A Punjab survey links teaching and university culture with student engagement

Updated Sep 07, 2026

Updated, 7 September 2026: We checked the complete paper and its main regression table. The previous headline overstated causation and generalisability. We have also clarified the model statistic and disclosed inconsistencies in the paper’s sample accounting.

What the study examined

Swarnima Sharma and Mamta Garg report a survey of 553 second-semester postgraduate students at four public and private universities in Punjab, India. Their analysis relates self-reported engagement to institutional and personal variables.

In the final stepwise model, teaching delivery, lifestyle, achievement motivation, organisational culture and perceived curriculum relevance were associated with engagement. Table 2 reports R²=.311 and adjusted R²=.305. The latter is the 30.5% figure used in the abstract. Neither is a measured improvement caused by changing those factors.

The authors’ approximately 21% institutional and 10% personal breakdown reflects variables entering this particular model. It is not a universal ranking of causes. Most demographic comparisons were non-significant, while gender differences and a behavioural-engagement difference between university types were reported. Non-significance does not establish that background never matters.

The source also contains numerical inconsistencies. Its stated initial sample of 1048 minus 447 excluded cases does not equal the reported final 553. We could not reconcile that discrepancy. We therefore retain the reported analytic sample with this qualification and avoid relying on its exclusion totals or fine-grained estimates.

Questions for interpreting engagement data

Our suggested starting point is to describe what your engagement measure captures and whose responses it includes. Attendance, self-reported effort, belonging and demonstrated learning are related questions, but they should not become interchangeable measures.

Read comments about teaching, support or curriculum relevance as accounts to investigate. They may suggest changes worth testing without establishing which condition caused a score or how much improvement to expect. Consider missing responses and other explanations before comparing groups.

This survey does not validate a UK intervention or a comment-analysis product. A local evaluation should specify the change, intended outcome and comparison. The NSS methodology documents a separate analysis context; it should not be treated as equivalent to this study’s engagement instrument.

Reference

Sharma, S. and Garg, M. (2026). What drives Student Engagement in Indian higher education? Exploring key demographic, institutional and personal variables. Student Engagement in Higher Education Journal, 7(3), 137–158. The journal landing-page title omits “Indian”; this citation follows the paper. Open paper.