100 MMPI-2 protocols were obtained from the files of an outpatient clinic. Distributions of all possible split-half combinations were computed for selected Harris-Lingoes subscales with few items. As expected, max r reliability estimates were significantly greater than alpha estimates and distributions of correlations were negatively skewed. Inspection of the distributions suggested that the Amorality scale contains an ambiguous item which does not appear to belong in this grouping. Serious problems were noted with the Hy5 subscale. Alpha reliabilities were comparable to previously reported estimates, and max r reliability estimates closely resembled reported test-retest estimates.
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, 485–494.
2.
GrahamJ. R. (1990) MMPI-2: Assessing personality and psychopathology. New York: Oxford Univer. Press.
3.
MillerH. R.StreinerD. L. (1985) The Harris-Lingoes subscales: Fact or fiction?Journal of Clinical Psychology, 41, 45–51.
4.
WagnerE. E.AdairH. E.FoerstnerS. (1990) Maximized reliability estimates for some research scales of the MMPI. Educational and Psychological Measurement, 50, 775–783.
