Attrition rate
Student progress rate
Graduate full-time employment
Graduate full-time study
Graduate starting salaries
Course Experience Questionnaire (CEQ) overall satisfaction
Attachment
These indicators show the drop out rate from institutions which represents one dimension of the effectiveness of the delivery of educational services. Students who complete their course and then transfer to another course within the institution are defined to be retained. Students simply completing a course are excluded from both the numerator and denominator of the attrition rate. That is, students simply completing a course are not included among students dropping out (attrition).
The 1999 attrition rate indicator is calculated as follows:
Data:
1999 and 2000 students unit record enrolment file
2000 student completions unit record file
Filters:
Data definitions for the 1999 attrition rate:
Total students in 1999:
Students are only included once for a major course. Thus students only included where E331 Concurrent/Major course indicator takes the value 1 or 2 (IF NOT (E331 IN (1 2)) THEN DELETE).
Students in enabling, cross institutional and non-award levels are excluded, that is, where E310 Course type code takes the value of 30, 40, 41, 42 or 50.
Students at the Open Learning Agency of Australia are excluded, that is, where E306 Institution code takes the value of 3037.
Students in 2000
Filter:
Students are only included once for a major course. Thus students only included where E331 Concurrent/Major course indicator takes the value 1 or 2 (IF NOT (E331 IN (1 2)) THEN DELETE).
Students in enabling, cross institutional or non-award courses are excluded, that is, where E310 Course type code takes the value of 30, 40, 41, 42 or 50.
Students at the Open Learning Agency of Australia are excluded, that is, where E306 Institution code takes the value of 3037.
Completing students in 1999 (from the 2000 student completions unit record file)
Only students with a primary record are included, that is, where E089 Primary record indicator takes the value of 1.
Students at the Open Learning Agency of Australia are excluded, that is, where E306 Institution code takes the value of 3037.
Indicator:
The crude attrition rate is defined as follows:
Student attrition in 2000 / Base students in 1999
where Base students = Total students in 1999 - Completing students in 1999
Note students completing in 1999 are obtained from the 2000 student completions unit record file.
Student attrition in 2000 is obtained by matching base students in 1999 with students in 2000 using student identification numbers. Students completing a course in 1999 are excluded from the measure of base students. However, where a student completes one course but is retained in another course, that student is included in the count of retained students, or rather, excluded from the count of student attrition. These students are not included in the count of completing students in 1999 used to derive the base number of students. Only students simply completing a course (and not proceeding to another course) are excluded from the number of base students.
The attrition rate indicators are shown for commencing undergraduate, non-commencing undergraduate and postgraduate coursework students.
Undergraduate students are defined where element E310 takes the value of 8, 9,10, 13, 20, 21, 22 and postgraduate coursework students are defined where element E310 Course type code takes the value of 4,5,6,7 or 11. Commencing students are defined where element E922 takes the value of 1 and non-commencing students where E922 takes the value of 2.
We conduct an Ordinary Least Squares regression with the attrition rate as the dependent variable (if the graduate has dropped out, the dependent variable takes the value of 1 and 0 otherwise).
Independent variables
Independent variables are defined as follows:
Students unit record enrolment file
Gender from element E315, entered as a dummy variable where 1=Female, 0=Male.
Age from element E997 sorted into age groups, less than 19, 20 to 24, 25 to 29 and 30 plus. Entered as a set of dummy variables.
NESB – from element E348, entered as a dummy variable, where 1=from a non- English speaking background (for values of E348 within the range 1000 to 9998) and 0= otherwise (for values of E348 = 0001, English speaking background).
Indigenous – from element E316, entered as a dummy variable, where 1 = Indigenous and 0 = non-Indigenous.
Disability - from element E386 where graduates are categorized as disabled if the value in the first field of this element is 1. Entered as a dummy variable, where 1 = disabled, 0 = non-disabled.
Broad field of study – from element E311 using the standard broad fields of study, agriculture, architecture, arts/humanities, business, education, engineering, health (with nursing separately identified), law, science and veterinary science, entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular broad field of study and 0 for remaining broad fields of study.
Level of course – from element E310.
For the undergraduate regressions, the following courses are included
8 = Bachelor’s Graduate Entry
9 = Bachelor’s Honours
10 = Bachelor’s Pass
13 = Associate Degree
20 = Advanced Diploma
21 = Diploma
22 = Other award course
For the postgraduate coursework regression, the following courses are included
4 = Master’s by Coursework
5 = Postgraduate Qualifying
6 = Graduate Diploma/Postgraduate Diploma in ‘new area’
7 = Graduate Diploma/Postgraduate Diploma in ‘existing area’
11 = Graduate Certificate
In each regression, the level variables are entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular level of study and 0 for remaining levels of study.
Mode of attendance (internal/external) - from element E329, entered as a dummy variable, where 1 = Internal and 0 = External.
Type of attendance (full-time/part-time) - from element E330, entered as a dummy variable, where 1 = Full-time and 0 = Part-time.
New to higher education – variable only used for the undergraduate commencing students regression. From element E924, entered as a dummy variable taking the value of 1 where element E924 = 2 (new to higher education) and the value of 0 otherwise.
Basis of admission and Tertiary Entrance Score – variable only used for the undergraduate commencing students regression. This was a set of dummy variables constructed using the Basis for admission element E327 and Tertiary Entrance Score element E369.
Basis of admission entered as a set of dummy variables for each basis of admission for the following codes 11, 12, 16, 17, 18, 19, 20, 21, 22 and 29 (and code 13 in 1999 and its equivalent of combining codes 21 and 22 in 2000), taking the value of 1 for the relevant basis of admission and 0 otherwise.
