See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/334096687

Does teaching experience increase teacher effectiveness? A review of US

research

Article  in  Journal of Professional Capital and Community · June 2019

DOI: 10.1108/JPCC-12-2018-0032

CITATIONS READS

179 12,967

3 authors:

Anne Podolsky Tara Kini

Stanford University Learning Policy Institute
8 PUBLICATIONS   340 CITATIONS    4 PUBLICATIONS   246 CITATIONS   

SEE PROFILE SEE PROFILE

Linda Darling-Hammond
Stanford University
337 PUBLICATIONS   40,199 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

SCOPE Case Studies, Stanford University View project

"A gestão da sala de aula Puan." Preparando os Professores para um Mundo em Transformação: O Que Devem Aprender e Estar Aptos a Fazer, L. Darling-Hammond,J.
Bransford, Editör, Penso, Sao-Paulo, ss.327-357, 2019 View project

All content following this page was uploaded by Anne Podolsky on 18 July 2019.

The user has requested enhancement of the downloaded file.

 The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/2056-9548.htm

Teacher
Does teaching experience effectiveness
increase teacher effectiveness?
A review of US research
Anne Podolsky, Tara Kini and Linda Darling-Hammond
Learning Policy Institute, Palo Alto, California, USA
Received 8 December 2018
Revised 11 April 2019
Accepted 7 May 2019
Abstract
Purpose – The purpose of this paper is to summarize the key findings from a critical review of relevant US
research to determine whether teachers, on average, improve in their effectiveness as they gain experience in
the teaching profession.
Design/methodology/approach – This paper is based on the authors’ review of 30 studies published since
2003 that analyze the effect of teaching experience on student outcomes in the USA.
Findings – The authors find that: teaching experience is positively associated with student achievement
gains throughout much of a teacher’s career; as teachers gain experience, their students are more likely to do
better on measures of success beyond test scores; teachers make greater gains in their effectiveness when
they teach in a supportive, collegial environment, or accumulate experience in the same grade, subject or
district; and more experienced teachers confer benefits to their colleagues.
Originality/value – A renewed look at this research is warranted due to advances in methods and data
systems that have allowed researchers to examine this question with greater sophistication.
Keywords Teacher development, Teacher quality, Professional environment, Returns to experience
Paper type General review

Introduction
In recent years, much research and policy making in the USA has been guided by an
assumption that teachers’ experience has little bearing on a teacher’s effectiveness after
about two or three years at the beginning of the career (Gates, 2009; Henry et al., 2011; Rice,
2013; Rivkin et al., 2005). For example, a policy brief published in 2010 summarizing the
benefits of teaching experience in US schools noted that “[t]eachers show the greatest
productivity gains during their first few years on the job, after which their performance
tends to level off” (Rice, 2013, p. 1). Quite often, this leads to the presumption that there is
little advantage to policy efforts that might seek to keep teachers in the classroom longer.
Although there is consensus in the research and policy communities that teachers improve
quickly early in their careers (Boyd, Lankford, Loeb, Rockoff and Wyckoff, 2008; Harris and
Sass, 2011; Kraft and Papay, 2014; Ladd and Sorensen, 2017; Staiger and Rockoff, 2010),
there is debate about whether or not teachers continue to learn as they gain additional
experience in the classroom. In contrast to the 2010 policy brief, a more recent study noted:
[W]e find that teachers experience rapid productivity improvement early in their careers. However,
we also find evidence of returns to experience later in the career, indicating that teachers continue
to build human capital beyond these first years. (Papay and Kraft, 2015, p. 105)
Advances in research methods and data sets help to explain these divergent findings.
Improvements in state and district longitudinal data systems have allowed researchers to
match student data with individual teachers and to track the effectiveness of individual
teachers throughout their careers. Such data systems have allowed researchers to look more

The technical report on which this manuscript is based (https://learningpolicyinstitute.org/product/

Journal of Professional Capital and
does-teaching-experience-increase-teacher-effectiveness-review-research) was funded in part by the Community
S.D. Bechtel, Jr Foundation. Core operating support for the Learning Policy Institute is provided by the © Emerald Publishing Limited
2056-9548
Ford Foundation, the William and Flora Hewlett Foundation, and the Sandler Foundation. DOI 10.1108/JPCC-12-2018-0032
JPCC closely at the effect of teaching experience on student outcomes for the same teachers over
time as they gain more experience. In contrast, earlier studies often could not track
individual teachers, and therefore used cross-sectional analyses, which compare groups of
teachers with different experience levels at one moment in time. Cross-sectional analyses do
not account for other differences that might exist among different groups of teachers, such
as differences in teaching abilities that can arise from changes in teacher credentialing
policies, labor market shifts or teacher attrition. Longitudinal studies help to control for
these differences between cohorts of teachers, thus allowing a more accurate understanding
of teacher development over time.
With significant additions to the literature on the effects of teaching experience in
recent years, a renewed look at the research is warranted. Moreover, policy makers
have debated whether to invest in developing and retaining talented teachers based on
their assumptions about whether teachers improve after the first few years in the
classroom. These questions take on additional importance given the increase in the share
of inexperienced teachers in the profession (Ingersoll et al., 2018), combined with their
disproportionate concentration in schools with high proportions of low-income students,
students of color and English learners (Betts et al., 2000; Boyd, Grossman, Lankford, Loeb
and Wyckoff, 2008; Boyd, Lankford, Loeb, Rockoff and Wyckoff, 2008; Goldhaber et al.,
2015; Sass et al., 2012). Accordingly, understanding the relationship between teacher
experience and effectiveness is critical for ensuring that policy makers equitably allocate
teachers across the education system.
We begin by providing an overview of the various methods used to measure the impact
of teaching experience on student outcomes and explain our approach to this review. After
that, we provide the results from our review of the research organized by three major
themes: the relationship between experience and student achievement; how contexts shape
returns to experience; and the effects of more experienced teachers on their colleagues and
schools. We then discuss the significance of our results, including implications for teachers’
professional learning and for school leaders. We conclude with the policy implications of our
findings and with suggestions for further research.

Theories of experience and effectiveness

The economic literature provides several models of human capital that may bear on the
relationship between teaching experience and effectiveness. Becker (1964) argued that more
experienced workers are more effective employees because they acquire more knowledge
about how to perform their work effectively over the course of their careers.
Ben-Porath’s (1967) model suggests that workers learn by spending time investing in
their skills and knowledge, which does not occur simultaneously with their production.
Under this model, the returns to learning may increase at the beginning of the career but
decrease with age because there are fewer opportunities to use new knowledge in the future.
However, the returns to producing increase with age, presumably because workers have
acquired greater skill and are more effective and efficient in their production. Accordingly,
Ben-Porath’s model predicts that workers spend less time investing in their learning as they
gain experience and more time producing. For teachers, this might mean that after teachers
accumulate a high level of experience, they may spend less time investing in their teaching
skills, which might cause their effectiveness at teaching students to increase for a period of
time based on their previous learning and skill development, but to plateau at some point in
time. In collaborative contexts, however, their greater knowledge and skill could increase
not only their own productivity, but also that of other more junior colleagues who can learn
from them.
Hargreaves and Fullan (2012) take these ideas further, building on established theories of
human capital development with the concept of professional capital in teaching, which is
comprised of three elements: human capital (investment in an individual teacher’s Teacher
knowledge and skills), social capital (investment in the relationships among educators and effectiveness
the quality and quantity of interactions that allow them to work productively together as a
group) and decisional capital (the wisdom and expertise to make sound judgments about
learners that are cultivated over many years). According to this theory, decisional capital is
built through experience and practice, “enabling [teachers] to make wise judgments in
circumstances where there is no fixed rule” (p. 94). Decisional capital is also enhanced when
it is mediated through interactions with colleagues (social capital) who bring their insights
and experiences to the process of forming judgments. In addition, the decisional capital of
experienced teachers contributes to the wisdom of the group as a whole, expanding social
capital and helping improve the capacity of novices to solve problems of practice, thus
strengthening their human capital.
Research about the relationship between experience and effectiveness in other
professions has found the type of non-linear – sometimes curvilinear – relationship
predicted by the Ben-Porath model. However, the findings have been mixed, depending on
the type of experience being studied (e.g. experience doing a specific task, in a specific job or
in a specific organization), the complexity of the job and the measure of effectiveness (e.g.
productivity or a supervisor’s evaluations). Generally, less complex jobs have been found to
have an inverted U-shaped relationship between experience at a specific job or in a specific
organization and performance. For jobs of higher complexity, the relationship between
experience and performance is stronger, but still non-linear (Sturman, 2003).
A meta-analysis of 350 studies found a curvilinear relationship between experience at a
given organization and performance (as measured by employee evaluations, reports of
compliance with safety practices and injuries, and self-reports of creativity), with the
strength of the relationship decreasing as employees gain experience (Ng and Feldman,
2010). The relationship between experience and performance was stronger “for younger
workers, for women, for non-Caucasians, and for college-educated workers” (p. 1220).
These findings, considered in tandem with those of Sturman, might suggest that in
teaching, which is a job requiring greater complexity and a college education, we might
expect a relatively strong relationship between teaching experience and teachers’
effectiveness at raising student achievement, but one that is non-linear, with some
attenuation of the relationship at some point in the career. Many studies have, indeed,
modeled the effects of teaching experience on performance with this in mind, and have
found a quadratic relationship. A key question for this review and for education policy
makers is at what point in the career – under what circumstances – this attenuation occurs.

