there are
important limitations that should bear caution. First and foremost, study data were crosssectional and prevented not only an assessment of the temporal relationships between
variables, but also a temporal look at the unfolding of truancy risk. Further, little data was
available on truancy and its correlates outside of the self-report measurement paradigm.
Although there are some limitations of self-report data in terms of over or underreporting
problematic behaviors, Khatapoush & Hallfors (2000, cited in Hallfors et al, 2002) found
that students reporting of truancy more closely approximated truancy rates in school records,
thus giving us greater confidence in the prevalence estimates of truancy from this self-report
survey. Our approach could have underestimated the prevalence of school truancy, because
youth who have already dropped out or are chronically absent for large swaths of time (e.g.,
weeks or months), may not be included in the analysis. It would also have been useful to
distinguish between whether truancy prevalence is similar or different with respect to public
or private school attendance. We were not able to do so due to data constraints and lack of
detailed information about school experiences of youth. Another limitation is the lack of
data on such ecological characteristics as neighborhood and community disadvantage, which
may uniquely contribute to truancy via such mechanisms as fear or loosening of social
controls and supports around families. In line with this, one of the weaknesses of this dataset
is a lack of data on normative developmental challenges as it relates to truancy. Future
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 10
complimentary studies capable of assessing contextual and situational risk of identified
correlates are a natural extension of the present investigation.
NIH-PA Author Manuscript
Despite these limitations, this study provides an empirical examination of the prevalence and
correlates of truancy in a nationally representative sample. We find that truancy is robustly
associated with an increased probability of reporting externalizing behaviors, less parental
involvement, and engagement and lower grades in school suggesting the importance of the
interplay between externalizing and disengaging from school.
Acknowledgments
The authors are grateful for support from the Meadows Center for Preventing Educational Risk, the Greater Texas
Foundation, the Institute on Educational Sciences grants (R324A100022 & R324B080008) and from the Eunice
Kennedy Shriver National Institute of Child Health and Human Development (P50 HD052117). The content is
solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy
Shriver National Institute Of Child Health and Human Development or the National Institutes of Health.
References
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Archambault I, Janosz M, Fallu JS, Pagani LS. Student engagement and its relationship with early high
school dropout. Journal of Adolescence. 2009; 32:651–670. [PubMed: 18708246]
Attwood G, Croll P. Truancy in secondary school pupils: Prevalence, trajectories and pupil
perspectives. Research Papers in Education. 2006; 21:467–484.
Best D, Manning V, Gossop M, Gross S, Strang J. Excessive drinking and other problem behaviors
among 14–16 year old schoolchildren. Addictive Behaviors. 2006; 31:1424–1435. [PubMed:
16442742]
Caspi A, Moffitt TE, Newman DL, Silva PA. Behavioral observations at age 3 years predict adult
psychiatric disorders: Longitudinal evidence from a birth cohort. Archives of General Psychiatry.
1996; 53:1033–1039. [PubMed: 8911226]
Cohen, J. Statistical power analysis for the behavioral sciences. 2nd ed.. Hillsdale, NJ: Lawrence
Earlbaum Associates; 1988.
Davies JD, Lee J. To attend or not to attend? Why some students choose school and others reject it.
Support for Learning. 2006; 21:204–209.
Dick DM, Latendresse SJ, Lansford JE, Budde JP, Goate A, Dodge KA, Pettit GS, Bates JE. The role
of GABRA2 in trajectories of externalizing behavior across development and evidence of
moderation by parental monitoring. Archives of General Psychiatry. 2009; 66:649–657. [PubMed:
19487630]
Donovan JE, Jessor R. Structure of problem behavior in adolescence and young adulthood. Journal of
Consulting and Clinical Psychology. 1985; 53:890–904. [PubMed: 4086689]
Donovan JE, Jessor R, Costa FM. Adolescent health behavior and conventionality-unconventionality:
An extension of problem-behavior theory. Health Psychology. 1998; 10:52–61. [PubMed: 2026131]
Dynarski, M.; Gleason, P. How can we help? Lessons from federal dropout prevention programs.
