The Revolving Door

Jenny D posted on school finance reform in New York City and has prompted a spirited dialogue in her comments section on the effectiveness of additional funding and teacher retention effects. It brings to mind an essay by Erik Hanushek, the late John Kain, and Steven Rivkin titled "The Revolving Door: Factors Affecting Teacher Turnover." This appeared both in Developments in School Finance: 2003 and in Education Next. I believe a more technical write-up appeared elsewhere. It serves as a good marker for what we know, and don't know, about teacher turnover.

Hanushek, Kain, and Rivkin used data from the Texas Schools Project to examine movement among elementary school teachers from 1993 to 1996. It's worth remembering that these are very skillful investigators, using a relatively rich data set, studying a level where specialization and competition from other job sectors is less than 7-12. Even so, their findings include several qualifications.

The authors found that financial factors had a stronger effect on male teachers than female teachers and that salary increases had a stronger influence on teacher movement in the first few years of their career than in later years. Working conditions, defined for this study's purposes as student achievement and demographics, were stronger factors. The authors did not have data to capture some of the variables you often see referenced on blogs such as administrative leadership, professional community, order and discipline, parental involvement, and adequate materials and supplies. These unmeasured factors may correlate with achievement and demographics, but which way do the arrows go?

Teachers consistently moved from districts with low achievement and high poverty to districts with higher achievement and less poverty. The same was true for movement within districts between schools. White teachers moved from schools with high minority populations to schools with lower minority populations. The net effect was that as teachers gained experience they would migrate to school settings with more advantaged students, with the result that schools with higher disadvantaged populations needed to hire inexperienced instructors every year.

The authors cited previous work to note that 10% of improvement in teacher effectiveness may be attributed to experience, so much could be gained by improving teacher quality through active professional development. They left unanswered whether high need schools would be investing in teachers that would be hired away by lower need schools, perpetuating the cycle.

It's impossible to do justice to the essay in a posting and I recommend those with an interest in teacher retention in high need schools read it in its entirety.