 |
|
|
| 
|
The Mentor: An Academic Advising Journal Predictors of Success In Exercise Physiology Scott R. Collier, Syracuse University; Todd M. Manini, University of Florida; and John W. Tillotson, Syracuse University Introduction Once a student enters higher education, there are few tools available to the academic adviser to aid in predicting future success or upcoming course difficulties for the undergraduate. Especially difficult to judge is the future success of a student who wishes to take course work in a different curriculum and work toward the completion of a degree that was not of original intention. Mathematical models could be developed to predict future success or identify possible trouble areas that may need attention. To capture a large percentage of would-be change-of-major students, these models would include first-semester course work that is attempted by most of the university population, such as general biology. Also, a general indicator of a student's aptitude would be helpful to determine the student's current level of understanding. General indicators of academic success, such as SAT scores, are a major criterion for admission into higher education. The SAT-I is a standardized test used to gauge the development of mathematical and verbal reasoning abilities of students; it is usually taken near the conclusion of a student's junior year in high school. The SAT-I is a multiple-choice and short-essay test, developed by a board of experts in the specific content areas of reading, mathematics, and composition. While Basten, Cole, Maestas, & Mason (1997) have shown that SAT scores are the strongest predictor for an institution's selection of students, Rooney and Schaeffer (1998) have shown that almost 300 colleges in the United States have eliminated the policy of using the SAT as a criterion in their decision-making process. Within the last few years, college administrators have begun discontinuing the use of SAT scores as an indication of future college performance (Castaneda v. The Regents of the University of California, 2003). Correlation predictors kept by the Department of Education from 1980 to 2004 have shown individual or combined SAT scores to be weak to medium predictors for future academic success of undergraduates (Kidder & Rosner, 2002). Moffatt (1993) showed that combined SAT scores were moderately correlated with success in college (r = .56), which was higher than published data from Baron and Norman (1992) (r = .20) and Willingham (1985) (r = .43). When combined SAT verbal and math scores were used during the years 1986 to 1993, the correlations were slightly better (r = .36). However, individual weighted averages for SAT verbal and math scores between the years 1986 and 1993 were slightly higher (Baron & Norman, 1992; Willingham, 1985) (r = .40 and .41 respectively), suggesting individual scores may be a better predictor of future success than a combination of scores. Some college administrators have stated that until there is a better measure of normative data across high schools, the SAT will have to be used (Gose & Selingo, 2001). At the collegiate level, curriculum frameworks are established to assist educators in addressing standards and helping students accumulate proper background knowledge to prepare for increased levels of competency in higher-level courses. Colleges expect academic departments to have well-designed curricula providing meaningful context for knowledge and skills with a balance of content and process. For example, courses such as introductory biology lay the foundation for more advanced courses in physiology taken later in students' academic careers. With the use of prediction analyses, advisers would be armed with more information regarding the future success of their students, which may increase the likelihood of future success and decrease the risk of failure. Also, academic advisers would be able to use an additional tool to determine future program success within their course curriculum for a student wishing to transfer into a different program. Even with the best intentions, the use of prediction equations cannot be adjusted for intrinsic drive that a student may have in a career path. With this in mind, one should use prediction equations not to “profile” a student, but rather as a method to determine potential pitfalls that may be avoided by additional assistance in the area of concern. One goal of academic advising should be to give students the best chance to succeed in their program choice by transitioning them from past to future course work. It has been shown that students are more likely to attain higher grades in courses that are complementary to their field. Shoemaker (1986) examined SAT scores from 296 engineering majors and compared their math and