For the basis of admission 14 and 15, that is admitted on the basis of satisfactory completion of secondary education at school or TAFE, these data were further disaggregated to allow for the influence of TER score. Within each basis of admission, a series of seven dummy variables were constructed for each TER decile score where element E369 was greater than or equal to 30 and less than 40 and so on with the last decile being greater than or equal to 90 and less than or equal to 100.
Size and field of study – Number of students by institution and broad field of study (see above for broad fields of study) derived from element E929. Entered as a continuous variable.
Institution – a set of institutional dummy variables where each dummy variable takes the value of 1 for a particular institution and 0 for remaining institutions.
1996 Census of Population and Housing
Socio-economic status -
- Index of education and occupation as defined by parents’ education and occupation background. This is an index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
- Index of economic resources as defined by income. Index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
Socioeconomic indexes by postcode from the 1996 Census are matched with the student’s postcode as given by element E320. Overseas students (without a postcode) are assigned an index value of zero.
Locality status –
- Rural status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census.
- Isolated status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census.
Entered as a set of dummy variables, taking the value of 1 for the particular locality and 0 for all other localities.
top
These indicators provide a measure of educational achievement and the effectiveness of educational delivery. The student progress rate measures successful student subject load.
We show student progress rates for commencing undergraduate, non-commencing undergraduate and postgraduate coursework students separately. Research students are excluded (including Bachelor’s Graduate Entry and Bachelor’s Honours level students) on the basis that the notion of success in subjects may be problematic for these students. Note that in the 1998 edition of performance indicators the scope of the student progress rate indicator was restricted to Bachelor Pass level students and students in lower level courses.
Student load relating to industrial experience is not included in the calculation of the student progress rate indicator, as was the case in the previous edition of performance indicators released in 1998. The basis for this exclusion is to avoid any potential bias in the student progress rate indicator among institutions with a disproportionate concentration of industrial experience load.
The student progress rate has been calculated for all students, non-overseas and overseas. Note that this coverage is broader than the calculation of student progress rates in DETYAPAC which includes non-overseas students only. The focus of DETYAPAC is on the experience of equity groups and hence the concern with the subset of non-overseas students.
The 1999 and 2000 progress rate indicators are calculated as follows
Data:
1999 and 2000 unit of study completion details file
Filter:
Students at the Open Learning Agency of Australia are excluded, that is, where E306 Institution code takes the value of 3037.
Exclude industrial experience E337 industrial experience code equals 2.
Indicator:
The progress rate indicator is defined as follows :
EFTSU status is given by element E339 Total EFTSU value and E355 Unit of study completion status where 1 = withdrew, 2= failed, 3= success, 4=no result.
Total successful EFTSU/ Base EFTSU
where Base EFTSU = Total EFTSU - Total EFTSU result not determined (E355=4)
The progress rate indicators are shown for commencing undergraduate, non-commencing undergraduate and postgraduate coursework students.
Undergraduate students are defined where element E310 takes the value of 10, 13, 20, 21, 22 (levels 8 and 9, Bachelor’s Graduate Entry and Bachelor’s Honours are not included) and postgraduate coursework students are defined where element E310 Course type code takes the value of 4,5,6,7 or 11. Commencing students are defined where element E922 takes the value of 1 and non-commencing students where E922 takes the value of 2.
We conduct an Ordinary Least Squares regression with the progress rate as the dependent variable. The individual observations are the proportion of student load passed by the individual student. Therefore observations will range in value between 0 and 1.
Independent variables
Independent variables are defined as follows:
Students unit record enrolment file
Gender from element E315, entered as a dummy variable where 1=Female, 0=Male.
Age from element E997 sorted into age groups, less than 19, 20 to 24, 25 to 29 and 30 plus. Entered as a set of dummy variables.
NESB – from element E348, entered as a dummy variable, where 1=from a non- English speaking background (for values of E348 within the range 1000 to 9998) and 0= otherwise (for values of E348 = 0001, English speaking background).
Indigenous – from element E316, entered as a dummy variable, where 1 = Indigenous and 0 = non-Indigenous.
Disability - from element E386 where graduates are categorized as disabled if the value in the first field of this element is 1. Entered as a dummy variable, where 1 = disabled, 0 = non-disabled.
Broad field of study – from element E311 using the standard broad fields of study, agriculture, architecture, arts/humanities, business, education, engineering, health (with nursing separately identified), law, science and veterinary science, entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular broad field of study and 0 for remaining broad fields of study.
Level of course – from element E310.
For the undergraduate regressions, the following courses are included
10 = Bachelor’s Pass
13 = Associate Degree
20 = Advanced Diploma
21 = Diploma
22 = Other award course
For the postgraduate coursework regression, the following courses are included
4 = Master’s by Coursework
5 = Postgraduate Qualifying
6 = Graduate Diploma/Postgraduate Diploma in ‘new area’
7 = Graduate Diploma/Postgraduate Diploma in ‘existing area’
11 = Graduate Certificate
In each regression, the level variables are entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular level of study and 0 for remaining levels of study.
Mode of attendance (internal/external) - from element E329, entered as a dummy variable, where 1 = Internal and 0 = External.
Type of attendance (full-time/part-time) - from element E330, entered as a dummy variable, where 1 = Full-time and 0 = Part-time.
New to higher education – variable only used for the undergraduate commencing students regression. From element E924, entered as a dummy variable taking the value of 1 where element E924 = 2 (new to higher education) and the value of 0 otherwise.
Basis of admission and Tertiary Entrance Score – variable only used for the undergraduate commencing students regression. This was a set of dummy variables constructed using the Basis for admission element E327 and Tertiary Entrance Score element E369.
Basis of admission entered as a set of dummy variables for each basis of admission for the following codes 11, 12, 16, 17, 18, 19, 20, 21, 22 and 29 (and code 13 in 1999 and its equivalent of combining codes 21 and 22 in 2000), taking the value of 1 for the relevant basis of admission and 0 otherwise.
