Reports on an experiment, conducted on 63 students of English, which sought to determine whether certain students within heterogeneous population groups can be expected to make significantly more rapid gains than others, and whether scores on a standardized proficiency test can identify students who will make greater or lesser gains. (Author/AM)