verbal scores to predict undergraduate grade-point averages. Correlations for their verbal scores were low (r = .21), while math was moderately associated with their undergraduate grade-point average (r = .43). Since the engineering curriculum has many math-based courses, SAT math scores are a better predictor of academic success than verbal scores. Likewise, knowledge in physiology gained by athletes through participation in sports may increase their performance in Exercise Physiology courses, as it is postulated that personal significance is critical for successful learning (Newby, 1991). However, there is recent evidence that athletes fare no better than their non-athlete colleagues in retaining physiology knowledge gained through repeated usage from their engagement in sports (Clark, Webster, & Druger, 2006). After a student enters college, however, there are few studies citing whether the integrated curricula are an effective predictor of academic success (Brophy & Alleman, 1991). The extent to which a student can process concepts at a higher level may be indicative of the student's aptitude, not the course content. Therefore, a general measure of knowledge such as the SAT score may be a better predictor than collegiate course work (Lucas, Cox, Croudace, & Milford, 2002). Exercise science is a field of science that studies the effect of exercise on the human body. Exercise science majors are required to complete science-based course work in biology, anatomy, physiology, chemistry, and physics. This gives students the base knowledge and science background when they arrive on campus, yet due to the applied nature of the field, verbal skills are necessary for future academic success. Conversely, students in non-science majors such as physical education are required to take both biology and exercise physiology courses to complete requirements for their respective programs. Students with a natural appreciation of science may perform better in biology and physiology course work, yet attention to earlier classes may help students and their advisers make well-informed academic decisions prior to potential trouble in class. Therefore, the purpose of this study was to investigate whether success in first-year Biology 1 and 2 courses is related to success in a junior-level Exercise Physiology course, while accounting for general aptitude test (SAT) scores and background knowledge. Method Study Sample Grades of seventy-two students enrolled in an Exercise Physiology (EP) course in 2003, 2004, and 2005 at a medium-sized university were matched to their performance in Biology 1 and 2 and with their SAT-I scores. (Database grade queries were approved by the university's Institutional Review Board.) Biology 1 and 2 are first-year courses, while Exercise Physiology is a junior-level course. All final course grades and SAT scores were entered into an SPSS data sheet by an independent administrator who coded the entries and removed the names to eliminate potential investigator bias. Grade-point averages (GPA) were categorized as follows, using the standard grade-point average at Syracuse University: (0) = 0.0–1.0 GPA; (2) = 1.1–2.0 GPA; (3) = 2.1–3.0 GPA; and (4) = 3.1–4.0 GPA. Ordinal logistic regression was used to predict EP final grades from Biology 1 and Biology 2 and SAT scores for seventy-two exercise science and physical education majors. Scores from exercise science (ES, science field) versus physical education (PE, non-science field) were used to determine the odds ratio for predicting success in the Exercise Physiology course. Logistic regression estimates the probability of an event occurring and allows for the simultaneous comparison of more than one contrast. Unlike ordinary least squares regression, logistic regression does not assume linearity of relationship between the independent and dependent variables (Rogers & Swaminathan, 1993). Ordinal logistic regression does not require normally distributed variables and does not assume homoscedasticity, which in general makes this less stringent. Since the numbers used in this statistical modeling are not linear in nature, ordinal logistic regression was employed as the best way to analyze the data (Rogers & Swaminathan, 1993). Treatment of the Data Ordinal logistic models were developed in SPSS version 14. The ordinal logistic modeling function was used for both methods. The link function for ordinal logistic regression is binomial. The logistic regression analyses were performed on the group as a whole. The regression models included the following predictors: (1) Biology 1 grades; (2) Biology 2 grades; (3) SAT verbal and math scores; (4) Student group (ES vs. PE); and (5) Athlete status (yes vs. no). Results Table 1 displays the number of subjects by grouping variable in each GPA quartile. Table 1 Categorical descriptive characteristics of Exercise Physiology course grades earned by grouping variable of science vs. non-science students (grade-point average scale: 0 = F; 1 = D to C-; 2 = C to B-; 3 = B to A; SAT scores are presented as Mean ± SE)
