Each year, physical therapy education programs attempt to identify candidates with the highest potential for success in their doctoral-level physical therapy (DPT) program. Each DPT program develops their admissions requirements to identify those applicants with a history of success using criteria shown to be predictive in an effort to not only select those with the greatest likelihood of success but also to ensure that the DPT program achieves and maintains its Commission on Accreditation in Physical Therapy Education (CAPTE) accreditation. Admissions criteria for physical therapist education programs can differ widely between programs, owing in part to each program's specific educational requirements, mission, and goals, as well as each applicant's feeder program requirements.1 The limited predictive value of many admissions criteria and the variety of outcomes used to measure academic success contributes to the regulatory burdens placed on physical therapist education programs.2-9 Failure to produce successful graduates from these programs can have far-reaching consequences, which include increasing the health care burden, and can threaten the financial and job security of students and programs alike. Therefore, swiftly and efficiently identifying which applicants will have the greatest likelihood of academic and professional success is an important objective that is in the best interest of the applicant, as well as the DPT program to which the applicant is applying.
Review of Literature
Evidence from the Physical Therapy Centralized Application Service (PTCAS) (Liaison International, Watertown, MAUS) shows that most, if not all, admissions boards for physical therapist education programs require thresholds for grade point average (GPA) achieved at postsecondary “feeder” programs to determine which students qualify for admission (commonly a 3.0 or above on a 4.0 scale). These requirements are usually calculated based on either a cumulative grade point average (cGPA), which is a value representing the weighted average of the grade points achieved by a student throughout their entire undergraduate collegiate career, or a prerequisite grade point average (pGPA), which is a value representing the weighted average of the grade points achieved by a student across a select list of prerequisite coursework as determined by the individual DPT program. Jones et al10 noted that although health professions admissions processes “remain imperfect works in progress,” cognitive measures such as cGPA remain the greatest predictors of success in both physical therapist and physician assistant education programs, as well as on licensure examinations for each profession. Nuciforo, et al followed 3 consecutive PTCAS admissions cycles between 2010 and 2012 and discovered that undergraduate cGPA predicted physical therapist education program admission (r2 = 0.358–0.381, P < .05).5
Undergraduate GPA has repeatedly been found to predict academic success in physical therapist education programs. Cumulative grade point average has been correlated with cumulative GPA on graduation from a physical therapist education program (GGPA) for a Bachelor of Science in Physical Therapy degree (r = 0.63, P < .01),2 DPT first-year “basic science” GPA (r = 0.441, P < .001),6 MPT GGPA (r = 0.436, P < .01),8 (r2 = 0.11, P < .001),11 DPT GGPA (r2partial = 0.275, P < .001),4 and physical therapist licensure examination (NPTE) scores (r = 0.243, P = .05).8 Cumulative grade point average has also been shown to predict risk of academic difficulty. For students with an undergraduate cGPA of 3.15 or below, risk of academic difficulty was found to be greater (β = 0.357, P < .05), whereas for students with a cGPA of 3.51 or above, risk of academic difficulty was decreased (β = −1.113, P < .05).9 Another study also demonstrated that cGPA had predictive values for risk of academic probation in a DPT program (ROC curve area = 0.59, 95% CI = 0.52–0.66).1 However, an identically sized (n = 305) sample that observed this same program 5 years before showed that cGPA did not increase risk of academic probation.12 That being said, the authors indicated that the outcomes of their study should not be applied to other PT programs because of study limitations such as sample size and failure to account for ethnicity or rigor of undergraduate training.
