Association of MMPI Validity Scales and Therapist Ratings of Psychopathology in Adolescent Psychiatric Inpatients

Although some self-report instruments (e.g., the Minnesota Multiphasic Personality Inventory, MMPI) have specifically been designed to consider the effects of participants' response styles in their disclosure of psychopathology, many of the most commonly used instruments (e.g., the revised edition of the Symptom Checklist-90, SCL-90-R; the Beck Depression Inventory, BDI) do not. This may compromise the accuracy of these measures, particularly in patient populations that may be liable to biased self-reports. Although observer ratings are often preferred to self-report data, no effort has been made to determine whether or not participant response styles might influence ratings provided by clinicians. This study examined 235 adolescent psychiatric inpatients with both self-report and therapist ratings of depression and nondepressive psychopathology together with common measures of response styles. Confirmatory factor analysis revealed that MMPI scales sensitive to response style were strongly correlated with both self-report and therapist ratings of the severity of psychopathology. Both transparent self-report measures and therapist ratings of psychopathology were significantly higher in patients with high F scales, with both of these ratings significantly lower in individuals with high L or K scales. Despite arguments regarding the desirability of observer ratings to avoid the influence of response bias, these results suggest that they also share variance with measures of response style.