The Science Journal of the American Association for Respiratory Care

2008 OPEN FORUM Abstracts

FACTORS THAT PREDICT PERFORMANCE IN A RESPIRATORY CARE PROGRAM

Leonard D. Wittnebel1, Douglas Murphy1, David L. Vines1



Background: Our Respiratory Care baccalaureate program experienced a 47% attrition rate over a five-year period from 2001-2005. The purpose of this study was to examine the relationship between program completion and a set of predictors: proportion of prerequisites completed at a four-year college or university (P4), total proportion of prerequisites completed (TP), prerequisite grade point average (PGPA) and student's age (AGE).

Methods: An ex post facto review of records for students admitted between 2001-2005 was conducted to collect needed information. The sample did not include advanced standing students, students with missing data, and students at a distant education site. All data pertaining to PGPA was obtained from a pre-existing departmental database, and the 4P and TP values were calculated on a 0-1.0 scale based on student transcripts. In addition, descriptive characteristics including age, sex, and ethnicity were collected. Logistic regression analyses were conducted to identify predictors and ascertain their relationship with graduation status.

Results: Four predictors were entered in stepwise fashion into the logistic regression: (1) PGPA, (2) TP, (3) P4, and (4) age. This four-predictor model was significant, but neither P4 nor AGE contributed unique explanation of variance. The model was reduced to a two-predictor model (PGPA and TP), which was significant (X2(3, N = 139) = 41.97, p < .000), with R2 = .44. Contrary to expectations, TP was the better predictor of graduation status (R2 = .40); PGPA added a small but significant increment in explained variance. The model correctly classified graduation status for 83.3% of the sample. However, prediction for those not graduating was not better than chance (53.3%).

Conclusion: Based on results, it may prove useful to consider TP more important than PGPA when predicting graduation probability for potential admits. Prediction models incorporating age and type of prior institution attended (2-yr vs. 4-yr) was not useful in making admission decisions. Further research is needed to determine factors contributing to the remaining variance of the sample's graduation success and failure.