This follow-up study helped determine the predictive validity of parent and teacher ratings and scores from the Early Identification Assessment Battery. This assessment involved a random selection of 4-year-old children scheduled to enter a French immersion program in the Ottawa, Canada, school system. In the spring of grade 1, 63% of the original group were still in French immersion; these children were classified as high or low achievers based on their teachers' ratings and on tests of academic achievement in English and French. The data indicated that highly accurate predictions of progress in French immersion programs could be made on the basis of the 4-year-old assessment battery. The significant predictor variables included IQ measures, tests of spelling and word segmenting, naming of letters and numbers, auditory discrimination, and color naming. The test scores were far more accurate than background and family variables in predicting achievement. Sample questionnaires and test batteries are appended. (SW)