For the basis of admission 14 and 15, that is admitted on the basis of satisfactory completion of secondary education at school or TAFE, these data were further disaggregated to allow for the influence of TER score. Within each basis of admission, a series of seven dummy variables were constructed for each TER decile score where element E369 was greater than or equal to 30 and less than 40 and so on with the last decile being greater than or equal to 90 and less than or equal to 100.
Size and field of study – Number of students by institution and broad field of study (see above for broad fields of study) derived from element E929. Entered as a continuous variable.
Institution – a set of institutional dummy variables where each dummy variable takes the value of 1 for a particular institution and 0 for remaining institutions.
1996 Census of Population and Housing
Socio-economic status -
- Index of education and occupation as defined by parents’ education and occupation background. This is an index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
- Index of economic resources as defined by income. Index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
Socioeconomic indexes by postcode from the 1996 Census are matched with the student’s postcode as given by element E320. Overseas students (without a postcode) are assigned an index value of zero.
Locality status –
- Rural status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census.
- Isolated status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census.
Entered as a set of dummy variables, taking the value of 1 for the particular locality and 0 for all other localities.
top
These indicators show the proportion of graduates in full-time employment as a percentage of graduates available for full-time work. From a domestic perspective it seems more appropriate to focus on the employment outcomes of Australian graduates only. These indicators refer only to graduates who were previously full-time students to avoid any bias that might arise as a result of part-time and external students being more likely to be in continuing employment.
Data:
1998, 1999 and 2000 Graduate Destination Survey file (survey year).
Note 2000 GDS refers to graduates completing their studies in 1999.
Filter:
Only Bachelor Pass and Honours graduates are examined that is where LEVEL takes the value of 1 or 2.
Only graduates that attended full-time are included that is where ATTEND takes the value of 1. (Up to and including the 1998 GDS, the ATTEND variable had four values, 1 = wholly full-time, 2 = wholly part-time, 3= wholly external and 4 = some combination of the above. From the 1999 GDS onwards, the ATTEND variable had two values, 1 = full-time and 2 = part-time and the new MODE variable had two values, 1= internal, 2 = external. Before and after this change, all records where ATTEND does not take the value of 1 are deleted).
Only include Australian graduates that is where PERMRES takes the value of 1.
Graduates who were provided time-off in lieu by their employer are excluded, that is, where FYTOFF takes the value of 1.
Graduates who do not indicate a major field of study are excluded, that is where MAJ1=missing these records are deleted.
Indicator:
Graduates in full-time employment are defined where EMPTOT takes the value of 1 (see attachment for derivation of EMPTOT from the Graduate Destination Survey).
Graduates available for full-time employment are defined as graduates in full-time employment (see above) plus graduates seeking full-time work but not in full-time employment. The latter group includes two categories. First, graduates not engaged in full-time study and who are working part-time but seeking full-time work (STUDY is not equal to 1 and ACTIVITY equals 3). The second category includes graduates not in full-time study who are not working and are seeking full-time work only or any work (STUDY not equal to 1 and ACTIVITY equals 5 or 7).
The graduate full-time employment rate (employment rate for short) is defined as the number of graduates in full-time employment as a percentage of graduates available for full-time employment.
We conduct an Ordinary Least Squares regression with the graduate full-time employment rate as the dependent variable (if the graduate is in full-time employment, the dependent variable takes the value of 1 and 0 otherwise).
Independent variables
Independent variables are defined as follows:
GDS file
Sex – from the SEX variable in the GDS file, entered as a dummy variable where 1=Female, 0=Male.
Age from the AGE variable in the GDS file and sorted into age groups, less than 19, 20 to 24, 25 to 29 and 30 plus. Entered as a set of four dummy variables.
NESB – from the NESB variable in the GDS file, entered as a dummy variable, where 1=from an non- English speaking background and 0= other.
Indigenous – from the ATSI variable in the GDS file, entered as a dummy variable, where 1 = Indigenous and 0 = non-Indigenous.
Disability - from the DISAB variable in the GDS file where graduates are categorized as disabled if they respond to categories 1, 2 or 3 to the disability question, entered as a dummy variable, where 1 = disabled, 0 = non-disabled.
Broad field of study – from the MAJ1 variable in the GDS file using the standard broad fields of study, agriculture, architecture, arts/humanities, business, education, engineering, health (with nursing separately identified), law, science and veterinary science, entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular broad field of study and 0 for remaining broad fields of study.
Level of course – from the LEVEL variable in the GDS file, entered as a dummy variable, where 1=Bachelor Pass and 0=Bachelor Honours.
Type of enrolment – Full-time, part-time and external defined as follows. Full-time defined where ATTEND = 1 (full-time) and MODE = 1 (internal). Part-time defined where ATTEND = 2 (part-time) and MODE=1 (internal). External defined where MODE=2(external). Prior to 2000, full-time defined where ATTEND=1 (wholly full-time), part-time defined where ATTEND=2 (wholly part-time) and external where ATTEND = 3 (wholly external). Entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular type of attendance and 0 for all other types of attendance.
Size and field of study – Number of GDS respondents by institution and broad field of study (see above for broad fields of study) from GDS file. Entered as a continuous variable.
Institution – a set of institutional dummy variables where each dummy variable takes the value of 1 for a particular institution and 0 for remaining institutions.
1996 Census of Population and Housing
Socio-economic status -
- Index of education and occupation as defined parents’ education and occupation background. This is an index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
- Index of economic resources as defined by income. Index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
Socioeconomic indexes by postcode from the 1996 Census are matched with the graduates’ home postcode as defined by the PCODE variable from the GDS file.
