Abstract: | Within the context of a counterbalanced design, 102 students from either a high school or a large Southeastern university were administered two versions of the Minnesota Multiphasic Personality Inventory–Adolescent (MMPI–A): a computer‐administered version (CA) and a paper‐and‐pencil version (PAP). Time between testing sessions was approximately one week. Differences in individual scale means between the CA and PAP were calculated using paired t tests, with the Bonferroni correction procedure; no mean differences were statistically significant ( p > .05). To determine if the scale distributions were similar Hartley's homogeneity of variance tests were conducted; there were no differences in the shapes of the scale distributions ( p > .05). Pearson product‐moment coefficients were calculated to determine if the relative rankings were similar across administration formats; coefficients for every scale were positive and statistically significant ( p < .01). © 2005 Wiley Periodicals, Inc. Psychol Schs 42: 605–613, 2005. |