This paper presents a study that assessed whether attributional retraining is an effective method for enhancing academic achievement in at-risk college students. Attributional retraining was provided to college students (N=198) who had previously performed either poorly or well on a test. Attributional retraining and previous performance (low or high) were further combined with a teaching variable, namely instructor expressiveness. Students received either none, one, two, or three videotaped attributional retraining sessions which stressed the role of effort and strategies. They then viewed a videotaped lecture presented by either a low or high expressive instructor. One week later, they wrote an achievement test based on the lecture. Results indicated that retraining enhanced performance on the achievement tests, but only when combined with effective instruction. Moreover, only students who had previously performed poorly benefitted from attributional retraining. These results suggest that contextual factors, such as teaching, and individual differences have to be considered when developing attributional retraining programs. Contains 15 references. (Author/GLR)