The purpose of this study was to develop a measure of learned helplessness. Based on a review of the literature, 50 items were developed for inclusion in the proposed Learned Helplessness Scale (LHS). On the basis of independent reviews by three experts, 20 items were selected for inclusion in the LHS. An individual's response to each item was graded, using a Likert scale. A standardized alpha reliability coefficient of .85 was obtained for a sample of 241 healthy adults. There was a positive correlation between the LHS scores and Beck's Hopelessness Scale (HS) scores (r = .252) and a negative correlation between the LHS scores and Rosenberg's Self-Esteem Scale (SES) scores (r = -.622). These correlations were in the direction postulated by various theorists. A Varimax-rotated factor analysis of the LHS data yielded five factors. Three of these factors tapped content relevant to the attributional styles of learned helplessness. Clinical data were also obtained on samples of oncology, hemodialysis, and spinal cord patients. Because the alpha reliability coefficients of the LHS and the Pearson product moment correlation coefficients between the LHS and the HS and the LHS and the SES were in the same direction, the instrument is believed to have adequate internal consistency.