Methodology
We are interested in addressing the following two research questions:
RQ1. Do teachers continue to improve in their effectiveness as they gain experience in
the teaching profession?
RQ2. If so, for how long do teachers continue to improve in their effectiveness as they
gain experience in the teaching profession?
To answer these questions, we examined studies that analyzed the effect of teaching
experience on student outcomes in K-12 public schools in the USA, as measured by student
standardized test scores and non-test metrics when available. We identified studies that
examined teaching experience published in peer-reviewed journals and by organizations
with established peer-review processes since 2003[1], when the use of teacher fixed effects
methods – which allow researchers to compare a teacher with multiple years of experience
to that same teacher when he or she had fewer years of experience – became more prevalent.
JPCC We used three search engines to identify studies. We used Scopus, Web of Sciences and
Google Scholar to identify published studies or studies under review between 2003 and 2016,
using the following Boolean search terms: “teacher experience” & “student achievement”;
“teaching experience” & “student achievement”; “returns to teaching experience”; and “returns
to teacher experience.” Based on these search parameters, SCOPUS returned 70 results, Web
of Sciences returned 55 results and Google Scholar returned 24,540 results.
For searches with thousands of results, we reviewed the abstracts of the first 100 results
because Google Scholar reports research by relevance, including “where it was published,
who it was written by, as well as how often and how recently it has been cited in other
scholarly literature” (Google Scholar, n.d.). In addition to peer-reviewed articles, Google
Scholar’s search results include non-peer-reviewed papers, like theses, books and materials
from online repositories, web sites, and professional societies. Consequently, articles beyond
the first approximately 60 results for each of our searches were generally outside of our
domain because they did not specifically focus on returns to teaching experience or were not
from peer-reviewed sources. We also reviewed the reference lists of the peer-reviewed
articles from our searches that examined teaching experience to confirm that we did not
overlook any key studies. The process of using the three different search engines and
reviewing the reference lists helped us triangulate the results to verify that we included all
relevant studies that met our criteria for inclusion.
To select studies for our review, we included studies that met the following criteria:
publication in a peer-reviewed journal or by an organization with established peer-review
processes; a specific focus on returns to teaching experience, rather than having teaching
experience as an incidental variable; and sufficient explanation of methods, which included
adequate controls to provide a reasonable estimation of the returns to teaching experience.
A summary of our review for each study is in a table available on-line that identifies in some
detail the sample and methods used by each study, the range of years of experience studied
and the findings[2]. By aggregating the findings of studies on teaching experience in
relation to the quality of methods applied and the years of experience studied, we hope to
provide a more comprehensive understanding of the variation across these studies, and the
potential reasons for that variation.
In addition, the “Studies Not Meeting Criteria” section in that table contains studies
released since 2003 that met most, but not all, of our criteria, either because the study was
not published in a peer-reviewed journal or by an organization with established peer-review
processes, or the study did not specifically focus on returns to teaching experience
(i.e. teaching experience was an incidental variable in a study that focused on another issue).
We included some of these studies because they are frequently cited in the teaching
experience literature. Despite the limitations of these studies, based on our thorough review
of their methods and findings, we found that they generally supported the same findings as
those we report here.
Our reviewed sources are confined to studies conducted in the USA. Relationships
between teacher experience and educational effectiveness may be different in jurisdictions
where conditions are not the same as the USA in terms of teacher certification, school
assignment, professional learning, or time for joint planning and collaboration. For example,
in Singapore and Shanghai, senior teachers are supported in developing their expertise and
are then assigned in their schools to help mentor other teachers and lead action research
groups (Darling-Hammond, Hyler and Gardner, 2017; Darling-Hammond, Burns, Campbell,
Goodwin, Hammerness, Low, … Zeichner, 2017). In Canada, veteran teachers are supported
to lead professional development and inquiry projects, with resulting strong enthusiasm
about professional learning from their colleagues (Campbell et al., 2016). In most countries,
teachers receive significantly more hours to spend time planning and learning with
colleagues out of class, compared to the USA where teachers have one of the lowest
allocations of in-school time for collaboration and joint planning (Organization for Economic Teacher
Cooperation and Development, 2014). effectiveness
Methodological issues in studying teaching experience
Research examining the effects of teaching experience generally addresses two questions.
First, is the average experienced teacher more effective at raising student test scores than
the average inexperienced teacher? Second, do teachers continue to improve in their
effectiveness as they gain experience in the teaching profession?
Earlier studies generally only answered the first question. These studies usually used
cross-sectional analyses, which compare distinct cohorts of teachers with different
experience levels during a single school year (Betts et al., 2003; Clotfelter et al., 2006;
Dee, 2004; Huang and Moon, 2009; Nye et al., 2004). Some of these studies found that on
average, teaching quality did not appear to differ across experience levels, meaning that the
effectiveness of a novice teacher was similar to the effectiveness of a veteran (Aaronson
et al., 2007; Betts et al., 2003; Rivkin et al., 2005). However, as we describe below, because
these studies often do not account for other factors that shape the different cohorts and may
contribute to student outcomes, their findings may be less accurate than those of studies
with more complete controls (Papay and Kraft, 2015; Wiswall, 2013).

Challenges with early empirical analyses

Most early empirical studies that analyzed the relationship between teachers’ years of
experience and their students’ outcomes could not distinguish between two effects. The first
effect is the extent to which teachers grow on the job, which is sometimes called “returns to
teacher experience.” The second effect is the possibility that a cohort of teachers with more
experience is simply more or less able collectively than those with less experience at that
moment in time.
The second effect could arise if the teachers who leave the profession after only a few years
of teaching differ in their basic ability from those who choose to remain in the profession. For
example, if more effective teachers are more likely to leave the teaching workforce under
certain circumstances, then estimates of the effects of teaching experience using a cross-
sectional analysis would likely underestimate the returns to experience, and show a negative
relationship between teaching experience and student outcomes (Papay and Kraft, 2015).
Research is mixed on the relationship between teacher effectiveness and teacher attrition.
Some research shows that attrition is higher among more effective teachers (Wiswall, 2013;
Clotfelter et al., 2007), yet other literature shows that attrition is higher among less effective
teachers (Henry et al., 2011; Goldhaber et al., 2011; Hanushek et al., 2005). Some research
finds no relationship (Papay and Kraft, 2015; Rivkin et al., 2005), or mixed results depending
on whether attrition occurs after year 1 or years 2 or 3 (Boyd, Grossman, Lankford, Loeb
and Wyckoff, 2008).
These different findings may be in part a function of different labor market conditions –
including the availability of other job opportunities for skilled individuals outside of the
education system in relation to teacher salary levels or working conditions in different
geographic locations and time periods. Under different circumstances, more or less effective
teachers may leave teaching (Nagler et al., 2015) – and these factors may vary among
different teaching fields. Differential attrition may also relate to differences in the induction
support and evaluation policies beginning teachers experience (Smith and Ingersoll, 2004).
This is an important question for further study, ideally with more information about the
labor market and human resource policies pertaining in the contexts studied.
Alternatively, the second effect could arise when social processes and norms that
influence the quality of individuals entering the profession change (e.g. the adoption of more
stringent certification requirements or the advent of economic recessions that make
JPCC teaching a comparatively more attractive job). For example, if teacher certification
standards became more rigorous, then newer teachers might be more competent and
capable, on average, than more experienced teachers. For example, some research shows
that a set of US federal and state policies in the late 1990s to 2000s – including state
certification reforms and the highly qualified teacher requirements of the federal No Child
Left Behind Act – improved the quality of the teacher workforce (Gitomer, 2007). Another
study of New York State teachers found that the academic ability of teachers (as measured
by standardized test scores and the selectivity of undergraduate institutions) increased after
1998 when the state imposed more stringent regulations for teacher preparation and
certification (Lankford et al., 2014). As a result, it is possible that New York State teachers
who started teaching after 1998 may, as a cohort, be more effective than the cohort of
teachers who entered previously.
Other research has found that teachers who enter the profession during an economic
recession are more effective in raising student achievement than those who enter during
rosier economic times (presumably because teaching competes more favorably for talent
with other professions when the economy lags), although they also tend to have higher
attrition (perhaps because they leave teaching for other opportunities when the economy
improves) (Nagler et al., 2015).
These illustrations demonstrate the complex dynamics of the public education system
that researchers often try to account for to ensure their estimates are accurate and
minimally biased. Different researchers employ different strategies to control for student,
teacher and school dynamics that influence teacher effectiveness and student outcomes.
The researchers in this field most frequently use the strategies described below to reduce
the bias in their estimates of returns to teaching experience. Even though we privilege
teacher fixed effects for the reasons explained below, each method has advantages and
limitations that must be considered when interpreting a study’s findings, and the findings of
the literature as a whole.

