Calculating the Australian Tertiary

Admission Rank in New South Wales

A Technical Report – March 2015
Contents

Preface ii 2.3 The ATAR in NSW 7

Common terms and abbreviations iii 2.3.1 Background 7

2.3.2 Categorising ATAR courses 7
1 University selection and the ATAR –
2.3.3 Eligibility for an ATAR in NSW 7
an overview 1
2.3.4 Calculation of the ATAR in NSW 7
1.1 Introduction 1
2.3.5 Extension courses 7
1.2 The pre-quota era 1 2.3.6 The ATAR Advice Notice 7
1.3 The post-quota era 1
3 Calculating the ATAR 9
1.4 From TES to TER: 1967–98 2
1.5 The UAI 2 3.1 Overview 9

1.6 Consequences of the 2001 HSC reforms 3 3.2 Marks used in the ATAR calculations 10

1.7 The ATAR 4 3.3 Data verification 10

1.8 Implications of the change from 3.4 Combined courses 10

UAI to ATAR 4 3.5 Initial standardisation 10
1.9 The use of a single index 4 3.6 Calculating scaled means and
standard deviations 10
2 The NSW HSC and the ATAR: an overview
3.7 Setting maximum scaled marks 10
of current NSW requirements 5
3.8 Scaling individual marks 11
2.1 Overview 5
3.9 Calculating aggregates 11
2.2 The NSW HSC 5
3.10 Calculating ATARs 11
2.2.1 Eligibility for an HSC 5
3.11 Verifying ATARs 13
2.2.2 Satisfactory completion
of courses 6 3.12 Distributing ATARs 13
2.2.3 Reporting student achievement
in the HSC 6
2.2.4 Defining standards by
performance bands 6
2.2.5 Examination marks 6
2.2.6 School assessments 6
2.2.7 HSC marks 7

Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015 i
 Preface

This report describes the technical The NSWVCC agreed that these
details surrounding the calculation of three changes, which would ensure
the index employed by NSW universities comparability of the distribution of
to assist in the processing of school NSW students’ ranks with those of
leaver applicants for tertiary places. other state and territories, would be
From 1998 until 2008 this index was implemented from the 2009 Higher
called the Universities Admissions School Certificate.
Index (UAI) but the New South Wales Numerical data presented in this report
Vice-Chancellors’ Committee (NSWVCC) illustrate the processes underpinning
agreed that, from the 2009 Higher the calculation of the ATAR in NSW.
School Certificate (HSC), it would The actual results will vary from year to
be known as the Australian Tertiary year; details can be found in the annual
Admissions Rank (ATAR). Report on the Scaling of the NSW
The decision to change the name Higher School Certificate and other
followed the 2008 decision by the reports.
Australasian Conference of Tertiary Documentation and reports are
Admission Centres (ACTAC) to adopt a available on the Universities
common name, the Australian Tertiary Admissions Centre (UAC) website at
Admissions Rank (ATAR), which would www.uac.edu.au.
replace the existing names of the
different selection indices used by
states and territories. The purpose of
the name change was to emphasise
that students’ ranks were reported on a
common scale.
With the new name in 2009 there
came two additional changes. Firstly,
consistent with the practice in other
states and territories, the NSWVCC
agreed to truncate students’ percentiles
so that the maximum rank in NSW
would be 99.95. Secondly, NSW
students’ ranks would be reported
against the cohort of students who
entered Year 7 with them six years
before, including those who did not
complete Year 12.

ii Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015
Common terms and abbreviations

ACTAC BSSS
Australasian Conference of Tertiary Admissions Board of Senior Secondary School Studies
Centres
Committee of Chairs
ATAR The Committee of Chairs refers to the
Australian Tertiary Admission Rank Committee of Chairs of Academic Boards and
Senates in NSW and the ACT.
ATAR cohort
The ATAR cohort refers to students who
HSC
received an ATAR in a particular year. The Higher School Certificate
students may have accumulated courses over a
five-year period. HSC cohort
The HSC cohort refers to students who have
ATAR courses completed at least one ATAR course in a
ATAR courses are Board Developed courses particular year.
for which there are examinations conducted by
the Board of Studies, Teaching and Educational KLA
Standards NSW that yield graded assessments.
Key Learning Area
VET courses for which there are no written
examinations, English Studies and Life Skills
courses are not ATAR courses. MCEETYA
Ministerial Council for Education, Employment,
Board Training and Youth Affairs
The Board refers to the Board of Studies,
Teaching and Educational Standards NSW NSWVCC
(BOSTES) and its predecessors. New South Wales Vice-Chancellors’ Committee

Board Developed courses SC cohort

Board Developed courses are HSC courses for The SC cohort refers to students who
which syllabuses have been developed and completed the School Certificate tests in a
examined by the Board of Studies, Teaching and particular year.
Educational Standards NSW.
SCE
BOSTES
School Certificate Examination
The Board of Studies, Teaching and Educational
Standards NSW TCOS
BSS Technical Committee on Scaling

Board of Secondary Studies

Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015 iii
 TER VET examination courses
Tertiary Entrance Rank The VET Curriculum Frameworks are based on training
packages for which the assessment is competency based.
TES As competence-based assessment does not yield a mark
that can be used in the ATAR calculations, the Board of
Tertiary Entrance Score
Studies, Teaching and Educational Standards introduced,
for each VET Curriculum Framework, an additional course
UAC that includes an examination. If students wish to have
Universities Admissions Centre a VET course contribute to their ATAR, they must enrol
in the appropriate additional course and complete the
UAI examination. These additional courses are termed VET
examination courses. Students who do not want their VET
Universities Admission Index courses to contribute towards their ATARs are not required to
complete these optional examinations.

