Answers are grounded in this paper and cite their sources.
Purpose This study examines how Bangladesh’s 2021 General, Science, and Technology (GST) Integrated Universities Cluster Admission System reshaped administrative efficiency, cost distribution and institutional governance during its inception. Design/methodology/approach The study uses a mixed-methods design combining a survey of 324 first-year students at two public universities with semi-structured interviews of 10 administrators, faculty and policymakers, interpreted through New Managerialism and social reproduction theory. Findings Most students reported that GST reduced logistical duplication and travel costs, but expenditure varied sharply – almost a third (29.94%) spent over 8,000 Bangladeshi taka (BDT) versus 9.26% under 2,000 BDT – and a majority (60.19%) reported psychological stress. Administrators and faculty members described diminished institutional autonomy. In regression analysis (N = 275), these disparities were associated primarily with institution rather than with household income or rural/urban origin. The evidence suggests that efficiency gains and equity costs arose from the same centralizing mechanism, concentrated in the post-examination coordination phase rather than the exam itself. Research limitations/implications Data come from two of the 20 GST universities and from self-reported, cross-sectional measures; findings indicate associations, not causal effects and are not statistically generalizable. Practical implications Reform should target post-examination allocation: a binding admission calendar, fewer migration rounds, standardized fees and broader university representation in governance. Originality/value This is among the first multi-stakeholder evaluations of GST. It suggests that procedural uniformity does not resolve – and can obscure – the unequal costs of centralized admission. It extends New Managerialism and social reproduction theory to a resource-constrained, postcolonial context.
No takes yet. Share an insight, caveat, or question.
Arefin et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: