Association for Public Policy Analysis & Management

Currently, the modal years of teaching experience in the US is one year, and within 5 years of entry, 40-50% of teachers leave the classroom (Perda, 2013; Ingersoll & Smith, 2003). Teacher attrition is problematic in that constant churn negatively impacts school climate, the hiring process is expensive, new hires have less experience—a factor linked to teacher quality, and new teacher induction can disrupt instructional programs (Boe, Bobbitt, Cook, Whitener, & Weber, 1997; Goldhaber & Anthony, 2007; Henry, Bastian, & Fortner, 2011).

Beginning teachers praise teacher preparation programs (TPPs) where they spend extensive amounts of time in schools and feel better prepared as a result of more extensive pedagogical coursework (Kee, 2013; O’Neill & Stephenson, 2012; Levine, 2006). Coursework in pedagogy, specifically classroom management, may aid teachers in overcoming obstacles in their early careers, while field experiences may help teachers build efficacy which is associated with retention (Fry, 2009; Yost, 2006).

This research provides evidence that TPPs may play a role in how long a teacher remains in the classroom, but no study yet has directly investigated these relationships longitudinally. This paper seeks to understand the relationship between the structural features of TPPs and beginning teacher retention by asking, How do the features of beginning teachers’ teacher preparation programs predict attrition in middle and high school English/ Language Arts?

Using an administrative dataset combined with preparation program data collected from 15 universities, I use 5 years of employment data, from 2008-2009 to 2012-13 to estimate the relationships between the structural features of TPPs and middle and high school English/Language Arts teacher retention.

The primary structural features of TPPs are coursework and fieldwork. Coursework typically involves subject matter, pedagogy coursework, foundations coursework (e.g., educational psychology and the history of education), and other courses including technology. Fieldwork includes early field experiences, occurring prior to student teaching, and student teaching. Key aspects of student teaching include university supervision, the length of time a teacher assumes full classroom teaching responsibility, and whether a seminar meant to link theory with practice is required.

While survival analysis would be appropriate over a longer period of time, with only 5 opportunities to attrit, I use an ordinal logit model to estimate the relationship of structural features to teacher attrition. Selection bias is inherent in estimating the effects of TPPs from selection into programs, sorting of teachers into schools and non-random assignment of students to teachers, and I include a university indicator, as well as a rich set of covariates, to mitigate against such bias.

Preliminary results suggest that pedagogy coursework and a seminar during student teaching are associated with lower teacher attrition.