The purpose of my dissertation is to determine which job-related factors are most likely to explain teachers' sense of satisfaction with their current job. Based on previous research (Fairchild et al., 2012; Lee, Dedrick, & Smith, 1991; Perie & Baker, 1997), I hypothesized that teachers' perceptions of classroom autonomy, staff collegiality, and administrative support would have a greater impact on teachers' job satisfaction than demographic characteristics of either teachers or schools.

My study used a dataset of 19,130 teachers in 2,420 schools drawn from the 2007-08 Schools and Staffing Survey (SASS). The study employed a two-level hierarchical linear modeling (HLM) technique. I modeled the key constructs at both the individual and collective levels by conceptualizing autonomy, collegiality, and support as individual teachers' perceptions and as the average of all teachers' perceptions in the school.

My dissertation found that individual teachers' perceptions of classroom autonomy, staff collegiality, and administrative support had a greater impact on teachers' job satisfaction than demographic characteristics of schools and teachers. In addition, my study found that school-wide perceptions of classroom autonomy, staff collegiality, and administrative support had a significant effect on teachers' job satisfaction over and above the effects of individual teachers' perceptions. The study found significant negative effects on job satisfaction for teachers working in Catholic schools and for teachers working in schools where poverty was perceived to be a school-wide problem. Finally, my study determined that sufficient between-school variance existed in the outcome to justify future school-level policy interventions aimed at improving teachers' job satisfaction.

Based on my findings, I recommend that policy makers who are concerned about addressing low levels of job satisfaction among teachers should begin by improving factors related to classroom autonomy, staff collegiality, and administrative support, since these are likely to have a meaningful impact. I also recommend that future research should employ a narrow definition of the outcome focusing on teachers' satisfaction with their job rather than their satisfaction with teaching in general.