skip to main content skip to footer

Predictability and Intellectual Growth - Some Comments on the Degree and Interpretation of Growth Correlations SCAT STEP

Hilton, Thomas L.
Publication Year:
Report Number:
ETS Research Memorandum
Document Type:
Page Count:
Subject/Key Words:
Achievement Tests, Correlation, Elementary School Students, Grade Equivalent Scores, High School Students, Intellectual Development, Longitudinal Studies, School and College Ability Tests (SCAT), Sequential Tests of Educational Progress (STEP), Test Interpretation


Five possible interpretations are given of very high correlations between scores on successively administered ability tests in a longitudinal sample of approximately 7,000 public school students tested in grades 5, 7, 9, and 11. At each of the four grades, students were given the appropriate level of the Sequential Test of Education Progress (STEP) and the School and College Ability Test (SCAT). The correlation between grade levels of a verbal factor and a quantitative factor were: verbal factor, .94 (5th grade vs. 7th), .95 (7th vs. 9th), and .96 (9th vs. 11th); and for the quantitative factor, .90 to .93 to .95. The interpretations are: (1) During these two-year periods, U.S. students change intellectually very little; (2) The high correlations result from methods or from factors specific to each SCAT and STEP test; (3) The high correlations result from the tests' measuring general intellectual abilities which mature without being influenced by differential student experience; (4) Which school a student attends makes no difference; and (5) Each student's growth rate is set early in his or her life and remains constant thereafter. None of the five interpretations were found to be wholly acceptable. It is concluded that suitable measures of all variables related to the data analysis of each of the probable causal pathways involved in the growth process in question are needed.

Read More