Locality status –
- Rural status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census. Rural status by postcode is then matched with graduates’ home postcode as defined by the PCODE variable from the GDS file.
- Isolated status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census. Isolated status by postcode is then matched with graduates’ home postcode as defined by the PCODE variable from the GDS file.
Entered as a set of dummy variables, taking the value of 1 for the particular locality and 0 for all other localities.
Higher Education Unit Record Student Enrolment file
Average TER by institution and field of study – Average TER from Element E369 for commencing students with a TER score, by institution and by broad field of study. Entered as a continuous variable.
DEWRSB, Small Area Labour Market Statistics data
Labour market conditions - Unemployment rate for the June quarter of the relevant year at the postcode level. These are derived in the following manner. Unemployment rates by Statistical Local Area (SLA) are derived from DEWRSB, Small Area Labour Market Statistics. Unemployment rates by postcode are derived using an SLA-postcode conversion based on 1996 Census of Population and Housing population data. Unemployment rates by postcode are then matched with graduates’ postcode as defined by the PCODE variable from the GDS file. Entered as a continuous variable.
top
These indicators show the proportion of graduates proceeding to full-time study. From a domestic perspective it seems more appropriate to focus on the destinations of Australian graduates only. These indicators refer to Bachelor Honours and Bachelor Pass level graduates only.
Data:
1998, 1999 and 2000 Graduate Destination Survey file (survey year).
Note 2000 GDS refers to graduates completing their studies in 1999.
Filter:
Only Bachelor Honours and Pass graduates are examined that is where LEVEL takes the value of 1 and 2.
Only include Australian graduates that is where PERMRES takes the value of 1.
Graduates who do not indicate a major field of study are excluded, that is where MAJ1=missing these records are deleted.
Indicator:
The further study rate is defined as the number of graduates proceeding to full-time study as a percentage of all graduates.
Graduates proceeding to full-time study are defined where FTSTUDY takes the value of 1.
We conduct an Ordinary Least Squares regression with the graduate further study rate as the dependent variable (if the graduate is in further study, the variable takes the value of 1 and 0 otherwise).
Independent variables
Independent variables are defined as follows:
GDS file
Sex – from the SEX variable in the GDS file, entered as a dummy variable where 1=Female, 0=Male.
Age from the AGE variable in the GDS file and sorted into age groups, less than 19, 20 to 24, 25 to 29 and 30 plus. Entered as a set of four dummy variables.
NESB – from the NESB variable in the GDS file, entered as a dummy variable, where 1=from an non- English speaking background and 0= other.
Indigenous – from the ATSI variable in the GDS file, entered as a dummy variable, where 1 = Indigenous and 0 = non-Indigenous.
Disability - from the DISAB variable in the GDS file where graduates are categorized as disabled if they respond to categories 1, 2 or 3 to the disability question, entered as a dummy variable, where 1 = disabled, 0 = non-disabled.
Broad field of study – from the MAJ1 variable in the GDS file using the standard broad fields of study, agriculture, architecture, arts/humanities, business, education, engineering, health (with nursing separately identified), law, science and veterinary science, entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular broad field of study and 0 for remaining broad fields of study.
Level of course – from the LEVEL variable in the GDS file, entered as a dummy variable, where 1=Bachelor Pass and 0=Bachelor Honours.
Type of enrolment – Full-time, part-time and external defined as follows. Full-time defined where ATTEND = 1 (full-time) and MODE = 1 (internal). Part-time defined where ATTEND = 2 (part-time) and MODE=1 (internal). External defined where MODE=2(external). Prior to 2000, full-time defined where ATTEND=1 (wholly full-time), part-time defined where ATTEND=2 (wholly part-time) and external where ATTEND = 3 (wholly external). Entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular type of attendance and 0 for all other types of attendance.
Size and field of study – Number of GDS respondents by institution and broad field of study (see above for broad fields of study) from GDS file. Entered as a continuous variable.
Institution – a set of institutional dummy variables where each dummy variable takes the value of 1 for a particular institution and 0 for remaining institutions.
1996 Census of Population and Housing
Socio-economic status -
- Index of education and occupation as defined parents’ education and occupation background. This is an index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
- Index of economic resources as defined by income. Index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
Socioeconomic indexes by postcode from the 1996 Census are matched with the graduates’ home postcode as defined by the PCODE variable from the GDS file.
Locality status –
- Rural status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census. Rural status by postcode is then matched with graduates’ home postcode as defined by the PCODE variable from the GDS file.
- Isolated status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census. Isolated status by postcode is then matched with graduates’ home postcode as defined by the PCODE variable from the GDS file.
Entered as a set of dummy variables, taking the value of 1 for the particular locality and 0 for all other localities.
Higher Education Unit Record Student Enrolment file
Average TER by institution and field of study – Average TER from Element E369 for commencing students with a TER score, by institution and by broad field of study. Entered as a continuous variable.
DEWRSB, Small Area Labour Market Statistics data
Labour market conditions - Unemployment rate for the June quarter of the relevant year at the postcode level. These are derived in the following manner. Unemployment rates by Statistical Local Area (SLA) are derived from DEWRSB, Small Area Labour Market Statistics. Unemployment rates by postcode are derived using an SLA-postcode conversion based on 1996 Census of Population and Housing population data. Unemployment rates by postcode are then matched with graduates’ postcode as defined by the PCODE variable from the GDS file. Entered as a continuous variable.
top
These indicators show graduates’ mean starting salaries and this represents one important dimension of post-graduation outcomes. From a domestic perspective it seems more appropriate to focus on the salary outcomes of Australian graduates only. These indicators refer only to graduates who were previously full-time students and who are in their first full-time job to avoid any bias that might arise as a result of part-time and external students being more likely to be in continuing employment.
Data:
1998, 1999 and 2000 Graduate Destination Survey file (survey year).
