Mitchell's two-step model of leader interventions for poor performing group members was examined as a toolfor understanding leader behaviors in relation to members who are task effective but personally distressed Case studies were presented to 136 hospital nurses who, in the role of a head nurse, were asked to deal with a highly effective yet distressed group member. Results indicated that leaders (subjects) could overcome self-serving attributional biases and infer that they, the group leaders, could be the cause of the group members' distress. Subjects who made such attributions were more likely to change their own leader-ship style than subjects who inferred the distress was caused by the members 'efforts, ability, or personal andfamily problems-a conclusion that supports Mitchell's two-step model. © 1992, SAGE PUBLICATIONS, INC. All rights reserved.