Abstract
Truax (1966) had clinicians rate typewritten statements by a patient (and the preceding and subsequent therapist's statements) from a successful therapy case of Rogers. Non-zero correlations between ratings of patient statements and ratings of therapist statements and changes in the former over time were taken to support a reinforcement interpretation of Rogerian therapy. The present critique offers five arguments to counter the suggestions that the Truax study is relevant and crucial to the theoretical dispute between Rogers and Skinner. Truax asserts the correlations mean that Rogers' theory is wrong; however, the correlations represent a measure of how selective Rogers actually is in a particular case. Truax does not acknowledge that the obtained correlations may be (1) lower than correlations in non-therapy contacts or (2) lower than those in poor therapy. (3) An analysis of the logic of Truax's method suggests that the validity and appropriateness of the ratings are questionable. (4) Truax's analysis of changes over time does not consider the correlation-causation issue. (5) His treatment does not acknowledge that any outcome of his study would have been consistent with Skinner's position.
Get full access to this article
View all access options for this article.