NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED158674
Record Type: RIE
Publication Date: 1978-Mar-14
Pages: 23
Abstractor: N/A
Reference Count: 0
ISBN: N/A
ISSN: N/A
Effect of Selection Ratio on Correlations Between Admissions Criteria and Measures of Dental Student Success.
Yancey, John M.; Stewart, Arthur Van
Correlations between popular admissions indicies, including Dental Admission Test scores and pre-dental grade point average, and numerous measures of student success, such as National Board scores and grades, were compared for recently-graduated students from two dental schools with substantially different selection ratios. While the differences in retrospective predictive validity coefficients for some measures of success were in the direction most would predict (higher for the school with a greater-than-40-percent selection ratio than for the school with a less-than-10-percent selection ratio), the magnitude of the differences was not nearly as large as might be inferred from the most frequently published Taylor-Russell Tables, which assume either a .50 or .60 base rate (proportion of applicants who would be successful if the predictor in question were not used in the selection process). In view of the very low academic attrition in almost all dental schools and the fact that failure is defined as more than two standard deviations below the raw score mean on the norm-referenced National Board examinations, the more appropriate references to consult in estimating changes in predictive ability of admissions indicies as a function of relative applicant pool size are those seldom-published tables with a base rate of .80 or .90. One major implication of this study is that selection ratio does not affect the predictive validity of the most common dental school admissions measures as much as is gathered from reading the more popular texts on psychological testing. (Author/LBH)
Publication Type: Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Identifiers: N/A
Note: Paper presented at the AADS Annual Meeting (Washington, D.C., March 14, 1978); Some illustrations may not reproduce well