Fixed effects
Teacher fixed effects. When researchers want to most accurately estimate the extent to
which teachers improve as they gain more years of teaching experience, the standard
solution for reducing bias is to include teacher fixed effects in the model. The addition of
teacher fixed effects allows researchers to compare a teacher with multiple years of
experience to that same teacher when he or she had fewer years of experience. This
approach, sometimes referred to as a “within-teacher comparison,” eliminates the problem of
comparing different cohorts to one another while it also controls statistically for teacher
ability (Kraft and Papay, 2014). In other words, teacher fixed effects analyses account for
each teacher’s characteristics that are not likely to vary with time, such as verbal ability,
content knowledge or general temperament. As a result, this method improves the estimate
of the relationship between the gains teachers make in their ability to improve student
outcomes and their experience, often referred to as “within-teacher returns to experience”
(Papay and Kraft, 2015). This method eliminates the limitations created by selective attrition
and/or differences in cohort quality. This method has become more common, in part because
of the increasing availability of large longitudinal data sets in which students can be
matched to their specific teachers over time.
A recent study using data from North Carolina fifth-grade public school teachers from
1996 to 2005 explored how excluding teacher fixed effects data can negatively bias
estimates of the returns to teacher experience. This study found that teachers who left the
teaching workforce earlier in their careers appeared to have higher “innate teaching quality”
than those who remained in teaching for a longer period of time (Wiswall, 2013, p. 71).
Several studies have similarly found that the most effective teachers leave the profession Teacher
earlier (Boyd, Grossman, Lankford, Loeb and Wyckoff, 2008; Boyd, Lankford, Loeb, Rockoff effectiveness
and Wyckoff, 2008; Harris and Sass, 2011; Wiswall, 2013). Wiswall (2013) applied multiple
empirical models from prior literature to replicate the lower returns to experience found in
earlier studies. When the author included fixed effects in models that analyzed a wide range
of experience, he found “much higher returns to teaching experience” than in models
without teacher fixed effects (p. 75). This study suggests that many of the specifications in
previous models that did not include teacher fixed effects mask the trend of selective teacher
exits and, therefore, may be biased.
Although teacher fixed effects are particularly helpful in examining the question, we
pursue here – whether individual teachers become more effective as they gain experience –
they cannot, by themselves, account for the potential differentials in teacher learning gains
that may be associated with their initial levels of knowledge, skills, or effectiveness;
the contexts within which they teach; or measurements of effectiveness associated with the
students they teach. These latter differentials have been noted in the recent literature on
value-added approaches to teacher evaluation, which often show downwardly biased and
particularly unstable results for teachers who work with students at the tails of the
achievement distribution because of the inability of grade-level tests to measure their
learning well ( for a review, see Haertel, 2013).
School and student fixed effects. Another challenge that researchers encounter when
estimating returns to teaching experience is that teachers are generally not randomly
assigned to students. More experienced teachers have often been found to teach students
with higher ability or to migrate to schools with more advantaged students (Clotfelter et al.,
2006; Kalogrides et al., 2012). As such, the estimated effects of teacher experience on student
achievement could be biased upward if more experienced teachers typically teach in schools
or classrooms serving more high-performing students and those students are more likely to
show gains. As noted above, however, if those students are so high-performing as to achieve
at the very top of a test which has a low ceiling, their teacher may appear to show low
value-added gains from one year to the next because the test may not measure the higher
levels of achievement at which they are functioning.
The standard solution for addressing potential bias associated with non-random
assignments of teachers across schools is to add school fixed effects, which allows researchers
to compare teachers within, not across, schools. In other words, school fixed effects analyses
compare a teacher only to other teachers in the same school. Thus, these analyses account for
the variation of school-level factors that cannot be observed but may contribute to a teacher’s
returns to experience, such as the nature of the student body, the effectiveness of a school’s
administration, or the resources available for teaching (Kane et al., 2008).
However, school fixed effects analyses can be limited. First, this approach assumes there
are no meaningful differences in teacher quality across schools, which we know to be an
unrealistic assumption. For example, because inexperienced teachers are disproportionately
concentrated at the highest-need schools and also show the steepest returns to experience,
this method may limit the teacher comparison group and bias estimates of returns to
teaching experience. Another downside to this approach in estimates of returns to teaching
experience is that it may not address the possibility that even within schools, more
experienced teachers might be given the classrooms with the stronger students.
One solution researchers employ for this possibility is to control for student
characteristics, or to add student fixed effects.
Student fixed effects analyses compare a teacher only to other teachers who have taught
the same student. In the teacher experience literature, this means that models with student
fixed effects look at whether a given student’s performance is better (or worse) when that
JPCC student is taught by a teacher with more (or fewer) years of experience. The application of
student fixed effects analyses seeks to address the issue that teachers are not randomly
assigned to students or randomly distributed across classrooms, possibly as a result of
teacher assignment or preference, parent pressure or student tracking.
Evidence suggests that teachers with better training and more experience tend to teach
not only at schools serving more affluent and higher-achieving students, but also in
classrooms serving these more advantaged students within a given school (Clotfelter et al.,
2006; Kalogrides et al., 2012). Even though student fixed effects analyses can be beneficial
for investigating some relationships, this method can bias estimates of returns to teaching
experience because it restricts the comparison group. For example, a student in a poorly
resourced school with substantial teacher turnover may see a string of relatively
inexperienced teachers, so that the student’s potential learning gains with more experienced
teachers may be impossible to estimate. Conversely, it may be difficult to estimate the
differential learning gains associated with teaching experience for a student in an affluent
school that rarely hires beginning teachers.

Analysis of the range of teaching experience

Researchers adopt different methods for specifying teaching experience in their analyses,
including focusing only on the early years of teaching, with the later years capped; using
categorical variables to group teachers into ranges of experience; and applying individual
indicators by years of experience. Including indicator variables for each year of experience
for a teacher in a similar grade over time introduces challenges for researchers. Specifically,
the year variable (which controls for broad changes within the year, such as changes in
testing or standards that might influence student achievement) and the teaching experience
variable move together over time. As a result, these variables are perfectly correlated, which
makes it difficult for researchers to separate the experience effect from the year effect (Ladd
and Sorensen, 2017). To address this issue, researchers use a variety of methods for
including experience variables in their estimates. We describe the commonly used indicator
variables for teaching experience analyses below, including the potential bias the different
methods can introduce.
One way that researchers analyze experience is by only looking at returns to experience
during the first few years of a teacher’s career, using an indicator variable that includes all
experience above a specific threshold together. This method assumes that teachers do not
improve their effectiveness after the cut-off year of experience (Wiswall, 2013, p. 108). Thus,
this limits the inferences that can be drawn about experience past the cut-off year because
the variable fails to measure the possible differential effects of teacher experience
throughout this wide period of time.
Another way that researchers analyze returns to experience is by using indicator
variables that combine wide ranges of experience. For example, a researcher may look at the
gains a teacher’s students make during the 1–5 years of the teacher’s career, the 6–10, and
then from the 11th year through the end of the teacher’s career. This approach is limited
because it assumes that teacher productivity does not change within each of the ranges of
experience (Papay and Kraft, 2015, p. 108). To illustrate, assume a sample covered a
five-year period, and the range of the intervals of experience was 0–4, 5–12 and 13–20 years.
If a teacher entered the sample with five years of experience and then gained experience for
each year through the five-year sample period, it would appear as if the teacher had not
gained experience because five to ten years of experience is in the same interval. As such,
estimates may represent an average return to experience within each range, and could
therefore underestimate returns, especially during periods of teachers’ careers when they
are rapidly improving in their effectiveness (Papay and Kraft, 2015).
 Papay and Kraft (2015) demonstrated how indicator variables for cut-off years and wide Teacher
ranges of experience create significant downward bias to estimates of returns to experience. effectiveness
First, they found that the main assumption of this method – that teachers do not improve in
their effectiveness in later years or past the cut-off year – was violated across multiple
specifications that showed teachers continuing to make gains in their productivity later in
their careers, past the cut-off year (Papay and Kraft, 2015, pp. 112-113). Moreover, this study
found that “as the [intervals for experience] get narrower, the estimated returns to
experience grow steeper and the extent of later-career improvement increases […] suggest
[ing] a violation of the key assumption” of the model, by demonstrating that teachers
continue to improve throughout the ranges of experience (p. 113). In addition, this study
found that using a cut-off year indicator variable created “substantial downward bias that
understates the estimated returns to experience” when teachers continue to make gains in
their effectiveness after the cut-off year (p. 111). The authors also found that using indicator
variables to represent broad ranges of years of experience “substantially understates the
estimated returns to experience […] by as much as 68%” (p. 111).

