Title

Construct and Criterion Validity of a Shortened form of the Total Individual Progress Level I Pre-Kindergarten Screen

Date of Award

1987

Document Type

Dissertation

Degree Name

Doctor of Education (Ed.D.)

Institution Granting Degree

Ball State University

Cedarville University School or Department

Psychology

Keywords

Educational psychology, preschool tests, educational measurements

Abstract

The purpose of the study was to investigate the construct and criterion validity of a shortened form of the Total Individual Progress Level I Pre-kindergarten Screen (TIP-SF). The subjects for the analysis of construct validity were 342 students from a rural and suburban school district in east-central Indiana who were administered the TIP-SF prior to kindergarten entrance during the years 1985 and 1986. The subjects for the analysis of criterion validitywere a subset of 153 students who were administered the TIP and TIP-SF prior to kindergarten entrance in 1985, and were subsequently administered the PMARP prior to their first grade year in 1986.

Principal components analysis with varimax rotation was conducted to evaluate the underlying constructs of TIP-SF. From the 28 items entered as variables, 10 factors with eigenvalues greater than 1.00 emerged. One-, two-, three-, four-, and five-factor solutions were successively imposed and evaluated. The solution containing four factors, Scholastic Aptitude, Motor, Eye/Hand Coordination, and Communication, was selected as the "best" solution. The obtained factors were similar in number, but dissimilar in nature to the pre-established subscales.

Simple and step-wise multiple regression and canonical correlation analyses were used to evaluate the relative abilityof scores obtained from TIP and TIP-SF to predict performance on Primary Mental Abilities Readiness Profile (PMARP) subtests. In each of the four canonical analyses, one significant and meaningful correlation between the sets ofpredictor and criterion variables was obtained. Step-wise multiple regression analyses were used to determine more specifically the predictive ability of TIP and TIP-SF factors and pre-established subscales. Zero-order correlations were computed between the TIP and TIP-SF total scores and the PMARP subscales.

Overall, the results of these analyses indicate that the use of the TIP total score provides the best prediction ofPMARP subscales. While the TIP-SF total score reliability is comparable to that for the TIP total score, it was not generally as highly related to the PMARP subscale scores. The linear composites of the TIP and TIP-SF factor scores and TIP pre-established subscale scores were approximately equivalent to one another for the prediction of PMARP scores.

Share

COinS