Max r reliability estimates were significantly and consistently greater than corresponding alpha estimates for Harris-Lingoes subscales of the MMPI-2 containing 13 to 21 items. As previously reported with subscales of fewer items, Max r distributions were negatively skewed. Implications of these findings were discussed.
Get full access to this article
View all access options for this article.
References
1.
CronbachL. J. (1943) On estimates of test reliability.Journal of Educational Psychology, 34, 483–494.
2.
FeldtL. S., and BrennanR. I. (1989) Reliability. In LinnR. I. (Ed.), Educational measurement. (3rd ed.) New York: Macmillan. Pp. 105–146.
3.
KelchL. W., and WagnerE. E. (1992) Maximized split-half reliabilities and distributional characteristics for Harris-Lingoes subscales with few items: a reevaluation for the MMPI.Perceptual and Motor Skills, 75, 847–850.