Results
The relationship between experience and student achievement gains
We reviewed 30 studies examining the effects of teaching experience on student
achievement, as measured by standardized test scores (see Table I: further details can be
found in Table II available online (see footnote 2)). Of these 30 studies, 28 found that
teaching experience is positively and significantly associated with teacher effectiveness.
Approximately two-thirds of the studies analyze longitudinal data sets with teacher fixed
effects, and the method is preferred because it allows for the examination of “within-teacher”
returns to experience. Of these studies, 18 out of 18 found that teaching experience is
positively associated with teacher effectiveness.
All of the studies applying teacher fixed effects analyses found that teaching
experience is positively associated with gains in student achievement. Teachers make the
steepest gains in effectiveness during their first few years in the classroom, when they are
“greenest.” Numerous studies confirm the unremarkable finding that, on average, brand
new teachers are less effective than those with some experience (Boyd, Grossman,
Lankford, Loeb and Wyckoff, 2008; Boyd, Lankford, Loeb, Rockoff and Wyckoff, 2008;
Clotfelter et al., 2006, 2007; Harris and Sass, 2011; Kane et al., 2008; Ladd and Sorensen,
2017). Most of these studies also found that teachers show the greatest gains from
experience during their initial years in the classroom, but continue to make meaningful
improvement in their effectiveness past these initial gains (Koedel and Betts, 2007). One
study suggested that teachers who received little coursework or hands-on training prior to
entering the classroom – such as those who come through alternative routes to
certification without completing student teaching under the guidance of an accomplished
teacher – may experience the steepest gains in their initial years in the classroom as
they are starting from zero and taking their preparation courses while teaching
(Kane et al., 2008).
Among the studies applying teacher fixed effects analyses and examining the effects of
teaching experience on student achievement after a teacher accumulates seven or more
years of experience, all 15 found that teachers continue to improve in their effectiveness in
fostering student achievement beyond the first decade of a teacher’s career. These studies
consistently found a positive and significant relationship between teaching experience and
student gains on standardized tests. The most recent studies that analyze a wider range of
teaching experience and apply teacher fixed effects analyses – in both math and reading at
the elementary, middle and high school levels – have found significant returns to experience
into the second, and often the third, decade of a teacher’s career.
JPCC Found positive Included All studies that
relationship teacher fixed measured 7+ years
between teaching Included effects and of experience, with
experience and teacher measured 7+ and without
student fixed years of teacher fixed
S. No. Study achievement effects experience effects

1 Ladd and Sorensen (2017) Yes Yes Yes Yes

2 Blazar (2015) Yes Yes No No
3 Papay and Kraft (2015) Yes Yes Yes Yes
4 Kraft and Papay (2014) Yes Yes Yes Yes
5 Ost (2014) Yes Yes Yes Yes
6 Wiswall (2013) Yes Yes Yes Yes
7 Sass et al. (2012) Yes Yes Yes Yes
8 Chingos and Peterson (2011) Yes Yes Yes Yes
9 Harris and Sass (2011) Yes Yes Yes Yes
10 Clotfelter et al. (2010) Yes Yes Yes Yes
11 Buddin and Zamarro (2009) Yes Yes Yes Yes
12 Jackson and Bruegmann Yes Yes Yes Yes
(2009)
13 Kukla-Acevedo (2009) Yes Yes Yes Yes
14 Boyd, Lankford, Loeb, Yes Yes Yes Yes
Rockoff and Wyckoff (2008)
15 Kane et al. (2008) Yes Yes No No
16 Koedel and Betts (2007) Yes Yes Yes Yes
17 Hanushek et al. (2005) Yes Yes No No
18 Rockoff (2004) Yes Yes Yes Yes
19 Henry et al. (2012) Yes No No No
20 Henry et al. (2011) No No No No
21 Staiger and Rockoff (2010) Yes No No No
22 Huang and Moon (2009) Yes No No Yes
23 Clotfelter et al. (2007) Yes No No Yes
24 Boyd et al. (2006) Yes No No Yes
25 Clotfelter et al. (2006) Yes No No Yes
26 Jepsen (2005) Yes No No Yes
27 Rivkin et al. (2005) Yes No No No
28 Dee (2004) Yes No No Yes
29 Nye et al. (2004) Yes No No No
30 Betts et al. (2003) No No No Yes
No. and % of studies with 28/30 18/18 15/15 21/22
Table I.
Summary of analyses positive findings 93% 100% 100% 95%
of teaching experience Note: The detailed table of the sample and methods used, the range of experience studied, and the findings
& student for each study in our review is located at https://drive.google.com/file/d/1c6hrXexjwmjnxhevZ3yACDtYcjx
achievement WgAls/view?usp=sharing

The summary below highlights two recently published studies that use teacher fixed effects
analyses over ten or more years of teaching experience in order to illuminate the association
between the methodologies used and the nature and size of the gains in student outcomes
attributed to more experienced vs less experienced teachers over various lengths of time.
Ladd and Sorensen’s (2017) study of 250,000 middle school students in North Carolina
over a five-year period found returns to teaching experience in several outcome areas
through at least 12 years of experience in both math and English Language Arts (ELA).
With respect to student achievement, gains leveled off somewhat after 12 years of
experience in both math and ELA, with declines after the 28th year of experience (see
Figure 1). These findings were robust across specifications, with three of the preferred
models including teacher and grade-by-year fixed effects. In addition, one model included
 Test Score Results: Math Test Score Results: Reading Teacher

STUDENT READING TEST SCORE (STANDARD DEVIATIONS)

effectiveness
STUDENT MATH TEST SCORE (STANDARD DEVIATIONS)

0.20

0.15

0.10

0.05

0
0 10 20 30 0 10 20 30
YEARS OF EXPERIENCE Figure 1.
Note: In their analysis, the authors used administrative data on teachers and students in The relationship of
North Carolina from the North Carolina Education Research Data Center for grades 6–8 in teaching experience to
math and reading test
2006–2011 score gains
Source: Ladd and Sorensen (2017)

school fixed effects, and the other two models included student fixed effects, without
school fixed effects. The gains were still present but smaller in a non-preferred model that
did not use teacher fixed effects.
A teacher with 12 years of experience raised test scores from 0.08 standard deviations in
ELA to 0.18 standard deviations in math as compared to a teacher with no prior experience.
Although the returns level off to some extent after 12 years for math and ELA teachers, the
study found that math and ELA teachers with 21–27 years of experience were still 0.04
standard deviations more effective than when they had five years of experience.
This study also found returns to teacher experience with respect to student absenteeism
and other desirable student behaviors. With respect to student attendance, the authors note:
One year of experience enables an English Language Arts teacher to reduce the proportion of
students with high absenteeism by 2.0 percentage points, and these reductions increase as they
continue to gain experience. A teacher of given quality who obtains over 21 years of experience on
average reduces the incidence of high student absenteeism by 14.5 percentage points. (Ladd and
Sorensen, 2017, p. 263)
Two years of experience allowed a math teacher to reduce “the proportion of students with
high absenteeism by 3.8 percentage points, an effect that rises to an 11.5 percentage point
reduction for teachers with extensive experience” (Ladd and Sorensen, 2017, p. 267).
In addition, experience increased the ability of ELA teachers to encourage students to spend
more time reading for pleasure, and math teachers to promote positive classroom behavior.
These findings are policy relevant, given the strong evidence base linking high rates of
absenteeism with negative long-term educational outcomes (Balfanz and Byrnes, 2012;
Balfanz et al., 2007). Importantly, the study found that more experienced teachers provided the
greatest benefit to higher-risk, chronically absent students. For example, “ELA and math
teachers with 21–27 years of experience reduce the number of students with over three
absences by 5.6 and 4.4 percent respectively,” but reduced “the number of students with over
JPCC 17 absences by 18.8 and 12.2 percent” (p. 263). The authors suggest that as teachers gain
experience, they improve their skills in classroom management and motivating students.
Papay and Kraft (2015) developed four different models based on different
methodological theories to examine returns to experience in a large urban school district
using longitudinal data over nine years for more than 3,500 teachers and their students in
grades 4–8. The four models measuring gains to experience included teacher, school and
grade-by-year fixed effects: a “Censored Growth Model” (which buckets 20+ years of
experience together); a “Two-Stage Model” (which first model[s] productivity as a function
of both experience and year effects and then adds teacher fixed effects in the second stage);
an “Indicator Variable Model” (which uses dummy variables for 1–2, 3–4, 5–9, 10–14, 15–24
and 25+ years of experience); and a “Discontinuous Career Model” (which uses a sample of
teachers with discontinuous careers).
The study found large and statistically significant early-career (years 1–5) returns to
experience across models in both students’ mathematics and reading achievement. The
study also found consistent evidence of growth in later stages of the teaching career,
particularly in mathematics. As might be expected from other analyses, the authors found
larger and more extended returns to experience from the indicator variable model than the
censored growth model, due to the more precise measurement of experience.
From year 5 to 15 of teacher experience, in mathematics, the study found significant
improvements in teacher effectiveness between 0.033 and 0.051 standard deviations using
the censored growth model. The authors noted that teachers’ gain in effectiveness during
this later ten-year period is meaningfully large, representing 45–60 percent of the gains
teachers make in their first five years of experience. In the “Discontinuous Career Model,”
they found that elementary and middle school math teachers improved at the same rate
from years 29 to 30 as they did from years 2 to 3, suggesting that there is not an inevitably
sharp decline in teachers’ effectiveness as teachers age.
Although a number of studies have found that teachers continue to improve through the
cut-off year of the study’s analysis (Blazar, 2015; Chingos and Peterson, 2011; Jackson and
Bruegmann, 2009; Ost, 2014), other studies have found a non-linear relationship between
teacher experience and effectiveness. Specifically, teaching experience can be modeled as a
non-linear or curvilinear relationship to account for the trend that individuals might
improve for a certain amount of time, but then their increases in productivity may taper off
and eventually decline. Multiple studies we reviewed did suggest a non-linear relationship
between teacher experience and effectiveness. Importantly, these studies generally found
that the declines in effectiveness at raising student achievement occurred later in a teacher’s
career, after 20 or more years in the profession. For example, both Ladd and Sorensen (2017)
and Sass et al. (2012) found that math and ELA teachers in low-poverty North Carolina
elementary schools declined in their effectiveness after 28 or more years of teaching,
studying middle school and elementary school teachers, respectively.