Note 2000 GDS refers to graduates completing their studies in 1999.
Filter:
Only Bachelor Pass and Honours graduates are examined that is where LEVEL takes the value of 1 or 2.
Only include Australian graduates that is where PERMRES takes the value of 1.
Only graduates that attended full-time are included that is where ATTEND takes the value of 1. (Up to and including the 1998 GDS, the ATTEND variable had four values, 1 = wholly full-time, 2 = wholly part-time, 3= wholly external and 4 = some combination of the above. From the 1999 GDS onwards, the ATTEND variable had two values, 1 = full-time and 2 = part-time and the new MODE variable had two values, 1= internal, 2 = external. Before and after this change, all records where ATTEND does not take the value of 1 are deleted).
Only graduates in their first full-time job that is where FIRSTFT takes the value of 1.
Graduates who were provided time-off in lieu by their employer are excluded, that is, where FYTOFF takes the value of 1.
Graduates who do not indicate a major field of study are excluded, that is where MAJ1=missing these records are deleted.
Indicator:
The starting salary indicator refers to average salaries paid to graduates in each institution derived from the SALARY variable.
We conduct an Ordinary Least Squares regression with the graduate starting salary as the dependent variable entered as a continuous variable.
top
Independent variables
Independent variables are defined as follows:
GDS file
Sex – from the SEX variable in the GDS file, entered as a dummy variable where 1=Female, 0=Male.
Age from the AGE variable in the GDS file and sorted into age groups, less than 19, 20 to 24, 25 to 29 and 30 plus. Entered as a set of four dummy variables.
NESB – from the NESB variable in the GDS file, entered as a dummy variable, where 1=from an non- English speaking background and 0= other.
Indigenous – from the ATSI variable in the GDS file, entered as a dummy variable, where 1 = Indigenous and 0 = non-Indigenous.
Disability - from the DISAB variable in the GDS file where graduates are categorized as disabled if they respond to categories 1, 2 or 3 to the disability question, entered as a dummy variable, where 1 = disabled, 0 = non-disabled.
Broad field of study – from the MAJ1 variable in the GDS file using the standard broad fields of study, agriculture, architecture, arts/humanities, business, education, engineering, health (with nursing separately identified), law, science and veterinary science, entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular broad field of study and 0 for remaining broad fields of study.
Level of course – from the LEVEL variable in the GDS file, entered as a dummy variable, where 1=Bachelor Pass and 0=Bachelor Honours.
Type of enrolment – Full-time, part-time and external defined as follows. Full-time defined where ATTEND = 1 (full-time) and MODE = 1 (internal). Part-time defined where ATTEND = 2 (part-time) and MODE=1 (internal). External defined where MODE=2(external). Prior to 2000, full-time defined where ATTEND=1 (wholly full-time), part-time defined where ATTEND=2 (wholly part-time) and external where ATTEND = 3 (wholly external). Entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular type of attendance and 0 for all other types of attendance.
Size and field of study – Number of GDS respondents by institution and broad field of study (see above for broad fields of study) from GDS file. Entered as a continuous variable.
Institution – a set of institutional dummy variables where each dummy variable takes the value of 1 for a particular institution and 0 for remaining institutions.
1996 Census of Population and Housing
Socio-economic status -
- Index of education and occupation as defined parents’ education and occupation background. This is an index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
- Index of economic resources as defined by income. Index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
Socioeconomic indexes by postcode from the 1996 Census are matched with the graduates’ home postcode as defined by the PCODE variable from the GDS file.
Locality status –
- Rural status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census. Rural status by postcode is then matched with graduates’ home postcode as defined by the PCODE variable from the GDS file.
- Isolated status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census. Isolated status by postcode is then matched with graduates’ home postcode as defined by the PCODE variable from the GDS file.
Entered as a set of dummy variables, taking the value of 1 for the particular locality and 0 for all other localities.
Higher Education Unit Record Student Enrolment file
Average TER by institution and field of study – Average TER from Element E369 for commencing students with a TER score, by institution and by broad field of study. Entered as a continuous variable.
DEWRSB, Small Area Labour Market Statistics data
Labour market conditions - Unemployment rate for the June quarter of the relevant year at the postcode level. These are derived in the following manner. Unemployment rates by Statistical Local Area (SLA) are derived from DEWRSB, Small Area Labour Market Statistics. Unemployment rates by postcode are derived using an SLA-postcode conversion based on 1996 Census of Population and Housing population data. Unemployment rates by postcode are then matched with graduates’ postcode as defined by the PCODE variable from the GDS file. Entered as a continuous variable.
top
These indicators show graduates’ satisfaction with three aspects of their courses, overall satisfaction, good teaching and generic skills as measured by the Course Experience Questionnaire (CEQ).
Since the indicators refer to satisfaction with the delivery of educational services it seems appropriate to have a broad focus on all graduates and not just Australian graduates. The graduate satisfaction indicators refer to Course Experience Questionnaire (CEQ) scores for Bachelor Honours, Bachelor Pass and Undergraduate Diploma graduates.
From the 1997 GDS onwards, the survey asked graduates about their satisfaction in each course (major) where graduates undertook multiple courses (majors). Previously the CEQ only asked graduates to rate their satisfaction in their first course (major). Since the focus of this indicator is with students’ satisfaction and not course satisfaction, the performance indicator refers only to satisfaction recorded for the first course (major). Otherwise graduate satisfaction would be ‘double counted’ for graduates studying in more than one course (major).
Data:
1998, 1999 and 2000 Graduate Destination Survey file (survey year).
Note 2000 GDS refers to graduates completing their studies in 1999.
Filter:
Only Undergraduate Diploma, Bachelor Pass and Bachelor Honours graduates are examined that is where LEVEL takes the value of 9, 1 and 2 respectively.