Prerequisite GPA has been correlated with MPT program GPA (r = 0.509, P < .01)8 and has also been a predictor of academic probation in a DPT program (ROC curve area = 0.59, 95% CI = 0.52–0.66).1 Prerequisite grade point average has also been shown to predict NPTE scores (r2partial = 0.078, P < .001).4 Elsewhere, Ruscingno et al6 followed 2 consecutive cohorts of PT students in a single physical therapist education program (n = 63). They failed to show a significant correlation between pGPA and DPT first-year “basic science” GPA with the typically accepted levels of confidence (r = 0.183, P = .152).6 Another study found that 4 physical therapy education cohorts' pGPA was correlated with NPTE scores (r = 0.341, P < .05) and with first-year program GPA (r = 0.316, P < .01).3 Meiners and Rush found that for their cohort, first-year GPA was an even stronger predictor of NPTE scores (β = 0.572, P < .001) and did not show any relationship between cGPA and NPTE scores.13
The discrepancies between admissions criteria for physical therapist education programs, coupled with rising student debt and threats to programs' funding, accreditation, and environment, give admissions committees impetus to develop the most efficient methods for identifying successful applicants for admission into their programs. Although it is a generally accepted practice for some programs to limit the number of times students can retake prerequisite coursework, to the knowledge of the authors, no published research has investigated the potential relationship between repeated prerequisite coursework and academic success in a physical therapist education program. A preliminary survey performed by the researchers of this study examined 60 randomly selected public and private physical therapy programs nationwide and found that 23.3% (n = 14) of programs examined currently consider a student's retaking a prerequisite course as a factor when determining their admissions requirements. Of the 14 programs that look at prerequisite retakes, 57.14% (n = 8) were public institutions, and 42.86% (n = 6) were private institutions. Admissions requirements for the 60 DPT programs examined were obtained through publicly available admissions standards posted on their university's website. Physical therapy programs that either limited the number of prerequisite courses or did not take the highest grade received if a course was retaken were classified as using a prerequisite retake as an admissions consideration. With this knowledge, it raises the question that if using, whether or not a student repeats a prerequisite course, is a valid preadmission criterion on the part of DPT programs. We hypothesized that students who did not retake prerequisite courses (nrPR) versus those who did retake prerequisite courses (rPR) would score higher in the following areas: 1) first-year grade point average (1GPA), 2) second-year grade point average (2GPA), and 3) cumulative grade point average on graduation (GGPA) during their physical therapy education.
Deidentified admissions data of 159 students from 5 different graduating classes between 2012 and 2015 from Tennessee State University (n = 98) and Marshall University (n = 61) were included in our analyses. At the time of data collection, both programs required a minimum prerequisite GPA of 3.0 for both pGPA and cGPA. The students were categorized into 2 separate groups: students who rPR (n = 71) and students who did nrPR (n = 88). Participant characteristics are listed in Table 1.
Both Tennessee State University (TSU) and Marshall University (MU) have similar admissions requirements (Table 2) with the only differences in that TSU requires an applicant to meet a minimum GRE and only requires 40 observation hours, whereas MU does not require a minimum GRE and requires 60 observation hours. Tennessee State University uses a traditional curriculum model, whereas MU uses a modified case-based model. As a result, there are some variations in the curriculum between the 2 programs as shown in Table 3. No significant changes in the curriculum occurred during the time frame of this study that may have affected the data. Both DPT programs used the best grade attempt for all prerequisite courses when calculating cumulative and prerequisite GPAs. In addition, both programs use onsite interviews and an essay as a component of the admissions process. These 2 components are graded subjectively by the faculty and are used primarily to screen for red flags but because of the subjectivity of the score has not been incorporated in this study.
The following performance criteria were compiled to determine whether differences exist between the rPR and nrPR groups: 1) year-1 GPA (1GPA), 2) year-2 GPA (2GPA), and 3) cumulative GPA on graduation (GGPA).
Data were retrospectively collected using the PTCAS and internal records maintained within the individual physical therapy departments. Each program was accredited by CAPTE at the time of data collection. Graduates were excluded from the study if any data were unavailable during the time of data collection. All procedures of the study were approved and compliant with the Institutional Review Boards of Tennessee State University and Marshall University.
Effect size was calculated using Cohen's d. Data were analyzed using a 2-tailed independent samples t-tests to identify any statistically significant differences in DPT academic performance between students who repeated prerequisite courses and students who did not. Alpha level was adjusted using a Bonferroni correction to 0.016. All data were analyzed using SPSS statistical package (Version 24, Chicago, ILUS).