iv Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015
1. University selection and the
ATAR – an overview

1.1 Introduction Because matriculation was based on passing

five or more subjects there was no need to
Universities receive applications from a broad add examination marks together and rank
range of people: school leavers who have students, except for the purpose of awarding
just finished their Higher School Certificate, scholarships.
mature age applicants who may or may not
have any formal educational qualifications, In response to high university failure rates,
and applicants from interstate and overseas or quotas were introduced, firstly for Medicine at
with other tertiary qualifications. The current the University of Sydney, and then more widely.
profile of applicants presents universities with
the challenge of developing valid procedures to 1.3 The post-quota era
choose applicants who have a good chance of The introduction of quotas in 1964 changed
successfully completing their programs. the landscape for both universities and
This document focuses on the Australian applicants.
Tertiary Admission Rank (ATAR), which is used Universities had to decide whether to set a
to rank most school leaver applicants, and how minimum standard for each of their courses
it is calculated in NSW. as a way of limiting the number of available
This chapter provides a brief historical places, or to first specify the number of
background to the current procedures. available places and then choose the best
applicants to fill those places. Their decision
1.2 The pre-quota era at that time was to choose the second option
as it gave them greater control over their
For over 100 years Australian universities course numbers. Under this model applicants
have argued that the best preparation for for each course were ranked in order of
tertiary study is achievement in a broad range overall academic merit, with the highest
of subjects. Before 1964, when quotas were considered first. The development of a valid
introduced, as long as students performed and efficient way of ranking applicants was
‘well enough’ in the examination that marked then required.
the end of secondary schooling, they were said
to have matriculated and could enrol in any Applicants had to specify a set of preferences
university course. The only impediment was indicating the courses in which they wanted
whether they could afford the fees. to enrol and then do well enough to gain a
place in at least one of those courses. Their
In NSW, students were required to pass at least preferences were considered in the order in
five subjects, including English, in the Leaving which they had been listed and if all the places
Certificate in order to matriculate. In most in the course listed as their first preference
subjects there were just two passing grades, had been taken by higher ranked applicants,
A and B, and one failing grade, F. Students who their second preference was considered. The
passed a particular subject were regarded as process continued until an offer was made or
performing well enough in that discipline to be there were no more preferences. This remains
able to study that subject at university. the current practice.

Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015 01
 1.4 From TES to TER: 1967–98 In 1990 a decision was taken to replace the TES with
a number, the Tertiary Entrance Rank (TER). This rank
The first post-quota aggregate used to rank school leaver indicated a student’s position in relation to other Year 12
applicants was based on the performance in a student’s students by rounding students’ percentiles (based on
best five HSC courses and calculated by the Board of their aggregate marks) to the nearest 0.05. Students
Senior Secondary Schools Studies (BSSSS). Because with percentiles at or above the 99.975th percentile thus
of the way subjects were structured at that time, the received TERs of 100.00.
maximum possible aggregate depended on the particular
combination of courses chosen, which led to a skewing in In 1991 responsibility for scaling was given to an inter-
the way students selected their HSC courses. The perception university committee, the Technical Committee on Scaling
emerged that students were encouraged to attempt the (TCOS), which reported to the NSWVCC.
highest levels as a mark-gaining strategy, which resulted
in many students attempting levels in courses beyond 1.5 The UAI
their abilities. The distribution of students’ selection ranks depends on the
In response to this concern changes to the structure of the reference population. Before 1998, NSW, in common with
HSC were implemented in 1976 with the introduction of other states and territories, reported students’ positions
2-unit and 3-unit courses. The resultant selection aggregate with reference to the cohort of Year 12 students who were
mark, the Tertiary Entrance Score (TES), was based on a eligible for a TER. Although the principles underpinning
student’s best 10 units with a possible maximum value of the states’ selection indices were similar, because their
500 irrespective of the mix of courses. Raw examination participation rates differed, their selection indices were not
marks were scaled before they were aggregated. Scaling comparable.
was considered necessary as a student’s position in a In 1994 the Ministerial Council for Education, Employment,
course depended on his/her ability and also the abilities of Training and Youth Affairs (MCEETYA) set up a taskforce to
other students in that course. The purpose of scaling was to devise a methodology for generating a common scale to
estimate a student’s position in a course if all courses had report students’ ranks. The decision was taken to adopt a
the same candidature. full weighted age cohort as the reference population. This
Further changes to the HSC were made during the was done on the assumption that students in different
following decade. A common core for related 2-unit and states achieving at the same level in relation to the
3-unit courses was introduced to place the marks for reference population were likely to perform at university in a
these courses on the same scale. In addition, from 1984 similar way.
a moderated school assessment mark was included as a All states except NSW agreed to use a logistic model to
formal component of the reported HSC mark. It had the determine the distribution of their Year 12 students’ ranks
same weight as the examination mark. and to report these ranks at intervals of 0.05 by truncating
In 1984 the Board of Secondary Studies (BSS) made a students’ percentiles at intervals of 0.05 commencing
decision not to continue with the calculation of the TES, at 99.95.
arguing that the calculation of a selection aggregate was the NSW agreed to adopt the common scale but elected to use
responsibility of the university sector. The Board’s intention School Certificate (SC) test data rather than the logistic
was that HSC marks would provide a profile of achievement model to determine the distribution of students’ ranks. NSW
and no measure of overall achievement was required. This argued that these data provided accurate information about
decision took effect in 1987. the quality of students who were eligible for a TER relative
The University of Sydney took initial responsibility for to their SC cohort. The NSW reference population thus
development of an algorithm for calculating the TES. To comprised those students who completed the SC two years
provide continuity with the past, this scaling algorithm was previously.
based on the Board’s scaling procedure. The decision to continue rounding percentiles in NSW meant
Because the TES was an aggregate of scaled marks, and that students above the 99.975th percentile received a TER
HSC marks reported by the Board were on a different scale, of 100.00 rather than 99.95.
there was some confusion in the community. Scaled marks In 1998 all states adopted a common scale for reporting
used in the aggregate of the TES were reported on a scale their TERs. In NSW the change was also accompanied
with course means clustered around 50; in contrast, HSC by a change of name: the TER became the Universities
marks were reported on a scale with a median of 60 for all Admission Index (UAI).
2-unit courses. Consequently, students’ aggregates could
no longer be calculated directly from their HSC marks and One consequence of the change in reference cohort in
were generally less than the sum of their best 10 units of NSW was that the middle Year 12 student who was eligible
HSC marks. for a UAI received a UAI of approximately 67.00, which was

