COGNITIVE, COLLEGIATE, AND DEMOGRAPHIC PREDICTORS OF ATTRITION IN PHYSICAL THERAPY EDUCATION. Addison W. Andrews*1,2; Susan A. Chinworth1; Duane Akroyd2 1. Physical Therapy Education, Elon University, Elon, NC; 2. Adult & Community College Education, North Carolina State University, Raleigh, NC PURPOSE: One purpose of this study was to quantify attrition in the physical therapy program at an accredited institution and make comparisons to the attrition rates in other physical therapy programs. The primary purpose of this research was to identify cognitive, collegiate, and demographic predictors of attrition in physical therapy education for students in this program. BACKGROUNDS/SIGNIFICANCE: Attrition in physical therapy education has consequences for individual students, academic institutions, and the healthcare needs of society as a whole. Attrition and its causes have not been studied much at all in the field of physical therapy education, despite the importance of the topic. Numerous investigators have examined the influence of cognitive factors such as undergraduate GPA on different forms of academic success in physical therapy education (Balogun 1988; Kirchner, Holm et al. 1994; Templeton, Burcham et al. 1994; Hayes, Fiebert et al. 1997; Dockter 2001; Thieman, Weddle et al. 2003). However, some of these studies examined physical therapy students in baccalaureate programs, though all domestic programs are now at the graduate level. Furthermore, no one has examined whether the quality of the student's undergraduate institution further strengthens the ability of undergraduate GPA to predict success in physical therapy education. SUBJECTS: Subjects (n = 198) were all students who enrolled in an entry-level, graduate physical therapy program between 1998 and 2002. The average age of the participants ± the standard deviation was 24.91 ± 3.97 years. The median age was 23 years. Thus, the distribution of age was skewed toward the higher ages; most of the participants were in their early 20s, within a year or two of graduating from college. One hundred seventy-nine (90.4%) of the participants were White while the remaining participants (n = 19) were members of a minority race. Seventy-five (37.9%) of the participants were male; 123 (62.1%) were female. METHODS AND MATERIALS: Using secondary data, this study consisted of a longitudinal, structured review of academic records of students who enrolled in a graduate physical therapy program. All information utilized in this study was gleaned from student records in the physical therapy department or in the Registrar's Office. Subjects were placed in one of two groups. Those who failed to graduate with the class in which they entered were placed in the Attrition group. Those whose graduation was neither delayed nor denied were placed in the Retention group. Causes of attrition were documented according to the categories listed by Gabard and associates (1997). Predictors of attrition were chosen based on the longitudinal model of doctoral persistence developed by Tinto (1993). Cognitive predictors addressed in this study were cumulative undergraduate GPA, Math GRE, and Verbal GRE. The collegiate predictor chosen was undergraduate institution quality as denoted by the average SAT score for entering students at the undergraduate institution. Demographic predictors studied included age, race, and gender. ANALYSES: The attrition rate was calculated as the percentage of the subjects whose graduation was either delayed or denied. Predictors of attrition were determined using the logistic regression procedure. The level of significance chosen was 0.05. RESULTS: The attrition rate in the physical therapy program was 10%. Most causes of attrition (65%) were for academic difficulties. This attrition rate is higher than the attrition rate in other physical therapy programs. The model including all of the predictor variables was not significant in predicting attrition. However, once those who experienced attrition for personal reasons were deleted from the data set, the model was able to significantly predict attrition (likelihood ratio = 15.876; p = 0.044). Two of the predictor variables, undergraduate GPA (odds ratio = 0.040; 95% Confidence Interval = 0.003 - 0.566) and average SAT score for the undergraduate institution (odds ratio = 0.990; 95% Confidence Interval = 0.984 - 0.997), were independent, significant predictors of attrition. Thus, after controlling for the other independent variables, as the value of undergraduate GPA increased by a value of 1.0, the odds of experiencing attrition decreased by 96%. Likewise, after controlling for the other independent variables, as the value of average SAT score for the undergraduate institution increased by only 1.0 point, the odds of experiencing attrition decreased by 1%. CONCLUSIONS: The admissions committee in the physical therapy program at this university should continue to emphasize the cognitive predictors when making admissions decisions. In addition, admissions committee members should begin to consider the quality of the applicant's alma mater, as measured by the average SAT score of students entering the institution. These results need to be replicated in other physical therapy programs before the results of this study can be generalized more broadly. FUNDING SOURCE: None. KEYWORDS: admissions, Education: Physical Therapist Students, entry-level education, attrition, undergraduate Copyright 2009 by the American Physical Therapy Association. Requests for reprints should be directed to the corresponding author of the article. Educators, students, and other academic customers may receive permission to reprint copyrighted material from Physical Therapy (ISSN 1538-6724) by contacting the Copyright Clearance Center Inc, 222 Rosewood Dr, Danvers, MA 01923. Other types of customers who want permission to reprint should contact the APTA Editorial Office, Attn: Physical Therapy.