Princeton, NJ: Mathematica Policy Research; 1999.
Garry EM. Truancy: First step to a lifetime of problems. OJJDP: Juvenile Justice Bulletin, October.
1996:1–7.
Hallfors D, Vevea JL, Iritani B, Cho H, Khatapoush S, Saxe L. Truancy, grade point average, and
sexual activity: A meta-analysis of risk indicators for youth substance use. Journal of School
Health. 2002; 72:205–211. [PubMed: 12109176]
Heaviside, S.; Rowand, C.; Williams, C.; Farris, E. Violence and discipline problems in the U.S.
public schools 1996–1997. Washington, DC: U.S. Department of Education, Office of Educational
Research and Improvement, National Center for Educational Statistics; 1998.
Henry KL. Who’s skipping school: Characteristics of truants in 8th and 10th grade. Journal of School
Health. 2007; 77:29–35. [PubMed: 17212757]
Henry KL. Skipping school and using drugs: A brief report. Drugs: Education, Prevention, and Policy.
2010; 17:650–657.
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 11
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Henry KL, Huizinga DH. School related risk and protective factors associated with truancy among
urban youth placed at risk. Journal of Primary Prevention. 2007; 28:505–519. [PubMed:
18004658]
Hirschfield PJ, Gasper J. The relationship between school engagement and delinquency in late
childhood and early adolescence. Journal of Youth and Adolescence. 2011; 40:3–22. [PubMed:
20706779]
Hinshaw SP. Externalizing behavior problems and academic underachievement in childhood and
adolescence: Causal relationships and underlying mechanisms. Psychological Bulletin. 1992;
111:127–155. [PubMed: 1539086]
Jessor, R.; Jessor, SL. Problem behavior and psychosocial development: A longitudinal study of youth.
New York: Academic Press; 1977.
Jessor R. Risk behavior in adolescence: A pychosocial framework for understanding and action.
Journal of Adolescent Health. 1991; 12:597–605. [PubMed: 1799569]
Kearney CA. School absenteeism and school refusal behavior in youth: A contemporary review.
Clinical Psychology Review. 2008; 28:451–471. [PubMed: 17720288]
Kessler, RC. National Comorbidity Survey: Adolescent Supplement (NCS-A) 2001–2004 [Data file].
2001–2004. Retrieved from http://www.icpsr.umich.edu/icpsrweb/ICPSR/studies/28581/
documentation.
Khatapoush, S.; Hallfors, D. Active parental consent and sample bias in a school-based AOD survey.
Poster presented at the Addictions 2000 Conference; September 22–23, 2000; Cape Cod, Mass.
2000.
Kim, JS.; Streeter, CL. Increasing school attendance: Effective strategies and interventions. In:
Franklin, C.; Harris, MB.; Allen-Meares, P., editors. The school services sourcebook: A guide for
school-based professionals. New York: Oxford University Press; 2006. p. 397-404.
Krueger RF, Hicks BM, Patrick CJ, Carlson SR, Iacono WG, McGue M. Etiologic connections among
substance dependence, antisocial behavior, and personality: Modeling the externalizing spectrum.
Journal of Abnormal Psychology. 2002; 111:411–424. [PubMed: 12150417]
Krueger RF, Markon KE, Patrick CJ, Iacono WG. Externalizing psychopathology in adulthood: A
dimensional spectrum conceptualization and its implications for DSM-V. Journal of Abnormal
Psychology. 2005; 114:537–550. [PubMed: 16351376]
Krueger RF, Markon KE, Patrick CJ, Benning SD, Kramer M. Linking antisocial behavior, substance,
use, and personality: An integrative quantitative model of the adult externalizing spectrum. Journal
of Abnormal Psychology. 2007; 116:645–666. [PubMed: 18020714]
Li Y, Zhang W, Liu J, Arbeit MR, Schwartz SJ, Bowers EP, Lerner RM. The role of school
engagement in preventing adolescent delinquency and substance use: A survival analysis. Journal
of Adolescence. 2011; 34:1181–1192. [PubMed: 22118510]
Lochner L, Moretti E. The effect of education on crime: Evidence from prison inmates, arrests, and
self-reports. The American Economic Review. 2004; 94:155–189.
