Factors associated with academic performance among computer science undergraduates: a retrospective cross-sectional study of institutional administrative data at a Chinese public university

Background: Universities accumulate large volumes of administrative data through routine teaching, assessment, enrolment, and student-support activity. These records offer an opportunity to describe variation in academic attainment across entire cohorts, yet learning-analytics research has largely emphasised predictive accuracy rather than the substantive interpretation of recorded characteristics. This study examined which socio-demographic, personal, and engagement-related records were associated with academic performance among computer science undergraduates at a single Chinese public university.Methods: A retrospective cross-sectional analysis used de-identified administrative records for 1,613 undergraduates from three consecutive entry cohorts in one school of computer science. Records were drawn from the academic-affairs, student-affairs, psychological-warning, and innovation-and-entrepreneurship systems. Descriptive statistics, independent-samples t tests, one-way analysis of variance with Welch's correction where appropriate, K-means clustering of second-classroom credits, and stepwise multiple linear regression were performed.Results: Significant differences in grade point average (GPA) were observed for gender, province and prefecture of origin, ethnicity, family economic status, psychological-warning level, political affiliation, and second-classroom performance; urban versus rural origin was not significant. GPA increased monotonically across four second-classroom engagement clusters (2.848 to 3.513), with partial eta-squared of 0.245. In the regression model, second-classroom credits showed the strongest standardised association (beta = 0.377), followed by political affiliation (0.206) and gender (0.166), whereas psychological-warning level was negatively associated with GPA (beta = -0.065). The seven retained predictors accounted for 29.7% of GPA variance (adjusted R-squared = 0.294).Conclusions: Co-curricular engagement and psychological wellbeing are connected correlates of attainment within this institutional context. Institutional administrative data can help identify groups that may benefit from integrated academic and wellbeing support, provided that privacy safeguards are maintained. Because the data are cross-sectional and drawn from one institution, the findings are associations and do not establish causal effects.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-20
DOI
https://doi.org/10.21203/rs.3.rs-11792023/v1
Primary Topic
Online Learning and Analytics
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Factors associated with academic performance among computer science undergraduates: a retrospective cross-sectional study of institutional administrative data at a Chinese public university

Guangsheng Liu, wenqian Yu
Research Square
Online Learning and Analytics
preprint

Factors associated with academic performance among computer science undergraduates: a retrospective cross-sectional study of institutional administrative data at a Chinese public university

Guangsheng Liu, wenqian Yu
preprint en

Abstract

Background: Universities accumulate large volumes of administrative data through routine teaching, assessment, enrolment, and student-support activity. These records offer an opportunity to describe variation in academic attainment across entire cohorts, yet learning-analytics research has largely emphasised predictive accuracy rather than the substantive interpretation of recorded characteristics. This study examined which socio-demographic, personal, and engagement-related records were associated with academic performance among computer science undergraduates at a single Chinese public university.Methods: A retrospective cross-sectional analysis used de-identified administrative records for 1,613 undergraduates from three consecutive entry cohorts in one school of computer science. Records were drawn from the academic-affairs, student-affairs, psychological-warning, and innovation-and-entrepreneurship systems. Descriptive statistics, independent-samples t tests, one-way analysis of variance with Welch's correction where appropriate, K-means clustering of second-classroom credits, and stepwise multiple linear regression were performed.Results: Significant differences in grade point average (GPA) were observed for gender, province and prefecture of origin, ethnicity, family economic status, psychological-warning level, political affiliation, and second-classroom performance; urban versus rural origin was not significant. GPA increased monotonically across four second-classroom engagement clusters (2.848 to 3.513), with partial eta-squared of 0.245. In the regression model, second-classroom credits showed the strongest standardised association (beta = 0.377), followed by political affiliation (0.206) and gender (0.166), whereas psychological-warning level was negatively associated with GPA (beta = -0.065). The seven retained predictors accounted for 29.7% of GPA variance (adjusted R-squared = 0.294).Conclusions: Co-curricular engagement and psychological wellbeing are connected correlates of attainment within this institutional context. Institutional administrative data can help identify groups that may benefit from integrated academic and wellbeing support, provided that privacy safeguards are maintained. Because the data are cross-sectional and drawn from one institution, the findings are associations and do not establish causal effects.

Research Square
Central China Normal University (CN), Wuhan University of Science and Technology (CN)
Quality education, Reduced inequalities
Online Learning and Analytics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon