The text below is an excerpt from the complete document. Read the full paper in PDF format.
Abstract
Do charter schools draw good teachers from traditional, mainstream public schools? Using a panel dataset of all North Carolina public school teachers from 1997-2007, this research paper finds nuanced patterns of teacher quality flowing into charter schools. High rates of inexperienced and uncertified teachers moved to charter schools, but among certified teachers changing schools, the on-paper qualifications of charter movers were better or no different than the qualifications of teachers moving to comparable mainstream schools. Estimated measures of classroom performance for a subset of grade 3 - 5 teachers show that charter movers were more effective in math and reading instruction, relative to other mobile teachers. Charter movers compared less favorably, however, to non-mobile teachers and colleagues within their sending schools. The distribution of classroom performance among future charter teachers, adjusted for sampling error, was significantly lower than the distribution for exclusively mainstream teachers.
Introduction
Charter schools are independently operated public schools, free from most of the district and
state regulations faced by traditional, mainstream public schools. Forty states and the District
of Columbia have legislation outlining the establishment, operation, and accountability
of charter schools. Charter systems are designed to provide families with more choice in their
children's education, to provide teachers with more choice in their career paths, to promote
innovative instruction, and to target special populations of students that may be under-served
by traditional public schools. A charter program represents a new, competitive branch of publicly
funded education that entrusts each campus with a degree of autonomy rarely seen in
mainstream schools. Autonomy and flexible resource allocation in charters schools may draw
good teachers away from the mainstream. A growing body of research has characterized the
qualifications of the stock of charter teachers, who compare favorably to mainstream teachers
in some respects (college selectivity, for instance) but not others (experience, certification). I
complement and advance this literature by analyzing the qualifications and classroom performance
of the flow of North Carolina teachers moving from mainstream to charter schools over
the years 1998-2007.1
Charter schools, playing the role of competitive entrants in partially deregulated public education
markets, are expected to spur efficiency gains by decreasing industry concentration and
challenging incumbents (here, traditional public schools) to improve performance. Proponents
of charter schools, and school choice more generally, expect competition between traditional
and choice schools to drive up the quality of education overall. Friedman (1955, 1997) proposed
vouchers as one way to stoke school competition. Dee (1998), Hoxby (2003), and most
recently, Booker, Gilpatric, Gronberg, and Jansen (2008) offer empirical evidence that mainstream
student performance improves in light of competition from choice schools. Long-run
gains from competition will require charters to be formidable competitors, however, and the
jury is still out as to whether they actually increase student learning relative to mainstream
schools. The emerging consensus is that new charters have a negative impact on student
achievement growth, a penalty which fades as schools and students gain experience.2
Teacher quality is a profound factor in student achievement,3 and charters seeking to produce
high achievement (or at least, meet accountability standards) will value effective teachers.
Charter schools are heterogeneous by nature; some specialize in priming the gifted and collegebound,
while others target students at risk of failure. Recruiting and retaining high-quality
teachers will be more difficult for the latter group. The teacher mobility literature is thick with
evidence of teacher preferences for high-performing and socioeconomically advantaged school
environments.4
Charter schools may have an advantage in the teachers' labor market, regardless of their student
composition. "They are free to recruit the best teachers and to raise money from foundations,
corporations, and individuals" (Manuel, 2007). Charters are not generally bound by state
pay scales, they can allocate budgets as they see fit, and feasibly, they can pay higher teacher
salaries. One New York City charter school famously offers teacher compensation packages in
excess of $125,000 (Gootman, 2008). Nationwide, charter teacher salaries are more comparable
to mainstream salaries,5 but charter teachers in some states earn significantly less than
other public school teachers with similar qualifications (Malloy andWohlstetter, 2003). Even if
charter schools cannot outbid mainstream schools on salary alone, school leaders can influence
teachers' utility in non-pecuniary ways, by reducing their non-instructional duties, encouraging
collegiality among faculty, manipulating class size and composition, and granting teachers
more creative license and autonomy than they are afforded in mainstream schools. Early advocates
of the charter model stressed the professionalization and empowerment of teachers
as critical tenets of charter development (Budde (1988); Kolderie (1990)). High teacher satisfaction
rates in charter schools typically stemmed from greater autonomy ("freedom to teach
the way I want"), like-minded colleagues, and innovative teaching philosophies. Teachers who
were dissatisfied in charter schools cited low pay, lack of benefits, high workload, and insufficient
facilities (Malloy and Wohlstetter, 2003).