Loeber R, Farrington D. Young children who commit crime: Epidemiology, developmental origins,
risk factors, early interventions, and policy recommendations. Development and Psychopathology.
2000; 12:737–762. [PubMed: 11202042]
Markon KE, Krueger RF. Categorical and continuous models of liability to externalizing disorders: A
direct comparison in NESARC. Archives of General Psychiatry. 2005; 62:1352–1359. [PubMed:
16330723]
Maynard BR, McCrea KT, Pigott TD, Kelly MS. Indicated truancy interventions for chronic truant
students: A Campbell systematic review. Research on Social Work Practice. 2012 Advance online
publication.
Maynard BR, Salas-Wright CP, Vaughn MG, Peters KE. Who are truant youth? Examining distinctive
profiles of truant youth using latent profile analysis. Journal of Youth and Adolescence. 2012
Advance online publication.
National Center for Education Statistics. The condition of education 2006. Washington, DC: U.S.
Department of Education; 2006. Retrieved from http://nces.ed.gov/programs/coe/indicator_mis.asp
Perez A, Ariza C, Sanchez-Martinez F, Nebot M. Cannabis consumption initiation among adolescents:
A longitudinal study. Addictive Behavior. 2010; 35:129–134.
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 12
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Sheppard A. An approach to understanding school attendance difficulties: Pupils’ perceptions of
parental behavior in response to their requests to be absent from school. Emotional and
Behavioural Difficulties. 2007; 12:349–363.
Stahl AL. Petitioned status offense cases in juvenile courts, 2004. OJJDP Fact Sheet. 2008; 2:1–2.
StataCorp. Stata statistical software: Release 11. College Station, TX: StataCorp LP; 2009.
Substance Abuse and Mental Health Services Administration, Office of Applied Studies. Results from
the 2008 National Survey on Drug Use and Health: National Findings. Rockville, MD: Substance
Abuse and Mental Health Services Administration—2009; 2009.
U.S. Census Bureau. [Accessed January 10 2012] School enrollment in the United States: 2008. 2011.
Available at http://www.census.gov/prod/2011pubs/p20-564.pdf.
United States Department of Justice, Bureau of Justice Statistics. School Crime Supplement (SCS) to
the National Crime Victimization Survey (NCVS) [Data file]. 2007. Available from http://
nces.ed.gov/surveys/ssocs/index.asp
Vaughn MG, Beaver KM, DeLisi M, Wright JP. Identifying latent classes of behavioral risk based on
early childhood manifestations of self-control. Youth Violence and Juvenile Justice. 2009; 7:16–
31.
Vaughn MG, DeLisi M, Gunter T, Fu Q, Beaver KM, Perron BE, Howard MO. The severe 5%: A
latent class analysis of the externalizing spectrum in the United States. Journal of Criminal Justice.
2011; 39:75–80. [PubMed: 22942480]
Vaughn, MG.; Perron, BE.; Beaver, KM.; DeLisi, M.; Wexler, J. Transactional pathways in the
development of externalizing behaviors in a sample of kindergarten children with impaired selfcontrol. In: Massari, VG., editor. Control Theory and Its Applications. New York: Nova Science
Publishers; 2010.
Willms, JD. Student engagement at school: A sense of belonging and participation. Paris: Organisation
for Economic and Co-operation and Development; 2003.
NIH-PA Author Manuscript
J Adolesc. Author manuscript; available in PMC 2014 August 01.
NIH-PA Author Manuscript
NIH-PA Author Manuscript
15.15
Age in years (mean)
9.21
Male (n=6,824)
8.91
12.18
7.79
Black (n=1,792)
Hispanic (n=2,279)
Other (n=1,331)
10.52
8.37
7.91
$20,000–$49,999 (n=4,201)
$50,000–$74,999 (n=2,603)
$75,000 or More (n=4,538)
12.52
No father in HH (n=3,576)
J Adolesc. Author manuscript; available in PMC 2014 August 01.