How contexts matter in shaping returns to experience

There is a growing body of research analyzing whether different types of teaching
experience accelerate teachers’ rates of improvement over time. Two types of experience
have been addressed in the research: experience in a supportive professional working
environment and experience at the same grade level/subject area or in the same district.
This research has found that teachers appear to make greater gains in their effectiveness
when they teach in a supportive and collegial working environment, or accumulate
experience in the same grade level, subject or district.
Experience in a more collegial and supportive working environment. At least two studies
have found that teachers’ rate of improvement over time depends on the supportiveness of
their professional working environment. These studies often measure support using
measures of teachers’ perceptions of their school climate, the quality of the leadership within Teacher
the school and the instructional support that they receive. effectiveness
A longitudinal study of 3,145 teachers and 280,000 elementary math students over ten
years in the Charlotte-Mecklenberg School District examined how teachers’ improvement
over time was related to the type of school they worked in (Kraft and Papay, 2014).
The study used data focused on teachers early in their careers (years 1–10) and compared
teachers to themselves over time, using teacher fixed effects analyses. The study also used
grade-by-year fixed effects to control for factors across grades and years, such as “the
introduction of new policies in certain grades” (p. 481). The study then compared the
within-teacher returns to experience of teachers in schools with more supportive
professional environments to those of their peers in schools with less supportive
professional environments. As with many other studies, the study found positive returns to
experience generally (see Figure 2).
Of particular interest, the study also showed that teachers who work in schools with strong
professional environments improve in their effectiveness in teaching mathematics at much
faster rates than their peers working in schools with weaker professional environments. These
environments are characterized by a trusting, respectful, safe and orderly environment, with
collaboration amongst teachers, school leaders who support teachers, time and resources for
teachers to improve their instructional abilities, and teacher evaluation that provides
meaningful feedback. After ten years, teachers who work in schools with stronger (75th
percentile) working environments will have become more effective, by approximately one-fifth
of a standard deviation than teachers who work at schools with weaker (25th percentile)
working environments (Kraft and Papay, 2014, p. 487). The authors note that this difference
“represents over 30 percent of the average total improvement teachers make in their first
10 years on the job” (Kraft and Papay, 2014, p. 494).
A similar study of over 9,000 teachers in 336 Miami-Dade County public schools, which
used teacher fixed effects and controls for school characteristics, also found that teacher
collaboration led to greater rates of improvement and improved student outcomes in math
and reading (Ronfeldt et al., 2015).

0.14
STUDENT MATHEMATICS ACHIEVEMENT

0.12
(STANDARD DEVIATIONS)

0.1

0.08

0.06 75th Percentile: Quartic

Average: Quartic
25th Percentile: Quartic
0.04 75th Percentile: Indicators
25th Percentile: Indicators
0.02
Figure 2.
0 Returns to teaching
1 2 3 4 5 6 7 8 9 10+ experience for
YEARS OF EXPERIENCE prototypical teachers,
Note: In their analysis, the authors used administrative data on teachers and students in across school
professional
a large, urban school district in the Southern USA in 2000–2009 environments
Source: Kraft and Papay (2014)
JPCC Experience in same grade level, subject area or district. A few studies have examined the
relationship between the type of experience and its effects on student achievement. They
show that teachers with prior experience in the same grade level, subject area or district
show greater returns to experience than those with less relevant prior experience.
One recent study, looking at North Carolina elementary teachers in grades 3–5 over an
18-year period found significant returns to experience, with larger returns to experience for
those teaching at the same grade level. In math, students of teachers with more
grade-specific experience made greater progress than students who had a similarly
experienced teacher with less grade-specific experience, with the grade-specific experience
effect half as large as the general experience effect (Ost, 2014, p. 136). In reading, the study
showed returns to general teaching experience but did not find greater returns for
grade-specific experience as compared to general teaching experience. Perhaps this is
because reading objectives in North Carolina are constant across grades, whereas the math
objectives vary dramatically for each grade (Ost, 2014, pp. 129, 138).
The study found that the teachers who had recently taught their current grade showed a
greater benefit from specific experience in the same grade than those who had not. Finally,
the study found that teachers switch grade assignments frequently: 18 percent of teachers
switch to a new grade in their second year of teaching, and fewer than half teach the same
grade over the course of their first five years of teaching. A similar study using ten-year
data from a large urban school district in California reached a similar conclusion: elementary
teachers are frequently required to switch grades, particularly in low-achieving,
high-minority schools, and grade switching is associated with smaller returns from
experience and higher rates of turnover among teachers (see Figure 3; Blazar, 2015). These
studies suggest that reducing the frequent reassignment of teachers, especially for less
experienced teachers, would help to improve the substantial returns to teaching experience.
A smaller study examining benefits of experience for teachers of elementary reading at
53 Title I schools in a mid-Atlantic state also found large and significant benefits where
teachers had prior experience at the same grade level covered in the study (Huang and
Moon, 2009). Importantly, the study found that:
[…][t]he effect size for seasoned grade level teachers [effect size ¼ 0.27] is three times larger than
the effect size of economic status (effect size ¼ −0.09) and nearly as large as the effect of minority

Average Returns to experience Grade Switchers

STUDENTS ACHIEVEMENT IN MATH

STUDENT ACHIEVEMENT IN ELA

(STANDARD DEVIATIONS)

(STANDARD DEVIATIONS)

0.3 0.3

0.2 0.2

0.1 0.1

Figure 3. 0 0
Returns to experience 1 2 3 4 5 6 7 1 2 3 4 5 6 7
in math and ELA on YEARS OF EXPERIENCE
average and for Note: In the analysis, the author used administrative records from a large urban school district in
teachers who switch
to any grade California for elementary school teachers and students in 2002–2012
Source: Blazar (2015)
 status (effect size ¼ −0.33). This suggests that experienced teachers teaching at a specific grade Teacher
level could have a large effect in countering the effects of the widening achievement gaps. (Huang effectiveness
and Moon, 2009, p. 226)
The study also found that:
Teachers constantly improved teaching effectiveness until the 21st year […] The most effective
teachers had 19-24 years of experience at grade level and were associated, holding all other
variables constant, with increased student reading achievement […] (effect size ¼ 0.40). (p. 227)
These highly experienced teachers were twice as effective as teachers with at least more
than five years of experience (effect size ¼ 0.20). The study did not find significant returns
to experience when prior experience did not occur at the same grade level.

The effects of more experienced teachers on their colleagues and schools

Some research indicates that teachers whose colleagues are more experienced are more effective
than those whose colleagues are less experienced, suggesting that more experienced teachers
provide important additional benefits to their school community beyond increased learning for
the students they teach. A study using data from third through fifth-grade students and their
teachers in North Carolina over an 11-year period found that teachers whose peer teachers had
more experience tended to have improved student outcomes ( Jackson and Bruegmann, 2009).
For example, the study found that teachers improve in their ability to raise their students’ test
scores by over 0.02 standard deviations in math and over 0.01 standard deviations in reading
when their peer teachers (those teaching at the same grade level) have at least four years of
experience. The study also found that novice teachers benefit most from having more
experienced teachers as peers, and that the quality of a teacher’s peers has ripple effects for that
teacher’s students’ achievement beyond the current school year. That is, the quality of a teacher’s
peers the year before, and even two years before, affects her current students’ achievement.
Finally, a study of Florida and North Carolina elementary schools from the 2000–2001 to
2004–2005 school years found that the returns to experience for teachers in the
lowest-poverty schools tended to be greater than for teachers in the highest-poverty schools
(Sass et al., 2012). In addition, the authors tracked teachers who left their school, and found
that the mobility of teachers at low- and high-poverty schools did not drive their different
rates of returns to experience. Instead, the authors asserted that differences in effectiveness
were likely the result of the quality and experience of peer teachers and the challenging
environment of many high-poverty schools that leads to “burn out.”

