ArticlePDF Available

Recent Trends in the Characteristics of New Teachers, the Schools in Which They Teach, and Their Turnover Rates

Authors:

Abstract and Figures

Background Over the past couple of decades, new teachers have become a pronounced focus of policy makers. This attention is a result of demographic shifts in the teacher labor market, increased attention to the quality of teachers assigned to historically underserved student populations, and high rates of new teacher turnover. Purpose Our goal is to understand how the characteristics of new teachers and the schools in which they teach have changed over time, including changes in the characteristics of teachers in schools that enroll a majority of economically disadvantaged and minority students. We examine how these broad changes relate to new teacher turnover, as well as the induction supports that may improve retention. Research Design We draw on nationally representative data from all seven waves of the Schools and Staffing Survey from the 1987–1988 to 2011–2012 school years to better understand the extent to which the characteristics of beginning teachers, the schools in which they teach, and their turnover rates have changed over time. We draw on data regarding teachers’ demographic characteristics, education and credentials, and school characteristics and working conditions, and turnover. Data Analysis We first describe how characteristics of new teachers and the schools in which they teach have changed. We then examine how these characteristics vary systematically across schools that enroll a majority of students who identify as racial/ethnic minorities and schools with fewer racial/ethnic minorities. We examine the extent to which new teachers are more likely than more experienced teachers to turn over from underserved schools and how organizational supports such as administrative support and induction programs predict lower rates of turnover. Findings Beginning teachers are now more likely to be certified, to have earned a graduate degree, and to have graduated from a selective college. They are also much more likely to begin their careers in schools that enroll more racially diverse students than in previous decades. Teachers are more likely to turn over from schools where the majority of students identify as racial/ethnic minorities, although the presence of supportive colleagues and administrators and induction supports are associated with lower turnover rates. Conclusions That the majority of new teachers now begin their careers in schools serving a majority of economically disadvantaged and minority students has implications for the preparation and induction of new teachers. Our results point to the continued importance of the provision of supports that integrate teachers into the social and professional culture of a school.
Content may be subject to copyright.
1
Teachers College Record Volume 122, 070311, July 2020, 36 pages
Copyright © by Teachers College, Columbia University
0161-4681
Recent Trends in the Characteristics of New
Teachers, the Schools in Which They Teach,
and Their Turnover Rates
CHRISTOPHER REDDING
University of Florida
TUAN D. NGUYEN
Kansas State University
Background: Over the past couple of decades, new teachers have become a pronounced focus
of policy makers. This attention is a result of demographic shifts in the teacher labor market,
increased attention to the quality of teachers assigned to historically underserved student
populations, and high rates of new teacher turnover.
Purpose: Our goal is to understand how the characteristics of new teachers and the schools in
which they teach have changed over time, including changes in the characteristics of teachers
in schools that enroll a majority of economically disadvantaged and minority students. We
examine how these broad changes relate to new teacher turnover, as well as the induction sup-
ports that may improve retention.
Research Design: We draw on nationally representative data from all seven waves of the
Schools and Staffing Survey from the 1987–1988 to 2011–2012 school years to better under-
stand the extent to which the characteristics of beginning teachers, the schools in which they
teach, and their turnover rates have changed over time. We draw on data regarding teachers’
demographic characteristics, education and credentials, and school characteristics and work-
ing conditions, and turnover.
Data Analysis: We first describe how characteristics of new teachers and the schools in which
they teach have changed. We then examine how these characteristics vary systematically across
schools that enroll a majority of students who identify as racial/ethnic minorities and schools
with fewer racial/ethnic minorities. We examine the extent to which new teachers are more
likely than more experienced teachers to turn over from underserved schools and how orga-
nizational supports such as administrative support and induction programs predict lower
rates of turnover.
Findings: Beginning teachers are now more likely to be certified, to have earned a graduate
degree, and to have graduated from a selective college. They are also much more likely to begin
their careers in schools that enroll more racially diverse students than in previous decades.
Teachers College Record, 122, 070311 (2020)
2
Teachers are more likely to turn over from schools where the majority of students identify as
racial/ethnic minorities, although the presence of supportive colleagues and administrators
and induction supports are associated with lower turnover rates.
Conclusions: That the majority of new teachers now begin their careers in schools serving
a majority of economically disadvantaged and minority students has implications for the
preparation and induction of new teachers. Our results point to the continued importance
of the provision of supports that integrate teachers into the social and professional culture
of a school.
Over the past couple of decades, new teachers have become a pronounced
focus of policy makers. First, in an attempt to move past the “sink or
swim” mentality of the first year of teaching, whereby new teachers man-
age the demands of teaching with little professional support (Johnson &
Birkeland, 2003; Kardos & Johnson, 2007), districts and states set up new
teacher induction and mentoring programs with the shared goals of im-
proving teacher quality and retention (Smith & Ingersoll, 2004). More
recently, new teachers’ effectiveness has been placed under enhanced
scrutiny, with districts and states experimenting with policies linking new
teachers’ tenure and employment with job performance (Dee & Wyckoff,
2015; Sartain & Steinberg, 2016).
This recent policy attention toward new teachers is a result of both
demographic shifts in the teacher labor market and increased attention
to the quality of teachers assigned to historically underserved student
populations. Beginning in the mid-1990s, the National Commission on
Teaching and America’s Future (1996) warned of the need to recruit
scores of new teachers to compensate for increased teacher retirement
and student enrollment. Although these warnings of teacher shortages
appear to have been overstated (Aaronson & Meckel, 2008; Ingersoll et
al., 2014), it provoked national attention on teacher recruitment as the
number of newly hired increased throughout the 1990s and 2000s. This
trend progressed until the 2007–2008 school year, when first-year teach-
ers became the modal teacher—a sharp contrast to 1987–1988, when
the modal teacher had 15 years of experience (Ingersoll et al., 2014).
Besides the distributional changes in teacher experience, Ingersoll and
colleagues (2014) noted that there were nearly 3 times more first-year
teachers in 2007–2008 compared with 1987–1988—roughly 84,000 com-
pared with 239,000.
The recent focus on new teachers has also arisen because of concerns
about the inequitable distribution of inexperienced teachers. Historically
underserved students are assigned to new teachers at higher rates than
White, affluent students (Clotfelter et al., 2005; Goldhaber et al., 2018;
Lankford et al., 2002), a pattern of sorting that is driven by differences
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
3
not only across districts and schools but also within schools (Kalogrides
& Loeb, 2013). This feature of the teacher labor market is generally at-
tributed to how more experienced teachers opt to teach in schools and
classrooms with higher performing students and more positive working
conditions, leaving less “desirable” vacancies to be filled by new teach-
ers. Although evidence of these gaps is important, it is also valuable to
understand how the school composition has changed over time, changing
the context in which new teachers begin their careers. In addition, it has
recently become less clear the extent to which the teacher experience gap
is related to the systematic sorting of the least experienced teachers to
underserved students, or new teachers’ preference to work in schools that
enroll children from low-income families and traditionally underserved
racial/ethnic groups (Ronfeldt et al., 2016).
Pivotal to beginning teachers’ shifting preferences for where to teach
has been the addition of new pathways into teaching for service-oriented
individuals who want to work in underserved schools. These pathways in-
clude Teach For America (TFA), TNTP’s Teaching Fellows program, and
other alternative certification programs aimed at filling positions in un-
derserved schools, as well as urban teacher residency and graduate pro-
grams focused on urban education. To the extent to which more teachers
now begin their careers in underserved schools, historic patterns of access
to inexperienced teachers may be exacerbated, although factors driving
this pattern would be distinct.
Evidence of such a trend would be concerning for two main reasons.
First, early-career teachers have sharp increases in their effectiveness,
particularly in their first year (Atteberry et al., 2015). When low-income
students or underrepresented minorities are disproportionately assigned
to new teachers, they are more likely to be taught by a teacher who is in
the process of developing classroom management strategies and instruc-
tional routines rather than prepared to deploy ambitious instructional
practices (Grossman, 1990). That the students assigned to new teachers
also tend to be lower performing could also create more challenging
classroom conditions for new teachers (Clotfelter et al., 2005; Kalogrides
& Loeb, 2013). Second, new teachers turn over at high rates. Previous
reports of beginning teacher turnover indicate that between 25% and
30% of teachers turn over after their first year, with an average of 10%
of first-year teachers leaving the profession every year (Boyd, Grossman,
et al., 2008; Smith & Ingersoll, 2004). High levels of new teacher turn-
over can adversely affect student achievement (Henry & Redding, 2020;
Ronfeldt et al., 2013) and also reduce the overall effectiveness of the
teacher workforce when new, inexperienced teachers must continually
be hired (Feng, 2010).
Teachers College Record, 122, 070311 (2020)
4
To better understand trends in the characteristics of new teachers, the
schools in which they teach, and their turnover rates, we draw on nation-
al data from all seven waves of the Schools and Staffing Survey (SASS)
from the 1987–1988 to 2011–2012 school years. We present evidence
on the ways in which the characteristics of new teachers, particularly
their education and credentials, and the schools in which they teach
have changed substantially over time and how these characteristics vary
systematically across schools with high levels and low levels of economic
disadvantage. We also examine the extent to which new teachers are
more likely to turn over from underserved schools, and how organiza-
tional supports, such as administrative support and induction programs,
may differentially influence teachers leaving the teaching profession or
switching schools.
LITERATURE REVIEW
Recent evidence from the SASS point to some notable changes in the
composition of the teacher workforce (Ingersoll & Merrill, 2017). For in-
stance, fewer teachers identified as White in the 2011–2012 school year
than in the 1987–1988 school year (82.7% vs. 87.6%), a change driven in
large part by an increased share of Hispanic teachers. Fewer teachers iden-
tified as female in 2012 than in 1988 (71.4% vs. 76.1%). Existing evidence
does not indicate, however, the degree to which these changes have been
influenced primarily by changing demographics of new teachers or differ-
ential exit from teaching among certain groups—a limitation addressed
in the current study’s focus on beginning teachers.
Ingersoll and Merrill’s (2017) work also illustrates how the majority of
teachers now begin their careers in schools with substantially different
student bodies. For instance, in 1987, 60% of teachers worked in schools
with less than a third of students living in poverty, compared with only
33% in 2012. While the overall changes in the student population have
changed the context of teaching for many teachers, we contend that these
changes are most consequential for new teachers, who have historically
been most likely to work in underserved schools (Lankford et al., 2002).
In the remaining sections of the literature review, we elaborate on why
new teachers are more likely to begin their career in underserved schools,
how the characteristics of new teachers in underserved schools differ from
those of new teachers in more affluent schools, and how more challenging
working conditions in these schools tend to exacerbate turnover, possibly
for beginning teachers.
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
5
WHY ARE NEW TEACHERS MORE LIKELY TO BEGIN THEIR CAREERS
IN UNDERSERVED SCHOOLS?
Explanations as to why beginning teachers are most likely to teach tradi-
tionally underserved student subgroups relate to patterns of teacher sort-
ing, the recruitment and hiring process, and preferences of new teachers.
Within teachers’ universal pay scale, it is argued that better working condi-
tions (i.e., smaller class sizes, better facilities and resources, teacher collegi-
ality, effective administrators, etc.) function as a nonpecuniary character-
istic that more experienced teachers seek out (Horng, 2009; Lankford et
al., 2002). In other words, because teachers generally receive no additional
compensation to work in schools with difficult working conditions—condi-
tions that are most likely to arise in schools that enroll children from low-
income families and traditionally underserved racial/ethnic groups—they
are more likely to transfer from these schools. Some contend that this sort-
ing of more experienced teachers away from underserved schools is exac-
erbated by seniority-based transfer rules, whereby more senior teachers are
given preference when filling vacancies in the district (Moe, 2005); howev-
er, the evidence has been inconsistent (Anzia & Moe, 2014; Koski & Horng,
2007; Moe, 2005). Experienced teachers also use social networks to access
information about vacancies and desirable teaching positions. Because new
teachers often have poor information about the characteristics of schools
and the hiring process (Cannata, 2010; Liu & Johnson, 2006), such informa-
tion asymmetries also facilitate teacher experience gaps.