14.13
Yes (n=182)
13.74
Yes (n=1,166)
12.22
Yes (n=628)
Lifetime anxiety
9.18
No (n=12,428)
Lifetime depression
11.36
No (n=4,148)
Worked in past year
9.28
No (n=13,209)
Stay in jail/detention
8.30
Father in HH (n=9,850)
Father in household
11.63
Less than $20,000 (n=2,094)
Annual family income
8.69
White (n=8,034)
Race / Ethnicity
9.43
Female (n=6,612)
Sex
%
Demographic Characteristics
[9.25,15.97]
[8.52,9.89]
[11.10,16.89]
[10.15,12.69]
[9.34,20.82]
[8.63,9.97]
[11.12,14.07]
[7.59,9.06]
[6.93,9.01]
[7.08,9.87]
[9.30,11.89]
[9.90,13.62]
[5.89,10.25]
[10.41,14.19]
[7.35,10.76]
[7.92,9.52]
[8.31,10.19]
[8.54,10.40]
[15.03,15.27]
95% CI
Moderate
4.84
1.53
2.25
2.32
5.58
1.65
2.62
1.45
1.16
1.27
1.56
4.08
1.82
2.63
1.79
1.43
1.74
1.72
15.53
%
[3.11,7.46]
[1.28,1.83]
[1.40,3.60]
[1.78,3.00]
[2.48,12.07]
[1.40,1.96]
[2.04,3.37]
[1.17,1.79]
[0.84,1.60]
[0.79,2.02]
[1.16,2.08]
[3.06,5.42]
[1.02,3.23]
[1.85,3.72]
[1.19,2.69]
[1.15,1.77]
[1.39,2.18]
[1.35,2.19]
1.43
1.00
1.24
1.00
1.69
1.00
1.60
1.00
0.63
0.67
0.87
1.00
0.89
1.48
1.03
1.00
0.97
1.00
1.30
OR
[1.04, 1.98]
[0.94, 1.64]
[1.04, 2.73]
[1.36, 1.89]
[0.50, 0.79]
[0.52, 0.87]
[0.69, 1.09]
[0.65, 1.23]
[1.21, 1.81]
[0.82, 1.3]
[0.83, 1.14]
[1.23, 1.37]
95% CI
Moderate
Number of days skipped
[15.33,15.72]
95% CI
High
3.41
1.00
0.99
1.00
3.74
1.00
1.93
1.00
0.26
0.29
0.37
1.00
1.27
1.95
1.26
1.00
1.01
1.00
1.52
OR
[2.07, 5.60]
[0.57, 1.74]
[1.57, 8.91]
[1.37, 2.70]
[0.17, 0.41]
[0.17, 0.51]
[0.24, 0.56]
[0.68, 2.38]
[1.28, 2.96]
[0.79, 2.02]
[0.72, 1.41]
[1.38, 1.68]
95% CI
High
Demographic and Mental Health Characteristics of Moderate and High Truancy Youth in the U.S.
0.85
1.00
1.19
1.00
1.09
1.00
1.37
1.00
0.76
0.79
0.83
1.00
0.96
1.31
1.01
1.00
1.10
1.00
1.31
AOR
[0.51, 1.41]
[0.90, 1.58]
[0.56, 2.11]
[1.04, 1.81]
[0.53, 1.09]
[0.54, 1.16]
[0.60, 1.14]
[0.63, 1.48]
[0.97, 1.78]
[0.71, 1.43]
[0.87, 1.38]
[1.13, 1.53]
95% CI
Moderate
2.88
1.00
1.03
1.00
1.49
1.00
1.08
1.00
0.34
0.54
0.46
1.00
1.58
1.36
0.89
1.00
1.30
1.00
1.35
AOR
[1.54, 5.39]
[0.59, 1.79]
[0.47, 4.68]
[0.59, 1.96]
[0.14, 0.81]
[0.21, 1.36]
[0.25, 0.85]
[0.69, 3.58]
[0.64, 2.90]
[0.43, 1.85]
[0.79, 2.15]
[1.01, 1.8]
95% CI
High
NIH-PA Author Manuscript
Table 1
Vaughn et al.
