A German version of Broadbent's Cognitive Failures Questionnaire was developed because there was no comparable German instrument. As external validity criteria observable failures in everyday situations were used. The positive correlations were significant but moderate in size. For construct validation, inventories of trait anxiety, coping styles and action control were used. Most of the correlations were significant and in the expected direction. A maximum likelihood factor analysis of the questionnaire suggested that the utility of the total score may be doubtful. The unidimensionality of the construct requires further investigation.