Differences in the hiring practices in more and less affluent schools
can also channel beginning teachers to less affluent schools. First, schools
that enroll more children from low-income families and traditionally un-
derserved racial/ethnic groups tend to have more vacancies as a result
of more teachers leaving teaching and transferring out of underserved
schools (Engel et al., 2014). These schools also tend to have fewer appli-
cants (Boyd, Lankford et al., 2011). Second, underserved schools tend to
hire candidates later in the summer and even into the school year (Engel,
2012). This pattern can result in schools losing more experienced appli-
cants to schools in more affluent districts, leaving more beginning teach-
ers in underserved schools’ hiring pools (Engel, 2013; Liu & Johnson,
2006). Third, principal preferences also influence where teachers begin
their careers. Research suggests that principals “mix and match” when
filling vacancies, looking to hire experienced teachers and inexperienced
teachers, depending on the composition of their school (Harris et al.,
2010). New teachers are particularly appealing for some principals be-
cause of their enthusiasm, pliability, and the fact they are untenured and
would be easier to remove (Engel, 2013).
Teachers College Record, 122, 070311 (2020)
6
More recent research indicates that aspiring teachers’ student teaching
experience may serve as an additional mechanism that channels qualified
candidates out of underserved schools. Although aspiring teachers who
conduct their student teaching in easier-to-staff schools are more likely
to remain in teaching (Ronfeldt, 2012), they are less likely to begin their
career in underserved schools, a pattern that is strongest for the most
qualified prospective teachers (Krieg et al., 2016). Although this research
helps to explain one mechanism driving qualified candidates away from
underserved schools, these studies do not consider teachers in alternative
certification programs and how they are more likely to work in under-
served schools (Redding & Smith, 2016).
Ronfeldt and colleagues (2016) began to address this issue in their study
of beginning teachers’ preferences to work in schools with historically un-
derserved student populations. The authors measured student teachers’
preferences to work with low-achieving, high-poverty, racially diverse, and
English language learner students before and after their student teach-
ing. They found that teachers’ preferences to teach underserved students
predicted the characteristics of the schools in which individuals com-
pleted their student teaching but not the schools where they began their
careers. The authors suggested that although preferences to work with
underserved students may be important—particularly for teachers of col-
or—they are contingent on larger labor market conditions and the hiring
preferences of principals in underserved schools.
THE CHARACTERISTICS OF NEW TEACHERS IN
UNDERSERVED SCHOOLS
One of the most marked differences in the characteristics of teachers in
underserved schools is their racial diversity. In schools that enrolled fewer
than one third low-income students in the 2011–2012 school year, only
9.2% of teachers identified as non-White or Hispanic; in contrast, 37.1%
of teachers in schools with three-quarters or more low-income students
identified as non-White or Hispanic (Ingersoll & Merrill, 2017). These
patterns may be changing, because the teacher workforce has become
slightly more racially/ethnically diverse in the past decades. Yet, although
the share of Hispanic and Asian teachers has increased, Black teachers
now make up a smaller share of the teacher workforce. It is unclear the
extent to which this trend is driven by Black teachers’ higher attrition
rates or lower rates of entry into teaching, a hypothesis we will consider in
this article.
The teachers in underserved schools also tend to be less qualified
and effective, whether measured by experience, licensure exam score,
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
7
certification, education level, college selectivity, or value-added (Goldhaber
et al., 2018). Yet, recent evidence indicates that the academic ability of
new teachers in the schools with the most economically disadvantaged
students is improving (Boyd, Lankford, et al., 2008; Lankford et al., 2014;
Master et al., 2018). In New York City, Lankford and colleagues found that
the percentage of new teachers to graduate from the most competitive
colleges increased from 13.4% in 1999 to 17% in 2005. The percentage
of new teachers to graduate from a competitive college increased from
24.9% in 1999 to 28% in 2005. Their findings are also notable in that
the improvements in academic ability of beginning teachers were greatest
in schools with the most students living in poverty. They attributed these
changes to state and federal policies, including No Child Left Behind’s
(NCLB) Highly Qualified Teacher (HQT) clause, and the expansion of
alternative certification programs that have brought renewed focus on
the academic ability of new teachers. It remains to be seen the extent to
which this pattern might generalize to a nationally representative sample
of teachers; evidence from earlier cohorts of teachers has shown a gradual
decline in the academic ability of new teachers since the 1960s (Bacolod,
2007; Corcoran et al., 2004; Goldhaber & Walch, 2013).
THE WORKING CONDITIONS FACED BY NEW TEACHERS IN
UNDERSERVED SCHOOLS
To this point, we have generally characterized schools that enroll histori-
cally underserved students based on the characteristics of the students, as
opposed to the conditions of the schools. Previous research indicates that
schools serving more traditionally underserved racial/ethnic groups tend
to have more challenging working conditions, which are predictive of
teacher turnover (Boyd, Grossman, et al., 2011; Grissom, 2011; Ingersoll,
2001; Kraft et al., 2016; Ladd, 2011). In particular, schools with more sup-
portive administrators and staff collegiality and fewer student discipline
problems have been shown to predict lower teacher turnover (Redding &
Henry, 2019; Simon & Johnson, 2015).
With more difficult working conditions in underserved schools, orga-
nizational supports, such as induction programs and common planning
time, can play an important role in reducing turnover for beginning teach-
ers in underserved schools (Bettini & Park, in press). Beginning teachers
working in high-poverty schools have reported needing additional profes-
sional development in classroom management and instructional strategies
(Anderson & Olsen, 2006), but whether they are more likely to receive the
supports they need, and the quality of these supports, is unclear. Because
previous studies have generally found participation in induction programs
Teachers College Record, 122, 070311 (2020)
8
to be associated with lower teacher turnover (Ingersoll & Strong, 2011;
Redding & Henry, 2019; Ronfeldt & McQueen, 2017; Smith & Ingersoll,
2004), evidence on how these supports are distributed across schools, as
well as how their availability has increased over time, is important.
CONTRIBUTION AND RESEARCH QUESTIONS
The chief contribution of this study comes from our examination of long-
term trends in the characteristics of a nationally representative sample
of beginning teachers. Previous research has examined changes in the
demographic characteristics or academic ability among all teachers
(Ingersoll et al., 2014), recent college graduates (Master et al. 2018), or
teachers in a specific district or state (Goldhaber et al., 2018; Lankford
et al., 2014). Focusing specifically on a nationwide sample of beginning
teachers allows us to explore changes in who is entering teaching, changes
that have implications for students and school systems. Embedded in this
broad analysis is the recognition that new teachers are increasingly likely
to begin their careers in underserved schools. Because school conditions
have a strong bearing on new teachers’ decisions to remain in their school
or the teaching profession, we believe that it is of critical importance to
document not only how the characteristics of new teachers’ schools have
changed over time, but also how various organizational supports have also
shifted as a way to improve new teacher retention, particularly in under-
served schools.
With this background, the following research questions motivate
this study:
(1) To what extent have the background characteristics of beginning
teachers changed between 1987 and 2012? To what extent have
the characteristics of the schools in which they teach changed over
this period?
(2) How has the distribution of new teachers in high-minority and high-
poverty schools changed over time?
(3) To what extent do the characteristics differ for teachers who begin
their careers in high- versus low-poverty schools, or high- versus low-
minority schools?
(4) To what extent are beginning teachers more likely to turn over than
more experienced teachers? To what extent are new teachers more
likely to turn over from underserved schools? How has this relation-
ship changed over time?
(5) To what extent are organizational supports associated with lower
new teacher turnover?
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
9
DATA
We used data from the SASS and its supplement, the Teacher Follow-Up
Survey (TFS). SASS, administered by the National Center for Educational
Statistics, consists of nationally representative samples of schools, princi-
pals, and teachers for public schools. These surveys include comprehen-
sive data on teacher characteristics, teacher preparation, organizational
supports, and school characteristics, making it an invaluable data source
for describing differences among first-time teachers and the factors as-
sociated with turnover. For this study, we used all seven iterations of
SASS and the last four iterations of TFS where teacher turnover data for
all teachers in the SASS base years can be generated. More specifically,
we use the 1987–1988, 1991–1992, 1993–1994, 1999–2000, 2003–2004,
2007–2008, and 2011–2012 SASS waves for the descriptive analyses, and
the 1999–2000, 2003–2004, 2007–2008, and 2011–2012 TFS waves for
the turnover analyses. We employ sampling weights to make the results
nationally representative.1
Although the core of the SASS has remained fairly consistent over time,
there have been changes to the questions that were asked. For instance,
the percent of students with individualized education plans (IEPs) was
not on the survey until the 1999–2000 wave. Likewise, the majority of
the variables in teacher preparation and support are not available in the
earlier waves. Consequently, not all variables are available for all years.
Fortunately, all the variables of interest do exist and are consistent for the
last four waves of SASS where turnover data can be generated. As such,
the sample sizes for different analyses vary. The overall sample size is more
than 264,000 observations. The sample size with turnover data is 145,780
observations, with 8,200 beginning teacher observations.
MEASURES OF ATTRITION
We provide a more complete definition of the variables used in this study
in Table 1 in the supplemental materials (see https://tuan-d-nguyen.
github.io/TCR_newteachers_Supplemental%20Materials_ce.pdf). In the
next two sections, we provide an overview of these variables. The main de-
pendent variable for this study comes from the principal report of teach-
ers’ employment status in the follow-up year following the baseline survey
year. We categorize teacher status into one of three categories: stayers,
movers, and leavers. Stayers are teachers who remained in the same school
in the baseline year, movers are teachers who moved to a new school, and
leavers are teachers who left the teaching profession.
Teachers College Record, 122, 070311 (2020)
10
MEASURES OF TEACHER CHARACTERISTICS, PREPARATION AND
SUPPORT, AND SCHOOL CHARACTERISTICS
We include a number of teacher characteristics such as gender; race/eth-
nicity; graduate degree(s); whether the teacher teaches math, science, or
special education; certification status; undergraduate college selectivity
using Barron’s Admissions Competitiveness Index; and reported annual
salary. College selectivity has been used in previous studies as a proxy for
academic ability or to indicate high-quality teaching candidates (Cohen-
Vogel & Smith, 2007; Lankford et al., 2014).2 Additionally, we also con-
sider beginning teachers’ preparation and support that can influence
their experience in the classroom and turnover decisions. We include
indicators of whether the teacher has practice teaching before teaching
and their report of how prepared they felt on classroom management,
instructional method, subject content matter, technology use, and student
assessment. Previous works have also shown the importance of additional
supports for beginning teachers. As such, we also include binary indica-
tors of whether a new teacher participated in an induction program, had a
reduced teaching schedule, had common planning with teachers in their
subject, received supportive communication from the principals or other
administrators, and whether they had a mentor teacher.
In terms of school context, we considered several important character-
istics: school’s urbanicity, enrollment size, secondary or elementary level,
the percent of students with free and reduced-price lunch (FRPL) eligibil-
ity, percent minority, percent IEP, and percent limited English proficiency.
We also characterized school working conditions with principal reports of
the level of student disciplinary problems, and teacher-reported support
from the administrators and cooperative effort among the staff.
Because of our interest in new teachers who begin their careers in un-
derserved schools, we followed Ingersoll (2001) in operationalizing un-
derserved schools in two different ways: (1) with a majority-minority indi-
cator for teachers in schools with 50% or more non-White students, and
(2) with a majority FRPL indicator for teachers in schools with 50% or more
students eligible for FRPL.3 We hope to capture how the distributions of
these two indicators change over time, which is important for all teachers,
but particularly so for new teachers. As an alternative, we compare teachers
in the lowest and highest quartile for percent FRPL and percent minority.
The results using this alternate operationalization are generally larger and
substantively similar than the majority indicator method, although some-
times less significant, given the reduction in sample size. Our preference is
to use the majority indicator approach because it provides an absolute level
of comparison that is easy to interpret across waves, compared with the low-
est–highest comparison that changes from wave to wave.
