The relationship between type of reading instruction (whole language or basal) and reading comprehension as measured by the California Achievement Test (CAT) was investigated. Nineteen third- and fourth-grade teachers from school 5 districts in North Carolina took part in the project with 215 students in the basal group and 200 students in the whole language group. Multiple regression analysis was used to predict the comprehension scale score on the spring 1992 administration of the CAT. Predictor variables included intelligence quotient (IQ), reading instruction group (whole or basal), race, and CAT pretest score. Results indicate that the dependent variable is not significantly related to the method of reading instruction, and there is no interaction effect between reading method and IQ in predicting the CAT comprehension scale score. Two tables present study data. (Author/SLD)