This study is a non-experimental quantitative design exploring teacher retention in Oklahoma by looking at variables including district level decisions and characteristics. The population in the study included 521 public school districts. This study has been broken into two distinct data analysis phases using hierarchical linear modeling to measure the relationship between the study predictive variables and teacher retention. The first analysis was conducted using time-varying variables at the district-level. The second phase of the analysis was conducted using time-invariant variables. Data were collected over a five-year time period, 2013-2017. This study found statistically significant relationships between higher teacher salaries, instructional and support resources, and proximity of the district to teacher retention. In addition, the study indicated a four-day workweek, administrative expenditures, and percent of economically disadvantaged students were significantly and inversely related to teacher retention.