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
11
METHODS
Our analysis consists of two main parts: (1) a description of the charac-
teristics of new teachers and the schools in which they teach, and (2) an
analysis of first-year teacher turnover. We report descriptive differences in
the characteristics of new teachers, their preparation and organizational
supports, and characteristics of the schools in which they teach. We then
compare teacher and school characteristics for those in a school with
50% or more FRPL or minority students with those with less than 50%.
Building from this analysis, we then test the extent to which new teachers
are more likely than more experienced teachers to begin their career in
schools with a majority of FRPL or minority students. To test this hypoth-
esis, we estimate the following model:
(1)
where working in a school with a majority of either minority or FRPL stu-
dents is a function of being new teacher i in survey year t and the interac-
tion of these two variables, a state fixed effect (
γs)
and an error term (
eit
).
In this model,
β1
represents the difference in the probability of working
in a majority school for a new teacher compared with a more experienced
teacher;
β2
captures changes in the prevalence of majority schools over
time; and
β3
allows for the probability of a new teacher working in a ma-
jority school to change across survey years. For ease of interpretation, this
model is estimated as a linear probability model. Moreover, because most
of the predicted values fall in the center of the data, they are less likely to
be biased in a linear model than a logistic regression model.
We employ state fixed effects to account for unobserved heterogeneity
across states that may be associated with teachers sorting into majority-
FRPL or -minority schools. However, there are limitations to the use of
state fixed effects because they do not account for important unobserved
school or district characteristics that may influence where teachers work.
There are, however, drawbacks to the use of either school or state fixed
effects. First, the use of school and district fixed effects reduces the gener-
alizability of the findings because the analysis is conducted on a restricted
sample. Second, the sampling design does not guarantee that the same
schools or even districts will appear from wave to wave. Third, the use of
school fixed effects will then exclude any time-invariant observable school
characteristics that are relevant in subsequent models. As such, we used
state fixed effects in the models because the SASS sampling design allows
for state representativeness at the state level. Moreover, this allowed us to
interpret our results as within state results.
Teachers College Record, 122, 070311 (2020)
12
The second part of our analysis focuses on new teacher turnover. First,
we compared the turnover (i.e., moving schools or leaving teaching) of
new teachers to that of more experienced teachers. This multinomial lo-
gistic regression model can be expressed as:
(2)
where the probability of moving schools (k = 1) or leaving teaching (k = 2)
for teacher i from school j in year t is a function of being a new teacher
(new teacheri)a vector of teacher background characteristics (Ti), a vector
of school characteristics (
Si )
a state fixed effect (
γs
)and a year fixed ef-
fect (
λt
). To test the extent to which new teachers are more likely to turn
over from high-poverty and majority-minority schools, we extended this
model by interacting the new teacher indicator with the majority school
indicators. We included the year fixed effects to account for time-specific
heterogeneity in the data.
To examine the extent to which organizational supports are predictive
of turnover, we limited the sample to beginning teachers and estimated
the same model, adding a vector of organizational supports. In addition to
focusing our analysis on the focal group of this article, this model provides
a straightforward analysis of the relationship between organizational sup-
ports and teacher turnover.
RESULTS
THE CHARACTERISTICS OF NEW TEACHERS
Our first research question asked about the extent to which the character-
istics of new teachers or the schools in which they first taught changed be-
tween 1987 and 2012. Table 1 reports on the characteristics of seven cohorts
of new teachers and the schools in which they work. The gender composi-
tion has changed slightly, with 23% of new teachers identifying as male in
1988, compared with 26% in 2012. New teachers are less likely to be White
(87% in 1988 vs. 80% in 2012), although there has been little change in
the share of teachers of color since 2000. Overall, incoming teachers are
slightly more likely to identify as a person of color (20%) compared with
the full teacher workforce (17%; Ingersoll & Merrill, 2017). The increased
diversity among new teachers has been driven almost entirely by the steady
increase in the share of new teachers identifying as Hispanic.
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
13
Table 1. Descriptive Statistics of First-Time Teachers
(1) (2) (3) (4) (5) (6) (7)
Wave:
1988
Wave:
1991
Wave:
1994
Wave:
2000
Wave:
2004
Wave:
2008
Wave:
2012
Teacher Characteristics
Female 0.77 0.77 0.73 0.73 0.74 0.75 0.74
Black 0.06 0.07 0.07 0.10 0.10 0.09 0.07
Asian 0.02 0.01 0.01 0.02 0.02 0.02 0.02
American Indian 0.01 0.01 0.01 0.01 0.01 0.02 0.01
Hispanic 0.04 0.07 0.08 0.07 0.09 0.10 0.11
White 0.87 0.84 0.83 0.80 0.79 0.79 0.80
Under 30 0.68 0.59 0.60 0.69 0.64 0.67 0.72
Graduate degree 0.09 0.11 0.13 0.16 0.17 0.21 0.22
Teaches STEM subjects 0.11 0.09 0.13 0.17 0.17 0.15 0.20
Teaches SPED 0.04 0.07 0.05 0.10 0.13 0.11 0.13
No certification 0.08 0.10 0.11 0.20 0.06 0.06 0.04
Most selective college 0.03 . 0.05 0.05 0.12 0.11 0.11
Very selective college 0.09 . 0.18 0.15 0.22 0.19 0.24
Salary ($1,000) 34.6 34.8 34.9 37.2 38.8 38.9 37.3
Satisfy w/ salary (std) 0.33 . 0.19 0.22 0.31 0.24 0.13
Union member . . 0.61 0.68 0.67 0.65 0.52
Preparation and Support
Practice teaching . . . 0.88 0.81 0.81 0.90
Preparedness (std) . . . 0.10 0.07 0.04 0.18
1st yr-induction prog . . . 0.67 0.71 0.79 0.83
1st yr-reduced sched . . . 0.09 0.09 0.15 0.11
1st yr-common planning . . . 0.48 0.57 0.62 0.60
1st yr-supportive comm . . . 0.81 0.83 0.87 0.81
1st yr-mentor . 0.57 0.59 0.70 0.75 0.84 0.74
School Characteristics
Urban school 0.26 0.27 0.26 0.27 0.32 0.27 0.28
Northeast region 0.14 0.13 0.15 0.18 0.19 0.14 0.13
Midwest region 0.24 0.23 0.22 0.24 0.21 0.20 0.20
South region 0.41 0.42 0.47 0.38 0.44 0.47 0.48
West region 0.21 0.23 0.16 0.21 0.17 0.19 0.19
Teachers College Record, 122, 070311 (2020)
14
(1) (2) (3) (4) (5) (6) (7)
Wave:
1988
Wave:
1991
Wave:
1994
Wave:
2000
Wave:
2004
Wave:
2008
Wave:
2012
K–12 enrollment 705 673 760 812 850 812 811
Secondary school 0.33 0.30 0.34 0.34 0.32 0.31 0.32
Combined elem-sec 0.07 0.04 0.04 0.03 0.06 0.06 0.10
Percent FRPL students 0.32 0.37 0.35 0.37 0.46 0.46 0.52
Majority FRPL 0.20 0.29 0.28 0.33 0.44 0.44 0.53
Percent minority students 0.29 0.33 0.35 0.39 0.45 0.46 0.47
Majority minority 0.24 0.29 0.31 0.36 0.42 0.43 0.44
Percent IEP . . . 0.12 0.13 0.14 0.13
Percent LEP . . . 0.07 0.08 0.09 0.08
Student discip (std) 0.02 -0.06 -0.13 -0.12 -0.01 -0.02 -0.02
Admin support (std) 0.17 . 0.23 0.23 0.21 0.17 0.24
Teacher coop (std) 0.05 . 0.14 0.08 0.13 0.08 0.13
Observations 1,000 1,740 2,080 2,260 2,000 2,140 1,800
Note. Nationally representative weights are employed. Sample sizes weighted to the
nearest 10 in accordance with NCES nondisclosure rule. FRPL = free and reduced-
price lunch; IEP = individualized education plan; LEP = limited English proficient
We found increases in the education levels, certification, and college se-
lectivity of entering teachers. New teachers are increasingly likely to have a
graduate degree (9% in 1988 vs. 22% in 2012). They are also less likely to
be uncertified (20% in 2000 vs. 4% in 2012). As part of the NCLB’s HQT
clause, which mandated that students be taught by a certified teacher, the
timing of this decrease mirrors the implementation of NCLB; this suggests
that this legislation was, in part, linked to fewer uncertified teachers enter-
ing teaching. We also report promising trends in the improved academic
ability of new teachers, as measured by college selectivity. In 1988, only
12% of new teachers graduated from a most or very selective college. By
2012, the percentage of entering teachers who graduated from a most or
very selective college had nearly tripled, increasing to 35%.
Other notable changes relate to beginning teachers’ salaries and their
rates of unionization. Adjusting for inflation, new teacher salaries in-
creased by $2,700 between 1988 and 2012, although salaries did not in-
crease from 2000 compared with 2012—a trend that corresponds to a
larger pattern of stagnating wages for middle-class families (Mishel et al.,
2015). Slightly more than half of new teachers were union members in
Table 1. Descriptive Statistics of First-Time Teachers (continued)
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
15
2012, a decrease from 61% in 1994. We attribute this decrease to the re-
gional sorting of new teachers—with 48% of new teachers working in the
South in 2012, where teacher collective bargaining is more likely to be
restricted (Swain & Redding, in press)—but it may also have resulted from
new teachers opting to work in nonunionized charter schools. It is also
worth emphasizing that this decrease in union membership mirrors the
broader trend of decreasing public-sector union membership (Dunn &
Walker, 2016).
We see that beginning teachers are more likely to receive additional
induction supports over time. In 1994, 57% of new teachers reported
working with a mentor their first year. By 2008, 84% of new teachers said
that they worked with a mentor, but this number decreased to 74% in
2012. Entering teachers in 2012 were also more likely to participate in an
induction program than in previous years. A total of 83% of new teach-
ers participated in an induction program in 2012, compared with 67% in
2000. Finally, a larger share of new teachers reported meeting for com-
mon planning time with other teachers in their grade or subject (48% in
2000 compared with 60% in 2012).
THE CHARACTERISTICS OF NEW TEACHERS’ SCHOOLS
Our second research question asked about the characteristics of the
schools in which new teachers began their careers, as well as the degree
to which new teachers became increasingly likely to work in underserved
schools over time. New teachers now begin their careers in schools that
enroll much different student populations than in 1988. In 1988, in the
schools in which new teachers worked, 32% of enrolled students were eli-
gible for FRPL, and 29% identified as a racial/ethnic minority. Further,
only a fifth of beginning teachers worked in a school where the majority of
students were enrolled in FRPL. By 2012, more than half of new teachers
worked in a school where the majority of students were enrolled in FRPL.
In 2012, in the schools in which new teachers worked, 52% of enrolled
students were eligible for FRPL, and 47% identified as a racial/ethnic mi-
nority. Next, we test the extent to which this trend is driven by changes in
the demographic makeup of American schools or new teachers’ increased
probability of working in schools with a majority of students enrolled in
FRPL or identifying as racial/ethnic minorities.
In Figure 1, we plot the linear predictions from Model 1. This figure
makes clear that both new and more experienced teachers are more likely
to work in schools enrolling a majority of FRPL students and racial/eth-
nic minorities. Notably, first-year teachers became significantly more likely
than more experienced teachers to work in racially concentrated schools
Teachers College Record, 122, 070311 (2020)
16
in 2000, and in majority-FRPL schools in 2004. From 2000 to 2008, we
found a 7-percentage-point difference in the probability of new teachers
working in a majority-minority school compared with more experienced
teachers. We found a similar gap in 2004 and 2008 for schools with a ma-
jority of FRPL students. In 2012, the first survey after the Great Recession,
the probability that a new teachers would work in majority-FRPL or -mi-
nority school was still 4 percentage points greater than for more experi-
enced teachers, but the difference was no longer statistically significant.4,5
Figure 1. Linear predictions of employment in a school enrolling a
majority of free and/or reduced-price lunch students or racial/ethnic
minorities
Notes. Linear predictions obtained from a regression model predicting a new or
more experienced teacher would work in a school enrolling a majority free and/
or reduced price lunch (FRPL) students or racial/ethnic minorities, conditioning
on state fixed effects. Regression results reported in Table 2.