02 Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015
higher than the TER of 50.00 previously awarded to the „„The marks of students in 3-unit (additional) courses were
corresponding student. Between 1998 and 2000 there was determined using common-person equating to put their
thus some correspondence between HSC marks reported 3-unit marks on the same scale as their marks in the
to students and their UAIs: the middle student in 2-unit corresponding 2-unit courses. An equipercentile method
courses received an HSC mark of 60 and the middle Year 12 was used.
student received a UAI of approximately 67.00. „„The marks of students enrolled in 4-unit Mathematics
In 1997 a review was made of the NSW HSC, which led to were scaled using common-person equating to put their
major changes in curriculum structure and the reporting of 4-unit marks on the same scale as their marks in the
student achievement. The changes were to take effect from 3-unit Mathematics course. An equipercentile method
2001. was used.
One recommendation was removal of the key learning area The result of these processes was that the marks provided
coverage requirement for award of the HSC. In order to ensure by the Board were on scales determined by the 2-unit only
that the aggregate included a verbal component, the TCOS students in the various courses, and the marks of 3-unit
made a decision to require that from 2001 at least two students in the same subject (in both their 2-unit and 3-unit
units of English were included in calculating the aggregate courses) were reported on the same scale.
underpinning the UAI. These marks, reported to students by the Board, were then
A further recommendation was that, from 1998, distribution used as the input for the calculation of the UAIs. The scaled
of the UAIs was the responsibility of the Universities means and standard deviations for the 2-unit courses were
Admissions Centre (UAC) and confidentiality provisions first determined as described in Chapter 3. To ensure that
prevented the reporting of UAIs to schools, the Board or the the same raw mark, whether a 2-unit or a 3-unit mark,
press. The major changes in reporting are described in the resulted in the same scaled mark, the scaling parameters
following section. used by the TCOS for 3-unit courses were the same as those
used to scale the corresponding 2-unit courses.
In response to the perception that academically able
students were being encouraged to study what were seen as The 2001 HSC, based on a new curriculum structure
very easy courses in order to increase their UAIs, the TCOS as recommended in the 1997 HSC Review, changed
decided in 1999 that the maximum marks in courses would the reporting of achievement in individual HSC
be determined on the basis of the academic quality of the courses. It moved from a norm-referenced model to
course candidatures. This decision was regarded as being a standards‑referenced model, using predetermined
consistent with the principles underpinning the calculation standards specified by six performance band descriptors.
of scaled means and standard deviations and took effect As performance band descriptors were not the same
from 2001. for different courses, the distributions of aligned marks
reported to students were, and are, not necessarily the
1.6 Consequences of the 2001 same for different courses. The percentages of students in
the performance bands may also vary from year to year.
HSC reforms
Since 2001 school assessments in a course have been
Prior to 2001 marks provided by the Board were moderated first moderated against the raw examination marks as
to satisfy the following conditions: before, and then aligned against the same standards as the
„„Assessments provided by schools were first transformed examination marks in that course.
so that moderated assessments in a course within a
Contrary to previous practice, marks in the extension
school had the same mean and standard deviation of
courses that replaced the previous 3-unit courses were
examination marks in that course within that school.
aligned against their own specific standards. No attempt
„„Examination marks in 2-unit courses were scaled by the has been made to place them on the scales used to report
Board such that: performance on the corresponding 2-unit courses.
——students who only completed the 2-unit course had
For the 2001 HSC the TCOS had two sets of marks that
their marks scaled by a multi-linear transformation so
could be used as input for UAI calculations:
that their median was 60, the maximum mark was
100, 1–2% of students received marks above 90, 25% „„the raw examination marks and school assessments
received marks above 80, 80% of students received moderated against the raw examination marks
marks above 50, no more than 1% of students received „„the aligned examination marks and aligned school
marks less than 20 and a mark of 0 was set at 0 assessments.
——students who also completed the corresponding The TCOS decided to use the two raw marks as input, rather
3-unit course had their 2-unit marks scaled by the than the marks reported to students, in order to preserve
same multi-linear transformation. the discrimination and relative differences between
students in a course as determined by the examiners.

Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015 03
 In 2001, the marks for the extension courses were not, the NSWVCC agreed that these three changes would be
as previously, on the same scales as the marks for their implemented together for the 2009 Higher School Certificate.
corresponding 2-unit courses. Therefore, the scaling
procedures had to be modified to determine a common 1.8 Implications of the change from
scale for the 2-unit and extension course scaled marks.
UAI to ATAR
The Board’s decision to have the boundary between Band 1
The change in reporting student ranks via the ATAR rather
and Band 2 set at 50 resulted in highly skewed distributions
than the UAI can be compared to the change in temperature
of marks, with most course means lying between 70 and 80.
scales, where once we reported a temperature in degrees
As the scaling algorithm remained the same, from 2001 to
Farenheit, but now we use degrees Celsius: the actual
2008 there was a difference between HSC marks reported
temperature has not changed, just the name and scale have
to students and their UAIs: the middle student in most
been changed.
courses received HSC marks in the range [70, 80] and the
middle UAI was in the vicinity of 67.00. The rules for the ATAR are exactly the same as those for
the UAI.
1.7 The ATAR Inclusion of the early leavers in the reference cohort meant
In 2008 a decision was taken by the Australasian that almost all ATARs were greater than the corresponding
Conference of Tertiary Admission Centres (ACTAC) to adopt UAIs. At the very top, because the maximum rank was 99.95
a common name, the Australian Tertiary Admission Rank rather than 100.00, some ATARs were less by 0.05, but the
(ATAR), which would replace the existing names of the actual number changing was very small.
different selection indices used by states and territories. The For students the effect had no practical consequences. The
name change was not to suggest that there was a common scaling process was the same, the rank order of students
method for calculating the selection indices in the different was the same, and the same applicants were selected for
jurisdictions, but to emphasise that students’ ranks from the same courses. Any change in an ATAR compared to a UAI
different jurisdictions were reported on a common scale. resulted in an equivalent change in the course cut-off.
The NSWVCC agreed to this change of name in 2008.
In 2008 the NSWVCC also agreed to truncate students’ 1.9 The use of a single index
percentiles, which is consistent with the practice in other
At different times it has been argued that the UAI, or
states and territories, so that the maximum rank in NSW
any single index, is a blunt instrument and that different
would be 99.95. One of the common mistakes people
indices should be used for selection for different
made was to think a score of 100 meant a ‘perfect score’.
university courses. Despite the apparent attractiveness
Prior to 2009, students who received a UAI of 100.00 were
of this view there is little empirical evidence in its favour.
those who performed better than 99.975% of their peers,
The choice of a university course, with all other factors
and the aggregate marks of this top group had ranged from
being equal, is likely to be determined by a student’s
480 to 498. There has never, in the past 30 years, been a
knowledge, interests and skills, so that future applicants
student with a ‘perfect’ aggregate of 500. For the ATAR the
for a particular course will have their ATARs based on
top group is comprised of students who have performed
HSC courses that provide the academic background
better than 99.95% of their peers, which means that the
required for that course. Students with ATARs based on
number who receive the maximum rank increased from
different patterns of HSC courses are likely to apply for
18–23 to approximately 46.
different university courses.
The minimum school leaving age in NSW in 2009 was
In recent years there has been an increase in the range of
15 years. In 2011 it increased to 17 years. A decision
alternative selection methods for school leavers as well as
was taken by the NSWVCC in 2009 to pre-empt the change
for other applicants.
in the nature of the HSC cohort by using the appropriate
Year 7 cohort as the reference group for calculating the
admission rank. This change came into effect with the
move to the ATAR.
Rather than have the changes occur in a piecemeal fashion,

04 Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015
2. The NSW HSC and the ATAR:
An overview of current NSW
requirements

2.1 Overview takes into account differences in the abilities

of course candidatures and produces marks
The base data for reporting student that can be deemed equivalent across
achievement are two sets of marks: courses. A student’s scaled mark in a course
assessment marks provided by schools and indicates their position within a common
HSC examination marks. From these two candidature. ‘Equivalence’ here is defined
sets of marks the Board produces Records of in a comparable sense: marks in different
Achievement for students showing how they courses are deemed to be equivalent if
have performed against set standards, which the students awarded those marks would
are described as Performance Bands, and have achieved the same position in these
the Technical Committee on Scaling produces different courses if the candidatures of the
numbers that indicate how students have courses were identical.
performed in relation to other students.
Details of the Board’s procedures are available on
To obtain the marks reported by the Board its website at www.boardofstudies.nsw.edu.au.
and the scaled marks that contribute to the
aggregate on which the ATAR is based, the two
sets of marks undergo three transformations to
2.2 The NSW HSC
take account of factors that are independent of The Higher School Certificate (HSC) is an exit
the students themselves. certificate awarded and issued by the Board. It
1. A student’s school assessment mark marks the completion of 13 years of schooling,
depends not only on the developed ability is the gateway to further study and employment,
of the student but also on the nature of and presents a profile of student achievement
the school assessments and the standards in a set of courses.
used by the school in the courses the
student has completed. Assessments 2.2.1 Eligibility for an HSC
are first moderated by the Board to take To qualify for an HSC in NSW, students
account of differences between schools. must complete a pattern of Preliminary and
The moderated marks indicate the HSC courses containing at least 12-units of
achievements of students in their courses Preliminary courses and at least 10 units of
as if there were only one school. HSC courses.
2. A student’s examination mark depends
These HSC courses must include at least:
not only on the developed ability of
the student but also on the difficulty of „„six units of Board Developed courses
the examination itself and the severity „„two units of a Board Developed course in
of marking. To take account of these English
factors and to produce marks that can be „„three courses of two unit value or greater
compared across time, the Board calibrates (either Board Developed or Board Endorsed
or aligns the examination marks in a course courses)
against the published standards for that
„„four subjects.
course. The aligned marks indicate the
standards students have achieved, rather Further details about HSC eligibility and
than the marks gained in a particular HSC courses can be found in the Assessment,
examination paper. Certification and Examination Manual and in
3. A student’s position in a course depends the Rules and Procedures for Higher School
not only on the student’s developed ability Certificate Candidates booklet, which are
but also on the abilities of the students with published annually by the Board, and available
whom he/she is compared. Scaling, carried on its website at www.boardofstudies.nsw.edu.au.
out by the Technical Committee on Scaling,

Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015 05
 2.2.2 Satisfactory completion of courses The range of reported marks for the bands is as follows:
Students are deemed to have satisfactorily completed a 2-unit courses
course if, in their principal’s view, there is sufficient evidence
Band 1 2 3 4 5 6
to show that the students have:
Mark range 0–49 50–59 60–69 70–79 80–89 90–100
„„followed the course of study prescribed by the Board
„„applied themselves with diligence and sustained effort
to the set tasks and experiences provided by the school Extension courses (except Mathematics Extension 2)
„„made a genuine attempt at assessment tasks that
Band E1 E2 E3 E4
total more than 50 per cent of the available school
Mark range 0–24 25–34 35–44 45–50
assessment marks for that course
„„achieved some or all of the course outcomes.

Students also need to make a serious attempt at the Mathematics Extension 21

examination for the courses. Receiving a mark for a course
Band E1 E2 E3 E4
on a Record of Achievement is an indication that the student
has satisfactorily completed that course. Mark range 0–49 50–69 70–89 90–100
1. Mathematics Extension 2 students have their achievement reported using
2.2.3 Reporting student achievement in four bands but the mark range is out of 100 rather than 50.

the HSC
For most ATAR courses, the Board reports student
2.2.5 Examination marks
achievement against published standards by:
„„an examination mark The examination mark reported on a student’s Record of
Achievement indicates the standard a student has attained
„„a school assessment
in that examination. If, for example, a student’s performance
„„an HSC mark in the Society and Culture examination is at the standard
„„a performance band. described for Band 3, the examination mark reported on
their Record of Achievement for that course will lie between
These results are shown on a student’s Record of
60 and 69.
Achievement. For most Board Developed courses, a
Course Report is also provided. The report describes, using In general this mark, termed the aligned examination mark,
performance bands, the standard achieved in the course will differ from the mark the student actually gained on the
and provides a graph indicating the student’s position in the examination (the raw examination mark).
course candidature.
The aligned marks indicate the standards reached
by students and their relative positions in a band. For
2.2.4 Defining standards by performance bands
example, a mark of 62 means that, while the student has
Standards in a course are described in terms of the content, demonstrated a Band 3 standard, their achievement is
skills, concepts and principles relevant to the course and closer to Band 2 standard than Band 4 standard.
represent the range of achievement expected of students
completing the course. Performance band descriptors, 2.2.6 School assessments
which describe typical achievement at different standards
To enable school assessments from different schools
(bands), have been developed for each course. There are six
to be compared, marks submitted by schools (raw
bands for 2-unit courses and four for extension courses.
assessments) are first moderated using the raw
The percentage of students in any band depends only on examination marks gained by their students and then
how students enrolled in that course perform at the standard aligned to course standards. The school assessments
specified by the band descriptor. There are no predetermined reported on a student’s Record of Achievement are the
percentages of students to be placed in the bands. aligned assessments.
It follows that, although the standards described by the Although school assessments are moderated and then
bands in a particular course will be the same from year aligned against standards, a school’s rank order of
to year, different courses will have different standards students in a course is maintained.
because they are based on different criteria. It is likely that
the percentage of students in each band will vary across
courses, and may vary from year to year.

06 Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015
2.2.7 HSC marks are not regarded as satisfactory in themselves, but their
contribution to a selection index is regarded as adequate
For each course, the Board’s aligned examination and if the other courses included in the aggregate are more
school assessment marks, rounded to whole numbers, academically demanding.
are released to students together with an HSC mark.
This HSC mark is the (rounded) average of the (rounded) All ATAR courses are reviewed on a regular basis by the
aligned examination mark and (rounded) aligned school Committee of Chairs and the categorisations can change
assessment and determines a student’s performance band over time. Current details can be found on the UAC’s website
for the course. at www.uac.edu.au.
Further details about the Board’s processes can be 2.3.3 Eligibility for an ATAR in NSW
found in Board Bulletins, in The Media Guide, which is
produced annually, and on the Board’s website at To be eligible for an ATAR a student must have satisfactorily
www.boardofstudies.nsw.edu.au. completed at least 10 units of ATAR courses, which included
at least:
2.3 The ATAR in NSW „„eight units of Category A courses
„„two units of English
2.3.1 Background „„three Board Developed courses of two units or greater
The Australian Tertiary Admission Rank (ATAR) is a numerical „„four subjects.
measure of a student’s overall academic achievement in
the HSC in relation to that of other students. The ATAR is 2.3.4 Calculation of the ATAR in NSW
reported as a number between 0 and 99.95 in increments
of 0.05. This measure allows the overall achievement of The ATAR is based on an aggregate of scaled marks in 10
students who have completed different combinations of HSC units of ATAR courses comprising:
courses to be compared. „„the best two units of English
„„the best eight units from the remaining units, which can
In NSW, ATARs indicate the positions of students relative to
their Year 7 cohorts. That is, students who receive an ATAR include up to two units of Category B courses.
of 80.00, for example, have performed better than 80% of Marks to be included in the ATAR calculations can be
their Year 7 cohort, assuming that all these Year 7 students accumulated over a five‑year period but if a course is
were eligible for an ATAR six years later. repeated only the last satisfactory attempt is used in the
The ATAR is calculated solely for use by tertiary institutions, calculation of the ATAR. For students accumulating courses
either on its own or in conjunction with other criteria, to towards their HSC, scaled marks are calculated in the year
rank and select school leavers for admission. Calculation the courses are completed.
of the ATAR is the responsibility of the TCOS on behalf of
the NSWVCC.
2.3.5 Extension courses
Extension courses do not have to be completed at the
Students who indicate on their HSC entry forms that
same time as the corresponding 2-unit courses; they can
they wish to be notified of their ATAR will receive an ATAR
be completed in a different year. However, the marks of
Advice Notice from UAC. ATARs are also made available to
extension courses will not be counted towards the ATAR
institutions for selection purposes.
calculation unless the corresponding 2-unit course has been
2.3.2 Categorising ATAR courses satisfactorily completed.