Page 13
14.27
Yes (n=375)
[10.33,19.40]
[8.53,9.89]
95% CI
Moderate
4.98
1.59
%
[2.60,9.33]
[1.34,1.88]
95% CI
High
1.72
1.00
OR
[1.18, 2.51]
95% CI
Moderate
3.46
11.00
OR
[1.72, 6.97]
95% CI
High
1.97
1.00
[1.13, 3.44]
95% CI
Moderate
AOR
2.47
1.00
AOR
[0.98, 6.24]
95% CI
High
Note: Reference group: students who did not report missing any school. Adjusted models include age, sex, race/ethnicity, annual income, and lifetime depression and anxiety. Odds ratios in bold are
statistically significant.
9.19
No (n=12,681)
NIH-PA Author Manuscript
%
NIH-PA Author Manuscript
Demographic Characteristics
NIH-PA Author Manuscript
Number of days skipped
Vaughn et al.
Page 14
J Adolesc. Author manuscript; available in PMC 2014 August 01.
NIH-PA Author Manuscript
8.10
8.21
Always/sometimes (n=10,612)
8.71
Very/somewhat important (n=11,823)
8.17
Very/somewhat interesting (n=10,348)
9.83
6.98
1–2 (n=6,754)
3 or more (n=4,487)
J Adolesc. Author manuscript; available in PMC 2014 August 01.
8.61
Always/sometimes (n=10,654)
14.15
13.48
C (n=2,783)
D or lower (n=758)
[10.51,17.12]
[12.46,16.03]
[6.88,8.32]
[7.92,9.36]
[10.68,13.94]
[6.08,8.00]
[8.90,10.85]
[10.74,14.54]
[7.49,8.91]
[11.82,15.10]
[8.05,9.43]
[11.75,16.16]
[7.54,8.93]
[11.90,15.44]
[7.44,8.81]
[13.07,16.88]
95% CI
8.00
2.92
0.89
1.41
2.98
0.59
1.84
3.69
1.52
2.49
1.60
2.70
1.38
3.14
1.24
3.96
%
[5.55,11.41]
[2.25,3.77]
[0.68,1.16]
[1.16,1.71]
[2.21,4.02]
[0.37,0.92]
[1.47,2.30]
[2.79,4.87]
[1.23,1.87]
[1.93,3.22]
[1.32,1.93]
[1.99,3.67]
[1.12,1.69]
[2.42,4.07]
[1.00,1.53]
2.08
2.07
1.00
0.66
1.00
0.51
0.75
1.00
0.57
1.00
0.59
1.00
0.56
1.00
0.49
1.00
OR
[1.53, 2.81]
[1.73, 2.47]
[0.56, 0.79]
[0.40, 0.63]
[0.61, 0.91]
[0.48, 0.68]
[0.48, 0.72]
[0.47, 0.66]
[0.41, 0.58]
95% CI
Moderate
Number of days skipped
[3.08,5.08]
95% CI
High
10.53
3.63
1.00
0.45
1.00
0.14
0.47
1.00
0.57
1.00
0.55
1.00
0.40
1.00
0.28
1.00
[6.51, 17.02]
[2.48, 5.32]
[0.31, 0.64]
[0.08, 0.25]
[0.33, 0.69]
[0.40, 0.80]
[0.38, 0.80]
[0.29, 0.57]
[0.20, 0.39]
95% CI
High
1.83
1.83
1.00
0.69
1.00
0.57
0.80
1.00
0.58
1.00
0.62
1.00
0.58
1.00
0.50
1.00
AOR
[1.33, 2.52]
[1.51, 2.22]
[0.58, 0.83]
[0.45, 0.72]
[0.65, 0.99]
[0.49, 0.69]
[0.50, 0.77]
[0.48, 0.7]
[0.41, 0.60]
95% CI
Moderate
8.80
3.23
1.00
0.52
1.00
0.19
0.55
1.00
0.61
1.00
0.63
1.00
0.48
1.00
0.26
1.00
AOR
[5.31, 14.60]
[2.17, 4.81]
[0.36, 0.77]
[0.11, 0.34]
[0.37, 0.81]
[0.43, 0.87]
[0.42, 0.94]
[0.33, 0.69]
[0.18, 0.37]
95% CI
High
Note: Reference group: students who did not report missing any school. Adjusted models include age, sex, race/ethnicity, annual income, and lifetime depression and anxiety. Odds ratios in bold are
statistically significant.
