The Use of Quantitative Admission Variables to Predict NPTE Success: A Pilot Study Justine Bolton, SPT; Alison Parks, SPT; Kara Yeager, SPT; Ally Walker, SPT Faculty Mentor: Kelly Meiners, PT, MPT, Ph.D., ATC
Introduction: Having a prediction tool to assist PT education programs in identifying students who are more likely to be successful on the National Physical Therapy Examination (NPTE) can also help to identify students who may benefit from extra support in their education to improve their performance on the exam. Improved performance on the NPTE will benefit the student, the program, the profession of physical therapy, and even future patients.
Purpose: The purpose of this study was to examine which preadmission and professional program variables best predict first-time NPTE scores. Additionally, the effect of coursework volume completed at a two-year institution on NPTE scores was examined.
Methods: Deidentified, archived qualitative data of 309 physical therapy (PT) students from two private Midwestern universities was analyzed retrospectively. Data gathered included undergraduate GPA (UGPA), undergraduate science GPA (USGPA), quantitative GRE score (QGRE), verbal GRE score (VGRE), number of undergraduate community college credit hours (CCH), number of physical therapy program prerequisite community college credit hours (CCPH), first-year physical therapy program GPA (PT1GPA), and overall physical therapy program GPA (PTGPA) as independent variables and the first-time NPTE score (NPTE) as the dependent variable. A stepwise multiple regression analysis was performed on the combined data. The study examined the effect of each independent variable on the predictability of first-time NPTE scores as well as predictive equations combining multiple variables.
Results: Multiple regression analysis including all eight independent variables predicted 42% of the variance seen in NPTE scores. Stepwise multiple regression analysis revealed that PT1GPA and PTGPA were the only significant predictors of first-time NPTE scores. PT1GPA predicted 37% of the variance seen in NPTE scores when analyzed independently (β=.61, p>.001) and variance prediction increased to 40% when including PT1GPA (β=.45, p>.001) and PTGPA (β=.22, p=.016). The stepwise analysis revealed the other preadmission variables including VGRE, QGRE, UGPA, USGPA, CCH and CCPH were not significant predictors of NPTE.
Discussion/Conclusion: This study found that PT1GPA and PTGPA are indicative of success on the NPTE, while UGPA, USGPA, QGRE, and VGRE were not significant in predicting NPTE scores. Additionally, this study concludes that CCH and CCPH were also not significant predictors of success on the NPTE. This study supports the equation of NPTE score = 134.089 + 93.373 (PT1GPA) + 56.535 (PTGPA) to be used in determining a student’s likely success on the NPTE. These findings aid in advancing knowledge of which preadmission and professional program variables are most predictive of future success on the NPTE and how to further identify students in need of educational support.
Acknowledgements: This research would not have been completed without Jamie Dehan or her students Jenna Humphrey, Cameron Zimmerman, Natalie Greene, and Katlin Hickerson of the University of Saint Mary.