With evidence of changes in the characteristics of new teachers and a
distinct likelihood of working in schools that enroll a majority minority or
FRPL student populations, in the next section, we examine differences in
teacher and school characteristics for new teachers employed at schools
with less than 50% racial/ethnic minorities (“low minority”) and more
than or equal to 50% (“high minority”).
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
17
COMPARING THE CHARACTERISTICS OF NEW TEACHERS WORKING
IN HIGH- VERSUS LOW-MINORITY SCHOOLS
With evidence that the characteristics of beginning teachers and the
schools in which they teach have changed over time, it is important to
understand specifically how the characteristics of new teachers in high-
and low-minority schools have departed or converged over time. To an-
swer our third research question, we present the characteristics of new
teachers and their schools in the 1988 and 2012 school years in Table 2.
The most notable patterns in Table 2 are in terms of certification status
and education. From 1988 to 2000, new teachers in high-minority schools
were more likely to be uncertified than those in low-minority schools. This
trend peaked in 2000, when 30% of new teachers in high-minority schools
were uncertified, compared with 14% in low-minority schools. By 2012,
there were no significant differences in certification. In other words, over-
all improvements in the certification rates of incoming teachers tended to
be concentrated in high-minority schools. Increases in the education level
were more equally dispersed across low- and high-minority schools; as the
fraction of new teachers with a graduate degree increased, they were just
as likely to have begun their career in a low- versus high-minority school.
We found no significant differences in terms of college selectivity across
low- and high-minority schools.
Table 2. Descriptive Statistics of First-Time Teachers by Majority
Minority School Status
Wave: 1988 Wave: 2012
Low
Min.
High
Min.
Diff Low
Min.
High
Min.
Diff
Teacher
Characteristics
Female 0.77 0.77 -0.01 0.72 0.77 0.05*
Black 0.03 0.12 0.09* 0.03 0.12 0.09**
Asian 0.01 0.05 0.05* 0.01 0.03 0.02
American Indian 0.01 0.01 0.00 0.01 0.01 0.01
Hispanic 0.02 0.13 0.11** 0.03 0.20 0.17**
White 0.93 0.69 -0.24** 0.93 0.64 -0.3**
Under 30 0.69 0.65 -0.04 0.75 0.67 -0.08
Graduate degree 0.07 0.13 0.06+ 0.21 0.24 0.03
Teaches STEM 0.11 0.12 0.02 0.17 0.23 0.06*
Teaches special ed 0.04 0.04 0.00 0.13 0.12 -0.01
Teachers College Record, 122, 070311 (2020)
18
Wave: 1988 Wave: 2012
Low
Min.
High
Min.
Diff Low
Min.
High
Min.
Diff
No certification .07 .12 .05+ 0.02 0.05 0.03
Most sel. college 0.03 0.03 0.00 0.09 0.13 0.05
Very sel. college 0.09 0.10 0.01 0.24 0.25 0.02
Salary ($1,000) 33.37 38.31 4.94** 35.18 39.97 4.79**
Satisf. w/
salary (std)
0.30 0.41 0.1 0.16 0.08 -0.08
Union member . . . 0.52 0.52 0.00
Preparation
and Support
Practice teaching . . . 0.93 0.87 -0.06**
Preparedness (std) . . . 0.26 0.08 -0.17*
1st yr
induction program
. . . 0.84 0.83 -0.01
1st yr
reduced schedule
. . . 0.11 0.11 -0.01
1st yr
common planning
. . . 0.58 0.62 0.03
1st yr
supportive comm
. . . 0.87 0.75 -0.12**
1st yr mentor . . . 0.76 0.70 -0.06+
School
Characteristics
Urban school 0.19 0.49 0.3** 0.14 0.45 0.31**
Northeast region 0.13 0.20 0.07 0.13 0.13 0.00
Midwest region 0.28 0.09 -0.19** 0.28 0.11 -0.18**
South region 0.40 0.44 0.03 0.43 0.53 0.1
West region 0.19 0.27 0.09 0.16 0.23 0.07
K–12 enrollment 651 871 220* 744 895 150*
Secondary school 0.32 0.34 0.02 0.35 0.29 -0.06
Combined elem-sec 0.07 0.05 -0.02 0.09 0.10 0.01
Percent FRPL 0.25 0.55 0.3** 0.38 0.70 0.31**
Majority FRPL 0.10 0.54 0.44** 0.33 0.79 0.46**
Percent minority 0.14 0.78 0.64** 0.21 0.81 0.6**
Table 2. Descriptive Statistics of First-Time Teachers by Majority
Minority School Status (continued)
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
19
Wave: 1988 Wave: 2012
Low
Min.
High
Min.
Diff Low
Min.
High
Min.
Diff
Percent IEP . . . 0.13 0.13 0.00
Percent LEP . . . 0.03 0.15 0.12**
Student discip (std) -0.09 0.35 0.44** -0.11 0.09 0.2*
Admin
support (std)
0.22 0.03 -0.18* 0.36 0.08 -0.28**
Teacher coop (std) 0.12 -0.15 -0.27* 0.24 -0.03 -0.27*
Observations 1,000 1,800
Note. Nationally representative weights are employed. Sample sizes weighted to the
nearest 10 in accordance with NCES nondisclosure rule. FRPL = free and reduced-
price lunch; IEP = individualized education plan; LEP = limited English proficient.
+p < .10. *p < .05. **p < .01.
New teachers in high-minority schools are less likely to have practiced
teaching before entering and report feeling less prepared to enter the
classroom. In addition, they report receiving less supportive communica-
tion from their administrators and are less likely to be assigned a mentor.
With fewer supports, they also must negotiate more challenging working
conditions than teachers who begin their career in schools with less ra-
cial/ethnic diversity. Compared with those in low-minority schools, new
teachers in high-minority schools face more student discipline problems,
less administrative support, and less teacher cooperation, and these differ-
ences have remained consistent over time.
NEW TEACHER TURNOVER
Thus far in our analysis, we have found that new teachers, relative to more
experienced teachers, have become more likely to work in high-minority
schools, even when accounting for trends in demographic changes in pub-
lic school students. At the same time, the availability of induction sup-
ports has increased for new teachers. Still, new teachers in high-minority
schools still fall behind their colleagues in less racially diverse schools in
terms of assignment to a mentor and receipt of regular, supportive com-
munication from an administrator. The final set of analyses considers how
some of these broad trends in the composition of the beginning teacher
workforce relate to new teacher turnover (our fourth research question).
In Table 3, we show the turnover rates have been relatively stable, even
decreasing slightly in 2008 and 2012 compared with the early 2000s. Still,
Teachers College Record, 122, 070311 (2020)
20
new teachers turn over at quite high rates. Across these four survey waves,
an average of 26% of new teachers turned over each year, compared with
14% of more experienced teachers. In addition, these turnover rates were
not equally distributed across schools. In Table 7 in the supplemental ma-
terials (link provided in the Data section), we show that 28% of new teach-
ers in high-minority schools turned over annually, compared with 24% of
new teachers in less racially diverse schools.
Table 3. Rate of Attrition for First-Time and Experienced Teachers
Teacher status (1) (2) (3) (4) (5)
Wave: 2000 Wave: 2004 Wave: 2008 Wave: 2012 Pooled
Panel A: More experienced teachers
Stayer 85.60 84.71 86.78 86.53 85.92
Mover 6.93 7.60 6.58 6.30 6.85
Leaver 7.47 7.69 6.64 7.17 7.24
Observations 36,910 37,240 31,270 32,170 137,580
Panel B: First-time teachers
Stayer 73.70 73.58 74.93 75.97 74.47
Mover 15.14 14.13 14.50 13.71 14.40
Leaver 11.17 12.29 10.57 10.32 11.13
Observations 2,260 2,000 2,140 1,800 8,200
Note. Nationally representative weights are employed. Sample sizes weighted to the
nearest 10 in accordance with NCES nondisclosure rule. Stayers are teachers who
remain in the school where they taught in the previous year. Movers are teachers
who remain in teaching but have moved to a different school. Leavers are teachers
who left teaching altogether.
Next, we examine the extent to which new teachers are more likely than
more experienced teachers to move schools or leave teaching.6 Table 4
reports the coefficients from a series of multinomial logistic regression
models, controlling for teacher background characteristics, school char-
acteristics, and state and year fixed effects. New teachers are consistently
more likely to move schools and leave teaching compared with more ex-
perienced teachers, holding all else constant. To test the extent to which
new teachers are more or less likely to turn over from schools that serve
high concentrations of low-income students or racial/ethnic minorities,
Columns 2 and 4 include interactions between the new teacher indica-
tor and the majority-FRPL/minority indicator. For both moving and leav-
ing, we found no evidence that new teachers move from underserved
schools or leave teaching at differentially higher rates than more experi-
enced teachers.
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
21
Table 4. Multinomial Logit of Teacher Turnover and Majority FRPL/
Minority Status
(1) (2) (3) (4)
Majority
FRPL
Majority
FRPL with
interaction
Majority
minority
Majority
minority with
interaction
Panel A: Mover
New teacher 0.344** 0.355** 0.348** 0.428**
(0.056) (0.069) (0.056) (0.067)
Majority indicator 0.044 0.046 0.175** 0.193**
(0.059) (0.058) (0.039) (0.040)
New teacher* -0.022 -0.191
Interaction (0.108) (0.121)
Panel B: Leaver
New teacher 0.295** 0.264** 0.302** 0.253**
(0.075) (0.077) (0.075) (0.080)
Majority indicator -0.012 -0.016 0.304** 0.296**
(0.033) (0.034) (0.041) (0.044)
New teacher* 0.066 0.096
Interaction (0.077) (0.105)
Observations 145,780 145,780 145,780 145,780
Note. Nationally representative weights are employed. Sample sizes weighted to the
nearest 10 in accordance with NCES nondisclosure rule. Heteroskedastic-robust
state-level clustered standard errors are in parentheses. All models control for
teacher race/ethnicity, gender, age under 30, graduate degree, subjects taught,
certification, salary, urbanicity, school enrollment, and school level, along with
state and year fixed effects. Models 1 and 2 also control for percent minority, and
Models 3 and 4 control for percent FRPL. FRPL = free and reduced price lunch.
+ p < .10. * p < .05. ** p < .01.
With overall evidence of higher turnover rates for first-year teachers, we
now examine the teacher and school characteristics and organizational
supports that are predictive of moving schools and leaving teaching. Table
5 reports the results from a series of multinomial logistic regression with
a sample restricted to new teachers. The marginal effects are reported
in Table 9 in the supplemental materials (link provided in the Data sec-
tion). When interpreting our results, we focus on the ways that teacher
education or organizational supports are associated with a reduced risk
of turnover.