In unadjusted analysis (Table 2a) and when compared to ES students, PE students were 83% less likely to have higher grades in Exercise Physiology [odds ratio (OR): 0.17, 95% confidence interval (CI): .06 to .48, p=0.001]. Table 2a Ordinal Logistic Regression of group variables (0 = science track, 1 = non-science track), biology grades (Biology 1 and Biology 2), SAT-I scores (SAT verbal and math), and athletic status (athlete); unadjusted odds of Exercise Physiology performance
This effect changed little after adjusting for performance in Biology 1 and 2, SAT-Verbal, and SAT-Math scores (Table 2b) [odds ratio (OR): 0.25, 95% confidence interval (CI): .07 to .85, p=0.027]. These data show that an exercise science student is 83% more likely to do better in Exercise Physiology than a physical education student. Table 2b Ordinal Logistic Regression of group variables (0 = science track, 1 = non-science track), biology grades (Biology 1 and Biology 2), SAT-I scores (SAT verbal and math), and athletic status (athlete); odds ratios adjusted for all other covariates
In the unadjusted analysis, Biology 1 scores were strongly related to performance in Exercise Physiology (OR: 9.5, 95% CI: 3.68 to 24.54, p<0.001). Adjustment for Biology 2, major (ES or PE), SAT-Verbal, and SAT-Math scores did not augment this relationship. This is interpreted to indicate that for every grade level a student increases in Biology 1, they will have nine times greater odds of having a grade increase in Exercise Physiology. In the unadjusted analysis, Biology 2 was moderately related to performance in Exercise Physiology (OR: 4.2, 95% CI: 1.77 to 10.31, p=0.001). This is interpreted to indicate that for every grade level a student increases in Biology 2, they will have four times greater odds of having a grade increase in Exercise Physiology. Adjustment for Biology 1, major (ES or PE), SAT-Verbal, and SAT-Math scores did not augment this relationship. When both verbal and math SAT-I scores were entered into the model, the unadjusted analysis odds ratio was 1.02. Adjustment for all other variables in the model did not change this relationship. This is interpreted as no prediction power for SAT-I scores on a junior-level Exercise Physiology class. When athlete status was entered into the model, it was shown that Exercise Physiology scores may be worse for this condition. Full-model scores are displayed in Table 2 above. Discussion Our results indicate that Biology 1 is the best predictor for continued success in Exercise Physiology, while Biology 2 holds some predictive power. Further, exercise science students consistently score higher in Exercise Physiology than physical education students. Additionally, having an athletic background did not increase the odds of earning a higher grade in Exercise Physiology, which was not as hypothesized. These results remain significant after adjusting for the shared variances between majors. Also, the SAT-I verbal and math scores do not predict future success in Exercise Physiology. Many college students take Biology 1 and 2 as science courses in their first year of undergraduate work. There are increasing enrollments in Exercise Physiology courses as a prerequisite for physical therapy, physical education, and exercise science degree requirements. These data imply that students with higher-than-average grades in Biology 1 and 2 will do better in Exercise Physiology than a student with grades that are below average. This is important since these data show that Biology 1 is an important predictor for success in Exercise Physiology, yet there were no physical education students achieving an above-average (3.0 GPA) grade in either biology course. This can be interpreted that physical education students may want to pay more attention to their introductory biology course content and work to derive the greatest understanding of these concepts, which lay a foundation for concepts taught later in Exercise Physiology. As an adviser, one can recommend to the non-science student who was below average in Biology 1 to seek extra help from the start of Exercise Physiology. It was hypothesized that students who have prior knowledge of human physiology and terminology due to participation as student-athletes may perform at a higher level than non-athletes. In agreement with Clark et al. (2006), athletic status did not contribute to increased knowledge in Exercise Physiology and, in this study, was a negative indicator of success (odds ratio of < 1.0). Athletes had less