Formulae for combining reliability coefficients from any number of samples are provided. These formulae produce the exact reliability one would compute if one had the raw data from the samples. Needed are the sample means, standard deviations, sample sizes, and reliability coefficients. The formulae work for coefficient alpha, KR-20, retest, alternate-forms, split-half, interrater (intraclass), Gilmer-Feldt, Angoff-Feldt, validity, and other coefficients. They may be particularly useful for meta-analytic and reliability generalization studies.
Get full access to this article
View all access options for this article.
References
1.
CharterR. A. (2001a) Formulas for combining summary statistics. Psychological Reports88, 1130.
2.
CharterR. A. (2001b) It is time to bury the Spearman-Brown “prophecy” formula for some common applications. Educational and Psychological Measurement, 61, 690–696.
3.
CharterR. A.AlexanderR. A. (1993) A note on combining correlations. Bulletin of the Psychonomic Society, 31, 123–124.
4.
FeldtL. S.CharterR. A. (2003a) Estimating the reliability of test split into two parts of equal or unequal length. Psychological Methods, 8, 102–109.
5.
FeldtL. S.CharterR. A. (2003b) Estimation of internal consistency reliability when test parts vary in effective length. Measurement and Evaluation in Counseling and Development, 36, 23–27.
6.
HensonR. K.ThompsonB. (2002) Characterizing measurement error in scores across studies: Some recommendations for conducting “reliability generalization” studies. Measurement and Evaluation in Counseling and Development, 35, 113–126.
7.
Vacha-HaaseT.HensonR. K.CarusoJ. C. (2002) Reliability generalizataion: Moving toward improved understanding and use of score reliability. Educational and Psychological Measurement, 62, 562–569.