Results are shown in Figure 1. An a priori power analysis revealed that to achieve a medium effect size using a Cohen's d = 0.5, statistical power level of 0.85, with an alpha value of 0.05, the minimum sample size was 146. No statistically significant differences were observed between the rPR group and the nrPR groups for any of the academic variables of interest. 1GPA did not significantly differ between the rPR group and the nrPR group, t(159) = 1.294, P = .198. 2GPA did not significantly differ between the rPR group and the nrPR group, t(159) = 0.876, P = .382, and GGPA did not significantly differ between the rPR group and the nrPR group, t(159) = 0.682, P = .492.
Analyses calculated a standardized effect size using Cohen's d14 and revealed that pGPA for the rPR group (3.34 ± 0.25) was slightly higher than pGPA for the nrPR group (3.27 ± 0.25) (d = 0.28, 95% CI = 0.22–0.33). As well, cGPA for the rPR group (3.30 ± 0.31) was slightly higher than cGPA for the nrPR group (3.16 ± 0.25) (d = 0.51, 95% CI = 0.43–0.56). However, all physical therapist education program GPAs were slightly, but not significantly higher for the nrPR group, 1GPA (d = 0.22, 95% CI = 0.16–0.28); 2GPA (d = 0.13, 95% CI = 0.04–0.24), and GGPA (d = 0.11, 95% CI = 0.02–0.22).
DISCUSSION AND CONCLUSION
Dockter found cGPA to be correlated with 1GPA (r = 0.316, P < .01) and 1GPA to be correlated with NPTE scores (r = 0.648, P < .01).3 She noted that cGPA accounted for 60% of the admissions criteria score that was used to rank applicants for admission into the program that was studied.3 Importantly, many studies suggest that the value of cGPA and pGPA for predicting academic success in a physical therapist education program increases meaningfully when combined with additional criteria, with GRE as a frequently reproduced predictor.8,9
The present study was unable to account for students who withdrew in our analysis, and there were a few subjects who were not included in the analysis because of difficulty accurately and/or completely retrieving those subjects' undergraduate academic records. Often, studies that attempt to predict academic success in physical therapist education programs remove students who drop out of the program from their final analyses. For example, Fell et al did not find cGPA and pGPA to differ significantly between students who withdrew (n = 13) and students who graduated (n = 290) from 2 different DPT programs over the course of 4 years (P > .05).4 Those authors noted that for 1 program, only one of the 5 students who withdrew in a 4-year period attributed the withdrawal to academic reasons, whereas in another program, 4 of the 8 withdrawals were attributed to academic reasons.4 This is likely helpful when correlating academic achievement in a physical therapist education program with postgraduate outcome measures, such as NPTE pass rates. It may be argued that removing these withdrawal cases may affect the validity and/or precision of the relationship between undergraduate GPA and outcome measures of academic success in physical therapist education programs, and therefore, including these cases may be useful for certain analyses in the future. However, this may require studies to tactfully account for withdrawals related to personal reasons and withdrawals related to academic failure. In the study by Day,11 the 8 students who withdrew did not have a significantly different program GPA than the 514 who graduated.
Because nuances such as grading scales and academic standards vary between programs, many authors agree that admissions committees should use caution when generalizing the predictive value of admissions criteria for 1 physical therapist education program to another.6,9,15 Grade point averages may occur on scales that have different precision (e.g. A+/A/A−). It is assumed that this balance will not affect calculated mean GPAs because it is assumed that there will be a proportional number of students “on the cusp” for all grades, regardless of the precision of the measurement scale or the groups to which they are assigned.
