Students completed surveys at the beginning and end of a sophomore-level course on research and statistics. We hypothesized that the course would produce advances in knowledge of research and statistics and that those changes would be accompanied by more favorable attitudes toward the subject matter. Results showed that knowledge did increase significantly, but 4 of 6 attitude measures showed no change. Two attitude measures (perceived utility of research and statistics) showed significant declines. These results demonstrate the independence of knowledge and attitudes and show that attitudinal change is not monolithic. We argue that students' misconceptions about research might underlie the declines in perceived utility of research and statistics. (Contains 2 notes and 3 tables.)