Abstract
Differences in teacher quality would appear to be the most likely reason for disparities in the quality of high-poverty and lower-poverty schools. However, the linkages between teacher quality and socio-economic-based disparities in student achievement are quite complex. Using data from North Carolina and Florida, this paper examines whether teachers in high-poverty schools are as effective as teachers in schools with more advantaged students. Bottom teachers in high-poverty schools are less effective than bottom teachers in lower-poverty schools. The best teachers, by comparison, are equally effective across school poverty settings. The gap in teacher quality appears to arise from the lower payoff to teacher qualifications in high-poverty schools. In particular, the experience-productivity relationship is weaker in high-poverty schools and is not related to teacher mobility patterns. Recruiting teachers with good credentials into high-poverty schools may be insufficient to narrow the teacher quality gap. Policies that promote the long-term productivity of teachers in challenging high-poverty schools appear key.
The text below is an excerpt from the complete document. Read the entire report in PDF format.
Introduction
Whether measured by student achievement or educational attainment, the divergence in student
performance between schools serving primarily low-income populations and those enrolling students from
more affluent families is stark. In 2009, only 14 percent of 4th grade students from high-poverty schools scored at or above the "proficient" level in reading on the National Assessment of Educational Progress (NAEP), whereas half the students in low-poverty schools met or exceeded the threshold for proficiency. Similarly, in math just 17 percent of students from high-poverty schools scored at the proficient level or above on the NAEP
while 60 percent of students from low-poverty schools performed at the proficient level or better.
The differences are even larger for students in large cities, where the proportion of students who are
eligible for free and reduced-price lunch (FRL) who earn a proficient score in either math or reading is lower
than for FRL students in the rest of the nation (Uzzell et al. 2010). Twelfth-grade students in high-poverty
schools are also less likely to earn a high school diploma than their counterparts from low-poverty schools (68
versus 91 percent) and are less likely to attend a four-year college (28 percent versus 52 percent) (Aud et al.
2010). Correspondingly, nearly 90 percent of so-called "dropout factories," where fewer than 60 percent of
high-school freshman are still attending the same school in grade 12, are schools serving large numbers of lowincome
students. The majority of these schools with low graduation rates are located in the nation's cites
(Balfanz and Legters 2005). In half of the 100 largest cities in the United States 50 percent or more of high
school students attend "dropout factories" (Balfanz and Legters 2004).
While student, parental and neighborhood factors undoubtedly contribute to the observed
performance differential between students in low and high-poverty schools, it is hard to deny that systematic
differences in school quality are partly to blame for the observed gaps in achievement and educational
attainment. The source of quality differentials between schools serving primarily low-income students and
those serving more affluent students, and hence the appropriate policies to ameliorate differences in school
quality, are less clear, however.
Differences in teacher quality would appear to be the most likely reason for disparities in the quality of
high-poverty and lower-poverty schools. Recent empirical evidence finds teachers to be the most important
schooling factor affecting student achievement. Rockoff (2004), Rivkin, Hanushek, and Kain (2005) and
Aaronson, Barrow, and Sander (2007) show that persistent measures of teachers' contributions to student
achievement or "value added" vary tremendously across teachers, and that the within-school variation in value
added is at least as large as between-school variation. Similarly, Figlio and Lucas (2004) demonstrate that
within-school variation in grading standards (and presumably other teacher behaviors as well) is nearly the
same as the overall variation in grading standards.
Previous research has also highlighted disparities in the qualifications of teachers in schools serving
primarily disadvantaged and minority students versus teachers in schools with more advantaged student
bodies (Clotfelter, Ladd, and Vigdor 2005; Goldhaber, Choi, and Cramer 2007; Lankford, Loeb, and Wyckoff
2002). However, while observed teacher characteristics (e.g. educational attainment, certification status,
years of experience beyond the first few years, etc.) vary across schools, these differences are only weakly
related to teacher performance (Harris and Sass 2007; Clotfelter, Ladd, and Vigdor 2007).
Several recent studies also clearly establish that schools serving disadvantaged students have more
difficulty hiring and retaining teachers. Teachers in general appear to prefer schools that serve students from
higher income families and who are higher-achieving, and teachers working in schools with more highly
disadvantaged students are more likely to leave their school district or transfer to a lower-needs school within
their district (Lankford, Loeb, and Wyckoff 2002; Hanushek, Kain, and Rivkin 2004; Boyd, Lankford, Loeb, and
Wyckoff 2005; Clotfelter, Ladd, and Vigdor 2005; Imazeki 2005; Scafidi, Sjoquist, and Stinebrickner 2007; Feng
2009). Accountability pressures can exacerbate the problems that schools serving low-achieving student
populations face in retaining high-quality teachers (Feng, Figlio, and Sass 2010). Teachers with better earning
opportunities are also more likely to leave teaching (Dolton and van der Klaauw 1999), which might make it
even harder for schools in economically disadvantaged area to hold on to good teachers. It may also be harder
for these schools to recruit teachers to begin with, as potential teachers tend to prefer to work in schools near
where they grew up (Boyd et al. 2005).
The combination of evidence on the importance of teacher quality, differences in observable
qualifications of teachers across schools, and the mobility patterns of teachers has led many observers to
conclude that the quality of teachers in high-poverty schools is generally inferior to that of teachers in lowerpoverty
schools. This view has fueled policy initiatives designed to encourage promising new teachers to teach
in high-poverty schools ("Teach for America") and provide incentives for existing teachers to move from lowerpoverty
schools to high-poverty schools. For example, the recent "Race-to-the-Top" competition graded
applicants in part on whether they had plans to provide financial incentives to teach in high-poverty schools.1
Similarly, the U.S. Department of Education has funded a set of experiments in seven school districts
throughout the country, known as the "Talent Transfer Initiative" that provide differential compensation to
highly effective teachers who agree to teach in high-poverty schools.
Despite the circumstantial evidence, in fact little is known about the relative productivity of teachers in
schools serving economically disadvantaged student populations and those enrolling students from more
affluent families. In this paper we seek to fill this void and inform the debate on teacher labor market policies
by addressing four related research questions:
- How does the average contribution of teachers in high-poverty elementary schools compare to that of
teachers in lower-poverty elementary schools in terms of student achievement gains in mathematics
and in reading/language arts?
- Are there differences in the variation of teacher effectiveness within schools serving largely students
from low-income families vis-à-vis the set of schools serving more affluent populations?
- To what extent do observed teacher characteristics (e.g., certification, experience, education) in highpoverty
elementary schools and in schools with lower-poverty levels explain differences in teacher
contributions to student learning?
- To what extent does teacher mobility contribute to differences in teacher value added across high and
lower-poverty schools?
Our analysis has important implications for public policy. Given the large variation in teacher
effectiveness, major improvements in student outcomes could be realized if schools were to identify the most
successful teachers and deploy them in the settings where they could make the most difference. Having
improved information about the likely contribution of teachers in high-poverty schools and in other schools,
and the degree to which these differences can be explained by the types of factors that are observable ex ante,
may help frame the magnitude of the potential problem of staffing schools serving disadvantaged students.
And understanding the variation of measured teacher effectiveness within and across high-poverty schools
and other schools can offer insight into the potential scope for and design of teacher compensation policies
aimed at attracting and retaining highly effective teachers in the most challenging schools. It may also have
implications for the design of performance accountability systems, in particular the balance targeted to the
school level and individual teacher level.
End of excerpt. The entire report is available in PDF format.)
Latest Research
PDF
Printer-friendly summary
