The potential of a cognitive model task analysis scheme (CMS) that specifies features of test problems shown by research to affect performance is explored. CMS describes the general skill area and the generic task or problem type. It elaborates features of the problem situation and required responses found by research to influence performance. Stimulus features specify particular concepts and procedures necessary to answer a question and place limits on the structure and range of content or examples that can illustrate a problem. Response requirements indicate the mode as well as solution procedures and operations. The task content describes time constraints and the purpose, function, or audience of the task. CMS was used to classify and compare the contents of a Stanford Achievement Test and a criterion referenced test, both administered in 1982. Content analyses indicated the tests measure different aspects of reading and with different degrees of emphasis. Response patterns implied that CMS may be a promising tool for describing, analyzing, and interpreting test performance. The detailed partitioning of skill requirements used in the CMS appears to present a clear and interpretable test analysis tool. (Author/PN)