Discussion
These research findings show the benefits of more experienced teachers for both students and
schools. As teachers gain experience – both within their first few years in the classroom as well
as later in their careers – they are better able to foster student learning. Taken as a whole, the
large and growing body of research, especially when it applies methods that include teacher
fixed effects, provides support for the conclusion that teachers become better able to support
student learning as they gain experience, and that gains from experience can continue well
into the second and, often third decades of their career. The assumption alluded to earlier that
teachers reach a plateau soon after their initial years in the classroom, after which additional
experience has no benefit for student achievement, finds little support in newer research,
particularly among those studies that examine teaching experience beyond seven years and
apply methods to examine “within-teacher” returns to experience.
These results reflect the research in other professions about patterns of employee growth
over time. As described earlier, the curvilinear results – teachers continue to improve in their
ability to raise student achievement through the second and often third decade of teaching,
JPCC but then level off or decline in their effectiveness later in their career – are in line with the
Ben-Porath (1967) human capital model from economic literature. The patterns in the
teaching profession also appear to mirror findings that more complex professions requiring
a college education, like teaching, tend to have a strong relationship between experience in
the profession and effectiveness in that profession, with a longer period of time before
effectiveness levels off.
Consistent with Hargreaves’ and Fullan’s theory of professional capital, more robust
returns to experience appear most strongly when teachers are working in supportive and
collegial school environments. A high level of collaboration among teachers within a school
is associated with increased gains in student achievement and faster rates of improvement
for teachers (Ronfeldt et al., 2015). In addition, a more stable and experienced teaching staff
appears to benefit students across the entire school, as more experienced teachers are better
able to support their less experienced colleagues in producing student achievement. A more
stable and experienced teaching workforce also allows for teachers to develop shared goals
and beliefs around student learning over time (Donohoo, 2017). When teachers believe they
can work together to overcome challenges and achieve goals (sometimes referred to as
collective efficacy), they are more likely to be effective at improving student achievement
(Donohoo, 2017; Donohoo et al., 2018). Importantly, retention is higher in collaborative school
environments, creating a virtuous cycle in which supportive and collegial schools are better
able to attract and retain excellent, experienced teachers, who are the ones best positioned to
contribute to school-wide learning and greater student achievement. In other words, schools
with lower attrition rates, and often more experienced teachers, tend to have a collegial
culture rooted in teachers’ shared knowledge and practice (Simon and Johnson, 2015).
In contrast, schools with large proportions of inexperienced teachers (often the
highest-poverty schools) have limited numbers of experienced mentor teachers to support
the development of new teachers (Loeb et al., 2005). In these types of settings where “the
blind are leading the blind,” returns to experience may be lessened because there simply are
not enough expert, experienced teachers to mentor and support novices, and the few who
could serve as mentors are stretched thin and feel overburdened by the needs of their
colleagues as well as their students (Darling-Hammond, 2010). Schools with fewer
experienced teachers, and often higher turnover rates, tend to have a disjointed knowledge
base within the school and a lack of coherence of instructional practice. High rates of teacher
turnover have been found to have a significant negative effect on student achievement that
extends beyond the classrooms of students whose teachers have left, particularly in schools
serving large populations of low-performing and black students (Ronfeldt et al., 2013).
Taken as a whole, the findings of this review suggest that investments in building an
experienced, highly-collaborative teacher workforce focused on continual learning are most
likely to result in greater student learning, while, at the same time, reducing teacher
attrition. Below, we discuss the implications of our findings for teachers’ professional
learning, as well as for school leadership, policy and future research.

Implications for professional learning

Our findings point to the importance of creating collaborative environments in which
teachers continue to grow. Integrating effectively structured professional learning
opportunities into collegial environments can contribute to stronger returns to experience
(Darling-Hammond, Hyler, and Gardner, 2017). For example, effective opportunities for
collaborative professionalism can include teachers working together in a lesson study,
curriculum planning or teacher-led professional learning communities (Hargreaves and
O’Connor, 2018).
These same principles can support stronger preparation for and entry into teaching.
Estimates suggest that between 17 and 30 percent of beginning teachers in the USA leave
the profession within the first five years (Gray et al., 2015; Darling-Hammond and Sykes, Teacher
2003). Ironically, just as teachers are making their steepest gains in effectiveness, many effectiveness
decide to leave the profession, leaving students – particularly low-income students and
students of color – with inexperienced, less effective teachers.
A growing body of evidence demonstrates that attrition is highest for those who enter
the profession without adequate preparation. While preparation was not the focus of our
study, during the course of our review, we noticed that preparation can influence the returns
to experience. For example, a study of North Carolina high school teachers found that
teachers prepared at more competitive institutions and those with regular (full) licensure
were associated with higher student achievement, compared to teachers prepared at less
competitive schools or those who were not fully certified (Clotfelter et al., 2010). Other
research shows that beginning teachers with extensive preparation – those who had
practice teaching, received feedback on their teaching, and completed coursework on
specific aspects of teaching – are about twice as likely to remain in the classroom as
compared to teachers with little or no preparation (Ingersoll et al., 2014).
A promising model that supports candidates to become part of a productive professional
community and to remain in the profession is the teacher residency program. These
programs provide talented college graduates with a year-long paid residency under the
guidance of an accomplished master teacher, coupled with coursework at a partnering
university that is closely intertwined with clinical practice. These programs typically train
cohorts of teachers in the subject areas most in demand in the sponsoring school district and
do so in partner schools organized to support a group of residents. In exchange for this
intensive preparation, which is substantially underwritten by government, district and
university sponsors, residents commit to teach in the district for several years. Rigorous
studies of teacher residency programs have found significantly higher retention rates for
graduates of these programs (e.g. Silva et al., 2015; Papay et al., 2012).
In addition, high-quality mentoring and induction programs for new teachers can
accelerate novice teachers’ learning and reduce teacher turnover. Providing mentoring and
induction support to novice teachers can lead to teachers who stay in the profession longer,
accelerated professional growth among new teachers and improved student learning
(Ingersoll and Strong, 2011).

Implications for school leadership

The quality of administrative support is often the top reason teachers identify for leaving or
staying in the profession (Kraft et al., 2016; Sutcher et al., 2016). In fact, several studies have
found that support from principals and other school leaders is one of the best predictors of
teacher attrition (Hughes et al., 2015; Torres, 2016; Grissom, 2011). Consequently, quality
school leaders play an important role in developing an experienced and stable teacher
workforce. High-quality principals work to include teachers in decision-making, foster
positive school cultures, and create learning communities that support teachers’ continual
development. The important role of principals in attracting, keeping and supporting
teachers highlights the need to invest in high-quality principals. Some studies have found
that effectively designed preparation and professional learning programs for principals can
reduce teacher turnover and, ultimately, be associated with improved student outcomes
(Sutcher et al., 2017).
To support these goals, school leaders should learn how to increase opportunities for
teachers to collaborate both because of the promise it holds for improved teacher retention
and because the benefits of experience are greater for teachers in strong professional
working environments (Kraft and Papay, 2014). School leaders should also be mindful of
the research, showing that teachers who have repeated experience teaching the same grade
level or subject area improve more rapidly than those whose experience is in another grade
JPCC level or subject (Ost, 2014). While many factors influence job assignment decisions,
including teachers’ desires for professional growth and new challenges as well as principals’
needs for flexibility in management, principals should take into account the increased
benefits of specific teaching experience when making decisions about teaching assignments.

Implications for policy makers

A more experienced teaching workforce offers numerous benefits to students and schools,
including greater individual and collective effectiveness in improving student outcomes as
well as greater stability and coherence in instruction and relationship building – the core
work of schools. Given the benefits of teachers’ experience for student learning and the
evidence that effectiveness can be increased over time in collegial settings, policy makers
should use a range of tools to build an experienced teaching workforce of well-prepared
individuals focused on continually learning. These policies would include financial supports
and program designs to ensure comprehensive preparation and mentoring that enable
teachers to stay in the profession long enough to gain further experience and build on their
knowledge base.
Strategies would also include policies to increase teacher retention, including competitive
salaries – sufficiently equitable across districts to ensure experienced teachers in all
communities – and working conditions that allow teachers to meet students’ needs and to
learn with and from one another (Carver-Thomas and Darling-Hammond, 2017; Podolsky
et al., 2016). Policy makers can also support principal development programs that enable
school leaders to learn how to create strong collaborative working environments that
support teacher learning and collective efficacy (Sutcher et al., 2017).

Implications for research

Based on our analysis, we recommend further research of several kinds. First, research
should continue to explore the circumstances of a teacher’s working conditions and collegial
environment associated with the greatest returns to teaching experience.
Second, research might seek to understand more fully how teachers build collective
expertise and efficacy within and across teams that comprise a mix of teaching experiences.
A third area for research would be to explore how teaching experience, coupled with
opportunities for ongoing collegial learning, influence teachers’ abilities to teach in the
more sophisticated ways required to develop the skills students need to succeed in the
twenty-first century, such as critical thinking and problem-solving, oral and written
communication, and an academic mindset. Evidence from the international TALIS surveys
suggests that teachers are more likely to feel a sense of efficacy and to implement
reform-oriented practices when they have more opportunities for collaboration with other
teachers (Organization for Economic Cooperation and Development, 2014). The ways in
which these skills and this sense of efficacy are developed in collegial settings is important
to understand.

Conclusion
The studies included in this review necessarily vary in their selection of target population
data (e.g. grade level, subjects, geography) and in their methods. However, this body of
studies – each of which analyzes the effects of teaching experience on student achievement
employing several different approaches – provides a more optimistic answer than was once
offered to the question of “Do teachers continue to improve in their effectiveness as they
gain experience in the teaching profession?”
The common refrain that teaching experience does not matter after the first few years in
the classroom appears no longer supported by the preponderance of the research,
particularly when teachers’ effectiveness is examined across the continuum of their Teacher
individual careers, rather than by comparing different cohorts of teachers with each other. effectiveness
Of course, not all experience is educative: some highly experienced teachers are not
particularly effective or have retired on the job, and some novice teachers are dynamic and
effective. However, by and large, a more experienced teaching workforce offers numerous
benefits to students and schools. A growing research base suggests that teachers’
effectiveness rises sharply in the first few years of their careers, and this upward trajectory
continues well into the second and often third decade of teaching, with a steeper slope when
teachers work in collegial settings. The effects of teaching experience on student
achievement are significant. As some analysts note, the compounded positive effect of
having a series of accomplished, experienced teachers for several years in a row may offer
the opportunity to reduce the achievement gap for low-income students and students of
color (Clotfelter et al., 2010; Huang and Moon, 2009; Wiswall, 2013). These effects are even
stronger in settings where teams of teachers have the opportunity to work together (Kraft
and Papay, 2014; Ronfeldt et al., 2015).