Teachers College Record, 122, 070311 (2020)
22
Table 5. Multinomial Logit of New Teacher Turnover
Moving Schools (vs. Staying) Leaving Teaching (vs. Staying)
(1) (2) (3) (4) (5) (6)
Female -0.150 -0.152 -0.150 0.001 -0.000 -0.002
(0.117) (0.117) (0.116) (0.075) (0.074) (0.075)
Black 0.156 0.161 0.155 -0.226 -0.220 -0.183
(0.174) (0.174) (0.166) (0.233) (0.231) (0.224)
Asian 0.158 0.165 0.157 0.000 0.004 0.017
(0.464) (0.466) (0.460) (0.263) (0.261) (0.261)
American
Indian
-1.112** -1.116** -1.113** -0.126 -0.128 -0.094
(0.414) (0.412) (0.412) (0.415) (0.413) (0.412)
Hispanic -0.091 -0.087 -0.093 0.260 0.261 0.297
(0.139) (0.139) (0.141) (0.200) (0.200) (0.209)
Age under 30 0.197 0.207 0.196 -0.154 -0.148 -0.140
(0.154) (0.155) (0.155) (0.151) (0.149) (0.149)
Grad degree 0.392** 0.401** 0.390** 0.333*0.338*0.340*
(0.124) (0.126) (0.125) (0.159) (0.159) (0.157)
STEM subjects -0.176 -0.177 -0.176 0.068 0.069 0.076
(0.177) (0.177) (0.178) (0.146) (0.147) (0.146)
Special
education
0.364*0.370*0.364*0.486+0.491+0.491+
(0.183) (0.183) (0.183) (0.271) (0.269) (0.269)
No certification -0.240 -0.247 -0.238 0.412*0.409*0.419*
(0.157) (0.158) (0.157) (0.202) (0.200) (0.194)
Salary ($1,000) -0.023** -0.022** -0.023** -0.048** -0.047** -0.047**
(0.007) (0.007) (0.007) (0.007) (0.007) (0.007)
Satisfy with
salary
-0.092 -0.092 -0.092 0.082 0.083 0.083
(0.063) (0.062) (0.063) (0.056) (0.056) (0.056)
Union -0.166+-0.171+-0.167+-0.510** -0.513** -0.505**
(0.094) (0.093) (0.094) (0.124) (0.125) (0.124)
Urban -0.192 -0.195 -0.196 -0.120 -0.122 -0.072
(0.130) (0.130) (0.126) (0.118) (0.117) (0.125)
School
enrollment
-0.254*-0.242*-0.253*-0.017 -0.009 -0.020
(0.112) (0.109) (0.112) (0.128) (0.122) (0.129)
Secondary -0.215 -0.203 -0.215 -0.003 0.009 0.002
(0.136) (0.138) (0.135) (0.170) (0.175) (0.173)
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
23
Moving Schools (vs. Staying) Leaving Teaching (vs. Staying)
(1) (2) (3) (4) (5) (6)
Combined
elem-sec
-0.643** -0.654** -0.642** -0.076 -0.083 -0.080
(0.187) (0.186) (0.188) (0.164) (0.167) (0.165)
Percent FRPL 0.153 0.158 0.071 0.123
(0.239) (0.230) (0.296) (0.283)
Percent
minority
-0.067 -0.173 0.834** 0.751**
(0.247) (0.266) (0.229) (0.196)
Majority FRPL 0.228+0.149
(0.138) (0.149)
Majority
minority
-0.053 0.513**
(0.135) (0.134)
Percent IEP 0.187 0.190 0.185 0.340 0.336 0.355
(0.337) (0.333) (0.338) (0.382) (0.402) (0.375)
Percent LEP 0.123 0.083 0.126 -0.884** -0.917** -0.813*
(0.506) (0.491) (0.533) (0.342) (0.350) (0.349)
Most selective -0.263 -0.254 -0.265 -0.054 -0.046 -0.032
(0.302) (0.299) (0.299) (0.196) (0.199) (0.202)
Ver y selective -0.040 -0.034 -0.042 0.052 0.057 0.070
(0.201) (0.202) (0.199) (0.149) (0.149) (0.154)
Barron missing 0.189 0.195 0.186 0.149 0.155 0.161
(0.204) (0.204) (0.208) (0.159) (0.160) (0.163)
Student
disc problem
0.024 0.023 0.024 0.088+0.087+0.090+
(0.052) (0.052) (0.052) (0.051) (0.052) (0.049)
Admin support -0.137+-0.137+-0.137+-0.258** -0.257** -0.256**
(0.076) (0.076) (0.076) (0.056) (0.057) (0.057)
Teacher coop -0.118+-0.118+-0.118+-0.185** -0.185** -0.186**
(0.070) (0.069) (0.070) (0.053) (0.054) (0.054)
Practice
teaching
-0.067 -0.071 -0.069 -0.289** -0.292** -0.287*
(0.149) (0.149) (0.147) (0.106) (0.105) (0.112)
Preparedness 0.017 0.016 0.017 0.062 0.062 0.060
(0.057) (0.058) (0.057) (0.094) (0.093) (0.093)
1st yr induction -0.205** -0.209** -0.204** -0.250 -0.253 -0.256
(0.078) (0.079) (0.078) (0.163) (0.162) (0.166)
1st yr re-
duc. schedule
0.480** 0.485** 0.479** 0.388*0.391*0.383*
(0.163) (0.163) (0.162) (0.198) (0.198) (0.194)
Teachers College Record, 122, 070311 (2020)
24
Moving Schools (vs. Staying) Leaving Teaching (vs. Staying)
(1) (2) (3) (4) (5) (6)
1st yr com-
mon planning
-0.206*-0.207*-0.206*-0.214 -0.216 -0.209
(0.095) (0.095) (0.093) (0.144) (0.144) (0.143)
1st yr sup-
portive
communication
-0.107 -0.111 -0.107 -0.174 -0.176 -0.173
(0.132) (0.133) (0.132) (0.147) (0.148) (0.149)
1st yr mentor 0.109 0.113 0.107 0.085 0.087 0.089
(0.152) (0.151) (0.153) (0.250) (0.251) (0.252)
_cons 0.139 0.133 0.134 0.379 0.367 0.449
(0.223) (0.220) (0.227) (0.319) (0.340) (0.314)
Note. Estimates presented as multinomial log-odds. Nationally representative
weights are employed. Heteroskedastic-robust state-level clustered standard errors
are in parentheses. All models employ state and year fixed effects. The total sam-
ple size for Tables 5 and 6 is 8,200 teacher observations. FRPL = free and reduced-
price lunch; IEP = individualized education plan; LEP = limited English proficient.
+p < .10. *p < .05. ** p < .01.
New teachers with a graduate degree have greater risk of moving schools
and leaving teaching. When exponentiating the logit coefficients, teach-
ers with a graduate degree have a 48% greater risk of moving schools and
40% greater risk of leaving teaching than teachers without a graduate de-
gree, holding all else constant. We found no evidence that new teachers
who are uncertified are more likely to move schools, but their risk of leav-
ing teaching is 54% greater than certified teachers, which translates to a
3.7% difference in the probability of leaving by certification status. We
found no evidence between college selectivity and either moving schools
or leaving teaching. Higher teacher salaries, however, were associated with
reduced risk of turnover for new teachers. A $1,000 increase in teachers’
starting salary was associated with a 2% decrease in the risk of moving
schools and a 5% decrease in the risk of leaving teaching.
Even when conditioning on teacher and school characteristics, the
percentage of minorities and the indicator of a high-minority school
are both predictive of much higher turnover rates for new teachers.
For instance, the risk of a new teacher leaving teaching after teaching
at a school with the majority of students identifying as a racial/ethnic
minority was 67% greater than for schools with fewer minorities. This
increased risk translates to a rate of leaving teaching that is 4.3 percent-
age points higher in more racially concentrated schools compared with
schools with less racial diversity.7
Table 5. Multinomial Logit of New Teacher Turnover (continued)
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
25
Our findings regarding the relationship between working conditions
and teacher turnover for first-year teachers mirror previous research on
the topic. More positive working conditions are related to lower turnover
rates. We found that a 1-standard-deviation increase in administrative sup-
port is associated with a 23% decrease in the risk of leaving teaching and a
13% decrease in the risk of moving schools, controlling for other variables
in the model. A 1-standard-deviation increase in teacher cooperation is
associated with a 17% decrease in the risk of leaving teaching and an 11%
decrease in the risk of moving schools.
Some organizational supports targeted specifically at new teachers are
also predictive of reduced risks of moving schools. Specifically, participa-
tion in an induction program and shared planning time is associated with
a 19% decrease in the risk of moving schools. We found no evidence that
mentoring is linked with lower turnover, which previous studies of the
2000 and 2008 SASS found to improve retention (Ronfeldt & McQueen,
2017; Smith & Ingersoll, 2004).
DISCUSSION AND CONCLUSION
Examining a nationally representative sample of beginning teachers al-
lows us to explore changes in who is entering teaching between 1987 and
2012, changes that have implications for students and school systems. In
describing these changes, we first outlined some of the broad changes
observed in this study before turning to the ways in which these changes
have differentially shaped the school context for teachers who begin their
careers in underserved schools.
Among demographic characteristics, there has been a mix of stasis and
change in terms of who is becoming a teacher. In 1988, only 4% of teach-
ers identified as Hispanic compared with 11% in 2012. Because only 7.5%
of teachers nationwide identify as Hispanic, our results suggest that the in-
creased share of incoming Hispanic teachers will help improve represen-
tation for Hispanic students over time, a positive shift given evidence of
improved student outcomes for Hispanic students assigned to a Hispanic-
race teacher (Buddin & Zamarro, 2009a, 2009b; Jennings & DiPrete,
2010; Redding, 2019). Yet, for Black teachers, we found no corresponding
evidence of such a shift, even as the share of Black adults who have com-
pleted college more than doubled between 1988 and 2012 (Pew Research
Center, 2013). In terms of gender identity, most incoming teachers still
identify as female (77% in 1988 vs. 74% in 2012).
More notable changes have been in regard to new teachers’ credentials
and education. By 2000, a fifth of all new teachers were uncertified. In the
following years, the rates of uncertified teachers quickly dropped, with
Teachers College Record, 122, 070311 (2020)
26
only 4% of teachers being uncertified in 2012. Although our research de-
sign prevents us from identifying what exactly caused this change, we sus-
pect that the increased certification levels of new teachers resulted from
NCLB requirements that teachers in core academic subjects have full state
certification to comply with the HQT clause. Regardless of the cause, this
shift toward increased teacher certification is promising, given evidence
showing that certification has a positive, albeit small, effect on student
achievement (Clotfelter et al., 2007, 2010; Kane et al., 2008) and evidence
from our study showing that uncertified teachers are predicted to leave
teaching at higher rates than certified teachers.
New teachers are also now more likely to enter teaching having gradu-
ated from a selective college and having earned a graduate degree. In
1988, 9% of new teachers had a graduate degree, and 12% graduated
from a most or very selective college. In 2012, 22% of new teachers had
a graduate degree, and 35% graduated from a most or very selective col-
lege. This finding—that an increased segment of new teachers graduated
from selective colleges—mirrors other work showing improvements in
the academic ability of new teachers (Lankford et al., 2014; Master et al.,
2018), although fewer teachers nationwide have attended a selective col-
lege than those reported in these studies.
New teachers now begin their careers in schools with drastically differ-
ent conditions than in prior years. In 2012, more than half of new teachers
worked in schools where the majority of students are eligible for FRPL, a
remarkable shift from the 1988 school year, when only one 1 in 5 teach-
ers worked in such schools. Relatedly, in 2012, new teachers were almost
twice as likely to work in schools with a majority of minority students as
compared with 1988. To a large degree, this trend reflects broader societal
trends, whereby the income segregation between schools has increased,
and the demographics of students served by the American public educa-
tion system have shifted; we are seeing an increase of students who are
eligible for FRPL and an increase in the number of minority students in
the United States (Owens et al., 2016; Snyder et al., 2018). Yet, even when
accounting for this trend in our regression analysis, throughout the 2000s,
we found that new teachers were more likely to work in high-minority and
high-FRPL schools than more experienced teachers.
Notably, this trend seems to have been anticipated by teacher educators,
who, in recent years, have moved to prepare a primarily White teacher
workforce to work with increasingly diverse students (Ladson-Billings,
1995; Sleeter, 2001). Traditional teacher preparation programs, as well
as urban teacher residencies and alternative certification programs, also
seem to be responding to an increased interest among service-oriented
individuals with a commitment to working in schools with historically
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
27
underserved students. We contend that this increased interest in working
in underserved schools contributed in part to the 7-percentage-point dif-
ference in the probability of new teachers working in a majority-minority
school compared with more experienced teachers beginning in 2000.
As new teachers have become increasingly likely to work in underserved
schools, differences in the characteristics of teachers in high- and low-
minority schools have shifted as well. For instance, the overall increases in
new Hispanic teachers have been concentrated in high-minority schools,
with a fifth of new teachers working in high-minority schools identifying as
Hispanic in 2012. Improvements have been made in terms of gaps in the
certification status of teachers in high-minority and low-minority schools.
