The effect of a modified test administration procedure on the performance of male and female African American inner-city school children on a group administered intelligence test

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)

First Committee Member

James D. Mckinney - Committee Chair


African-American children typically score lower than other children on group-administered tests, and are over-identified for placement in remedial and special education classes. Since disadvantaged African-American children also tend to be poor readers and standardized group-administered intelligence tests require proficient reading skills, such tests may underestimate the general ability of African-American children. The purpose of this study was to investigate the effects of an audio-taped oral test administration procedure on the performance of African-American 4th and 5 th grade children who were poor readers.The research design was a 2 (gender) x 2 (administration condition) factorial in which male and female students were randomly assigned to either the audio-taped treatment or the conventional treatment. The final sample contained 123 students with an overage reading comprehension level at the 26th percentile of the Stanford Achievement Test. The sample was drawn from three inner-city schools which served predominantly African-American children, with more than 96% of the students receiving free and reduced lunch.In the audio-taped condition, students were given an abbreviated version of the Otis-Lennon School Ability Test (OLSAT); these students read while they listened to a tape of the instructions, item questions, and multiple choice responses. Students in the control group were given the same test using the instructions specified in the test manual (the conventional condition).Preliminary analyses indicated that test performance was not significantly correlated with SAT scores; however, it was correlated with age at testing. Therefore, the intelligence test data were analyzed by a 2 (gender) x 2 (administration condition) analysis of covariance. The results indicated that students who received the audio-taped condition scored higher than those who had the conventional administration procedure. There were no significant gender differences and no gender by treatment condition interaction. Thus, the audio-taped administration procedure resulted in a more accurate assessment of the learning potential of African-American poor readers than the conventional procedure for group-administered tests.


Education, Tests and Measurements; Black Studies; Education, Elementary; Education, Educational Psychology

Link to Full Text


Link to Full Text