High-profile instances of workplace violence and increased pressure from competitors have threatened the viability of one of the nation's largest employers, the United States Postal Service (USPS). As a result, the USPS began a massive change effort in the early 1990's. One of the initiatives implemented to improve labor-management relations was a derivative of the self-managed work team known as the crew chief program. This study provides a mixed-methods and multi-level approach to understand the impact this unique program had on organizational attitudes. The first aim of this study was to investigate whether the crew chief program reduced employees' stress and strain and improved job and supervisory satisfaction and company and union commitment, while controlling for the nesting of employees within sites and employee demographic characteristics. The second aim was to replicate and extend the stressor-strain-outcome (SSO) model of stress and to determine whether employee perceptions of crew chief support moderated the relationships between stressors, strain, and outcomes. I evaluated Aim 1 using data from 177 mail processors from 27 units matched from baseline to one-year follow-up with hierarchical linear modeling. This was followed up with an implementation analysis of qualitative data to determine the extent to which the program was implemented compared to the original design. I evaluated Aim 2 using structural equation modeling from 538 mail processors who participated at follow-up. There was little quantitative support for Aim 1. However, the results of the implementation analysis suggested that the crew chief program was not functioning as conceived. Aim 2 received strong support, with almost all of the main effects of the SSO model replicated. However, there was no support for the moderator effects. Additionally, I found role ambiguity to have direct relationships with other organizational outcomes beyond the indirect effects via strain and that crew chief support was strongly related to stressors and outcomes. These findings reinforce the notion that employing both quantitative and qualitative methods can dramatically improve the quality of organizational research. Based on these findings, I describe several suggestions for improvements to this specific program and for improving future initiatives aimed at enhancing labor-management relations.