Why Do Charter Schools Have High Teacher Turnover?
by Halley Potter
Charter school advocates, like those featured in the new documentary Waiting for "Superman," cite the lack of teachers' union representation in most charter schools as a major advantage. The ability to fire teachers at will, and to pay some teachers more than others, is an important ingredient to the success of nonunionized charter schools, they argue. But critics point out that there is a serious downside to reducing teacher voice in charter schools. As my Century Foundation colleague Richard Kahlenberg points out in a just-released Reality Check on charter schools, the fact that almost 90 percent of charter schools are non-unionized is surely related to the high rate of teacher turnover in charter schools (25 percent compared with 14 percent in regular public schools).
Advocates of charter schools now have a response on the teacher turnover issue. In a study released this fall from the Center on Reinventing Public Education, researchers Betheny Gross and Michael DeArmond hope to "quiet some fears that charter schools are particularly susceptible to teacher turnover," as recent studies by researchers at Vanderbilt University and Western Michigan University suggest. In trying to explain why charter schools have higher turnover rates, Gross and DeArmond hope to show that the charter school model per se is not to blame. However, critical problems in their choice of data undermine the study's conclusions.
In the report, Parallel Patterns: Teacher Attrition in Charter vs. District Schools, Gross and DeArmond argue that higher teacher turnover rates in Wisconsin charter schools may be a result of teacher and school characteristics rather than the schools' status as charters. Based on data from 1997 to 2006, Wisconsin's charter school teachers were 40 percent more likely to move schools and 52 percent more likely to leave teaching than traditional public school teachers. However, when the authors controlled for teacher characteristics (academic degrees, age, gender, ethnicity, salary, and certification), the gap dropped to 32 percent more likely to move schools and 41 percent more likely to leave teaching. Most importantly, when they controlled for school characteristics as well (concentration of minority students, percent of students passing assessments, location, and grade span), there was no statistically significant difference between the turnover rates of charter school teachers versus traditional public school teachers. Thus, the authors conclude that "high turnover rates in Wisconsin's charter schools appear to be a disadvantaged school problem rather than a charter school problem per se."
The major flaw in the study, however, is that the percentage of charter school teachers covered by union contracts in the state the researchers chose to study is very high compared to most other states with sizable charter school programs. Wisconsin divides its charter schools into "instrumentalities" versus "non-instrumentalities," with individual school districts deciding each school's designation. If a charter school is an instrumentality—as 82 percent of Wisconsin charters are—then all personnel at the school are employed by the district. These charter schools are required to follow local school board policies, and their teachers are employed under the same union contract, including state retirement benefits, as traditional public school teachers. In non-instrumentality charter schools—which make up 18 percent of the state's charters—all personnel are employees of the charter school. The schools do not have to follow local school board policies, and teachers are not members of the local teachers' union or the state retirement plan. At all Wisconsin charter schools, teachers must be licensed by the state.
To get an idea of how Wisconsin's charter laws compare to those of other states, consider a 2010 report from the pro-charter Fordham Institute. The report examines the amount of autonomy granted to charter schools in 26 states that account for more than 90 percent of the country's charter schools, granting higher grades to those that allow more autonomy. Autonomy is defined as the ability to control staffing, instruction, culture, and finances/governance. The average autonomy grade for a state's charter laws was B+. Wisconsin (excluding the non-instrumentalities) received an F, tying with Maryland for last place. Maryland law requires that charter school teachers be covered by collective bargaining agreements.
Gross and DeArmond acknowledge that the stipulations on hiring and contracts make Wisconsin unusual, but they argue that it allows them to isolate "the effects of organizational and cultural differences." This may be, but in most states variations in teacher contracts, collective bargaining, and credentialing are key differences between employment at a public school and a charter school and should be considered in a study of teacher turnover. By choosing a state that requires most charters to use the same teacher contracts as public schools, the authors have controlled for a key variable that is likely to affect teacher turnover rates in charter schools elsewhere. Even in the case of Wisconsin, it would be helpful to compare turnover rates in non-instrumentality versus instrumentality charter schools. (Gross and DeArmond don't distinguish the two.)
By choosing to study a state with extremely unrepresentative labor policies for charter schools, the authors of Parallel Patterns undermine their attempt to quiet fears about high teacher turnover in charter schools. They do, however, offer two important insights. First, the study supports two patterns previously observed by researchers from Western Michigan University, the American Institutes for Research, and UCLA: (1) schools with higher concentrations of poverty and younger, less experienced teachers have higher rates of turnover, and (2) charter schools disproportionately have these characteristics. Far from letting charter schools off the hook, this finding underscores the fact that both charter schools and public schools need to consider how high concentrations of poverty and the characteristics of their teaching force affect their performance.
Second, the study highlights the need for more research on how the various components of charter models affect teacher turnover. Perhaps research on other states will show, as Gross and DeArmond suggest, that the charter model per se is not to blame. But additional research might instead show that teacher contracts, collective bargaining, and credentialing do impact teacher turnover in charter schools. Such a result would not be surprising. In fact, the final section of Gross and DeArmond's study provides national data from the U.S. Department of Education's Schools and Staffing Survey (SASS) and Teacher Follow-Up Survey (TFS) that is consistent with this hypothesis. Charter school teachers who moved to new schools were more likely than their public school counterparts to cite seeking greater job security and a better salary/better benefits as the reason for their move.
The bottom line is that if charter school advocates are going to highlight non-unionized environments as critical to providing administrators with freedom to make staffing decisions, they need also to acknowledge that loss of teacher voice may have serious negative consequences that are manifested in high teacher turnover rates.