Investigating the feasibility of using advanced statistical analysis to understand the impact of uncoordinated and overlapping student equity interventions
The final report from a ACSES Feasibility study undertaken by Griffith University.
The final report from a ACSES Feasibility study undertaken by Griffith University.The final report from a ACSES Feasibility study undertaken by Griffith University.
Author
Clancy Birrell, Muhammad Zia, Ruth Kamrowski, Roby Jose, Laura Sturmer, and Vivian Wu
Published on
November 9, 2026
Read time
60
min
This is a final report from a ACSES Feasibility Study exploring whether advanced statistical analyses could be applied to institutional observational data to better understand how multiple overlapping and uncoordinated equity interventions relate to key student outcomes.
The study focused on domestic undergraduate equity-group students and their participation in selected supports across 2023–2024. Intervention participation data were consolidated across multiple service units and linked with student administrative records. Three complementary analytical approaches were tested to assess feasibility and the types of insights that could be generated:
Survival analysis to model the timing of outcomes using survival/time-to-event models, allowing supports to be treated as changing over time and recognising multiple possible end outcomes (retention/attrition, completion).
Multilevel mixed-effects modelling to assess change in academic success for the same student over time (as well as annual retention patterns), recognising that students are grouped within broader contexts (for example, study areas) and that students may be observed multiple times across teaching periods.
Channel attribution analysis to examine how different supports may contribute together towards student outcomes, depending on the order and combination in which students receive them.
All three analytical approaches were feasible to implement using the available administrative and intervention participation data, and each generated complementary initial insights about overlapping supports.
While the study was not able to draw causal conclusions, largely because a credible counterfactual was not available, the feasibility work clarified certain methodological, data governance, and implementation conditions required to support future investigations into the impact of overlapping supports. The report makes key recommendations to enable future evaluation of overlapping, uncoordinated equity interventions.
