A factor analysis of the 24 unit activities contained in the Key Group Data Form was carried out using descriptions of 214 engineering positions in the General Electric Company. Supervisors ranked 24 unit activities in terms of their importance for each position. The intercovariances among these 24 activities were computed, and this intercovariance matrix was then factor analyzed. The analysis was complicated by the use of ranks in scoring the unit activities. A principal axis factor analysis was carried out on the matrix of intercovariances and seven dimensions emerged. In attempting to rotate this result to simple structure, it became apparent that a dimension had been lost as a result of a translation of the origin due to the use of the ranking procedure for obtaining scores. Accordingly, an additional factor of constants was introduced, and when this eight-factor structure was rotated, a clear simple structure resulted. An excellent correspondence was observed with an earlier study which utilized a different sample of engineering positions. It may be concluded that these results provide a useful method for describing engineering jobs and the differences between engineering jobs.