A quasi-experimental design using multiple levels of measurement was used to evaluate the effectiveness of a U.S. General Accounting Office communications training program. Results from a multivariate analysis of variance indicated that employees who underwent training reported more positive attitudes about communication skills and performed better on knowledge and recognition assessments than did control groups in the study. With the exception of employees' ability to apply knowledge of communication skills to simulated written problems, the relative amount of change on the measures was not very large. The course and its instructor were rated very favorable by the participants, though certain aspects of the course-such as the evaluation, written exercises, and booklets and handouts-were more negatively valued by course participants. The results from this project are comparable to those obtained elsewhere, though the direct benefit of such changes to the organization are less clear.Good interpersonal and communication skills have long been recognized as important in work organizations. Such skills likely contribute to better information exchange and transmission across levels of the organization as well as influence the motivation and satisfaction of T/tM