August 2016


What’s Happening

Retention, attrition, and

mobility among teachers and

administrators in West Virginia

Chad R. Lochmiller
Indiana University
Eishi Adachi
Westat
Colleen E. Chesnut
Indiana University
Jerry Johnson
University of West Florida

Key findings
Over 2008/09–2012/13:
•  On average 90.1 percent of teachers and 87.7 percent of administra­
tors stayed in the same West Virginia school district from one year to
the next.
•  On average 9.0 percent of teachers and 11.1 percent of administrators
left the West Virginia public school system each year.
•  On average 0.9 percent of teachers and 1.2 percent of administrators
moved to a different school district in the West Virginia public school
system each year.
•  Of teachers with zero years of experience who began teaching in the
West Virginia public school system in 2008/09, 32.0 percent had left
by 2012/13.

U.S.

Department

of

Education

At CNA

 U.S. Department of Education
John B. King, Jr., Secretary
Institute of Education Sciences
Ruth Neild, Deputy Director for Policy and Research
Delegated Duties of the Director
National Center for Education Evaluation and Regional Assistance
Joy Lesnick, Acting Commissioner
Amy Johnson, Action Editor
Felicia Sanders, Project Officer
REL 2016–161
The National Center for Education Evaluation and Regional Assistance (NCEE) conducts
unbiased large-scale evaluations of education programs and practices supported by federal
funds; provides research-based technical assistance to educators and policymakers; and
supports the synthesis and the widespread dissemination of the results of research and
evaluation throughout the United States.
August 2016
This report was prepared for the Institute of Education Sciences (IES) under Contract
ED-IES-12-C-0005 by Regional Educational Laboratory Appalachia administered by
CNA. The content of the publication does not necessarily reflect the views or policies of
IES or the U.S. Department of Education, nor does mention of trade names, commercial
products, or organizations imply endorsement by the U.S. Government.
This REL report is in the public domain. While permission to reprint this publication is
not necessary, it should be cited as:
Lochmiller, C. R., Adachi, E., Chesnut, C. E., & Johnson, J. (2016). Retention, attrition, and
mobility among teachers and administrators in West Virginia (REL 2016–161). Washington,
DC: U.S. Department of Education, Institute of Education Sciences, National Center for
Education Evaluation and Regional Assistance, Regional Educational Laboratory Appala­
chia. Retrieved from http://ies.ed.gov/ncee/edlabs.
This report is available on the Regional Educational Laboratory website at http://ies.ed.gov/
ncee/edlabs.

 Summary
Members of the West Virginia School Leadership Research Alliance partnered with
Regional Educational Laboratory Appalachia to study the average retention, attrition,
and mobility rates among teachers and administrators in the West Virginia public school
system. There is increasing evidence nationwide that low teacher and administrator reten­
tion rates adversely affect student academic outcomes, particularly in reading and math,
which are reform priorities in many states (Béteille, Kalogrides, & Loeb, 2012; Branch,
Hanushek, & Rivkin, 2012; Kane & Staiger, 2008; Ronfeldt, Lankford, Loeb, & Wyckoff,
2013). West Virginia policymakers and educators have thus expressed interest in increasing
teacher and administrator retention rates to improve student achievement.
This report provides descriptive information about retention, attrition, and mobil­
ity among teachers and administrators that can be used to inform policy and program
decisionmaking in West Virginia. The analyses were based on personnel data for teachers
and administrators provided by the West Virginia Department of Education for the aca­
demic years 2008/09–2012/13, as well as district information covering the same years from
the National Center for Education Statistics Common Core of Data. Unless otherwise
stated, the retention, attrition, and mobility rates are annual averages for the academic
years examined.
The study’s primary findings included the following:
•  On average 90.1 percent of teachers and 87.7 percent of administrators stayed in
the same West Virginia school district from one year to the next; 9.0 percent of
teachers and 11.1  percent of administrators left the West Virginia public school
system.
•  Female and male teachers left the system at similar rates (8.6  percent versus
10.1  percent); female administrators left the system at a lower rate than male
administrators did (8.8 percent versus 13.2 percent).
•  Teachers and administrators with fewer than 4 years or 15 or more years of
experience left the system at roughly double the rates that other teachers (10.4–
11.1 percent versus 4.3–5.6 percent) and administrators (12.2–13.0 percent versus
6.1–7.6 percent) did.
•  Teachers with a doctoral degree left the system at a higher rate (13.3 percent) than
did teachers with a master’s degree (9.7 percent) or a bachelor’s degree (8.2 percent).
Administrators with a doctoral degree left the system at about double the rate that
administrators with a bachelor’s degree (14.8 percent versus 7.8 percent) did and at
a higher rate than did administrators with a master’s degree (11.0 percent).
•  Teachers and administrators without National Board for Professional Teaching
Standards certification left the system at a higher rate than did teachers with
the certification (9.0 percent versus 6.9 percent among teachers and 11.3 percent
versus 4.7 percent among administrators).
•  Teachers earning the highest and lowest annual salaries left the system at the
highest rates, and administrators making the highest annual salaries left the
system at the highest rates.
•  Teachers in rural districts left the system (9.1  percent) and moved to different
districts (1.2  percent) at about the same rate, on average, as teachers working
in town (9.0  percent and 1.0  percent), suburban (9.3  percent and 0.6  percent),
and city (8.2  percent and 0.6  percent) school districts. Administrators in rural
i

