This paper reports on a short version of the Student Jenkins Activity Survey (JAS), a multiple choice questionnaire that measures Type A "coronary-prone" behavior in assessing subjects' A/B types. The primary objective was to determine if the short and long forms of the student JAS represent similar measurement instruments. A secondary objective was to determine if the short and/or the long JAS results in comparable factor patterns and covariance structures for As and Bs. The short Student JAS gives factors that are more independent than those of the long JAS, and provides a more comparable structure for Type As and Bs. The present results suggest that researchers should still use the short form since its structure most closely resembles that hypothesized by the underlying theory. An important finding is that neither the Glass orthogonal nor the oblique two-factor model fit the data well in an absolute sense: less than half of the common variance was explained by any model, on either form. Examination of the residual correlation matrices indicated that two classes of large residuals (self versus perceived others' ratings and actions versus feelings), which were largely responsible for the significantly poor fits, were observed on every LISREL computer run. (PN)