Association for Public Policy Analysis & Management

A strong body of research has sought to better understand what makes highly qualified or effective teachers decide to leave a school or exit the profession altogether (e.g. Clotfelter, Ladd, & Vigdor, 2011; Scafidi, Sjoquist, & Stinebrickner, 2007; Feng 2010; Boyd, Lankford, Loeb, & Wyckoff, 2008; Feng, Figlio, & Sass, 2010). Several recent studies have found promising results when estimating the effects of teacher bonuses to promote retention at high-need, low-performing schools (Clotfelter et al. 2008; Glazerman et al. 2013; Springer, Swain, & Rodriguez, 2015). However, extant literature tells us relatively little about the impact of these retention-oriented policies on their ultimate goal—improving student academic outcomes. 

This study estimates the effect of selective retention bonuses for highly effective teachers (teachers receiving the highest ratings on the Tennessee teacher evaluation) on low-performing schools’ ability to elevate student test scores. The theory of action is simple: selective retention bonuses result in greater numbers of highly effective teachers at participating schools, who subsequently drive larger student gains than the teachers who would otherwise fill their positions. To examine whether this theory of action holds true, this study uses rich longitudinal administrative data from the state of Tennessee, where an experimental retention bonus program for highly effective teachers in Priority Schools (the bottom 5% of schools in the state based on a composite measure of state test scores and graduation rates) was recently implemented and found to keep top-performing teachers in their schools. 

In prior work, we exploited the threshold for eligibility for teachers to receive the bonus to estimate the effect on their retention (Springer, Swain, & Rodriguez, 2016). Here, we instead use the threshold in eligibility for schools to offer bonuses to estimate the effects on the teaching pool at the school and subsequent student achievement. Although the estimated effects of the retention bonus on teacher retention were relatively modest in magnitude, and the highly effective teachers represented a relatively small portion of the faculty in Priority Schools, the radical differences between the retained level 5 teachers’ estimated effectiveness and that of their likely replacements results in the equivalent to a profound intervention. The 361 teachers who accepted bonuses had overall teacher effectiveness ratings more than a full standard deviation above the state average, and the average teacher hired by Priority Schools was rated roughly two thirds of a standard deviation below the state average. Thus, for every teacher that was retained as a result of the bonus, students taught by that teacher, rather than the likely replacement, experience an increase in estimated teacher effectiveness of 1.7 standard deviations. 

Preliminary results indicate that the retention bonus program significantly elevated the average test score value-added of teachers in participating schools, all of which were low-performing by design, and almost entirely composed of low-income, non-white students. These early findings are of interest for policymakers and scholars interested in interventions that promote access to effective teaching in the context of racially and socioeconomically isolated schools.