Records are excluded where there is no response in terms of the overall satisfaction item, that is OALL1=missing. Similarly, missing values to more than two of the six items for the generic skills scale results in that record being deleted and likewise for the good teaching scale.
Graduates who do not indicate a major field of study are excluded, that is where MAJ1=missing these records are deleted.
Indicator:
Satisfaction is measured by the percentage of graduates that ‘broadly agree’ with statements in the CEQ. That is, graduates responding 3, 4 or 5 respectively to statements in the questionnaire as a percentage of all respondents.
‘Broad agreement’ for overall satisfaction is measured by the percentage of graduates responding, 3,4 or 5 to the overall satisfaction item, OALL1.
‘Broad agreement’ for the generic skills scale and good teaching scale is measured by the percentage of responses, 3,4 or 5 to the 6 constituent items in the generic skills scale (course2, course5, course9, course10, course11, course22) and the good teaching scale (teach3, teach7, teach15, teach17, teach18, teach20).
We conduct an Ordinary Least Squares regression where graduates form the individual observations. For overall satisfaction, the graduate is recorded as being either ‘broadly satisfied’ (the dependent variable takes the value of 1) or not ‘broadly satisfied’ (where the dependent variable takes the value of 0). For the generic skills and good teaching scales we record for each graduate the proportion of responses to constituent items in each scale where the graduate is ‘broadly satisfied’.
Independent variables
Independent variables are defined as follows:
GDS file
Sex – from the SEX variable in the GDS file, entered as a dummy variable where 1=Female, 0=Male.
Age from the AGE variable in the GDS file and sorted into age groups, less than 19, 20 to 24, 25 to 29 and 30 plus. Entered as a set of four dummy variables.
NESB – from the NESB variable in the GDS file, entered as a dummy variable, where 1=from an non- English speaking background and 0= other.
Indigenous – from the ATSI variable in the GDS file, entered as a dummy variable, where 1 = Indigenous and 0 = non-Indigenous.
Disability - from the DISAB variable in the GDS file where graduates are categorized as disabled if they respond to categories 1, 2 or 3 to the disability question, entered as a dummy variable, where 1 = disabled, 0 = non-disabled.
Broad field of study – from the MAJ1 variable in the GDS file using the standard broad fields of study, agriculture, architecture, arts/humanities, business, education, engineering, health (with nursing separately identified), law, science and veterinary science, entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular broad field of study and 0 for remaining broad fields of study.
Level of course – from the LEVEL variable in the GDS file, entered as a dummy variable, where 1=Bachelor Pass and 0=Bachelor Honours.
Type of enrolment – Full-time, part-time and external defined as follows. Full-time defined where ATTEND = 1 (full-time) and MODE = 1 (internal). Part-time defined where ATTEND = 2 (part-time) and MODE=1 (internal). External defined where MODE=2(external). Prior to 2000, full-time defined where ATTEND=1 (wholly full-time), part-time defined where ATTEND=2 (wholly part-time) and external where ATTEND = 3 (wholly external). Entered as a set of dummy variables, where each dummy variable takes the value of 1 for a particular type of attendance and 0 for all other types of attendance.
Size and field of study – Number of GDS respondents by institution and broad field of study (see above for broad fields of study) from GDS file. Entered as a continuous variable.
Institution – a set of institutional dummy variables where each dummy variable takes the value of 1 for a particular institution and 0 for remaining institutions.
1996 Census of Population and Housing
Socio-economic status -
- Index of education and occupation as defined parents’ education and occupation background. This is an index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
- Index of economic resources as defined by income. Index defined at the postcode level from ABS, 2039.0, 1996 Census of Population and Housing, Information Paper, Socioeconomic Indexes for Areas. Entered as a continuous variable.
Socioeconomic indexes by postcode from the 1996 Census are matched with the graduates’ home postcode as defined by the PCODE variable from the GDS file. Overseas students (without a postcode) are assigned an index value of zero.
Locality status –
- Rural status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census. Rural status by postcode is then matched with graduates’ home postcode as defined by the PCODE variable from the GDS file.
- Isolated status by postcode as defined by Australian Institute of Health and Welfare, February 2000, Rural, Remote, Metropolitan Areas Classification using data from the 1996 Census. Isolated status by postcode is then matched with graduates’ home postcode as defined by the PCODE variable from the GDS file.
Entered as a set of dummy variables (RURAL and ISOLATED), taking the value of 1 for the particular locality and 0 for all other localities.
Higher Education Unit Record Student Enrolment file
Average TER by institution and field of study – Average TER from Element E369 for commencing students with a TER score, by institution and by broad field of study. Entered as a continuous variable.
top
Derivation of EMPTOT variable from the Graduate Destination Survey
Graduates in full-time employment are those where the variable EMPTOT takes the value of one.
The EMPTOT variable is derived from the STUDY and ACTIVITY variables. However, there are a number of consistency checks which synthesise the responses from the STUDY and ACTIVITY variables. The variables EMPTYPE1, EMPLOYER, INDUSTRY, DEST and DEST1 perform this task. DEST1 produces the following variables GOVT, PINC, HEALTH, EDUC and OTHER and these variables combined generate estimates of full-time employment, that is, where EMPTOT takes the value of 1.
The SPSS code used in the Graduate Destination Survey to generate the EMPTOT variable is shown below.
* FTWORK - aggregate Q4c
- 1 GOVT 2 PRIV 3 EDUC 4 OTHER
if (((activity eq 1) or (activity eq 2)) and (study ne 1)) ftwork = emptype
if (((activity eq 1) or (activity eq 2)) and (study ne 1)
and (emptype eq 0)) ftwork = 9
if (((activity eq 1) or (activity eq 2)) and (study ne 1)
and (missing(emptype))) ftwork = 9
recode ftwork (1 thru 3 = 1) (6, 7 = 2) (4, 5 = 3) (8, 9 = 4).
value labels ftwork 1 'GOVT' 2 'PRIV' 3 'EDUC' 4 'OTHER'.
variable labels ftwork 'WORK FT'.