By 2000, close to a third of all new teachers in high-minority schools lacked
certification. Although more teachers are still uncertified in high-minor-
ity schools compared with low-minority schools (5% vs. 2%), these gaps
have narrowed substantially, and the certification rates are no longer sig-
nificantly different from one another. That said, students in high-minority
schools may still be more likely to be taught by underprepared teachers.
Other evidence from the SASS shows that the decline in the proportion of
uncertified teachers corresponded with increases in alternatively certified
teachers (Redding & Smith, 2016).
Unlike other studies that have examined long-term trends in teacher
academic ability (Lankford et al., 2014; Master et al., 2018), we found no
evidence that overall increases in the selectivity of a teachers’ baccalau-
reate institution are linked to shrinking gaps between high-poverty and
low-poverty schools or high-minority and low-minority schools. That the
increased share of new teachers to have graduated from a selective col-
lege is not driven by schools enrolling historically underserved students
suggests that this trend is not solely driven by programs, such as TFA, that
tend to be associated with recruitment of teachers from selective colleges.
New teachers in high-minority schools also face more difficult working
conditions and do not always receive the same induction supports as teach-
ers in less racially diverse schools. Compared to low-minority schools, new
teachers in high-minority schools consistently face more student discipline
problems, less administrative support, and less teacher cooperation—dif-
ferences that have remained consistent over time. Even though these new
teachers also report feeling less prepared to enter the classroom, they are
less likely to be assigned a mentor or receive supportive communication
from their administrators, differences that have emerged more recently.
Despite these broad changes in the characteristics of the schools in
which teachers begin their careers, the new teacher turnover rates have
been relatively stable over time and even decreased in the two final years
for which we have data. In our regression analysis, we found that new
Teachers College Record, 122, 070311 (2020)
28
teachers are more likely to move schools and leave teaching than more ex-
perienced teachers, but not at substantially different rates in underserved
schools. Organizational supports, such as administrative support, teacher
cooperation, and induction programs, are associated with reduced risks
of turnover.
Even as our work contributes to this important topic, unanswered ques-
tions emerged from the limitations of our work that motivate future re-
search on this topic. Although we showed evidence of improved certifica-
tion and education levels of incoming teachers over time, the descriptive
research design precludes us from identifying the factors that drove this
shift. Researchers could apply more rigorous research designs to examine
the policies or programs that successfully improve the academic ability
or racial/ethnic diversity of incoming teachers. For instance, research-
ers could examine how changes in state-level policies regarding the cer-
tification of teachers influence the composition of who enters teaching
across states.
Second, the data used for this study only include one survey after the
Great Recession. In 2012, the gap between new and more experienced
teachers working in majority-FRPL or minority schools narrowed to 4 per-
centage points and is no longer statistically significant. Without additional
years of data, it is unclear the extent to which new teachers will still be
more likely to enter schools that enroll a majority of FRPL and minority
students after this period of slowed hiring of new teachers.
Third, our results do not indicate whether this pattern is ultimately posi-
tive for students. New teachers, as they begin their career, are on average
less effective than more experienced teachers. In this light, the inequitable
distribution of inexperienced teachers into schools that enroll historically
underserved student populations would be detrimental. However, there is
significant variation in new teachers’ effectiveness, and some new teach-
ers can immediately make important contributions to student learning
(Atteberry et al., 2015; Henry et al., 2014). Future research could more
carefully examine the conditions under which certain new teachers are
more effective than others. Such an approach may require education re-
searchers to adopt research methods that have not conventionally been
applied to research of teachers, such as latent class analysis. This approach
may help to more clearly identify and recruit service-oriented teachers ca-
pable of making meaningful differences in their students’ academic and
behavioral development.
Fourth, even when service-oriented teachers are able to make positive
contributions in their students’ lives, the overall sustainability of this ap-
proach to staffing hard-to-staff subjects in hard-to-staff schools remains
debatable given the high levels of turnover. Programs that channel new
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
29
teachers into underserved schools at worst risk exacerbating historic in-
equities in underserved students’ assignment to inexperienced teachers
and high turnover within underserved schools. In our study, we show that
new teachers are much more likely to leave teaching after beginning in
their career in schools with more racial/ethnic minorities. An average
of 14.03% first-year teachers leave teaching from high-minority schools,
compared with 9.10% from low-minority schools. Even after controlling
for other teacher background and school characteristics, the increased
risk of turnover persists. That said, counter to our expectation, new teach-
ers were no more likely to turn over from high-poverty or high-minority
schools than more experienced teachers. On the one hand, this finding
could indicate that the provision of induction supports has equalized the
turnover of new teachers with more experienced teachers. On the other
hand, the success of mentoring and induction programs may depend on
broader social conditions of a school—conditions that influence all teach-
ers to turn over from underserved schools at high rates. Future research
that more carefully identifies the working conditions and social supports
associated with higher new teacher retention is of continued importance.
Fifth, broader questions remain about the supply of new teachers mov-
ing into the next decade (Sutcher et al., 2016). Recent policies that place
more scrutiny on new teachers and stagnating beginning salaries all have
the potential to make teaching a less desirable profession and reduce the
number of aspiring applicants. Future research can begin to better under-
stand the factors leading to declining enrollment in teacher preparation
programs and possibly in the numbers of individuals entering teaching.
Although the differences we describe in this article generalize to a
national population of new teachers over a 25-year period, the results
likely overlook important variation across states (Goldhaber et al., 2018).
Further, other studies that use administrative data have been able to make
more careful comparisons regarding teacher quality gaps than we were
able to (Goldhaber et al., 2015; Kalogrides & Loeb, 2013). In addition,
when we focus on school-level differences, we focus on schools that have
a majority of either racial/ethnic minorities or FRPL students. Although
we supplement this analysis with comparisons between schools at the top
and bottom quartile of these demographic characteristics, the exact cut
point remains somewhat arbitrary. Finally, the measures we can access to
measure teacher quality—certification, education level, and college selec-
tivity—are weakly related to student achievement.
Still, findings from this study have implications for both policy and prac-
tice. In terms of practice, our study presents sharp differences in the char-
acteristics of the schools in which recent new teachers are beginning their
careers. In 1988, fewer than a quarter of teachers worked in a school where
Teachers College Record, 122, 070311 (2020)
30
the majority of students identified as a racial/ethnic minority or were liv-
ing in poverty. In 2012, approximately half of all new teachers worked in
these schools, schools in which teachers also reported facing more difficult
working conditions. Although a service orientation likely motivates many
teachers to work in underserved schools, the lack of resources, induction
supports, and stable, supportive colleagues and administrators can make
the transition into teaching more difficult for these teachers (Bettini &
Park, in press; Ronfeldt et al., 2016). Relatedly, our findings suggest that
school-level organizational factors are associated with reduced odds of
turnover. In particular, increasing administrative support and teacher co-
operation may reduce the likelihood of new teachers leaving the teaching
profession, while induction and common planning may keep new teach-
ers in the schools where they began their teaching career. New teachers’
awareness of the characteristics of schools and the differences in working
conditions in which they are increasingly likely to teach is critical because
they have important implications for teacher development and retention
(Bettini & Park, in press).
In terms of policy, our results speak to the continued importance of high-
quality induction supports for all beginning teachers, particularly those
in underserved schools. We also emphasize that supporting new teachers
requires the provision of both formal and informal supports that integrate
teachers into the social and professional culture of a school (Youngs et
al., 2012). Moreover, our work can inform our ongoing and developing
understanding of the status of the teaching profession by presenting evi-
dence on how new teacher characteristics, particularly their credentials
and education. Improvements in the education level and certification
rates are arguably indicative of a more professionalized teacher workforce.
NOTES
1. Because charter schools are a part of the sampling frame for SASS, we have
included them in our analysis to keep our results nationally representa-
tive. Our analyses are substantively similar when we drop charter schools
from the sample (about 4%–5% of the sample size in the last four waves of
SASS and TFS).
2. It is important to note some colleges have increased their ranking or
have been reclassified as top-tier institutions over the past 20 years with-
out substantial evidence of improvements in instructional quality (Hess &
Hochleitner, 2012).
3. A limitation of using free and reduced-price lunch eligibility as a proxy
for poverty over time relates to changes in the eligibility criteria. The most
notable change over the course of our study is the Community Eligibility
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
31
option, whereby districts could opt for all students to receive a free or sub-
sidized lunch if more efficient. Snyder and Musu-Gillette (2015) described
how 38% of students were eligible for FRPL in the 2000–2001 school year
compared with 50% in 2012–2013. Over this same period, the percentage
of public school students living in poverty increased from 17% to 23%.
4. Figure 1 and Table 3 in the supplemental materials (link provided in the
Data section) replicate this analysis for underrepresented minorities. We
found evidence of similarly sized teacher experience gaps, but they were
only marginally significant.
5. With evidence that first-year teachers are increasingly likely to work in
schools with more than 50% minorities or FRPL students, we tested for
trends at other points of the distribution of the percentage of FRPL and
minority students using quantile regression. Because of sparse data in the
bottom quartile of minority student enrollment in the first three survey
waves, we were unable to consistently estimate this model. In terms of the
percentage of FRPL students, we found new teachers to be significantly
more likely to work in schools above the median. Although the direction
and magnitude of coefficients on the interaction between new teacher and
survey year support a similar interpretation as displayed in Figure 1, the
estimates are rarely significant when testing for trends over time. Results
available on request.
6. We tested for changes in new teacher turnover over time but found no
evidence to suggest that new teachers were more likely to turn over than
more experienced teachers between 2000 and 2012. As a result, we present
results from a model pooling all years of data.
7. When predicting moving, the lack of significant evidence between the per-
cent FRPL or minority student enrollment was not driven by the moder-
ate to high correlation between these variables. When these variables were
included separately in the models, we found no evidence that they were
associated with moving schools or leaving teaching for new teachers.
Teachers College Record, 122, 070311 (2020)
32
REFERENCES
Aaronson, D. & Meckel, K. (2008, September). The impact of baby boomer retirements on
teacher labor markets. Chicago Fed Letter, 254.
Anderson, L., & Olsen, B. (2006). Investigating early career urban teachers’ perspectives on
and experiences in professional development. Journal of Teacher Education, 57(4), 359–377.
Anzia, S. F., & Moe, T. M. (2014). Collective bargaining, transfer rights, and disadvantaged
schools. Educational Evaluation and Policy Analysis, 36(1), 83–111.
Atteberry, A., Loeb, S., & Wyckoff, J. (2015). Do first impressions matter? Predicting early
career teacher effectiveness. AERA Open, 1(4), 1–23.
Bacolod, M. P. (2007). Do alternative opportunities matter? The role of female labor markets
in the decline of teacher quality. Review of Economics and Statistics, 89(4), 737–751.
Bettini, E., & Park, Y. (in press). Novice teachers’ experiences in high-poverty schools: An
integrative review. Urban Education.
Boyd, D., Grossman, P. L., Ing, M., Lankford, R. H., Loeb, S., & Wyckoff, J. (2011). The
influence of school administrators on teacher retention decisions. American Educational
Research Journal, 48(2), 303–333.
Boyd, D., Grossman, P., Lankford, H., Loeb, S., & Wyckoff, J. (2008). Who leaves? Teacher
attrition and student achievement (NBER Working Paper No. 14022). National Bureau of
Economic Research.
Boyd, D., Lankford, H., Loeb, S., Rockoff, J., & 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, 27(4), 793–818.
Boyd, D., Lankford, H., Loeb, S., Ronfeldt, M., & Wyckoff, J. (2011). The role of teacher
quality in retention and hiring: Using applications-to-transfer to uncover preferences of
teachers and schools. Journal of Policy and Management, 30(1), 88–110.
Buddin, R., & Zamarro, G. (2009a). Teacher qualifications and middle school student achievement
(WR-671-IES). RAND Corporation.
Buddin, R., & Zamarro, G. (2009b). Teacher effectiveness in urban high schools (WR-693-IES).
