This study was designed to determine whether the scores obtained from the Kindergarten Diagnostic Instrument (KDI) Test given to children prior to their enrollment in kindergarten would allow educators to identify those children who would eventually qualify for a first grade reading intervention program. A total of 346 children were randomly divided into 2 groups, and the test scores for each group were analyzed with logistic regression models. The results for each group were cross-validated by determining the model's ability to correctly classify the children in the other group. The logistic regression analyses indicated that when the KDI Test scores were used as predictor variables, the models were not able to meet the statistical and practical criteria set in the study. Both models classified too many students as qualifying for the reading intervention who in fact did not qualify. (Six tables of data are included and 25 references are attached.) (Author/SR)