* DEST VARIABLE - PRESENT DESTINATIONS AS A SINGLE VARIABLE
* First create a new variable - employer - for all respondents
* to mirror previous employer codes.
* make everyone equal 'other, NI'.
compute employer = 120.
if (emptype eq 1) employer = 1.
if (emptype eq 2) employer = 4.
if (emptype eq 3) employer = 6.
if ((emptype eq 4) or (emptype eq 5)) employer = 110.
if ((emptype eq 6) or (emptype eq 7)) employer =8.
* Now fix up anomalies.
* Fed govt/hi ed to Hi ed.
if ((emptype eq 1) and ((industry eq 126) or (industry eq 127)))
employer = 100.
* Fed govt/schools to Pub ed.
if ((emptype eq 1) and ((industry ge 121) and (industry le 125)))
employer = 90.
* Fed govt/other ed to Other ed.
if ((emptype eq 1) and (industry eq 120)) employer = 110.
* State govt/hi ed to Hi ed.
if ((emptype eq 2) and ((industry eq 126) or (industry eq 127)))
employer = 100.
* State govt/schools to Pub ed.
if ((emptype eq 2) and ((industry ge 121) and (industry le 125)))
employer = 90.
* State govt/other ed to Other ed.
if ((emptype eq 2) and (industry eq 120)) employer = 110.
* Priv Sect/hi ed to Hi ed.
if ((emptype eq 6) and ((industry eq 126) or (industry eq 127)))
employer = 100.
* Priv Sect/schools to Pub ed.
if ((emptype eq 6) and ((industry ge 121) and (industry le 125)))
employer = 90.
* Priv Sect/other ed to Other ed.
if ((emptype eq 6) and (industry eq 120)) employer = 110.
* Priv Sect/ed to Priv Ed.
if ((emptype eq 6) and ((industry ge 120) and (industry le 125)))
emptype = 5.
if ((emptype eq 6) and ((industry eq 126) or (industry eq 127)))
employer = 100.
* Self Emp/ed to Other Ed.
if ((emptype eq 7) and ((industry ge 120) and (industry le 127)))
employer = 110.
if ((employer eq 90) and ((industry eq 126) or (industry eq 127)))
employer = 100.
if ((employer eq 90) and (industry lt 120)) employer = 110.
if ((employer eq 90) and (industry gt 127)) employer = 110.
if ((employer eq 90) and (industry eq 120)) employer = 110.
if ((employer eq 120) and ((industry eq 126) or (industry eq 127)))
employer = 100.
if ((emptype eq 120) and ((industry ge 120) and (industry le 125)))
employer = 90.
if ((employer eq 120) and (industry lt 120)) employer = 110.
if ((employer eq 120) and (industry gt 127)) employer = 110.
if ((employer eq 120) and (industry eq 120)) employer = 110.
* Split education up on basis of industry code & duties code.
if ((employer eq 110) and ((industry eq 126) or (industry eq 127)))
employer = 100.
if ((employer eq 110) and ((industry ge 121) and (industry le 125)))
employer = 90.
if ((employer eq 110) and (duties eq 074)) employer = 100.
if ((employer eq 110) and ((duties ge 070) and (duties le 073)))
employer = 90.
if (employer eq 110) industry = 120.
compute green = 2.
if ((industry ge 120) and (industry le 127)) green = 1.
if ((emptype eq 5) and (green eq 2)) industry = 120
* create Defence category.
if (industry eq 116) employer = 2.
* these 4 transformations pull prof priv pract out of ind/comm.
if ((emptype eq 6) and ((industry ge 101) and (industry le 104)))
employer = 7.
if ((emptype eq 7) and ((industry ge 101) and (industry le 104)))
employer = 7.
if ((emptype eq 6) and ((industry eq 132) or (industry eq 133)))
employer = 7.
if ((emptype eq 7) and ((industry eq 132) or (industry eq 133)))
employer = 7.
if ((missing(emptype)) and ((industry ge 10) and (industry le 100)))
employer = 8.
if ((missing(emptype)) and ((industry ge 101) and (industry le 104)))
employer = 7.
if ((missing(emptype)) and ((industry eq 132) or (industry eq 133)))
employer = 7.
if ((missing(emptype)) and (industry eq 111)) employer = 1.
if ((missing(emptype)) and (industry eq 112)) employer = 4.
if ((missing(emptype)) and ((industry ge 121) and (industry le 125)))
employer = 90.
if ((missing(emptype)) and ((industry eq 126) or (industry eq 127)))
employer = 100.
if ((missing(emptype)) and (industry eq 120)) employer = 110.
if ((missing(industry)) and (emptype eq 1)) employer = 1.
if ((missing(industry)) and (emptype eq 2)) employer = 4.
if ((missing(industry)) and (emptype eq 3)) employer = 6.
if ((missing(industry)) and (emptype eq 4)) employer = 110.
if ((missing(industry)) and (emptype eq 5)) employer = 110.
if ((missing(industry)) and (emptype eq 1)) employer = 1.
if (((emptype eq 1) or (emptype eq 2) or (emptype eq 3)) and
((industry eq 130) or (industry eq 131))) employer = 80.
if ((emptype eq 6) and ((industry eq 130) or (industry eq 131)))
employer = 81.
if (((emptype eq 7) or (emptype eq 8) or (emptype eq 9) or
(missing(emptype))) and ((industry eq 130) or (industry eq 131)))
employer = 81.
if ((employer eq 120) and ((industry eq 130) or (industry eq 131)))
employer = 81.
if ((employer eq 90) and (emptype eq 5)) employer = 91.
value labels employer
1 'Aust. Govt' 4 'State Govt' 2 'Defence' 6 'Local Govt'
7 'Prof. Practice.' 8 'Ind.\Comm.'
