Psychometric evaluation of the Personal Problem-Solving Inventory for adolescents (PPSI)
Abstract
The purpose of this study was to develop and validate psychometrically a problem-solving instrument for adolescents. The PPSI is a 62-item multidimensional measure of interpersonl problem solving. Each item is rated on a Likert-type scale from 0 (not at all true of me) to 4 (extremely true of me). The items are divided into three scales: Automatic Processing, Problem Orientation, and Problem-Solving Skills. The Problem Orientation scale is further divided into three subscales: (a) Cognition, (b) Emotion, and (c) Behavior. The Problem-Solving Skills scale is divided into four subscales: (a) Identifying Problems, (b) Listing Alternatives, (c) Weighing Consequences, and (d) Implementing/Evaluating/Reorganizing. Psychometric data for the PPSI was collected in two phases. The first phase consisted of administering the 80-item instrument to 394 14 and 15 year-old male and female subjects; the second phase included administering the 65-item revision of the PPSI to 360 male and female subjects of the same age. The results indicated that the PPSI was internally consistent, r =.94. Reliability coefficients for the three scales were r =.81,.92, and.94, respectively; and.74,.88,.86,.82,.85,.84, and.87, respectively, for the seven subscales. The total PPSI was shown to be stable over a two-week interval, r (279) =.82, p $<$.0001. Test-retest correlations for the three scales were r (325) =.67, r (311) =.78, and r (304) =.77, and significant at p $<$.0001; correlations for all seven subscales were also significantly correlated. Convergent evidence for construct validity was based on a correlation with another measure of problem solving, the Problem-Solving Inventory, r (316) =.82, p $<$.0001 and with a measure of personal problems, the Personal Problems Checklist, r (318) = -.32, p $<$.001. Discriminant validity for construct validity was.40, p $<$.001, based on a correlation with grade point average. Factorial validity supported the problem-solving model as items loaded on nine factors. Criterion validity was based on the inverse relationship with a measure of distress, the Brief Symptom Inventory, r -.32, p $<$.001. Application and investigation of the utility of the instrument appear warranted. Future research with this instrument should focus on the validation process and the connections between problem solving and stress, as well as other health parameters.
Degree
Ph.D.
Advisors
Black, Purdue University.
Subject Area
Educational evaluation|Health education|Psychological tests
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.