Dear Tipsters,
There have been many studies of the predictive validity of the GRE, the ones
conducted independently of ETS being of most interest. There are also reports
about the construction of the GRE by our colleagues who have been involved
(e.g., Kalat and Matlin, 2000).
Kuncel et al. (2001) conducted a meta-analysis and found the following 90%
credibility intervals for the predictive validity of the GRE and various
criteria:
Predictor Criterion
Graduate GPA Comp. exams Faculty ratings Degree
attainment
Verbal .09 to .59 .19 to .69 .19 to .65
overlaps 0
Quantititative .19 to .45 .16 to .36 .40 to .54
overlaps 0
Analytical .26 to .46 .35
overlaps 0
Subject .30 to .52 .51 .20 to
.80 .11 to .67
Undergrad GPA .12 to .48 .12 .12 to .58
overlaps 0
In a special study that had no restriction of range because of an open
admissions policy, Huitema and Stein (1993) found the following a validity
coefficients for psychology:
Predictor Criterion
Advanced Stats Behavioural Assessment
Overall courses
Course
Total GRE .60 .70
.63
Verbal+Quantitative
References:
Kalat, J. W., & Matlin, M. W. (2000). The GRE psychology test: A useful but
poorly understood test. Teaching of Psychology, 27, 24-27.
Kuncel, N. R., Hezlett, S. A., & Ones, D. S. (2001). A comprehensive
meta-analysis of the predictive validity of the Graduate Record Examinations:
Implications for graduate student selection and performance. Psychological
Bulletin, 127, 162-181.
Huitema, B. E., & Stein, C. R. (1993). Validity of the GRE without restriction
of range. Psychological Reports, 72, 123-127.
___________________________________________
Stuart J. McKelvie, Ph.D.,
Department of Psychology,
Bishop's University,
Route 108 East,
Borough of Lennoxville, Sherbrooke,
Québec J1M 1Z7,
Canada.
E-mail: [EMAIL PROTECTED]
Phone: (819)822-9600, Extension 2402
Fax: (819)822-9660
Bishop's Psychology Department Web Page:
http/:www.ubishops.ca/ccc/dev/soc/psy
__________________________________
---
You are currently subscribed to tips as: [email protected]
To unsubscribe send a blank email to [EMAIL PROTECTED]
