A Decision Algorithm for Admitting Students to Advanced Practice Programs in Nursing

The academic records of 738 master's-level nursing students were reviewed to develop an optimal compound decision rule for selecting candidates who would be successful in advanced practice programs. Chi-square analysis was used to establish the association between various combinations of independent variables and dismissal status. Independent variables included cumulative grade point average (CUMGPA), undergraduate nursing grade point average (NSGGPA), and Graduate Record Examination (GRE) scores (quantitative and verbal). Results indicated that the combination of any two independent variables was as predictive of success as was any combination of three independent variables and that the combination of a CUMGPA of 3.25 and NSGGPA of 3.0 predicted success in 99% of the cases. The addition of the GRE scores added no additional predictive value. On the basis of these analyses, we waived the GRE requirement for applicants who met the GPA criteria and developed decision rules for admitting those who did not.