In practice, the intangible benefits of working in a charter school may be too low to offset
low pay and other resource limitations. Common charter finance models allocate each school
a per-pupil rate roughly equal to the surrounding district's average per-pupil cost, excluding
the cost of buildings. If a district enjoys substantial economies of scale in variable cost, its perpupil
expenses will be less than a charter school's average cost. Charters with competing uses
for limited resources may sacrifice some teaching talent in favor of administrative and capital
improvements if doing so maximizes their objectives (student achievement, enrollment, and
budget size being likely objectives). Furthermore, many states allow charters to employ a
high rate of uncertified teachers. This permits charters to attract teachers from outside the
traditional pipeline, but also increases the supply of low-cost, low-skilled individuals eligible
to work in charter schools, including uncertified mainstream teachers nearing the expiration of
temporary licenses. Recently, Wisconsin raised subject-based certification requirements for its
charter teachers, prompting school leaders to argue that they could not afford to hire teachers
meeting the new standard (Borsuk, 2008). Charter licensure requirements vary across states,
and little is known about the qualifications of uncertified teachers in charter schools, or the
impact of relaxed licensure standards on student performance in charter schools.
Much of the developing research on charter teacher quality examines the qualifications,
workload, and job satisfaction of the stock of charter teachers nationwide or within particular
states. Podgursky and Ballou (2001) surveyed teachers in seven states, and found that charter
teachers were less likely to be certified, more likely to be inexperienced, and more likely
to have merit pay than mainstream teachers. Hoxby (2002), using a 1998 national survey of
teachers, showed that charter teachers had typically taken more math and science courses in
college, were more likely to have graduated from a good college, and logged more extracurricular
hours. Interestingly, charters paid a premium for these qualities, but not for certification
or master's degrees. Taylor (2005) also failed to find a premium for advanced degrees in Texas
charter schools, and showed that teachers realized a 7.5 percent pay cut upon moving to a
charter school.
While a picture of teacher quality in charter schools is emerging, little is known about the
flow of teaching talent between mainstream and charter schools, or the classroom performance
of individual charter teachers. Here, I fully characterize the resume qualifications of all North
Carolina public school teachers who moved to the charter sector between 1998 and 2007. For a
subsample of elementary teachers, I characterize their classroom performance as well. North
Carolina is a rare setting where passively collected administrative data have recorded longitudinal
school assignments for all charter and mainstream personnel over a period exceeding
ten years. Furthermore, the data link some teachers directly to their students, allowing me to
estimate measures of instructional effectiveness. By analyzing the flow of teachers from one
sector to another, I determine whether charter schools were "cream skimming" good teachers
from mainstream schools. If highly qualified and effective teachers were voting with their feet
in favor of charter schools, their migration is a favorable signal of the decentralized model's
appeal, and mainstream schools may need to emulate charter features to retain faculty. If
charters were drawing less qualified and less effective teachers, whether because of low pay,
poor organization, or relaxed licensure standards, the charter model is unlikely to fulfill its
promise as a revolutionary vehicle for the improvement of public schools.
In this study, I evaluate the resume qualifications of North Carolina charter movers against
the qualifications of teachers moving between mainstream schools, controlling for receiving
school profiles. Charter movers were less experienced than other moving teachers on average,
but were also more likely to have at least twenty-five years' experience. Charter schools were
attracting teachers with high licensure test scores, but only among certified, regularly licensed
teachers. Uncertified teachers moving to charter schools, a large minority, substantially attenuated
the average qualifications of all charter movers. Resume qualifications are, at best,
incomplete signals of teacher quality. For a subset of elementary grade teachers, I also evaluate
their classroom performance directly, using estimated teacher fixed effects on student end-of-grade
math and reading exam scores. Charter movers were low in sending school distributions
of classroom performance, relative to their colleagues, but compared favorably to mainstream
teachers moving to similar schools. I complement these estimated mean differences in teacher
quality with analyses of the variance and distribution of teacher quality, dissected from the
variance in sampling error. Quality distributions for future charter teachers largely overlapped
quality distributions for exclusively mainstream teachers, but were centered at a significantly
lower figure.
These findings neither affirm nor reject the effectiveness with which North Carolina's current
charter model draws good teachers from mainstream schools. The system attracted highly
qualified, certified teachers more effective than teachers moving to comparable mainstream
schools, but low licensure requirements attracted uncertified, less qualified teachers who may
have had few career options in the mainstream sector. The paper is organized as follows. Section
II reviews pertinent details of the North Carolina charter system and describes the data.
Section III outlines the analytic methodology and discusses results. Section IV concludes.
(End of excerpt. The entire report is available in pdf format.)
Latest Research
PDF
Printer-friendly summary
