Higher Education Attrition Rates 1994-2002: A Brief Overview

• The 2002 attrition rate for all domestic students and all international students is the lowest since 1994.
  • The attrition rate for international students has declined to a greater extent than that for domestic students over the period.
• Attrition rates vary considerably across institutions and student groups.
  • School leavers commencing in undergraduate courses have a lower attrition rate than other undergraduate commencers
  • Domestic students generally have a higher attrition rate than their international counterparts
  • Postgraduate students have a higher attrition rate than undergraduate students
• Rates in the first year after commencement of a course are around double that of those in the second year.

Attrition rates are one of a range of indicators of outcomes used to measure performance of institutions in the higher education sector. As such they should be viewed in conjunction with other indicators to obtain a full picture of performance¹. Attrition rates provide a measure of the proportion of students who ‘drop out’ of an award course at an institution each year.

It is important for policy and program development to know if particular groups or categories of students are more inclined to drop out of higher education than others, and whether attrition is greater at particular institutions, in particular fields of education or in particular course types than in others. However, the interpretation of attrition rates is not simple. The benefits that flow from completing a university degree are well known. Participating in higher education and completing units of study may also have possible positive benefits for individuals. Whether this is actually the case, and the type and level of benefit for individuals who partially complete a course, is not known.

Attrition rates (first year) measure the proportion of students in a particular year who neither graduate nor continue studying in an award course at the same institution in the following year. That is, they provide one measure of the proportion of students who ‘drop out’ from one year to another. As an example, an attrition rate of 16% for the University of Sydney in 2002 implies that 16% of students who were enrolled in an award course at the University of Sydney in 2002 were not enrolled at that university in an award course in 2003, and their non-enrolment was not due to completion of their course in 2002. Using this method of calculation, the attrition rate plus the retention rate plus the completion rate for a given student population in a given year will equal 100 percent.

The attrition rates provided in this paper are simple measures of attrition at an institution level. The calculation of the rates is reliant on matching a student Identification Code which is unique only within a university. Thus students who leave a course at one university and enrol the next year at another university are included in the attrition number.

In addition, those students who leave university without completing their course, but who return later to the same university, are also counted as part of attrition.

Attrition in the second year following the reference year has also been calculated for undergraduate commencing students. The crude attrition rate for the second year measures the proportion of students in a year who neither graduate nor continue studying at the same institution in the year after the following year (independent of what the student did in the following year). For example, the 2001 figure refers to the proportion of 2001 students who neither graduated in 2001 or 2002 nor were enrolled at the same university in 2003. They may or may not have been enrolled in 2002. The purpose of this data is to facilitate the identification of the most common patterns of attrition.

The aim of this paper is to provide a starting point for examining attrition from higher education institutions by providing crude attrition rates for a number of broad groups of students. It highlights the variations in attrition rates within the student population by providing separate rates for a number of student groups, including commencing students, commencing new to higher education students, undergraduates, postgraduates, domestic and international students, school leavers, sex and for a number of age categories.

By examining first and second year attrition rates it also provides a clear indication that students who do ‘drop out’ of a higher education institution do so early in their course.

This paper does not seek to explain the differences in attrition rates across student population groups or institutions. Nor does it explore the main factors that drive attrition rates and how these factors interact. These are quite complex issues that require different statistical techniques and additional analyses to address properly.

The crude attrition rates are provided in Appendix 1 and Appendix 2. These should be examined in conjunction with the Notes and definitions provided in Appendix 3. The main points to note from the tables are:

• The 2002 attrition rate for all domestic students was 18.5% and for international students, 17.7%. For these groups these are the lowest rates since 1994.
   
• There is considerable variability in the level of attrition across institutions and the different student populations analysed. (see Figure 1). Age of the student, citizenship status, level of course and whether the student is new to higher education or not seem to affect attrition rates. Student population mix, differences in the mobility of populations of the centres where institutions are located and course types offered contribute to variability across institutions.
   
