"Ghost student" failure among equity cohorts: Towards understanding non-participating enrolments
Type of Publication: Research report
Lead Organisation: La Trobe University
Year Published: 2021
Lead Researcher: Bret Stephenson
Bret Stephenson, Beni Cakitaki and Michael Luckman
The “ghost student” phenomenon
Unit-level failure is a common experience among Australian university students, but it has not received adequate attention in the higher education research literature nor, until very recently, within government policy discussions. This oversight has many causes, but one result has been to leave unexamined important subtypes of student failure, such as those that are often described by academic staff as “ghost” or “zombie” failures at the unit level.
Every year, a significant percentage of students remain enrolled in one or more of their units yet exhibit no evidence of having engaged in learning or assessment activities. This most severe form of student disengagement and failure frequently goes unnoticed at institutional and national levels, as it is obscured by high-level metrics such as the all too binary “success rate”. The “ghosting” phenomenon does not, however, go unnoticed by academic teaching staff in Australia’s universities. What they frequently witness are students who are formally enrolled in their units, but who do not attempt any assessment tasks, and neglect to formally withdraw from the unit.
Ghosting behaviours have ramifications for students, institutions and the Australian government. The student receives a record of academic failure and an increased financial burden, institutions see a reduction in their published “success rate” and governments see little return for the allocation of Commonwealth Supported Place (CSP) funding. A report of this kind is also timely given the forthcoming government changes to CSP eligibility and the introduction of a “50 per cent pass rule” in 2022. As this report shows, ghosting behaviours are likely to make a significant contribution to student ineligibility under the new rules.
Non-participating enrolments (NPEs)
This report provides an in-depth examination of “ghost student” failure among commencing domestic bachelor students generally but focuses particularly on four student equity cohorts: regional and remote, low socioeconomic status (SES), Indigenous, and non-English speaking background (NESB) students. We term this type of unit failure a non-participating enrolment (NPE) and define it as a completed unit attempt that resulted in a failing grade and a numeric mark of zero on a 0-100 scale. NPE results can then be contrasted to what we term “non-zero failures”, or failures where a student has achieved any non-zero level of assessed credit for the unit.
This definition is intentionally tailored to the unit-level as an acknowledgement of the often selective nature of NPE behaviours which are frequently limited to individual unit enrolments. A student can also be considered an NPE over their total enrolment, or across their entire course, by exhibiting NPE behaviours in all attempted units — we have termed these “total NPE” results. Alternatively, a student may only have “partial NPE” results, characterised by receiving an NPE in one or some of their units, but otherwise showing evidence of participation in other units.
Utilising a large multi-year dataset of commencing domestic bachelor students sourced from a large public (Table A) Australian university, we seek to:
- quantify patterns of NPE results and contrast these with conventional unit failures
- examine the extent to which student equity categories are linked to a higher risk of NPE behaviour
- describe the relationship between NPE results and rates of student retention and completion, particularly for students from equity groups
- analyse, via multivariate regression, the extent to which potential mediating factors—particularly studying part-time, ATAR, age and field of study— influence the NPE rates of equity cohorts.
At the unit-level of analysis, we found that NPE results—or results of “0”—were far and away the most common numeric fail mark and accounted for more than a quarter of all fail grades. While 13.7 per cent of all units undertaken by the students in our sample resulted in a fail grade, 3.7 per cent resulted in an NPE result. At the student-level of analysis, we found that a third of students in our sample had failed at least one unit in their commencing year, and a tenth of students had at least one NPE result. Just 1.8 per cent of students had an NPE result in all of their enrolled units (total NPEs), while 8.9 per cent of students in our sample registered at least one NPE result while achieving higher marks in other units (partial NPEs). Crucially, only a tenth of students with an NPE result managed to achieve an overall average pass mark (>= 50). Students with a non-zero fail result as their lowest mark had much better overall academic achievement than NPE students. Despite having failed a unit, more than two thirds of these students still achieved an average mark of 50 or higher.
NPE as a leading indicator of retention and completion
We found that NPE is a leading indicator of student attrition and non-completion. Even after controlling for overall academic achievement (average marks), receiving an NPE result is one of the strongest predictors of both attrition and non-completion among all the variables we modelled for. Once NPE and other fail marks are accounted for, equity group membership has no statistically significant adverse association with retention and completion outcomes. On the contrary, low SES and NESB status were associated with a higher probability of retention, and students from regional and remote areas were associated with a higher probability of completion. Consistent with much of the existing literature, studying part-time had a large and statistically significant impact on retention and completion outcomes.
Equity and NPE
From an equity perspective, Indigenous students were shown to be at a high risk of receiving an NPE result, even after controlling for ATAR and study attendance mode. The probability of a low SES student achieving at least one NPE result was not statistically different than for high SES students. Conversely, students from regional and remote areas, and NESB students were actually at a lower risk of registering an NPE result than metropolitan and English-speaking students respectively.
While equity group membership was not a predictor of receiving an NPE result for most equity groups, equity students were overrepresented among course characteristics that were strong predictors. Students admitted with an ATAR below 60 were at a significant risk of having an NPE result. Yet Indigenous students, low SES students, and students from regional and remote areas had higher rates of students with an ATAR below 60 in our sample. Similarly, students studying part-time were at an elevated risk of receiving an NPE result, and Indigenous students, as well as students from non-English speaking backgrounds, were much more likely to study part-time.
This report further provides several important recommendations for university policy and practice in response to the NPE challenge. We also find that there is a need for greater conformity of NPE data collection across institutions and by the Department of Education, Skills and Employment’s Higher Education Information Management System (HEIMS) data collection processes. The report further contains several suggestions for future research aimed at further illuminating the mystery of NPE behaviours.
- Australian university planning and performance units should carefully track, report and utilise non-participating enrolment (NPE) statistics as an important measure of institutional quality and performance.
- Australian universities should work to particularly understand and address NPE amongst Indigenous students, who appear to have a substantially higher risk of receiving NPE results compared to non-Indigenous students. Additional advising or support to help address Indigenous NPE behaviour could lead to significant improvements in Indigenous student retention and completion outcomes.
- Australian universities should adopt grade scales that capture a full taxonomy of student failure types/grades and, at minimum, include a dedicated category for failure via NPE.
- Australian universities should review their grade reporting practices and definitions to ensure that multiple failure grade categories are being consistently applied and reported across academic areas and according to a clear rubric.
- Australian universities should create policies dedicated to NPE failures and make the tracking and remediation of NPE failures a central feature of their student success and retention, and student equity strategies.
- Australian universities should adopt academic progression and monitoring (APM), or academic probation, policies and practices that recognise and respond to differences in the relative severity and impact of different types of student failure, but particularly NPE failures as compared to, for example, marginal or “non-zero failures”.
- The Australian Department of Education, Skills and Employment should adopt an NPE definition and make the reporting of NPE results a regular feature of HEIMS data collection.
- Researchers of higher education should account for and include NPE failures in relevant research studies, but particularly those relating to the evaluation of student success and retention intervention efforts. Failing to account for the type of student disengagement that NPE results represent can, in some instances, dramatically skew research results.
- Researchers of higher education should seek to further extend the findings of this report by conducting qualitative research studies aimed at better understanding student motivations for, and responses to, NPE behaviours. We recommend that qualitative research be focused particularly on Indigenous student cohorts where better understanding of the behaviour is particularly needed.