Sponsored research versus graduating students? Intervening variables and unanticipated findings in public research universities

This article examines the relationship between sponsored research monies and the graduation of undergraduate students at 22 public research universities. Using institutional and student characteristics for 59,982 students at these universities, we conducted chi-square, cross-tabulation, correlation, and hierarchical nonlinear modeling (HNLM) analyses to determine the relationship between our principal independent and dependent variables, and among them and other individual and institutional variables, including SAT (individual and institutional mean), gender (individual and percentage), underrepresented minority (individual and percentage), and graduate academic program rating. In contrast to a basic premise of most policymakers and academic managers, our findings suggest that an institution's sponsored research expenditures are positively related to undergraduates' graduation. We also find that mean SAT has a powerful impact on student graduation, although at the individual level gender is a more powerful correlate of graduation than is SAT score.