Notes
1. In Harris and Sass’s (2011) review of research into teacher training and productivity, the authors
note that earlier research (published prior to 2003) about teaching experience and productivity has
been thoroughly reviewed.
2. The detailed table of the sample and methods used, the range of experience studied and the
findings for each study in our review is located at https://drive.google.com/file/d/1c6hrXexjwmjnx
hevZ3yACDtYcjxWgAls/view?usp=sharing

References
Aaronson, D., Barrow, L. and Sander, W. (2007), “Teachers and student achievement in the Chicago
Public High schools”, Journal of Labor Economics, Vol. 25 No. 1, pp. 95-135, doi: 10.1086/508733.
Balfanz, R. and Byrnes, V. (2012), The Importance of Being There: A Report on Absenteeism in the
Nation’s Public Schools, Johns Hopkins University Center for Social Organization of Schools,
Baltimore, MD, available at: http://new.every1graduates.org/wp-content/uploads/2012/05/
FINALChronicAbsenteeismReport_May16.pdf
Balfanz, R., Herzog, L. and Mac Iver, D.J. (2007), “Preventing student disengagement and keeping
students on the graduation path in urban middle-grades schools: early identification and
effective interventions”, Educational Psychologist, Vol. 42 No. 4, pp. 223-235, doi: 10.1080/004
61520701621079.
Becker, G. (1964), Human Capital: A Theoretical and Empirical Analysis with Special Reference to
Education, Columbia University Press, New York, NY.
Ben-Porath, Y. (1967), “The production of human capital and the life cycle of earnings”, The Journal of
Political Economy, Vol. 75, No. 4, Part 1, pp. 352-365.
Betts, J.R., Reuben, K.S. and Danenberg, A. (2000), Equal Resources, Equal Outcomes? The Distribution
of School Resources and Student Achievement in California, Public Policy Institute of California,
San Francisco, CA.
Betts, J.R., Zau, A.C. and Rice, L.A. (2003), “Determinants of student achievement: new evidence from
San Diego”, Public Policy Institute of California, San Francisco, CA.
Blazar, D. (2015), “Grade assignments and the teacher pipeline: a low-cost lever to improve student
achievement?”, Educational Researcher, Vol. 44 No. 4, pp. 213-227, doi: 10.3102/0013189X15580944.
Boyd, D., Grossman, P., Lankford, H., Loeb, S. and Wyckoff, J. (2006), “How changes in entry
requirements alter the teacher workforce and affect student achievement”, Education Finance
and Policy, Vol. 1 No. 2, pp. 176-216, available at: doi.org/10.1162/edfp.2006.1.2.176
JPCC Boyd, D., Grossman, P., Lankford, H., Loeb, S. and Wyckoff, J. (2008), Who Leaves? Teacher Attrition
and Student Achievement (No. w14022), National Bureau of Economic Research, Cambridge, MA.
Boyd, D., Lankford, H., Loeb, S., Rockoff, J.E. and Wyckoff, J. (2008), “The narrowing gap in New York
City teacher qualifications and its implications for student achievement in high-poverty schools”,
Journal of Policy Analysis and Management, Vol. 27 No. 3, pp. 793-818, doi: 10.1002/pam.20377.
Buddin, R. and Zamarro, G. (2009), “Teacher qualifications and student achievement in urban elementary
schools”, Journal of Urban Economics, Vol. 26 No. 2, pp. 103-115, doi: doi.org/10.1016/j.jue.2009.05.001.
Campbell, C., Osmond-Johnson, P., Faubert, B., Zeichner, K., Hobbs-Johnson, A., with Brown, S.,
DaCosta, P., Hales, A., Kuehn, L., Sohn, J., and Steffensen, K. (2016), The State of Educators’
Professional Learning in Canada, Learning Forward, Oxford, OH.
Carver-Thomas, D. and Darling-Hammond, L. (2017), Teacher Turnover: Why it Matters and What we
can do about it, Learning Policy Institute, Palo Alto, CA.
Chingos, M.M. and Peterson, P.E. (2011), “It’s easier to pick a good teacher than to train one: familiar
and new results on the correlates of teacher effectiveness”, Economics of Education Review,
Vol. 30 No. 3, pp. 449-465, doi: 10.1016/j.econedurev.2010.12.010.
Clotfelter, C.T., Ladd, H.F. and Vigdor, J.L. (2006), “Teacher-student matching and the assessment of teacher
effectiveness”, Journal of Human Resources, Vol. 41 No. 4, pp. 778-820, doi: 10.2307/40057291.
Clotfelter, C.T., Ladd, H.F. and Vigdor, J.L. (2007), “Teacher credentials and student achievement:
longitudinal analysis with student fixed effects”, Economics of Education Review, Vol. 26 No. 6,
pp. 673-682, doi: 10.1016/j.econedurev.2007.10.002.
Clotfelter, C.T., Ladd, H.F. and Vigdor, J.L. (2010), “Teacher credentials and student achievement in high
school”, The Journal of Human Resources, Vol. 45 No. 3, pp. 655-681, doi: 10.1016/j.econedurev.2007
.10.002.
Darling-Hammond, L. (2010), “Recruiting and retaining teachers: turning around the race to the bottom
in high-need schools”, Journal of Curriculum and Instruction, Vol. 4 No. 1, pp. 16-32.
Darling-Hammond, L., Hyler, M.E. and Gardner, M. (2017), Effective Teacher Professional Development,
Learning Policy Institute, Palo Alto, CA.
Darling-Hammond, L., Burns, D., Campbell, C., Goodwin, A.L., Hammerness, K., Low, E.L. McIntyre, A.,
Sato, M. and Zechner, K. (2017), Empowered Educators: How High-Performing Systems Shape
Teaching Quality around the World, John Wiley & Sons.
Darling-Hammond, L. and Sykes, G. (2003), “Wanted, a national teacher supply policy for education:
the right way to meet the ‘highly qualified teacher’ challenge”, Education Policy Analysis
Archives, Vol. 11 No. 33, doi: doi.org/10.14507/epaa.v11n33.2003.
Dee, T.S. (2004), “Teachers, race, and student achievement in a randomized experiment”, Review of
Economics and Statistics, Vol. 86 No. 1, pp. 195-210, doi: 10.1162/003465304323023750.
Donohoo, J. (2017), “Collective teacher efficacy research: implications for professional learning”, Journal
of Professional Capital and Community, Vol. 2 No. 2, pp. 101-116.
Donohoo, J., Hattie, J. and Eells, R. (2018), “The power of collective efficacy”, Educational Leadership,
Vol. 75 No. 6, pp. 40-44.
Gates, B. (2009), “Mosquitos, malaria, and education”, TED Talk, available at: www.ted.com/talks/bill_
gates_unplugged?language=en (accessed November 5, 2018).
Gitomer, D.H. (2007), “Teacher quality in a changing policy landscape: improvements in the teacher
pool”, Policy Information Report. Educational Testing Service, Princeton, NJ.
Goldhaber, D., Gross, B. and Player, D. (2011), “Teacher career paths, teacher quality, and persistence in
the classroom: are public schools keeping their best?”, Journal of Policy Analysis and
Management, Vol. 30 No. 1, pp. 57-87.
Goldhaber, D., Lavery, L. and Theobald, R. (2015), “Uneven playing field? Assessing the teacher quality
gap between advantaged and disadvantaged students”, Educational Researcher, Vol. 44 No. 5,
pp. 293-307, doi: 10.3102/0013189X15592622.
Google Scholar (n.d.), “About”, available at: https://scholar.google.com/intl/en/scholar/about.html Teacher
(accessed November 5, 2018). effectiveness
Gray, L., Taie, S. and O’Rear, I. (2015), Public School Teacher Attrition and Mobility in the First Five Years:
Results from the First through Fifth Waves of the 2007-08 Beginning Teacher Longitudinal Study,
National Center for Education Statistics, US Department of Education, Washington, DC.
Grissom, J.A. (2011), “Can good principals keep teachers in disadvantaged schools? Linking principal
effectiveness to teacher satisfaction and turnover in hard-to-staff environments”, Teachers
College Record, Vol. 113 No. 11, pp. 2552-2585.
Haertel, E. (2013), Reliability and Validity of Inferences about Teachers Based on Student Test Scores.
14th Angoff Lecture, Educational Testing Service, Princeton, NJ.
Hanushek, E.A., Kain, J.F., O’Brien, D.M. and Rivkin, S.G. (2005), “The market for teacher quality”,
National Bureau of Economic Research Working Paper Series No. 11154, Cambridge, MA.
Hargreaves, A. and Fullan, M. (2012), Professional Capital: Transforming Teaching in Every School,
Teachers College Press, New York, NY.
Hargreaves, A. and O’Connor, M.T. (2018), Collaborative Professionalism: When Teaching Together
Means Learning for All, Corwin, Thousand Oaks, CA.
Harris, D.N. and Sass, T.R. (2011), “Teacher training, teacher quality and student achievement”, Journal
of Public Economics, Vol. 95 Nos 7-8, pp. 798-812, doi: 10.1016/j.jpubeco.2010.11.009.
Henry, G.T., Fortner, C.K. and Bastian, K.C. (2012), “The effects of experience and attrition for novice
high-school science and mathematics teachers”, Science, Vol. 335 No. 6072, pp. 1118-1121,
doi: doi.org/10.1126/science.1215343.
Henry, G.T., Bastian, K.C. and Fortner, C.K. (2011), “Stayers and leavers: early-career teacher effectiveness
and attrition”, Educational Researcher, Vol. 40 No. 6, pp. 271-280, doi: 10.3102/0013189X11419042.
Huang, F.L. and Moon, T.R. (2009), “Is experience the best teacher? A multilevel analysis of teacher
characteristics and student achievement in low performing schools”, Educational Assessment,
Evaluation and Accountability, Vol. 21 No. 3, pp. 209-234, doi: 10.1007/s11092-009-9074-2.
Hughes, A.L., Matt, J.J. and O’Reilly, F.L. (2015), “Principal support is imperative to the retention of teachers
in hard-to-staff schools”, Journal of Education and Training Studies, Vol. 3 No. 1, pp. 129-134.
Ingersoll, R.M. and Strong, M. (2011), “The impact of induction and mentoring programs for beginning
teachers: a critical review of the research”, Review of Educational Research, Vol. 81 No. 2, pp. 201-233.
Ingersoll, R.M., Merrill, L. and May, H. (2014), What are the Effects of Teacher Education and Preparation
on Beginning Teacher Attrition?, Consortium for Policy Research in Education, Philadelphia, PA.
Ingersoll, R.M., Merrill, E., Stuckey, D. and Collins, G. (2018), “Seven trends: the Transformation of the
teaching force: Updated October 2018”, Consortium for Policy Research in Education,
Philadelphia, PA.
Jackson, C.K. and Bruegmann, E. (2009), “Teaching students and teaching each other: the importance
of peer learning for teachers”, American Economic Journal: Applied Economics, Vol. 1 No. 4,
pp. 85-108, doi: 10.1257/app.1.4.85.
Jepsen, C. (2005), “Teacher characteristics and student achievement: evidence from teacher surveys”,
Journal of Urban Economics, Vol. 57 No. 2, pp. 302-319, doi: doi.org/10.1016/j.jue.2004.11.001.
Kalogrides, D., Loeb, S. and Beteille, T. (2012), “Systematic sorting: teacher characteristics and class
assignments”, Sociology of Education, Vol. 86 No. 2, pp. 103-123, doi: 10.1177/0038040712456555.
Kane, T.J., Rockoff, J.E. and Staiger, D.O. (2008), “What does certification tell us about teacher
effectiveness? Evidence from New York City”, Economics of Education Review, Vol. 27 No. 6,
pp. 615-631, doi: 10.1016/j.econedurev.2007.05.005.
Koedel, C. and Betts, J.R. (2007), “Re-examining the role of teacher quality in the educational production
function”, Working Paper No. 2007-03, National Center on Performance Incentives, Peabody
College of Vanderbilt University, Nashville, TN, available at: https://my.vanderbilt.edu/
performanceincentives/files/2012/10/200703_KoedelBetts_ReExamTeacherQual1.pdf (accessed
November 5, 2018).
JPCC Kraft, M.A. and Papay, J.P. (2014), “Can professional environments in schools promote teacher
development? Explaining heterogeneity in returns to teaching experience”, Educational
Evaluation and Policy Analysis, Vol. 36 No. 4, pp. 476-500, doi: 10.3102/0162373713519496.
Kraft, M.A., Marinell, W.H. and Shen-Wei Yee, D. (2016), “School organizational contexts, teacher
turnover, and student achievement: evidence from panel data”, American Educational Research
Journal, Vol. 53 No. 5, pp. 1411-1449.
Kukla-Acevedo, S. (2009), “Do teacher characteristics matter? New results on the effects of teacher
preparation on student achievement”, Economics of Education Review, Vol. 28 No. 1, pp. 49-57,
doi: 10.1016/j.econedurev.2007.10.007.
Ladd, H.F. and Sorensen, L.C. (2017), “Returns to teacher experience: student achievement and
motivation in middle school”, Education Finance and Policy, Vol. 12 No. 2, pp. 241-279.
Lankford, H., Loeb, S., McEachin, A., Miller, L.C. and Wyckoff, J. (2014), “Who enters teaching?
Encouraging evidence that the status of teaching is improving”, Educational Researcher, Vol. 43
No. 9, pp. 444-453, doi: 10.3102/0013189X14563600.
Loeb, S., Darling-Hammond, L. and Luczak, J. (2005), “How teaching conditions predict teacher
turnover”, Peabody Journal of Education, Vol. 80 No. 3, pp. 44-70.
Nagler, M., Piopiunik, M. and West, M.R. (2015), “Weak markets, strong teachers: recession at career start
and teacher effectiveness”, preliminary version, NBER Working Paper Series, Cambridge, MA.
Ng, T.W.H. and Feldman, D.C. (2010), “Organizational tenure and job performance”, Journal of
Management, Vol. 36 No. 5, pp. 1220-1250.
Nye, B., Konstantopoulos, S. and Hedges, L.V. (2004), “How large are teacher effects?”, Educational
Evaluation and Policy Analysis, Vol. 26 No. 3, pp. 237-257, doi: 10.3102/01623737026003237.
Organization for Economic Cooperation and Development (2014), TALIS 2013 Results: An
International Perspective on Teaching and Learning, OECD, Paris.
Ost, B. (2014), “How do teachers improve? The relative importance of specific and general human capital”,
American Economic Journal: Applied Economics, Vol. 6 No. 2, pp. 127-151, doi: 10.1257/app.6.2.127.
Papay, J.P. and Kraft, M.A. (2015), “Productivity returns to experience in the teacher labor market:
methodological challenges and new evidence on long-term career improvement”, Journal of
Public Economics, Vol. 130, pp. 105-119, doi: 10.1016/j.jpubeco.2015.02.008.
Papay, J.P., West, M.R., Fullerton, J.B. and Kane, T.J. (2012), “Does an urban teacher residency increase
student achievement? Early evidence from Boston”, Educational Evaluation and Policy Analysis,
Vol. 34 No. 4, pp. 413-434.
Podolsky, A., Kini, T., Bishop, J. and Darling-Hammond, L. (2016), Solving the Teacher Shortage: How to
Attract and Retain Excellent Educators, Learning Policy Institute, Palo Alto, CA.
Rice, J.K. (2013), “Learning from experience? Evidence on the impact and distribution of teacher
experience and the implications for teacher policy”, Education and Finance Policy, Vol. 8 No. 3,
pp. 332-348.
Rivkin, S.G., Hanushek, E.A. and Kain, J.F. (2005), “Teachers, schools, and academic achievement”,
Econometrica, Vol. 73 No. 2, pp. 417-458, available at: www.jstor.org/stable/3598793
Rockoff, J.E. (2004), “The impact of individual teachers on student achievement: evidence from panel
data”, American Economic Review, Vol. 94 No. 2, pp. 247-252.
Ronfeldt, M., Loeb, S. and Wyckoff, J. (2013), “How teacher turnover harms student achievement”,
American Educational Research Journal, Vol. 50 No. 1, pp. 4-36, doi: 10.3102/0002831212463813.
Ronfeldt, M., Farmer, S.O., McQueen, K. and Grissom, J.A. (2015), “Teacher collaboration in
instructional teams and student achievement”, American Educational Research Journal, Vol. 52
No. 3, pp. 475-514, doi: 10.3102/0002831215585562.
Sass, T.R., Hannaway, J., Xu, Z., Figlio, D.N. and Feng, L. (2012), “Value added of teachers in
high-poverty schools and lower poverty schools”, Journal of Urban Economics, Vol. 72 No. 2,
pp. 104-122, doi: 10.1016/j.jue.2012.04.004.
Silva, T., McKie, A. and Gleason, P. (2015), “New findings on the retention of novice teachers from Teacher
teaching residency programs”, NCEE Evaluation Brief, NCEE 2015-4015, National Center for effectiveness
Education Evaluation and Regional Assistance, Washington, DC.
Simon, N.S. and Johnson, S.M. (2015), “Teacher turnover in high-poverty schools: what we know and can
do”, Teachers College Record, Vol. 117 No. 3, pp. 1-36.
Smith, T.M. and Ingersoll, R.M. (2004), “What are the effects of induction and mentoring on beginning
teacher turnover?”, American Educational Research Journal, Vol. 41 No. 3, pp. 681-714.
Staiger, D.O. and Rockoff, J.E. (2010), “Searching for effective teachers with imperfect information”,
Journal of Economic Perspectives, Vol. 24 No. 3, pp. 97-118, doi: 10.1257/jep.24.3.97.
Sturman, M.C. (2003), “Searching for the inverted u-shaped relationship between time and performance:
meta-analysis of the experience/performance, tenure/performance, and age/performance
relationships”, Journal of Management, Vol. 29 No. 5, pp. 609-640.
Sutcher, L., Darling-Hammond, L. and Carver-Thomas, D. (2016), Coming Crisis in Teaching?
Teacher Supply, Demand, and Shortages in the U.S. (research brief ), Learning Policy Institute,
Palo Alto, CA.
Sutcher, L., Podolsky, A. and Espinoza, D. (2017), Supporting Principals’ Learning: Key Features of
Effective Programs, Learning Policy Institute, Palo Alto, CA.
Torres, A.C. (2016), “Is this work sustainable? Teacher turnover and perceptions of workload in charter
management organizations”, Urban Education, Vol. 51 No. 8, pp. 891-914.
Wiswall, M. (2013), “The dynamics of teacher quality”, Journal of Public Economics, Vol. 100, pp. 61-78,
doi: 10.1016/j.jpubeco.2013.01.006.

Corresponding author
Anne Podolsky can be contacted at: apodolsky@learningpolicyinstitute.org

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com

View publication stats