Author response.

We found Sandstrom's comments to be an insightful contribution to a dialogue as the profession grapples with how best to prepare professional physical therapist students for safe and effective practice. Academic physical therapy admissions committees have a daunting task, as Sandstrom implies in his commentary. Committees have to balance societal needs and program objectives with the knowledge that students must meet rigorous academic demands. As our studies (1,2 suggest, only a small portion of the variance in students' success, which we defined as the avoidance of academic difficulty or passing the National Physical Therapy Examination (NPTE) on the first attempt, was explained by the variables we studied. A majority of the variance for either favorable outcome was left unexplained despite the use of several demographic, quantitative academic- and program-level variables. There is no question that there are many challenges to making defensible admissions decisions. Given these challenges, we believe that our reports have expanded our understanding of the potential role that quantitative academic variables may play in predicting student performance.

Sandstrom discussed several issues in his commentary. Early on, the point was made that admissions committees may take solace in finding that our studies reaffirmed what was already known and generally accepted--that grade point average (GPA) and Graduate Record Examination (GRE) score are important predictors of success. Although there was weak evidence for these assumptions prior to our work, our studies are the first to demonstrate these predictions on a national scale. One reason we believe this to be important is because many programs, as of June 2007, do not require GRE scores as part of their admissions process. We examined Web sites of all current programs and determined that 31% of the 199 accredited programs listed in the Commission on Accreditation in Physical Therapy Education's (CAPTE's) online directory (3) do not require GRE scores from applicants. If there was doubt about the utility of GRE scores to assist in making important predictions about student performance, we hope that our studies put an end to this uncertainty. For programs that are not requiring the GRE, they are missing an opportunity to better understand the potential of students in their applicant pools.

When it comes to making predictions, GRE scores are at least as important and, in the case of the licensure examination, more important than GPA in explaining performance. This is true, in part, because both the verbal and quantitative GRE scores are independently predictive. The GRE scores essentially contribute 2 scores to the prediction, whereas the GPA contributes only a single score. A quick glance at the prediction rules (see Tab. 10 in the first article and Tab. 6 in the second article) support this argument.

The GRE and GPA scores were not the only predictive variables we found. Academic difficulty and NPTE performance also were related to student age and race or ethnicity. Program mission and societal need play a large role when making admissions decisions. To this end, older students and students of diverse racial and ethnic backgrounds are an extremely important and ever-growing part of the population of professional physical therapist students.

We see an urgent research need related to students of all ages and backgrounds who are at increased risk for academic or licensure examination difficulty. Using data from our studies, one can identify students who are at increased risk of difficulty, but we have no evidence for educationally based interventions designed to increase a student's chance for success. We believe that research should examine the effectiveness of educational interventions that will assist any student who may be at risk for academic difficulty.

Sandstrom also indicated that the "program" variable was an important predictor in both studies. That is, we found large variations in the proportions of students from each program who had academic difficulty or who did not pass the NPTE on the first attempt. Sandstrom suggested that faculty, institutions, and CAPTE should pay attention to this finding and potentially intervene to make changes. We suspect that faculty and parent institutions of those programs with higher proportions of unsuccessful students likely attempt changes to improve their success rates. Research is needed to explore the benefits of these changes. We found the suggestion that CAPTE potentially play a role to be provocative food-for-thought that the profession should discuss further.

Sandstrom concludes his commentary with the following sentence: "A balanced approach to admissions needs to consider these applicant aptitudes as well as academic record." We agree wholeheartedly with Sandstrom's admonition with one caveat. Just as our studies added to our understanding of the impact that GRE and GPA scores can have in making predictions about student performance, we urgently need evidence to guide our decisions about how to measure aptitude.

As Sandstrom points out, previous grades and GRE scores give us little information regarding a student's psychosocial characteristics. Sedlacek (4) argued that although measurements such as GPA and GRE scores are easy and relatively inexpensive to collect, they overlook important qualitative characteristics, such as culture, goals, and values. Many programs utilize qualitative measures, such as letters of recommendation, evaluations of personal statement essays, and interviews, during their admissions processes. A recent metaanalysis (5) regarding the utility of interviews suggests that they correlate weakly with academic performance of students in a variety of health care professions. More large-scale studies are needed to identify appropriate, reliable, and valid methods for measuring qualitative student characteristics.

In summary, we believe that our studies provide evidence for the use of quantitative data in admissions decisions for professional physical therapist education programs. Tradition and opinion have been the customary approach to admitting students, and it is time to rely more on evidence and less on tradition when making admissions decisions.

DOI: 10.2522/ptj.2006.0221.0222.ar

References

(1) Utzman RR, Riddle DL, Jewell DV. Use of demographic and quantitative admissions data to predict academic difficulty among professional physical therapist students. Phys Ther. 2007; 87: 1164-1180.

(2) Utzman RR, Riddle DL, Jewell DV. Use of demographic and quantitative admissions data to predict performance on the National Physical Therapy Examination. Phys Ther. 2007; 87: 1181-1193.

(3) Commission on Accreditation in Physical Therapy Education. CAPTE Accredited Physical Therapist Education Programs. 2007. Available at: http://www.apta.org/ AM/Template.cfm?section=PT_Programs& template=/aptaapps/accredltedschools/acc__ schools_map.cfm&process=3&type=PT. Accessed June 15, 2007.

(4) Sedlacek WE. Beyond the Big Test: Noncognitive Assessment in Higher Education. San Francisco, Calif: Jossey-Bass Inc Publishers; 2004.

(5) Goho J, Blackan A. Effectiveness of academic admission interviews: an exploratory meta-analysis. Medical Teacher. 2006: 28; 335-340.