RAND Corporation.
Cannata, M. (2010). Understanding the teacher job search process: Espoused preferences
and preferences in use. Teachers College Record, 112(12), 2889–2934.
Clotfelter, C. T., Ladd, H. F., & Vigdor, J. L. (2005). Who teaches whom? Race and the
distribution of novice teachers. Economics of Education Review, 24, 377–392.
Clotfelter, C. T., Ladd, H. F., & Vigdor, J. L. (2007). Teacher credentials and student
achievement: Longitudinal analysis with student fixed effects. Economics of Education,
26(6), 673–682.
Clotfelter, C. T., Ladd, H. F., & Vigdor, J. L. (2010). Teacher credentials and student
achievement in high school: A cross-subject analysis with student fixed effects. Journal of
Human Resources, 45(3), 655–681.
Cohen-Vogel, L., & Smith, T. M. (2007). Qualifications and assignments of alternatively
certified teachers: Testing core assumptions. American Educational Research Journal,
44(3), 732–753.
Corcoran, S. P., Evans, W. N., & Schwab, R. S. (2004). Women, the labor market, and
the declining relative quality of teachers. Journal of Policy Analysis and Management,
23(3), 449-470.
Dee, T. S., & Wyckoff, J. (2015). Incentives, selection, and teacher performance: Evidence
from IMPACT. Journal of Policy Analysis and Management, 34(2), 267–297.
Dunn, M., & Walker, J. (2016). Union membership in the United States. In Spotlight on
Statistics. Bureau of Labor Statistics.
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
33
Engel, M. (2012). The timing of teacher hires and teacher qualifications: Is there an
association? Teachers College Record, 114(12), 1–29.
Engel, M. (2013). Problematic preferences? A mixed method examination of what principals’
preference for teacher characteristics in Chicago. Educational Administration Quarterly,
49(1), 52–91.
Engel, M., Jacob, B. A., & Curran, F. C. (2014). New evidence on teacher labor supply.
American Educational Research Journal, 51(1), 36–72.
Feng, L. (2010). Hire today, gone tomorrow: New teacher classroom assignments and teacher
mobility. Education Finance and Policy, 5(3), 278–316.
Goldhaber, D., Quince, V., & Theobald, R. (2018). Has it always been this way? Tracing the
evolution of teacher quality gaps in U.S. public schools. American Educational Research
Journal, 55(1), 171–201.
Goldhaber, D., & Walch, J. (2013). Rhetoric versus reality: Is the academic caliber of the teacher
workforce changing? (CEDR Working Paper 2013-4). University of Washington, Seattle.
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, 113(11), 2552–2585.
Grossman, P. (1990). The making of a teacher. Teachers College Press.
Harris, D. N., Ingle, W. K., & Rutledge, S. A. (2014). How teacher evaluation methods matter
for accountability: A comparative analysis of teacher effectiveness ratings by principals
and teacher value-added measures. American Educational Research Journal, 5(1), 73–112.
Harris, D., Rutledge, S., Ingle, W., & Thompson, C. (2010). Mix and match: What principals
really look for when hiring teachers. Education Finance and Policy, 5, 228–246.
Henry, G. T., Purtell, K. M., Bastian, K. C., Fortner, C. K., Thompson, C. L., Campbell, S. L.,
& Patterson, K. M. (2014). The effects of teacher entry portals on student achievement.
Journal of Teacher Education, 65(1), 7–23.
Henry, G. T., & Redding, C. (2020). The consequences of leaving school early: The effects of
within-year and end-of-year teacher turnover. Education Finance and Policy, 15(2), 332–356.
Hess, F. M., & Hochleitner, T. (2012). College rankings inflation: Are you overpaying
for prestige? Education Outlook (No. 3). American Enterprise Institute for Public
Policy Research.
Horng, E. L. (2009). Teacher tradeoffs: Disentangling teachers’ preferences for working
conditions and student demographics. American Educational Research Journal, 46, 690–717.
Ingersoll, R. M. (2001). Teacher turnover and teacher shortages: An organizational analysis.
American Education Research Journal, 38(3), 499–534.
Ingersoll, R., & Merrill, L. (2017). A quarter century of changes in the elementary and secondary
teaching force: From 1987 to 2012. Statistical Analysis Report (NCES 2017-092). U.S.
Department of Education, National Center for Education Statistics.
Ingersoll, R. M., Merrill, L., & Stuckey, D. (2014). Seven trends: The transformation of the teaching
force. CPRE Research Report No. RR-80. Consortium for Policy Research in Education.
Ingersoll, R., & Strong, M. (2011). The impact of induction and mentoring for beginning
teachers: A critical review of the research. Review of Educational Research, 81(2), 201–233.
Jennings, J. L., & DiPrete, T. A. (2010). Teacher effects on social/behavioral skills in early
elementary school. Sociology of Education, 83, 135–159.
Johnson, S. M., & Birkeland, S. E. (2003). Pursuing a “sense of success”: New teachers explain
their career decisions. American Educational Research Journal, 40(3), 581–617.
Kalogrides, D., & Loeb, S. (2013). Different teachers, different peers: The magnitude and
effects of student sorting within schools. Educational Researcher, 42(6), 304–316.
Teachers College Record, 122, 070311 (2020)
34
Kane, T. J., Rockoff, J. E., & Staiger, D. O. (2008). What does certification tell us about
teacher effectiveness? Evidence from New York City. Economics of Education Review,
27(6), 615–631.
Kardos, S. M., & Johnson, S. M. (2007). On their own and presumed expert: New teachers’
experience with their colleagues. Teachers College Record, 109(9), 2083–2106.
Koski, W. S., & Horng, E. (2007). Facilitating the teacher quality gap? Collective bargaining
agreements, teacher hiring and transfer rules, and teacher assignment among schools in
California. Education Finance and Policy, 2(3), 262–300.
Kraft, M. A., Marinell, W. H., & Yee, D. S. (2016). School organizational contexts, teacher
turnover, and student achievement: Evidence from panel data. American Educational
Research Journal, 53(5), 1411–1449.
Krieg, J., Theobald, R., & Goldhaber, D. (2016). A foot in the door: Exploring the role of
student teaching assignments in teachers’ initial job placements. Educational Evaluation
and Policy Analysis, 38(2), 364–388.
Ladd, H. F. (2011). Teachers’ perceptions of their working conditions: How predictive
of planned and actual teacher movement? Educational Evaluation and Policy Analysis,
33(2), 235–261.
Ladson-Billings, G. (1995). Toward a theory of culturally relevant pedagogy. American
Educational Research Journal, 32(3), 465–491.
Lankford, H., Loeb, S., McEachin, A., Miller, L. C., & Wyckoff, J. (2014). Who enters
teaching? Encouraging evidence that the status of teaching is improving. Educational
Researcher, 43(9), 444–453.
Lankford, H., Loeb, S., & Wyckoff, J. (2002). Teacher sorting and the plight of urban schools:
A descriptive analysis. Educational Evaluation and Policy Analysis, 24, 37–62.
Liu, E., & Johnson, S. M. (2006). New teachers’ experiences of hiring: Late, rushed, and
information-poor. Educational Administration Quarterly, 42(3), 324–360.
Master, B., Sun, M., & Loeb, S. (2018). Teacher workforce developments: Recent changes
in academic competitiveness and job satisfaction of new teachers. Education Finance and
Policy, 13(3), 310–332.
Mishel, L., Gould, E., & Bevins, J. (2015). Wage stagnation in nine charts. Economic Policy
Institute. http://www.epi.org/
Moe, T. M. (2005). Bottom-up structure: Collective bargaining, transfer rights, and the education of
disadvantaged children [Unpublished manuscript]. Stanford University.
National Commission on Teaching and America’s Future. (1996). What matters most: Teaching
for America’s future.
Owens, A., Reardon, S. F., & Jencks, C. (2016). Income segregation between schools and
school districts. American Educational Research Journal, 53(4), 1159–1197.
Pew Research Center. (2013, August 22). King’s dream remains an elusive goal; many Americans
see racial disparities. https://www.pewsocialtrends.org/2013/08/22/kings - dream
- remains-an-elusive-goal-many-americans-see-racial-disparities/
Redding, C. (2019). A teacher like me: A review of the effect of student-teacher racial/ethnic
matching on teacher perceptions of students and student academic and behavioral
outcomes. Review of Educational Research, 89(4), 499–535.
Redding, C., & Henry, G. T. (2019). Leaving school early: An examination of novice teachers’
within- and end-of-year turnover. American Educational Research Journal, 56(1), 204–236.
Redding, C., & Smith, T. M. (2016). Easy in, easy out—Are alternatively certified teachers
turning over at increased rates? American Educational Research Journal, 53(4), 1086–1125.
Ronfeldt, M. (2012). Where should student teachers learn to teach? Effects of field placement
school characteristics on teacher retention and effectiveness. Educational Evaluation and
Policy Analysis, 34(1),3–26.
TCR, 122, 070311
Recent Trends in the Characteristics of New Teachers
35
Ronfeldt, M., Kwok, A., & Reininger, M. (2016). Teachers’ preferences to teach under-served
students. Urban Education, 51(9), 995–1030.
Ronfeldt, M., Loeb, S., & Wyckoff, J. (2013). How teacher turnover harms student
achievement. American Educational Research Journal, 50(1), 4–36.
Ronfeldt, M., & McQueen, K. (2017). Does new teacher induction really improve retention?
Journal of Teacher Education, 68(4), 394–410.
Sartain, L., & Steinberg, M. (2016). Teachers’ labor market responses to performance
evaluation reform: Experimental evidence from Chicago Public Schools. Journal of
Human Resources, 51(3), 615–655.
Simon, N. S., & Johnson, S. M. (2015) Teacher turnover in high-poverty schools: What we
know and can do. Teachers College Record, 117(3), 1–36.
Sleeter, C. E. (2001). Preparing teachers for culturally diverse schools: Research and the
overwhelming presence of Whiteness. Journal of Teacher Education, 52(2), 94–106.
Smith, T. M., & Ingersoll, R. M. (2004). What are the effects of induction and mentoring on
beginning teacher turnover? American Educational Research Journal, 41(3), 681–714.
Snyder, T. D., de Brey, C., & Dillow, S. A. (2018). Digest of Education Statistics 2016 (NCES
2017-094). National Center for Education Statistics, Institute of Education Sciences, U.S.
Department of Education.
Snyder, T., & Musu-Gillette, L. (2015). Free or reduced lunch: A proxy for poverty? https://nces.
ed.gov/blogs/nces/post/free-or-reduced-price-lunch-a-proxy-for-poverty
Sutcher, L., Darling-Hammond, L., & Carver-Thomas, D. (2016). A coming crisis in teaching?
Teacher supply, demand, and shortages in the U.S. Learning Policy Institute.
Swain, W., & Redding, C. (in press). Teacher union power in a budget crunch: Collective
bargaining and differential response to the Great Recession. Educational Policy.
Youngs, P., Frank, K., Thum, Y. M., & Low, M. (2012). The motivation of teachers to produce
human capital and conform to their social contexts. In T. M. Smith, L. M. Desimone, &
A. C. Porter (Eds.), Organization and effectiveness of induction programs for new teachers (pp.
248-272). New York, NY: National Society for the Study of Education.
Teachers College Record, 122, 070311 (2020)
36
CHRISTOPHER REDDING is an assistant professor in the Department
of Human Development and Organizational Studies in Education at the
University of Florida’s College of Education. His research focuses on teach-
er labor markets, teacher education and development, and school im-
provement. His work has recently been published in Educational Researcher,
the American Educational Research Journal, and AERA Open.
TUAN D. NGUYEN is an assistant professor in the Department of
Curriculum and Instruction at Kansas State University. His main research
interests include teacher leadership and school improvement, and teach-
er labor markets. His work has been recently published in Economics of
Education Review and AERA Open.