• The attrition rates for domestic students on the whole have remained relatively stable over the years 1994 to 2002, with a slight increase in the rate for domestic commencing postgraduate students balanced by a slight fall in the attrition rate for domestic commencing undergraduate students. The rates for international students on the other hand appear to have declined slightly overall, due mainly to a relatively large drop in the attrition rate of postgraduate students (see Figures 2 and 3).
   
• Current attrition rates for domestic students are generally higher than those for their international counterparts, particularly at the postgraduate level. For example, the 2002 rate for all domestic commencing undergraduate students was 21.2%% compared to 18.0%% for all international commencing undergraduate students. For commencing postgraduate students the respective rates were 27.4% and 19.7%.
   
• Attrition rates for postgraduate students are generally higher than those for undergraduate students. This is true for both domestic and international students, although the difference is much greater for domestic students. The higher attrition rates for postgraduate students may be due to age and age related factors (the attrition rate for older commencing undergraduates is similar to that for commencing postgraduate students). The fact that postgraduate students already have a degree and may be interested in undertaking modules as well as whole courses, greater mobility of postgraduate students across institutions, and differences in study patterns may also be factors that contribute to the difference in rates.
   
• Attrition rates for commencing undergraduate students (both domestic and international) are generally higher than for all domestic undergraduate students. This is consistent with other findings of a much higher attrition rate in the first year after commencement compared to second year. For example, for commencing undergraduate students (in total, new to higher education and not new to higher education) the first year attrition rate was over 20% for all years 1994 to 2002. For the second year the attrition rate for these students was around 10 to 11%.
   
• Young domestic commencing undergraduate students (17 to 20 years) have much lower attrition rates than their older counterparts, regardless of whether they are new to higher education or have previously enrolled in another course, and regardless of the sex of the student. 2002 attrition rates for 17-20 year old commencing undergraduates were just under 18%. The rates for those aged 21-24 were around 24%. The rates for those aged over 25 years ranged between 27.1% and 33.7% depending on age group and sex (Figure 4).
   
• The overall attrition rate for domestic commencing undergraduate students was very similar across the sexes, with females having a slightly lower rate (21.0% in 2002) than males (21.5%). The main difference in attrition rates for male and female students in this group were in the 30 to 39 year and 40 to 49 year age groups, with females having a lower attrition rate than males (Figure 4).
   
• School leavers commencing in undergraduate courses in 2002 had an attrition rate of 17.4% in 2002. This is compared with the overall rate for domestic commencing undergraduate students of 21.2%.
   
• The combination of lower attrition rates for younger students and a higher proportion of younger students in the population leads to a lower overall attrition rate for commencing students who are ‘new to higher education’ than students who have previously been enrolled in higher education. The overall attrition rate for domestic commencing students who had previously enrolled at university in another course in 2002 was 22.5% compared to 20.6% for those who had never previously been enrolled.
   
• However, taking the age of the student into account alters the picture. For both the younger group of students (aged 17 to 20 years) and the older group of students (the remainder) the attrition rates for those ‘new to higher education’ were higher than those of students who had previously been enrolled in another course. For 17 to 20 year olds the 2002 rate for new to higher education students was 18.3%. The rate for students of the same age who were not new to higher education was 16.2%. For older students the comparable rates were 28.6% for new to higher education students and 25.8% for those not new to higher education. Thus age of the student may be a more important contributor to attrition rates than whether the student was new to higher education or not.

Contact Maureen Maclachlan, Higher Education Analysis Section, SAEG, DEST at Maureen.maclachlan@dest.gov.au

Tatjana Lukic
Anne Broadbent
Maureen Maclachlan
Higher Education Analysis Section
Strategic Analysis and Evaluation Group
March 2004

1. Characteristics and Performance Indicators of Australian Higher Education Institutions, 2000, Occasional Paper Series 01-B, Higher Education Division, DEST

2. McMillan,J. Course Change and Attrition from Higher Education, Unpublished report from the ACER, February 2004

*Note:
You need Adobe Acrobat Reader to read the PDF files.
Free Adobe Acrobat PDF reader software is available for downloading from the Adobe website    
Tools to help make PDF files accessible to the vision impaired are available from http://access.adobe.com/