The More the Merrier? A Contextual Analysis of the Frequency of Teacher Mentoring

• In Event: 34.047 - Mentoring Matters, but How? Conceptualizing Mentoring Effectiveness From Preservice to In-Service Teaching

Abstract

The teaching force in the United States is more stable today than it was twenty years ago (Gray & Taie, 2015), when statistics estimated a 40-50 percent 5 year attrition rate for beginning teachers (Ingersoll, 2003). The observed improvement may be attributable, in part, to the provision of mentoring, which has become an increasingly routine support for those new to the teaching profession (Fideler & Haselkorn, 1999; Serpell, 2000; Goldrick, Osta, Barlin, & Burn, 2012). Empirical research has confirmed the potential of a well-implemented mentoring strategy for improving new teacher commitment, that is, their attitude toward, and identification and involvement with the school (Leithwood, Jantzi, & Steinbach, 1999; Mowday, Porter, & Steers, 1982) and alleviating teacher turnover (Gray & Taie, 2015; also see review of Ingersoll & Strong, 2011). As mentoring continues to gain stature as a critical means of supporting and retaining new teachers, the school context that influences its effects must be better understood. Central to this study is the impact of principal leadership, as one key example of school context, on the effects on mentoring. We ask: How does the frequency of formal mentoring affect beginning teachers’ commitment to schools? Do the effects differ by levels of principal leadership?

The primary data source for this study consisted of a district-wide survey from 1,046 elementary school teachers who were in their first or second year of teaching in a large, midwest, urban district in 2005. The survey included items about their mentoring experiences, attitudes towards their schools and school leaders. We analyzed the survey data along with administrative records, student test scores to examine our primary outcome of interest, beginning teachers’ commitment to schools using a two level model. Because the teachers are unlikely to be randomized to different levels of teacher mentoring, a nonparametric propensity score (PS) based weighting approach is used to remove selection bias associated with pretreatment characteristics of the teachers, classrooms, and schools. When examining the overall effects of teacher mentoring, we use the PS approach to approximate a randomized design in which new teachers are randomly assigned to different levels of mentoring frequency. To investigate the role of principal leadership, we then approximate a blocked design in which teachers within each level of principal leadership are randomized.

Our results indicate that in general new teachers who interacted with their mentors on a biweekly or more frequent basis may have higher commitment to school than those who received less or no mentoring (coefficient = .85, SE = .17, p < .001). Under strong principal leadership, the more frequent the mentoring, the higher beginning teachers’ commitment to their school. Under weak leadership, teachers benefited the most from biweekly mentoring; but there is no added benefit of having more frequent mentoring interactions. These findings suggest that simply having more frequent mentoring does not result in stronger commitments, and that the environment principals create can modify the effects of teacher mentoring.

Authors

• Yihua Hong, RTI International

• Kavita Kapadia Matsko, National Louis University