The impact of feeder school selectivity on predicting academic success in an allied health professional program

Admission committee members have the important task of selecting candidates who have the best chance of academic success in their professional programs. Graduate record examination (GRE) scores and grade point average (GPA) are typically given heavy consideration in the admissions process. The purpose of the present study is to assess the impact of feeder school selectivity on the ability of GRE scores and GPA to predict probationary status in entry level physical therapy students.

Three hundred five (249 female, 56 male) students, who graduated from an entry-level Doctor of Physical Therapy Program between 2001 and 2006, were included in this analysis. Feeder school selectivity was defined as either less selective/selective or more selective/most selective (US News and World Report 2005). Unpaired t-tests assessed differences in GRE scores and GPA between the less selective/selective and more selective/most selective groups. Receiver operating characteristic curve (ROC) analysis assessed the ability of GRE score and GPA to predict probationary status. Probation for at least one semester (Cumulative GPA <3.0 in the physical therapy program) was considered an event. Statistical tests with a p-value <0.05 were considered significant.

Verbal GRE scores [443.68 (+/-75.16) vs. 489.01 (+/-77.63)], Quantitative GRE scores [546.40 (+/-98.70) vs. 610.16 (+/-78.57)] and Total GPA [3.37 (+/-0.32) vs. 3.30 (+/-0.31)] were significantly different between the less selective/selective group and the more selective/most selective group (p<0.05). Math science GPA [3.29 (+/-0.34) vs. 3.19 (+/-0.37)] was not significantly different between the less selective/selective and more selective/most selective groups (p>0.05). One hundred fourteen of the subjects attended feeder schools that were considered less selective/selective. Eighty-four students (28%( were on probation for at least one semester. Verbal GRE (ROC curve area: 0.62, p=0.001), Quantitative GRE (ROC curve area: 0.71, p<0.001), total GRE (ROC curve area: 0.69, p<0.001), total GPA (ROC curve area: 0.59, p=0.01) and undergraduate program rigor (ROC curve area: .64, p<.001) classification schemes were all significant predictors of probation. The optimal threshold values for Verbal GRE, Quantitative GRE, total GRE, and total GPA were 455 (sensitivity 61%, specificity 52%), 575 (sensitivity 69%, specificity 65%), 1025 (sensitivity 67%, specificity 61%) and 3.25 (sensitivity 64%, specificity 56%) respectively. The optimal threshold value for undergraduate rigor was 4 (sensitivity 70%, specificity 56%).

The results of the present study indicate feeder school selectivity, GRE scores, and GPA are all independent predictors of probationary status. Assessing feeder school selectivity in addition to GREs and GPAs during the admissions process may therefore be warranted.