Although transactional analysis (TA) theory has been used by psychotherapists since its introduction by Eric Berne in the 1960s, the ego state functioning constructs, a seminal part of the theory, have not been adequately validated. Previous research has focused on whether therapy using TA methodology works. This study tested the TA ego state constructs by measuring client change occurring during psychotherapy and comparing those changes with predictions from the TA theory. Fifty-six subjects, who were clients at a university counseling center in the Rocky Mountain region of the United States, were tested before and after therapy using two standardized instruments, the Adjective Check List (ACL) and the Brief Symptom Inventory (BSI) and client and therapist global ratings of success of therapy. Critical Parent, Nurturing Parent, Adult, and Adapted Child ego state scores from the ACL all changed in the predicted directions and reached statistical significance. Free Child ego state score changes did not reach statistical significance. Ego state scores correlated with BSI Global Severity Index in predicted directions and all correlations except Critical Parent were statistically significant. Changes in ego state scores did not correlate with client and therapist ratings of success with one exception--Nurturing Parent was related to Client ratings of success. Changes in ego state scores did not correlate with subject pretest symptomatology, number of sessions, or the therapist's level of experience. Limitations of the study and recommendations for further research are discussed.