80 'Health, Public' 81 'Health, Priv\Oth'
90 'Schools, Pub.' 91 'Schools, Priv' 100 'Higher Ed.' 110 'Educ Other, NEI'
120 'Oth Emp or Not Ind'/
* compute dest = employer
if ((employer eq 8) and (industry eq 10)) dest = 10.
if ((employer eq 8) and (industry eq 20)) dest = 11.
if ((employer eq 8) and (industry eq 30)) dest = 12.
if ((employer eq 8) and (industry eq 40)) dest = 13.
if ((employer eq 8) and (industry eq 50)) dest = 14.
if ((employer eq 8) and (industry eq 60)) dest = 15.
if ((employer eq 8) and (industry eq 70)) dest = 19.
if ((employer eq 8) and (industry eq 140)) dest = 19.
if ((employer eq 8) and (industry eq 80)) dest = 16.
if ((employer eq 8) and (industry eq 90)) dest = 17.
if ((employer eq 8) and (industry eq 100)) dest = 18.
if ((employer eq 8) and (industry eq 150)) dest = 20.
if ((employer eq 8) and (industry eq 160)) dest = 20.
if (employer eq 8) dest = 20.
value labels dest
1 'Aust. Govt' 4 'State Govt' 2 'Defence' 6 'Local Govt'
7 'Prof. Practice.' 8 'Ind.\Comm.'
10 'Ag\Forest\Fish' 11 'Mining' 12 'Manufacturing' 13 'Elec\Gas\Water'
14 'Construction' 15 'W\sale, retail' 16 'Transp\Storage' 17 'Comm Servs'
18 'Business, Finance' 19 'Ent\Recreation' 20 'Personal\Other Servs'
80 'Health, Public' 81 'Health, Priv\Oth'
90 'Schools, Pub.' 91 'Schools, Priv' 100 'Higher Ed.' 110 'Educ Other, NEI'
120 'Oth Emp or Not Ind'/
value labels dest
1 'Aust. Govt' 2 'Defence' 4 'State Govt' 6 'Local Govt'
7 'Prof. Practice.' 8 'Ind.\Comm.'
10 'Ag\Forest\Fish' 11 'Mining' 12 'Manufacturing' 13 'Elec\Gas\Water'
14 'Construction' 15 'W\sale, retail' 16 'Transp\Storage' 17 'Comm Servs'
18 'Business, Finance' 19 'Ent\Recreation' 20 'Personal\Other Servs'
80 'Health, Public' 81 'Health, Priv\Oth'
90 'Schools, Pub.' 91 'Schools, Priv' 100 'Higher Ed.' 110 'Educ Other, NEI'
120 'Oth Emp or Not Ind'
121 'Wk pt not sk ft' 122 'Wk pt sk ft'
123 'Not wk seek ft' 124 'Not wk seek other'
125 'Hons' 126 'Higher Deg' 127 'Oth Deg\Dip'
128 'Teach Train' 129 'Other Study'
130 'Unavailable'/
* AGGREGATE DESTA INTO DEST VARIABLE.
compute dest1 = dest
recode dest1 (1, 2 = 50) (4 = 51) (6 = 52) (7 = 53) (10 thru 20 = 54)
(80 = 55) (81 = 56) (90, 91 = 57) (100 = 58) (110 = 59) (120 = 60)
(122 = 61) (123 = 62) (121 = 63) (124 = 64) (130 = 65)
(125 = 1) (126 = 2) (127 = 3) (128 = 4) (129 = 5)
value labels dest1 50 'Aust. Govt' 51 'State Govt' 52 'Local Govt'
53 'Prof. Practice.' 54 'Ind.\Comm.' 55 'Health Pub.' 56 'Health Priv.'
57 'Schools' 58 'Higher Ed.' 59 'Educ Other, NEI' 60 'Oth Emp or Not Ind'
61 'Work P\T, Seek F\T' 62 'Not Working, Seek F\T'
63 'Work P\T, Not Seek F\T' 64 'Not Work, Seek P\T Only' 65 'Unavailable'
1 'Honours' 2 'Higher Deg' 3 'Other Deg\Dip' 4 'Teacher Training'
5 'Other F\Study'/
if (dest1 eq 50) govt = 1.
if (dest1 eq 51) govt = 2.
if (dest1 eq 52) govt = 3.
value labels govt 1 'Aust. Govt' 2 'State Govt' 3 'Local Govt'/.
variable labels govt ''.
if (dest1 eq 53) pinc = 1.
if (dest1 eq 54) pinc = 2.
value labels pinc 1 'Prof. Practice' 2 'Ind.\Comm.'/.
variable labels pinc ''.
if (dest1 eq 55) health = 1.
if (dest1 eq 56) health = 2.
value labels health 1 'Health Pub.' 2 'Health Priv'/.
variable labels health ''.
if (dest1 eq 57) educ = 1.
if (dest1 eq 58) educ = 2.
if (dest1 eq 59) educ = 3.
value labels educ 1 'Schools' 2 'Higher Education' 3 'Educ. Other'/.
variable labels educ ''.
if (dest1 eq 60) other = 1.
value labels other 1 'Other Emp.\NEI'/.
variable labels other ''.
if (not missing (govt)) emptot = 1.
if (not missing (pinc)) emptot = 1.
if (not missing (health)) emptot = 1.
if (not missing (educ)) emptot = 1.
if (not missing (other)) emptot = 1.
variable labels emptot 'TOTAL FT EMPLOYMENT'/.
value labels emptot 1 ''.