... While other studies focused on AC teachers at the beginning of their careers, the present study examined AC teachers with different levels of teaching experience. Since the turnover rate is highest among beginning teachers (Redding & Nguyen, 2020) and even higher for AC teachers (Redding & Henry, 2019), it must be noted that the present study looked mainly at teachers who "survived" their first years of teaching and did not leave the profession. This aspect could possibly explain the results found in this study and must be kept in mind when interpreting them. ...
Article
Full-text available
This study investigates alternatively certified (AC) teachers' motives for teaching, their well-being, and their intention to stay in the profession. We conducted multivariate covariance analyses using a large-scale dataset of 446 traditionally certified (TC) teachers and 143 AC teachers at secondary schools in Germany. Findings show that AC teachers reported more frequently than TC teachers that they chose teaching due to social influences and because of more time for their family. Furthermore, AC teachers report significantly higher enthusiasm for teaching. No differences were found regarding emotional exhaustion or the intention to stay in the profession.
... Overcoming the teacher shortage also requires high retention of new teachers. Research suggests, however, that teacher retention is lowest among beginning teachers (Fantilli & McDougall, 2009;Ingersoll, 2001Ingersoll, , 2002Ingersoll, , 2003Redding & Nguyen, 2020), including AC teachers (Carver-Thomas & Darling-Hammond, 2019; Redding & Henry, 2019). School administrators in many countries are currently seeking to better understand how to meet the professional needs of AC teachers to increase new teacher retention and overcome acute staffing shortages. ...
Article
Full-text available
In this study, we investigated retention intention and job satisfaction of 238 first-year alternatively certified (AC) teachers. Drawing on Organizational Socialization Theory, we tested the hypothesis that AC teacher extraversion and perceived school support are positively related to the two variables and mediated by self-efficacy. To test our hypothesis, we applied structural equation modeling. Our results demonstrate that extraversion and perceived social support are positively related to retention intentions and job satisfaction. In addition, self-efficacy serves as a mediator. The findings could help school administrators to better understand how to support and retain AC teachers and thus address teacher shortages.
... Many of these factors are associated with increased risk of teacher turnover in different rural contexts. However, prior works suggest there are positive steps that can be taken to reduce turnover for novice and special education teachers and to enable administrators to be more supportive and encouraging to teachers to increase retention for all teachers (Billingsley & Bettini, 2019;Northrup, 2018;Redding & Nguyen, 2020). In many ways, this research highlights how teacher labor market and attrition in rural contexts, and by extension rural education, should not be lumped together with urban-suburban research. ...
Article
Full-text available
Using repeated cross-sectional nationally representative data, we demonstrate how the teacher labor markets for rural contexts are different from those in urban-suburban areas. We also show that teacher attrition is not uniform across various rural settings. In particular, novice teachers in rural schools in sparsely populated states are more likely to turn over than novice teachers in urban-suburban schools in sparsely populated states. We also examine how teacher and school characteristics are associated with turnover in different rural contexts. The findings indicate there should be a concerted effort to examine teacher attrition in various rural contexts and not simply as delineation from urban-suburban areas.
Article
Background Prior work examined the relationship between teacher preparation programs (TPPs) and teacher attrition and turnover. However, such research aggregated all types of non-traditional certification programs (non-TCPs) into a single non-TCP group. This practice masks some critical differences among non-TCPs, and could send a misleading message to stakeholders and policymakers. Purpose Our study builds on prior studies and empirically investigates the relationship between the types of TPPs and beginning teachers’ attrition and turnover behaviors in Texas public schools using a detailed non-TCP typology. Our study provides information about variation in attrition and turnover rates among non-TCPs and highlights the types of non-TCPs that exhibited different turnover and attrition patterns. Research Design Using longitudinal administrative data and detailed data on preparation program type obtained from the Texas Education Agency, we estimated a series of logistic regression models and discrete-time hazard models with district fixed effects, school random effects, and school-by-year random effects. Findings Non-TCP teachers were more likely to leave the profession or the district than TCP teachers for the first few years. Yet, by the end of the fourth year, attrition and turnover rates converged between TCP and non-TCP teachers. Among non-TCP teachers, teachers from district-administered programs and programs operated by private nonprofit organizations exhibited different attrition and turnover behaviors than other non-TCP teachers. Conclusion Given the differential turnover and attrition rates among non-TCP teachers, district and school administrators may design tailored activities and programs that could meet the varying needs of teachers from different types of preparation programs.
Chapter
The growing dilemma of teacher job turnover has severely handicapped the fundamental responsibility of school districts to maintain a committed and stable instructional force. Using the lens of organizational systems theory, this chapter explores ways in which building leaders can actively increase job embeddedness of teachers and staff by constructing collaborative perspectives of teaching and learning. A systems approach where disciplines are integrated and community culture is respected can result in teachers being more engaged in school-wide pedagogy, feeling less isolated, and developing a convivial sense of fidelity which can lead to job satisfaction and commitment to remain.
Article
Research suggests that work environments are associated with turnover patterns for teachers of Color. This study investigates variation in work environments using longitudinal administrative data from 20 large urban and suburban K-12 school districts. Results indicate that teachers of Color are more often employed in “hard-to-staff” work environments compared to their White colleagues. Despite this, findings demonstrate high relative levels of teacher experience in schools where teachers of Color, specifically Latinx educators, represent a majority of the teaching staff. These results have implications for educational policy and practice, specifically in the areas of teacher preparation, recruitment, hiring, and retention.
Article
Drawing on nationally representative data from the 1988 to 2018 school years, the authors provide an overview of some of the key changes in the characteristics of first-year teachers in the U.S., including racial/ethnic identity, education levels, subject area, and certification status. The data also show that new teachers have become increasingly more likely, compared to experienced teachers, to work in schools with a greater fraction of students of color, which has consequences for equitable teacher assignments and teacher attrition.
Article
h i g h l i g h t s Some school organizational characteristics can potentially influence teacher turnover. Better administrative support, stronger teacher cooperation and more effective principals may help to retain teachers. School factors have a more pronounced relationship with turnover in low-income schools than more affluent schools. These results are robust to a variety of modeling choices. a b s t r a c t Teacher labor markets have strong implications for learning outcomes and equity for students, and yet schools that need teachers the most often face high turnover. Using repeated cross-sectional nationally representative data, this study examines the relationships between school organizational characteristics and teacher turnover and identifies what may be done to positively influence the teacher workforce. I find school organizational characteristics may indeed reduce teacher turnover, but this varies based on the type of organizational characteristic as well as the specific form of turnover. Policy and practical implications from the study are discussed.
Article
Full-text available
We study the relationship between school organizational contexts, teacher turnover, and student achievement in New York City (NYC) middle schools. Using factor analysis, we construct measures of four distinct dimensions of school climate captured on the annual NYC School Survey. We identify credible estimates by isolating variation in organizational contexts within schools over time. We find that improvements in school leadership especially, as well as in academic expectations, teacher relationships, and school safety are all independently associated with corresponding reductions in teacher turnover. Increases in school safety and academic expectations also correspond with student achievement gains. These results are robust to a range of threats to validity suggesting that our findings are consistent with an underlying causal relationship.
Article
Considerable research has examined the positive educational experiences of students of color assigned to teachers of the same race or ethnicity. Underlying this research is the belief that the cultural fit between students and teachers has the potential to improve a child’s academic and nonacademic performance in school. This comprehensive review examines the extent to which Black and Latino/a students (1) receive more favorable ratings of classroom behavior and academic performance, (2) score higher on standardized tests, and (3) have more positive behavioral outcomes when assigned to a teacher of the same race/ethnicity. Assignment to a same-race teacher is associated with more favorable teacher ratings, although the relationship differs by school level. There is fairly strong evidence that Black students score higher on achievement tests when assigned to a Black teacher. Less consistent evidence is found for Latino/a students.
Article
This report utilizes the nationally representative Schools and Staffing Survey (SASS) to examine changes in the elementary and secondary teaching force in the United States over the quarter century from 1987-88 to 2011-12. The report focuses on three key demographic characteristics: the size of the teaching force, the level of teaching experience of the teaching force, and the racial/ethnic composition of the teaching force.
Article
Using unique administrative data from North Carolina that allow us to separate classroom teacher turnover during the school year from end-of-year turnover, we find students who lose their teacher during the school year have significantly lower test score gains (on average −7.5 percent of a standard deviation unit) than those students whose teachers stay. Moreover, the turnover of other teachers during the year lowers achievement gains, whereas end-of-year teacher turnover appears to have little effect on achievement. The harmful effects of within-year turnover cannot be explained by other extraneous shocks or the quality of departing teachers. Teachers who depart from December through April have the most harmful effects on achievement, although these vary somewhat by level of schooling and subject.
Article
Most prior research measures teacher turnover as an annual event, but teachers actually leave their positions throughout the school year. We use data from North Carolina to measure teacher turnover monthly throughout the entire year and conduct an analysis of their persistence to examine the differences in early career teacher turnover. Annually, 6% of early career teachers turn over during the school year. Teachers trained in traditional, university-based programs are most likely to move schools, and alternate entry and out-of-state prepared teachers are more likely to leave teaching, both during and at the end of the school year. We discuss the implications within-year turnover has on creating disruptive learning environments, particularly in underserved schools.
Article
There is mounting evidence of substantial “teacher quality gaps” (TQGs) between advantaged and disadvantaged students but practically no empirical evidence about their history. We use longitudinal data on public school students, teachers, and schools from two states—North Carolina and Washington—to provide a descriptive history of the evolution of TQGs in these states. We find that TQGs exist in every year in each state, and for all measures, we consider student disadvantage and teacher quality. But there is variation in the magnitudes and sources of TQGs over time, between the two states, and depending on the measure of student disadvantage and teacher quality.
Article
In the midst of discussions about improving education, teacher education, equity, and diversity, little has been done to make pedagogy a central area of investigation. This article attempts to challenge notions about the intersection of culture and teaching that rely solely on microanalytic or macroanalytic perspectives. Rather, the article attempts to build on the work done in both of these areas and proposes a culturally relevant theory of education. By raising questions about the location of the researcher in pedagogical research, the article attempts to explicate the theoretical framework of the author in the nexus of collaborative and reflexive research. The pedagogical practices of eight exemplary teachers of African-American students serve as the investigative "site." Their practices and reflections on those practices provide a way to define and recognize culturally relevant pedagogy.
Article
Policymakers have increasingly worked to combat teacher turnover by implementing induction programs for early-career teachers. Yet the existing evidence for the effects of induction on turnover is mixed. Drawing on data from the three most recent administrations of the Schools and Staffing and Teacher Follow-Up Surveys, as well as the Beginning Teacher Longitudinal Study, this study investigates whether different kinds of induction supports predict teacher turnover among nationally representative samples of first-year teachers. We find that receiving induction supports in the first year predicts less teacher migration and attrition, suggesting that using induction to reduce new teacher turnover is a promising policy trend. We also find that levels of induction support are fairly constant for different kinds of teachers and teachers in different kinds of schools. The exceptions are that teachers who are Black and who work in schools with more students who speak English as a second language report higher levels of induction supports.
Article
Retaining teachers in high-poverty schools is essential for ensuring students who live in poverty have equitable educational opportunities. Understanding novices’ experiences can help school leaders improve novices’ retention in high-poverty schools throughout their careers. This integrative review of studies investigates novices’ experiences teaching in high-poverty schools. We identified common themes in extant studies, as well as several important areas of research that are, to date, under-studied; most notably, more research is needed to explore differences in novices’ experiences of their working conditions in high- versus low-poverty schools, and the implications of those differences for teacher development and retention.
Article
Traditional teacher evaluation systems have come under scrutiny for not identifying, supporting, and, if necessary, removing low-performing teachers from the classroom. Leveraging the experimental rollout of a pilot evaluation system in Chicago, we find that, while there was no main effect of the pilot on teacher exit, the pilot system increased exit for low-rated and nontenured teachers. Furthermore, teachers who exited were lower performing than those who stayed and those who replaced them. These findings suggest that reformed evaluation systems can induce low-performing teachers to exit schools and may also improve the overall quality of the teacher labor force. © 2016 by the Board of Regents of the University of Wisconsin System.