Twenty-seven third-grade children in a northern Louisiana public school participated in a study which examined whether a minimum amount of direct instruction in sight word recognition, combined with drill of sight words, could result in significantly better performance for the targeted at-risk children on the story and unit reading tests. The control group consisted of 16 students in an at-risk self-contained class; the experimental group consisted of 11 at-risk students, 3 of whom were special education inclusion students. Both groups received 165 minutes of language arts instruction daily from their respective regular education teachers. Control group students received instruction in the basal reading series from their regular education teacher in a whole-group setting. The intervention used with the experimental group focused on expansion of direct instruction in vocabulary associated with each story and unit test. Data were collected over a 6-week period. Analysis revealed a significant difference between experimental and control group scores, in favor of the experimental group, on the cumulative unit test. The Mann Whitney Test revealed a significant difference in favor of the experimental group on the following unit test components: vocabulary, fact/opinion, and graphic aids. Data collected indicated that very brief direct instruction in sight word recognition combined with daily word drill assignments resulted in vocabulary and comprehension unit test scores of at-risk third graders significantly higher than those of control group students. (Contains 22 references and 4 tables of data.) (NKA)