7.56
A or B (n=9,370)
Grades for last semester
12.21
Seldom/never (n=2,754)
Teacher let youth know doing a good job
12.52
None (n=2,143)
No. of school based activities
13.37
Somewhat/very boring (n=3,064)
How interesting courses at school
13.81
Somewhat/very unimportant (n=1,585)
How important things learned
13.57
Seldom/never (n=2,790)
How often felt school work meaningful
14.87
Liked a lot/Kind of liked (n=10,949)
%
Didn’t like very much/hated (n=2,472)
How felt overall about going to school
School Engagement
Moderate
OR
NIH-PA Author Manuscript
Comparisons of School Engagement for Moderate and High Truancy Youth in the U.S.
NIH-PA Author Manuscript
Table 2
Vaughn et al.
Page 15
NIH-PA Author Manuscript
NIH-PA Author Manuscript
13.84
17.65
12.13
3–5 times (n=439)
6 or more times (n=218)
15.34
1 or more times (n=457)
24.10
1 or more times (n=462)
22.20
27.60
1 or 2 times (n=462)
3 or more times (n=183)
J Adolesc. Author manuscript; available in PMC 2014 August 01.
17.79
1 or more times (n=944)
12.52
Sometimes/always (n=4,515)
12.08
Sometimes/always (n=5,246)
6.13
16.51
No (n=9,227)
Yes (n=4,209)
Alcohol
7.61
Never/seldom (n=8,023)
Get a real kick out of doing dangerous things
7.74
Never/seldom (n=8,826)
Like to test yourself by doing risky things
8.71
None (n=12,457)
Youth attacked with intent to seriously harm
8.68
0 times (n=12,756)
Youth stole/tried to steal
8.85
None (n=12,948)
Youth sold illegal drugs
9.14
None (n=12,939)
Youth carried handgun
8.03
1–2 times (n=2,167)
%
0 times (n=10,557)
Serious fight at school/work
Externalizing Behaviors
[15.00, 18.13]
[5.53, 6.80]
[10.92, 13.33]
[6.87, 8.42]
[11.28, 13.88]
[7.01, 8.53]
[14.67, 21.42]
[8.07, 9.40]
[19.39, 37.67]
[17.35, 27.96]
[8.04, 9.36]
[19.03,30.02]
[8.21,9.53]
[11.34,20.42]
[8.49,9.83]
[7.40,19.26]
[13.44,22.82]
[11.99,15.93]
[7.36,8.76]
95% CI
Moderate
3.76
0.84
2.37
1.32
2.47
1.36
5.11
1.44
9.69
4.59
1.52
8.91
1.51
5.97
1.59
7.78
5.40
2.64
1.28
%
[3.07,4.60]
[0.64,1.10]
[1.90, 2.95]
[1.03, 1.69]
[1.96,3.12]
[1.08,1.72]
[3.43,7.56]
[1.21,1.73]
[5.22,17.28]
[2.91,7.17]
[1.26,1.82]
[6.11,12.81]
[1.26,1.81]
[3.72,9.44]
[1.34,1.89]
[3.94,14.78]
[3.21,8.94]
[1.91,3.64]
[1.03,1.58]
95% CI
High
3.14
1.00
1.69
1.00
1.73
1.00
2.38
1.00
4.56
3.14
1.00
3.64
1.00
1.90
1.00
1.71
2.59
1.87
1.00
OR
[2.68, 3.68]
[1.44, 1.98]
[1.43, 2.08]
[1.86, 3.04]
[2.83, 7.34]
[2.28, 4.33]
[2.65, 5]
[1.33, 2.73]
[0.98, 2.99]
[1.85, 3.63]
[1.55, 2.27]
95% CI
Moderate
5.24
1.00
1.92
1.00
1.94
1.00
4.13
1.00
9.15
3.71
1.00
7.90
1.00
4.25
1.00
6.89
4.97
2.24
1.00
OR
Number of days skipped
[3.70, 7.41]
[1.37, 2.69]
[1.39, 2.71]
[2.62, 6.51]
[4.52, 18.52]
[2.22, 6.21]
[5.02, 12.43]
[2.50, 7.23]
[3.23, 14.68]
[2.76, 8.95]
[1.51, 3.33]
1.00
95% CI
High
Comparisons of Externalizing Behaviors (past year) for Moderate and High Truancy Youth in the U.S.