The results of the present study did not find any significant differences in 1GPA, 2GPA, or GGPA between students in 2 physical therapist education programs who repeated prerequisite courses and students in those same programs who did not. These results suggest that restricting the repetition of prerequisite coursework of applicants for physical therapist education programs does not increase the likelihood of academic success in these programs. It should be noted that as the mean GPAs of the study cohorts are relatively high (3.16 ± 0.25 to 3.30 ± 0.31), the findings may not correlate as well to those who enter a DPT program on the lower spectrum of mean GPA and may demonstrate a different outcome with regard to program academic success.
Future studies should investigate whether different types of degrees have different predictive ability for physical therapy school success (e.g. Bachelor of Science degrees compared with Bachelor of Arts degrees) and whether the location of the prerequisite courses (e.g. Carnegie classification15 or 2-year vs 4-year and accelerated programs6) affect GGPA. For students in a Doctor of Pharmacy Education Program, obtaining a BS was a significant predictor of GGPA (r = 0.187, P < .001) and of graduating on time.16 Previous field-related coursework (biology) has also been predictive of graduating on time for doctor of pharmacy students.16 Similarly, anatomy and physiology course GPA has predicted GGPA17 and NPTE scores18 for students in physical therapist assistant education programs. Further research may be able to clarify the relationships between GPA for keystone courses, such as anatomy and physiology, and predictors of academic success in a DPT program, such as GGPA.
1. Wheeler E, Arena R. The impact of feeder school selectivity on predicting academic success in an allied health professional program. J Allied Health. 2009;38:e79–e83.
2. Balogun JA, Karacoloff LA, Farina NT. Predictors of academic achievement in physical therapy. Phys Ther. 1986;66:976–980.
3. Dockter M. An analysis of physical therapy preadmission factors on academic success and success on the national licensing examination. J Phys Ther Educ. 2001;15:60.
4. Fell N, Mabey R, Mohr T, Ingram D. The preprofessional degree: Is it a predictor of success in physical therapy education programs? J Phys Ther Educ. 2015;29:9.
5. Nuciforo M, Litvinsky Y, Rheault W. Variables predictive of admission to US physical therapist education programs. J Phys Ther Educ. 2014;28:8.
6. Ruscingno G, Zipp GP, Olson V. Admission variables and academic success in the first year of the professional phase in a doctor of physical therapy program. J Allied Health. 2010;39:138–142.
7. Shiyko MP, Pappas E. Validation of preadmission requirements in a doctor of physical therapy program with a large representation of minority students. J Phys Ther Educ. 2009;23:29–36.
8. Thieman TJ, Weddle ML, Moore MA. Predicting academic, clinical, and licensure examination performance in a professional (entry-level) master's degree program in physical therapy. J Phys Ther Educ. 2003;17:32–37.
9. 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.
10. Jones PE, Simpkins S, Hocking JA. Imperfect physician assistant and physical therapist admissions
processes in the United States. J Educ Eval Health Prof. 2014;11:11.
11. Day JA. Graduate record examination analytical scores as predictors of academic success in four entry-level master's degree physical therapy programs. Phys Ther. 1986;66:1555–1562.
12. Jewell DV, Riddle D. A method for predicting a student's risk for academic probation in a professional program in allied health. J Allied Health. 2005;34:17–23.
13. Meiners KM, Rush DK. Clinical performance and admission variables as predictors of passage of the national physical therapy examination. J Allied Health. 2017;46:164–170.
14. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41:3–13.
15. Riddle DL, Utzman RR, Jewell DV, Pearson S, Kong X. Academic difficulty and program-level variables predict performance on the national physical therapy examination for licensure: A population-based cohort study. Phys Ther. 2009;89:1182–1191.
16. McCall KL, Allen DD, Fike DS. Predictors of academic success in a doctor of pharmacy program. Am J Pharm Educ. 2006;70:106.
17. Kabiri LS, Voight M, Gleeson PB, Mitchell K. Predicting academic and licensure examination success among physical therapist assistant students. J Phys Ther Educ. 2017;31:29–34.
18. Desmarais L, Woble-Valenski MA, Oestmann E. Factors influencing physical therapist assistant licensure examination success. J Phys Ther Educ. 2011;25:36–41.