NotesFAQContact Us
Search Tips
ERIC Number: ED202420
Record Type: RIE
Publication Date: 1981-Apr
Pages: 17
Abstractor: N/A
Reference Count: 0
Predicting Admissions Committee Behavior in a Medical School.
Wergin, Jon F.
The decisions made by admissions committee members of the Medical College of Virginia were studied to determine the criteria used to arrive at value judgments and to analyze variations in predicted ratings based on these criteria. All 983 applicants to the 1980-81 entering class of the medical school who underwent file review evaluations (the total applicant pool was 2,315) were studied. The first study phase consisted of a series of stepwide multiple regression equations for each rater, using file review ratings and interview ratings as criterion variables. Predictor variables for these regressions included age, sex, minority status, presence of graduate degrees, previous applications to medical school, scores on the Medical College Admission Test, and undergraduate grade point average both in science courses and overall. Regressions were calculated both for individual committee members and the committee as a whole, and residents and nonresidents were analyzed separately. In the second study phase, each committee member's file review regression equation was applied to a random sample of 50 residents and 50 nonresidents drawn from the pool of 983 files. Results were then analyzed for the difference between the highest and lowest pairs of predicted ratings and for the placement of the actual ratings within the distribution of predicted ratings. As expected, the prediction formulas suffered from a lack of precision, due in part to the existence of certain qualitative criteria (e.g., quality of undergraduate school) not entered into the analysis. The amount of ratings variance was generally higher for individual members than for the committee as a whole. For file reviews, ratings assigned to resident applicants tended to be more predictable on the basis of admissions characteristics than those assigned to nonresidents. Additional findings and policy implications are considered. (SW)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Identifiers: Medical College of Virginia
Note: Paper presented at the Annual Meeting of the American Educational Research Association (Los Angeles, CA, April 13-17, 1981).