The term "burnout" represents a significant perspective on how people respond to their work, but the attention paid to this phenomenon has largely been clinical and often anecdotal. In this article, the authors seek to expand the analysis of burnout in ways that permit comparative analysis, especially in large populations. This study specifically addresses three questions. First, does a paper-and-pencil instrument isolate domains of burnout that are relatively consistent between people-intensive work and the broader range of activities found in a commercial enterprise? Second, can we develop phases of progressive burnout? Third, can we test the efficacy of the burnout phases by searching for regularities in a panel of 22 variables commonly thought to tap the important facets of the work site? The author's analysis shows that we can answer these three central questions affirmatively, though occasionally with complex and potentially significant qualifications. The results of the analysis provide further evidence of the usefulness of a convenient instrument for measuring burnout and also suggest that behavioral scientists will find valuable a phase model that distinguishes regular and robust covariation by using a panel of variables thought to tap the important aspects of organizational life.