2.51
1.00
1.57
1.00
1.63
1.00
2.34
1.00
3.68
2.75
2.97
1.00
1.84
1.00
1.59
2.63
1.99
1.00
AOR
[2.11, 2.98]
[1.33, 1.86]
[1.38, 1.92
[1.82, 3.02]
[2.23, 6.06]
[1.99, 3.8]
[2.14, 4.12]
[1.28, 2.64]
[0.87, 2.92]
[1.86, 3.70]
[1.63, 2.44]
95% CI
Moderate
3.48
1.00
1.75
1.00
1.80
1.00
3.83
1.00
5.73
2.98
5.39
1.00
3.73
1.00
6.27
3.92
2.46
AOR
[2.36, 5.13]
[1.22, 2.51]
[1.24, 2.61]
[2.28, 6.42]
[2.66, 12.35]
[1.72, 5.16]
[3.35, 8.70]
[2.14, 6.51]
[2.60, 15.12]
[2.06, 7.46]
[1.64, 3.69]
95% CI
High
NIH-PA Author Manuscript
Table 3
Vaughn et al.
Page 16
NIH-PA Author Manuscript
26.04
[20.96, 31.86]
[8.05, 9.35]
[18.28, 22.91]
[6.47, 7.73]
95% CI
Moderate
9.49
1.45
5.43
0.99
%
[6.84,13.03]
[1.20,1.75]
[4.44,6.61]
[0.76,1.29]
95% CI
High
4.18
1.00
3.60
1.00
OR
[3.10, 5.65]
[3.03, 4.27]
95% CI
Moderate
9.13
1.00
6.79
1.00
OR
[6.05, 13.76]
[4.83, 9.54]
95% CI
High
3.00
1.00
2.71
1.00
[2.20, 4.08]
[2.25, 3.27]
95% CI
Moderate
AOR
5.19
1.00
4.33
1.00
AOR
Includes cocaine or crack, ecstasy, heroin, hallucinogens. Odds ratios in bold are statistically significant.
1
Note: Reference group: students who did not report missing any school. Adjusted models include age, sex, race/ethnicity, annual income, and lifetime depression and anxiety.
Yes (n=530)
No (n=12,839)
8.68
20.50
Yes (n=2,396)
Illicit drug use1
7.07
No (n=11,040)
Marijuana
%
NIH-PA Author Manuscript
Externalizing Behaviors
NIH-PA Author Manuscript
Number of days skipped
[3.31, 8.16]
[2.95, 6.36]
95% CI
High
Vaughn et al.
Page 17
J Adolesc. Author manuscript; available in PMC 2014 August 01.
