ScholarLens Back to search
1973•The Journal of Educational ResearchRequires access

The Predictive Validity of the Lorge-Thorndike Intelligence Tests at the Kindergarten Level

Glen E. Mendels

Open publisher page 5 citations

Abstract

The purpose of this study was to investigate the predictive validity of the Lorge-Thorndike Intelligence Tests (LTIT) at the kindergarten level. Seventy-nine Ss were administered the LTIT at the end of kindergarten. In mid-first grade the Ss were tested with the Metropolitan Achievement Tests. In addition, teacher grades and reading level data were collected for each S. Correlations between the LTIT and the academic criteria ranged from .46 to .62. While stepwise regression analyses revealed that the demographic variables of sex, school ability, age, and father’s occupation increased the power of prediction (multiple R’s ranged from .64 to .78), the LTIT alone was judged to be a valid predictor of academic achievement.

About this research paper

What this paper is about

The purpose of this study was to investigate the predictive validity of the Lorge-Thorndike Intelligence Tests (LTIT) at the kindergarten level. Seventy-nine Ss were administered the LTIT at the end of kindergarten. In mid-first grade the Ss were tested with the Metropolitan Achievement Tests. In addition, teacher grades and reading level data were collected for each S. Correlations between the LTIT and the academic criteria ranged from .46 to .62. While stepwise regression analyses revealed that the demographic variables of sex, school ability, age, and father’s occupation increased the power of prediction (multiple R’s ranged from .64 to .78), the LTIT alone was judged to be a valid predictor of academic achievement.

Why it matters

OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The purpose of this study was to investigate the predictive validity of the Lorge-Thorndike Intelligence Tests (LTIT) at the kindergarten level. Seventy-nine Ss were administered the LTIT at the end of kindergarten. In mid-first grade the Ss were tested with the Metropolitan Achievement Tests. In addition, teacher grades and reading level data were collected for each S. Correlations between the LTIT and the academic criteria ranged from .46 to .62. While stepwise regression analyses revealed that the demographic variables of sex, school ability, age, and father’s occupation increased the power of prediction (multiple R’s ranged from .64 to .78), the LTIT alone was judged to be a valid predictor of academic achievement.

Key concepts: Predictive power, Predictive validity, Stepwise regression, Psychology, Academic achievement, Regression analysis, Developmental psychology, Test (biology)

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
The Predictive Validity of the Lorge-Thorndike Intelligence Tests at the Kindergarten Level — Research Paper | ScholarLens