Subjects in a stress condition were led to believe that they had failed an important achievement test while subjects in a non-stress condition were not led to believe that they had failed. Projection strategies were manipulated by encouraging subjects either to attribute the cause of their poor performance to the examiner instead of to themselves (blame projection) or to estimate how poorly their friends would have scored on the test had they taken it and not performed well on it (similarity projection). Measures of subjective anxiety, anger, and depression indicated that (a) the failure feedback was effective in increasing stress, (b) blame projection was effective in controlling stress, and (c) similarity projection was partially effective in controlling stress. The results provide evidence that blame projection can be effective in controlling stress and also have implications concerning the conditions under which the use of projection will and will not be effective in coping with stress. The importance that situational factors may have on the effectiveness of cognitive coping strategies was emphasized. (Author)