Publication Date


Advisor(s) - Committee Chair

Carl Martray, Doris Redfield, Julia Roberts

Degree Program

Department of Psychology

Degree Type

Master of Arts


The purpose of this study was to investigate the relationship, among gifted students, between scores obtained on the Structure of Intellect-Learning Abilities (SOI-LA) test and two measures of achievement: teacher assigned grades and scores obtained on the Comprehensive Test of Basic Skills (CTBS). This study was based on the assertion that academic abilities should be linked to specific cognitive abilities measured by the SOI-LA subtests. Significant, positive relationships between academic abilities and SOI-LA subtest scores would imply that curricula based on the Structure of Intellect theory, in areas identified as deficient by the SOI-LA tests, may increase achievement among the gifted population.

One hundred fifty-seven academically gifted students enrolled in grades 5 through 7 during the 1984-1985 school year were selected for this study. All of the participants qualified for admittance into the Gifted and Talented (GAT) program in a south central Kentucky school district. The SOI-LA and CTBS were administered between November 1984 and April 1985 by one of two GAT teachers; classroom teachers additionally provided grades in reading, language arts, and mathematics.

Stepwise multiple regression analyses indicated that:

  1. Convergent Production of Semantic Systems (NSS) alone was the best predictor model for teacher assigned math grades.

  2. Cognition of Semantic Relations (CMR) and Divergent Production of Semantic Units (DMU) together provided the best predictor model for teacher assigned reading grades.

  3. None of the SOI-LA test variables proved significant predictors of teacher assigned language arts grades.

  4. The combination of Evaluation of Symbolic Classes (ESC), Cognition of Semantic Relations (CMR), Convergent Production of Figural Units (NFU), Memory of Symbolic Implications (MSI), Convergent Production of Symbolic Systems (NSS), and Cognition of Symbolic Relations (CSR) provided the best predictor model for CTBS math scores.

  5. The combination of CMR and CMU was shown to be the best predictor model for CTBS reading scores.

  6. The best predictor model for CTBS language scores was Convergent Production of Symbolic Transformations (NST), ESC, Memory of Symbolic Units-Visual (MSU-V), DMU, MSI, CMR, and NFU.

Pearson product-moment coefficients were additionally calculated to facilitate the interpretation of the multiple regression analyses.

An explanation for the SOI-LA test's relatively poor predictive power for teacher assigned grades, compared to standardized test scores, may be the subjective nature of the assigned grades rather than the validity of the SOI-LA test.


Education | Gifted Education | Psychology | Social and Behavioral Sciences