NIH-PA Author Manuscript
NIH-PA Author Manuscript
7.98
8.03
Always/sometimes (n=10,646)
8.73
Always/sometimes (n=11,441)
6.64
Always/sometimes (n=5,163)
8.18
Always/sometimes (n=9,406)
[7.48,8.95]
[10.92,13.75]
[5.81,7.59]
[10.25,12.12]
[8.05,9.46]
[10.94,14.75]
[7.35,8.76]
[12.54,16.01]
[7.31,8.71]
[12.92,16.42]
95% CI
1.44
2.49
1.04
2.16
1.36
3.80
1.39
2.99
1.42
2.91
%
[1.15,1.79]
[1.94,3.18]
[0.72,1.49]
[1.80,2.59]
[1.12,1.66]
[2.84,5.07]
[1.13,1.71]
[2.29,3.90]
[1.15,1.74]
[2.22,3.80]
95% CI
High
0.63
1.00
0.56
1.00
0.64
1.00
0.52
1.00
0.50
1.00
OR
[0.53, 0.74]
[0.47, 0.66]
[0.52, 0.77]
[0.44, 0.61]
[0.42, 0.59]
95% CI
Moderate
0.54
1.00
0.45
1.00
0.33
1.00
0.43
1.00
0.44
1.00
OR
Number of days skipped
[0.39, 0.76]
[0.3, 0.68]
[0.23, 0.48]
[0.30, 0.60]
[0.31, 0.63]
95% CI
High
0.71
1.00
0.67
1.00
0.75
1.00
0.66
1.00
0.62
1.00
AOR
[0.60, 0.84]
[0.56, 0.80]
[0.61, 0.91]
[0.56, 0.8]
[0.52, 0.74]
95% CI
Moderate
0.62
1.00
0.65
1.00
0.40
1.00
0.63
1.00
0.59
1.00
AOR
[0.43, 0.89]
[0.42, 1.00]
[0.27, 0.58]
[0.43, 0.91]
[0.41, 0.85]
95% CI
High
Note: Reference group: students who did not report missing any school. Adjusted models include age, sex, race/ethnicity, annual income, and lifetime depression and anxiety. Odds ratios in bold are
statistically significant.
12.27
Seldom/never (n=3,860)
Parents limit time out at night
11.15
Seldom/never (n=8,208)
Parents limit television viewing
12.72
Seldom/never (n=1,973)
Parents provide positive reinforcement
14.19
Seldom/never (n=2,747)
Parents help with homework
14.58
Always/sometimes (n=10,617)
%
Seldom/never (n=2,787)
Parents check homework
Parental Involvement
Moderate
Comparisons of Parental Involvement for Moderate and High Truancy Youth in the U.S.
NIH-PA Author Manuscript
Table 4
Vaughn et al.
Page 18
J Adolesc. Author manuscript; available in PMC 2014 August 01.
Vaughn et al.
Page 19
Table 5
NIH-PA Author Manuscript
Summary of Study Effect Sizes for Adjusted Analyses.*
Moderately truant youth
Highly truant youth
Small (1.01–2.00)
Small (1.01–2.00)
Mean age (1.31)
Mean age (1.35)
No father in household (1.37)
Enjoy risky things (1.80)
Less likely engaged at school (0.80–0.50)
Enjoy dangerous things (1.75)
6 or more serious fights at school/work (1.59)
Last semester grades were C’s (1.83)
Medium (2.01–4.00)
Last semester grades were D’s or lower (1.83)
Lifetime depression (2.88)
Youth carried handgun 1 or more times (1.84)
Less likely engaged at school (0.63–0.19)
Lifetime anxiety (1.97)
Last semester grades were C’s (3.23)
1–2 serious fights at school/work (1.99)
1–2 serious fights at school/work (2.46)
Enjoy risky things (1.63)
Youth stole or tried to steal 1–2 times (2.98)
Enjoy dangerous things (1.57)
Alcohol use (3.48)
Youth carried handgun 1 or more times (3.73)
NIH-PA Author Manuscript
Medium (2.01–4.00)
Violent aggression 1 or more times (3.83)
Violent aggression 1 or more times (2.34)
3–5 serious fights at school/work (3.92)
Alcohol use (2.51)
3–5 serious fights at school/work (2.63)
Large (4.01+)
Marijuana use (2.71)
Marijuana use (4.33)
Youth stole or tried to steal 1–2 times (2.75)
Illicit drug use (5.19)
Youth sold illegal drugs 1 or more times (2.97)
Youth sold illegal drugs 1 or more times (5.39)
Illicit drug use (3.00)
Stealing 3 or more times (5.73)
Stealing 3 or more times (3.68)
6 or more serious fights at school/work (6.27)
Last semester grades were D’s or lower (8.80)
*
Note: Although we organize these effect sizes based on Cohen (1988), we recognize that small effects can have greater practical significance than
large effects and effect size magnitudes should be interpreted within the context of a given phenomenon.
NIH-PA Author Manuscript
J Adolesc. Author manuscript; available in PMC 2014 August 01.