Almost 4000 principal wage‐earners were surveyed to assess the main and interactive effects of undesirable and desirable life‐events, social support, and perceived and objective job security on psychological symptoms. Events and support were measured for both the job and non‐job spheres. Three methods of analysis were used. Ordinary regression was used as a preliminary analysis and to provide comparison with the customary approach. A second analysis was based on dichotomizing the continuous symptom variable at a point approximating a ‘case’ vs. non‐case distinction. The third analysis used a small panel to control for time 1 symptoms. The different methods produced somewhat different results but, in general, provided support for certain main effects including desirable life‐events and job security but little or weak support for the moderating effects of such variables as social support.