Objectives Current theories regarding worry and generalized anxiety disorder ( GAD ) highlight the potential avoidance functions of worry, and it has been suggested that worry functions to avoid self‐conscious emotions in particular. Therefore, the present study examined the roles of proneness and aversion to self‐conscious emotions in worry and GAD . Design Cross‐sectional data from two samples were collected: (1) a sample of 726 undergraduates, and (2) a selected sample of 51 community members, 37.3% of whom met DSM ‐ IV criteria for GAD . Zero‐order correlations and hierarchical multiple regression analyses were used to examine associations of self‐conscious emotion constructs to worry and GAD . Method Proneness to guilt and shame (propensities for experiencing guilt and shame, respectively) were assessed via the Test of Self‐Conscious Affect‐3. Aversion to guilt and shame (perceptions of guilt and shame, respectively, as especially painful, undesirable emotions) were assessed using the Guilt Aversion Assessment and Shame‐Aversive Reactions Questionnaire, respectively. Worry was assessed using the Penn State Worry Questionnaire, and GAD was assessed via the Structured Clinical Interview for DSM ‐ IV ‐ TR Axis I Disorders. Results Correlations indicated positive associations between self‐conscious emotion constructs and worry/ GAD . However, in the selected community sample, regression analyses indicated that only shame aversion was positively associated with worry/ GAD , over and above all other self‐conscious emotion constructs and depression. Conclusions Results suggest a prominent role for an intolerance for shame in worry and GAD , which is broadly consistent with psychological models of worry. Future directions for research and clinical implications are discussed. Practitioner points Positive clinical implications Evidence supporting the theorized importance of self‐conscious emotions in worry and GAD. Specifically highlights the need to address intolerance for shame in treatment. Limitations Small sample size in Study 2. Use of cross‐sectional data.