Weakenings and repairs of the working alliance in counselor supervision

This exploratory study represents the first investigation of Bordin\u27s model of the working alliance applied to counselor supervision. The ten week study explored the development of the working alliance within supervision dyads via audiotapes and specially adapted instruments including Horvath\u27s Working Alliance Inventory and Stiles and Snow\u27s Session Evaluation Questionnaire. The focus of the study was concerned with perceptions of congruence of the alliance by supervisors and supervisees in the areas of goals, tasks, and bonds and weakening and repair of the alliance and its effect on outcome. Perceptions of congruence within dyads were found to fluctuate in dissimilar patterns. Supervisors responses as a group tended to be stable and less reactive to weakenings across sessions while supervisees demonstrated a fluctuating pattern of positive-negative-positive attachment to the alliance. These fluctuating patterns occurred in sequences of weakening and repair events somewhat similar to Lansford\u27s findings with counselor-client dyads and roughly paralleled Mann\u27s model which described fluctuating phases of attachment in short term therapies. The relationship of the initial strength of the alliance to outcome (as measured by the Working Alliance Inventory - Modified) was strong and quite similar to Horvath and Greenberg\u27s findings thus lending further support to both the reliability of the measure and its predictive value