ATAR courses are assessed by formal examinations 2.3.6 The ATAR Advice Notice
conducted by the Board and are deemed by the Committee
The ATAR Advice Notice includes:
of Chairs of Academic Boards and Senates in NSW and
the ACT to have sufficient academic rigour to be useful as „„the student’s ATAR
preparation for university study. „„a list of the ATAR courses which the student studied, the
categorisation of each course and the year the course
ATAR courses are classified as either Category A or
was completed
Category B courses. The criteria for Category A courses
are academic rigour, depth of knowledge, the degree to „„the number of units of each ATAR course that were
which the course contributes to assumed knowledge for actually included in the calculation of the ATAR.
tertiary studies and the coherence with other courses While ATARs are calculated for all ATAR-eligible students,
included in the ATAR calculations. Category B courses are only those students who indicate on their HSC entry forms
those whose level of cognitive and performance demands that they wish to be notified of their ATAR will receive an
ATAR Advice Notice from UAC.

Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015 07
 There are two cases where an ATAR will not be shown on the has not met the requirements for an ATAR, in which case the
ATAR Advice Notice. The first is when a student receives an statement ‘Not Eligible’ will appear.
ATAR between 0.00 and 30.00, in which case the ATAR will
An example of an ATAR Advice Notice is given below.
be indicated as ‘30 or less’. The second is when the student

An example of an ATAR Advice Notice

08 Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015
3. Calculating the ATAR

3.1 Overview Scaling first modifies the mean, the standard

deviation and the maximum mark in each
Tertiary institutions are concerned with ranking course. Adjustments are then made to the
school leaver applicants. From their perspective, marks of individual students to produce scaled
the importance of HSC marks is that they marks, which are the marks the students would
convey information about a student’s position in have received if all courses had the same
relation to other students. candidature.
With the exception of English, which is Although scaled marks are generally different
compulsory, students are free to choose their from the raw marks from which they are
courses of study. Consequently, individual derived, the ranking of students within a course
course candidatures vary in size and nature is not changed.
and there are many different enrolment
patterns. For example, in 2014 there were Once the raw marks have been scaled,
27,535 different enrolment patterns for ATAR- aggregates are calculated for ATAR-eligible
eligible students; only 202 of these 27,535 students. Percentiles, which indicate the
combinations were completed by 20 or more ranking of students with respect to other
students and 20,008 were taken by only one ATAR-eligible students, are then determined on
student. Given the choice available, it follows the basis of these aggregates. In most cases,
that a student’s rank in different courses will the ranking or order of merit based on these
not necessarily have the same meaning, as aggregates is quite different from the order of
good rankings are more difficult to obtain when merit using aggregates based on HSC marks.
the student is competing against students of The last step is to translate these percentiles
high academic ability. into ranks relative to the weighted age cohort
Because of the lack of comparability of HSC population based on Australian Bureau of
marks achieved in different courses, marks of Statistics data. This cohort effectively reflects
individual students are scaled before they are the Year 7 cohort from five years earlier. The
added to give the aggregates from which the population percentiles are truncated at intervals
ATAR is determined. of 0.05 starting at 99.95. These are the ATARs.

The scaling process is designed to encourage Each ATAR corresponds to a range of

students to take the courses for which they are aggregates and the number of students with
best suited and which best prepare them for each ATAR varies, depending in part on how
their future studies. The underlying principle is many candidates tie on the same aggregate.
that a student should neither be advantaged The scaling process, which does not assume
nor disadvantaged by choosing one HSC course that one course is intrinsically more difficult
over another. The scaling algorithm estimates than another or that the quality of the course
what students’ marks would have been if all candidature is always the same, is carried out
courses had been studied by all students. afresh each year.
The scaling model assumes that a student’s All students who complete at least one ATAR
position in a course depends on the student’s course in a given year are included in the
developed ability in that course and the scaling process for that year. Students who are
‘strength of the competition’. Since the ATAR accumulating courses towards their HSC have
is a rank that reflects academic achievement, their scaled marks calculated in the year the
‘strength of the competition’ is defined in courses are completed.
terms of the demonstrated overall academic
attainment of a course candidature.

Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015 09
 3.2 Marks used in the ATAR 3.6 Calculating scaled means and
calculations standard deviations
The marks used in the calculation of the ATAR are derived The model underpinning the scaling algorithm specifies
from the following two marks provided by the Board that the scaled mean in a course is equal to the average
„„a raw examination mark academic achievement of the course candidature
where, for individual students, the measure of academic
„„a raw moderated school assessment – these are school
achievement is taken as the average scaled mark in all
assessments that have been moderated using the raw
courses completed. The model specification leads to a set
examination marks.
of simultaneous equations from which the scaled means of
All marks are provided on a one-unit basis to one decimal 2-unit courses are calculated.
place. A student’s raw HSC mark in a course is the average
The scaled standard deviation for a 2-unit course is the
of their raw examination mark and their raw moderated
standard deviation of the measure of overall academic
school assessment.
achievement of the candidature of that course.
In the description of the scaling process that follows, to cater
For extension courses the scaled means and standard
for both 2-unit and extension courses, marks are described
deviations are determined by the performance of
on a one-unit basis.
the extension students on the corresponding 2-unit
courses. The exceptions are History Extension, which
3.3 Data verification can be completed by both Modern History and Ancient
Before any processing is carried out, an extensive number of History students, and the second Extension courses
data verification checks is completed to ensure the integrity in English and Mathematics: English Extension 2 and
of the marks received from the Board. Mathematics Extension 2.
Students who have marks for extension courses but no A scaled mean is determined for the Modern History
marks for the corresponding 2-unit courses will have their students in History Extension on the basis of their
extension course marks ignored. Likewise, students will performance in the 2-unit Modern History course. A scaled
not have their Mathematics Extension 2 marks included if mean for the Ancient History students in History Extension
they have not completed Extension 1, nor will their English is found in a similar manner. The scaled mean for History
Extension 2 marks be included unless they also have Extension is then set equal to the weighted average of these
English (Advanced) and English Extension 1 marks. These two scaled means. The scaled standard deviation is found in
courses may be completed in different years. a similar manner.
Students who have completed Mathematics and Scaled means and standard deviations for the Extension 1
Mathematics Extension 1, and subsequently complete courses in English and Mathematics are calculated
Mathematics Extension 2, will have their Extension 1 and as described above. The scaled mean and standard
Extension 2 marks included but their Mathematics marks deviation for the Mathematics Extension 2 course are
will be ignored. The unit value of Mathematics Extension 1 then determined by the performance of the Extension 2
will be changed from 1 to 2. students in the Mathematics Extension 1 course. For English
Extension 2, the scaled mean and standard deviation are
Students who repeat a course have their last satisfactory
determined by the performance of the Extension 2 students
attempt included in the calculations.
in English (Advanced). This option is not available for
Mathematics as the Extension 2 students do not complete
3.4 Combined courses the Mathematics 2-unit paper.
The Board places English (Standard) and English (Advanced)
raw marks on a common scale using the common Paper 1. 3.7 Setting maximum scaled marks
For scaling purposes these courses are combined and
The maximum possible scaled mark in a course is
scaled as a single course. In the Report on Scaling the
determined according to the academic quality of the course
courses are reported as separate courses in order to be
candidature in such a way that the maximum possible
consistent with the Board’s reporting practice.
scaled mark for the combined 2-unit English candidature is
50 on a one-unit basis.
3.5 Initial standardisation
In 2014 the maximum possible scaled mark in a
Before the scaling algorithm is implemented, a linear course was given by the smaller of 50 and the scaled
transformation is applied to the raw HSC marks in each mean + 2.48 times the initial scaled standard deviation,
course to set the top mark to a common value. The marks where the scaled mean and initial scaled standard deviation
in each course are then standardised to a mean of 25 and of the course are determined using the scaling algorithm.
standard deviation of 12 on a one-unit basis.

10 Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015
The number, 2.48, was determined to ensure that the Year 7 cohort if all students in this cohort completed Year 12
maximum possible scaled mark in the combined 2-unit and were eligible for an ATAR.
English course was 50. This number is calculated afresh
Up until 2013 the ATARs were estimated by common-
each year. It has varied from 2.47 to 2.49 over the past six
person equating using the School Certificate Examination
years.
(SCE) mark as the anchor variable. The common persons
For extension courses with small initial scaled standard were those students in the SC cohort who were eligible
deviations a variation exists to replace the initial scaled for an ATAR two years later; typically, they represented
standard deviation by 6 when calculating the maximum approximately 60% of the total SC cohort. These students
possible scaled mark. had both an aggregate mark gained in Year 12 and an SCE
mark gained in Year 10.
3.8 Scaling individual marks The last School Certificate tests were held in 2011 and
Once the scaled means and standard deviations are from 2014 translating ranks based on scaled aggregates
determined, individual raw marks are scaled using a into ATARs was completed using a two-parameter logistic
non‑linear transformation which preserves the scaled mean function, which is consistent with the technique used in
and standard deviation of a course and restricts the scaled other states without Year 10 data.
marks to­the range [0, Max]. Max is the maximum possible The logistic function approach was adopted by other
mark for the course calculated according to the method jurisdictions in 1998 based on the patterns observed in
described in section 3.7. NSW data. To illustrate the pattern, Figure 3.1 shows the
If the actual maximum scaled mark in a course is less proportions of the 2010 SC cohort who were eligible for
than the maximum possible scaled mark a further an ATAR two years later in 2012 plotted against the SCE
linear transformation is applied. The effect of this linear mark. Clearly, almost all of the most able students stayed
transformation is to increase the standard deviation so that on to Year 12 and applied for an ATAR and the proportion of
the actual maximum scaled mark in the course is changed ATAR‑eligible students decreased as the SCE total decreased.
to be the same as the maximum possible scaled mark. The The larger spikes at the extreme SCE marks are due to
transformation does not affect the scaled mean. the proportions being based on very small numbers of
candidates.
For some courses with very small candidatures the
non‑linear transformation is not always appropriate, in which The specific shape of the logistic function will depend on
case alternative transformations, which are consistent with the proportion of students in the target population who are
the principles of the scaling algorithm, are used. ATAR-eligible. This figure is the participation rate. In 2014
the participation rate in NSW (determined using ABS data)
3.9 Calculating aggregates was 59.6%, up from 58.9% in 2013.

