Teacher questions have been the subject of a substantial amount of research and analysis in recent years. Debate has centered around the appropriate use of this instructional and evaluative tool. A study examined the impact on student achievement of manipulating abstractness-concreteness, independent of cognitive level, in achievement test items. Subjects for the study were 66 students enrolled in a senior-level class at a major university in the southwest United States. The specially constructed test was administered to students during the course of a regularly scheduled quiz period, and there was no suggestion to subjects that this quiz was any different than others they had taken. Results revealed that manipulating the cognitive complexity of test items influenced students' ability to respond correctly. The data suggest that these differences have important impact on student achievement. The difficulty students had with higher-than-recall items suggest the need to consider further the usefulness of cognitive level, and abstractness-concreteness in explaining student achievement. (JMK)