Item and subscale validity of three personality scales for blacks, whites, males, females in three age groups (child, adolescent and adult) /
| Main Author: | |
|---|---|
| Other Authors: | , |
| Format: | Thesis Book |
| Language: | English |
| Published: |
1990.
|
| Subjects: | |
| Online Access: | ProQuest, Abstract Link to OAKTrust copy |
| Abstract: | The Behavior Assessment System for Children (BASC) was designed to provide a comprehensive picture of personality and behavior. As part of the test construction process, the present investigation was conducted to determine the construct validity of the three self-report questionnaires on the BASC. The Children's Personality Profile (CPP) and the Adolescent Personality Profile (APP) were designed for students aged eight through 11 years, and 12 through 19 years, respectively. These two questionnaires consist of clinical, adaptive, and validity scales. The Parent Personality Profile (PPP) measures parental variables that have been found to be related to childhood psychopathology. The first phase of the study involved the identification of items on the three questionnaires for which blacks and whites, and males and females provided non-equivalent responses. The partial correlation procedure was used for this purpose. The subjects were drawn from the standardization sample of the BASC, and each sample (CPP, APP, and PPP) was randomly split for cross-validation. The results of the item bias analyses are described by item relative to the total score on each scale. On the CPP, five of the 225 items were answered differently by blacks and whites. Males and females provided non-equivalent responses to 20 items. On the APP, blacks and whites responded differently to 31 of the 262 items, while males and females provided 62 non-equivalent responses. On the PPP, blacks and whites responded differently to 24 of the 157 items, while males and females provided non-equivalent responses for 14 items. The items for which non-equivalent responses were provided were approximately evenly distributed among blacks and whites, and among males and females. Factor analytic procedures were used to determine if the factors were invariant for blacks and whites, and for males and females. Invariant factors would indicate that the non-equivalent item responses reflected significant real differences among the groups. The first factor, which accounted for the largest proportion of the common score variance, was found to be constant for all cross/ethnic, and cross/gender comparisons for the CPP, and APP. The factor analytic results for the PPP were inconclusive due to possible effects of small sample size, and socioeconomic status. |
|---|---|
| Item Description: | Typescript (photocopy). Vita. "Major subject: School psychology." |
| Physical Description: | xiii, 252 leaves : illustrations ; 29 cm |
| Bibliography: | Includes bibliographical references. |