Once the final scaled marks have been calculated for The anchor frequency is the number, N, allocated to the
each course, students’ marks are brought together and 99.95 category. The top category should contain 1/2000th
aggregates of scaled marks calculated according to the rule of the target population as all the most able candidates
described in section 2.3.4 for each ATAR-eligible student. would study to Year 12 and apply for an ATAR. In 2014 this
These aggregates are calculated to one decimal place and target frequency was N = 47.
will lie in the range [0, 500]. The logistic model is log [pj /(1-pj)] = a + b xj, where Npj is
Students who have completed courses in a previous year the target frequency of students at ATAR xj, for xj less than
will have the scaled marks for those courses incorporated 99.95. To be consistent with recent NSW ATAR patterns
with the scaled marks from the current year to determine the minimum ATAR awarded is set at 8.00. The target
their aggregate, provided the course has not been repeated proportions are then rescaled to ensure they sum to 1.
in the current year or replaced by an alternative course The parameters in the logistic model were estimated using
(for example, replacing Mathematics General 2 with historical ATAR and participation rate data for NSW for
Mathematics). 2006–13.
Starting with the highest aggregate the candidates are
3.10 Calculating the ATARs progressively allocated to ATAR categories to achieve
The aggregates of 10 units of scaled marks can be used to the cumulative target frequencies. There is noise in the
rank all ATAR-eligible candidates. The percentiles within the allocation due to ties in the aggregates. The resulting
ATAR-eligible group (rounded to the nearest 0.05) are the pattern is shown in Figure 3.2
equivalent of the old TER scores. To calculate the ATARS the
The relationship between the ATAR and TER in 2014 is
percentiles are required relative to the appropriate weighted
shown in Figure 3.3.
age cohort determined using Australian Bureau of Statistics
data. This reference population is effectively the appropriate

Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015 11
 Figure 3.1: Proportions of the SC cohort who were eligible for an ATAR in 2012, by SCE mark
1.0
0.9
Proportion ATAR-eligible 0.8

0.7

0.6
0.5

0.4

0.3

0.2

0.1

0.0
0 50 100 150 200 250 300 350 400
SCE mark

Figure 3.2: Percentage of ATAR-eligible students in each ATAR truncated category, 2014.
2.0
1.8

1.6
Percent with ATAR truncated

1.4

1.2
1.0

0.8

0.6

0.4

0.2

0.0
0 10 20 30 40 50 60 70 80 90 100
ATAR truncated

allocated target

Figure 3.3: The relationship between ATAR and TER, 2014

100
90

80

70

60
ATAR

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80 90 100
TER

12 Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015
The relationship between the ATAR and aggregate in 2014 results with previous years to identify what might be
is shown in Figure 3.4. aberrant values, running consistency checks on individual
students and groups of students and checks on the internal
Each ATAR corresponds to a range of aggregate marks
consistency of the results when considered as a whole.
which is greatest in the extremes of the distribution of
aggregates and smallest near the middle of the distribution
of aggregates. 3.12 Distributing ATARs
The ATARS of individual students are distributed by UAC and
3.11 Verifying ATARs used in the processing of school leaver applicants.
The final step is to implement a broad range of data
verification checks to ensure the accuracy of the calculated
ATARs. These include running comparisons of the current

Figure 3.4 Relationship between aggregate and ATAR, 2014

100
90

80

70

60
ATAR

50

40

30

20

10

0
0 50 100 150 200 250 300 350 400 450 500
Aggregate

Calculating the Australian Tertiary Admission Rank in New South Wales: A Technical Report – March 2015 13
Contacting UAC © Universities Admissions Centre
(NSW & ACT) Pty Ltd 2015
UAC ACN 070 055 935
Locked Bag 112 ABN 19 070 055 935
Silverwater NSW 2128 Published May 2015
UAC is the owner of the copyright in this publication.
Quad 2, 8 Parkview Drive Apart from any fair dealing for the purpose of private
Sydney Olympic Park NSW 2127 study, criticism or review, or otherwise as permitted
under the Copyright Act, no part may be reproduced
For those travelling by train, UAC is 250 metres
by any process without UAC’s written permission.
from Olympic Park railway station.
Enquiries should be addressed to the Managing
Director, UAC.
8.30am–4.30pm UAC has no objection to schools and tertiary
Monday to Friday (Sydney local time) institutions reproducing the publication provided it
is for use only within their own institution and this
copyright statement is included.
1300 ASK UAC (1300 275 822)
from mobiles: (02) 9752 0200 If the document is not reproduced in full, sections
from overseas: +61 2 9752 0200 should not be copied out of context where information
could be incomplete and/or misleading. Schools and
tertiary institutions must ensure that this information
www.uac.edu.au is not transmitted to any other person or body without
prior permission from UAC.
This publication is available on UAC’s website at
email online enquiry form www.uac.edu.au/general/
www.uac.edu.au/general/contact.shtml About this publication
Calculating the Australian Tertiary Admission Rank in
New South Wales: A Technical Report describes the
www.facebook.com/universitiesadmissionscentre
technical details surrounding the calculation of the
ATAR and provides a brief historical background to the
current procedures.
http://twitter.com/UACinfo

www.youtube.com/user/UACinfo
Universities Admissions Centre (NSW & ACT) Pty Ltd
Quad 2, 8 Parkview Drive
Sydney Olympic Park NSW
Locked Bag 112, Silverwater NSW 2128
1300 ASK UAC (1300 275 822)
from mobiles: (02) 9752 0200 from overseas: +61 2 9752 0200
www.uac.edu.au
ABN 19 070 055 935 ACN 070 055 935