success than non-athletes in Exercise Physiology, which may also raise a red flag when advising student-athletes prior to their enrolling in Exercise Physiology following a below-average grade in Biology 1. However, if a student wishes to change his or her major, advisers can use predictive equations as tools to help predict how the student will do in his or her potential major by investigating the Biology 1 grade. One limitation of this data is that it was derived from a single institution and recorded only for a three-year period. While the same teachers have been instructing this class during that period, there have been different teaching assistants during that same period, which may have contributed to varying grades in biology. Due to faculty variability, varying knowledge depth of teaching assistants, and unknown transitional experiences of incoming first-year students, it is important not to use predictive equations as a means of ranking students into ability levels. The “profile” of students should consider not just one aspect of their collegiate career, but all of their previous course work, any external measures of drive and ambition, and their extracurricular activities, such as participation in sports. In conclusion, it has been shown that Biology 1 is a better predictor of grade outcome in Exercise Physiology, with exercise science majors usually outperforming non-science students. Future interventions for the student should consider how to help students in Exercise Physiology who have poor biology grades. Further, SAT-I scores, either verbal or math, do not have any predictive power for final grades in Exercise Physiology. Future research should apply ordinal logistic regression to various other prerequisite classes in various academic departments to evaluate their predictive power. References Baron, J., & Norman, M. F. (1992). SATs, achievement tests, and high-school class rank as predictors of college performance. Educational and Psychological Measurement, 52, 1047–1055. Basten, J., Cole, H., Maestas, R., & Mason, K. (1997, November). Redefining the virtuous cycle: Replacing the criterion of race with socioeconomic status in the admissions process in highly selective institutions. Paper presented at the 22nd Annual Conference of the Association for the Study of Higher Education, Albuquerque, NM. Brophy, J., & Alleman, J. (1991). A caveat: Curriculum integration isn't always a good idea. Educational Leadership, 49(2), 66. Castaneda v. The Regents of the University of California, C 99-0525 (SI 2003). Clark, B. C., Webster, C., & Druger, M. (2006). Knowledge retention of exercise physiology content between athletes and non-athletes. Journal of College Science Teaching (July/August), 29–33. Gose, B., & Selingo, J. (2001, October 26). The SAT's greatest test. The Chronicle of Higher Education. Retrieved November 15, 2006, from http://chronicle.com Kidder, W. C., & Rosner, J. (2002). How the SAT creates 'built-in headwinds': An educational and legal analysis of disparate impact. Santa Clara Law Review, 43. Lucas, U., Cox, P., Croudace, C., & Milford, P. (2002, April). “Who writes this stuff?”: Students' perceptions of their skills development. Paper presented at the Critical Perspectives on Accounting Conference, New York, NY. Moffatt, G. K. (1993, February). The validity of the SAT as a predictor of grade point average for nontraditional students. Paper presented at the 16th Annual Meeting of the Eastern Educational Research Association, Clearwater Beach, FL. Newby, T. (1991). Classroom motivation: Strategies of first-year teachers. Journal of Educational Psychology, 83, 195–200. Rogers, H., & Swaminathan, H. (1993). A comparison of the logistic regression and Mantel-Haenszel procedures for detecting differential item functioning. Applied Psychological Measurement, 17(2), 105–116. Rooney, C., & Schaeffer, B. (1998). Test scores do not equal merit: Enhancing equity & excellence in college admissions by de-emphasizing SAT and ACT scores. Cambridge, MA: National Center for Fair & Open Testing (FairTest). Shoemaker, J. S. (1986, April). Predicting cumulative and major GPA of UCI engineering and computer science majors. Paper presented at the American Educational Research Association, San Francisco, CA. Willingham, W. (1985). Success in college: The role of personal qualities and academic ability. New York: College Board. About the Author Scott R. Collier, Ph.D., is assistant professor, Department of Exercise Science, at Syracuse University. He can be reached at sccollie@syr.edu. Published in The Mentor on January 4, 2007, by Penn State's Division of Undergraduate Studies Available online at www.psu.edu/dus/mentor/ Privacy and Legal Statements | Copyright | © The Pennsylvania State University | All rights reserved | |