 (12.4  percent) and town (11.8  percent) districts left at higher rates than did
administrators in suburban (8.9  percent) or city (7.9  percent) districts; adminis­
trators in rural (1.9  percent) and town (1.4  percent) districts moved to different
districts at about the same rate as administrators in suburban (0.2  percent) and
city (0.5 percent) districts.
•  Very few teachers and administrators moved to a different school district in the
state system (0.9 percent of teachers and 1.2 percent of administrators, on average).
•  Of teachers with zero years of experience who began teaching in the West Vir­
ginia public school system in 2008/09, 32.0 percent had left within their first four
years of teaching; on average 19.5 percent of teachers with zero years of experience
who began teaching during each of the baseline years examined left after their
first year.

ii

 Contents
Summary

i

Why this study?

1

What the study examined

2

What the study found
Close to 90 percent of teachers and administrators stayed in their West Virginia school
district during the study period
Average retention, attrition, and mobility rates varied by district, but rates among
administrators varied more than rates among teachers
Teachers left the West Virginia public school system at different rates depending on their
years of experience, highest degree earned, national board certification status, and
annual salary
Districts with a larger proportion of students eligible for the school lunch program had
higher teacher attrition rates than did districts with a smaller proportion
Administrators left the West Virginia public school system at different rates depending on
their gender, years of experience, highest degree earned, national board certification
status, annual salary, and district- or school-level assignment
Administrators left the West Virginia public school system at different rates depending on
district locale, student enrollment, percentage of students who are racial/ethnic minority
students, percentage of students eligible for the federal school lunch program, and
percentage of students who are limited English proficient students
The average mobility rate among teachers in West Virginia school districts was less than
3 percent across all personal characteristics and varied depending on teachers’ years of
experience and annual salary
The average mobility rate among administrators in West Virginia school districts was
less than 6 percent across all personal characteristics and varied depending on
administrators’ years of experience and highest degree earned
Teachers moved to a different district at about the same rate regardless of the characteristics
of their baseline district
Administrators moved to a different district at varying rates based on the characteristics of
the students the district served
About a fifth of beginning teachers left the West Virginia public school system after their
first year of teaching, and a third left by the end of the fourth year

4

15

Implications of the study findings

16

Limitations of the study

17

Appendix A. Data and methodology

A-1

Appendix B. Average retention, attrition, and mobility rates among teachers and
administrators by West Virginia public school district

B-1

Appendix C. Cumulative attrition rates among beginning teachers in West Virginia public
school districts

C-1

iii

4
4

5
6

7

10

11

11
13
14

 Note

Notes-1

References

Ref-1

Boxes
1 Key terms
2 Data and methods

2
3

Maps
1 Average retention rates among teachers vary across the 55 West Virginia public school
districts, 2008/09–2012/13
2 Average retention rates among administrators vary across the 55 West Virginia public
school districts, 2008/09–2012/13
Tables
1 Average retention and attrition rates among teachers in West Virginia public school
districts, by teacher characteristic, 2008/09–2012/13
2 Average retention and attrition rates among teachers in West Virginia public school
districts, by district characteristic, 2008/09–2012/13
3 Average retention and attrition rates among administrators in West Virginia public
school districts, by administrator characteristic, 2008/09–2012/13
4 Average retention and attrition rates among administrators in West Virginia public
school districts, by district characteristic, 2008/09–2012/13
5 Average mobility rates among teachers in West Virginia public school districts, by
personal characteristic, 2008/09–2012/13
6 Average mobility rates among administrators in West Virginia public school districts,
by personal characteristic, 2008/09–2012/13
7 Average mobility rates among teachers in West Virginia public school districts, by
district characteristic, 2008/09–2012/13
8 Average mobility rates among administrators in West Virginia public school districts,
by district characteristic, 2008/09–2012/13
9 Percentage of beginning teachers who left the West Virginia public school system
after each year of teaching, 2008/09–2012/13
A1 Variables used in the retention, attrition, and mobility analyses
B1 Average retention, attrition, and mobility rates among teachers in West Virginia
public school districts, 2008/09–2012/13
B2 Average retention, attrition, and mobility rates among administrators in West Virginia
public school districts, 2008/09–2012/13
C1 Cumulative attrition rates among beginning teachers who left the West Virginia
public school system, by district, 2008/09–2012/13

iv

5
6

7
8
9
10
12
13
14
15
16
A-1
B-1
B-2
C-1

 Why this study?
Among the reasons policymakers and educators focus on raising retention rates among
teachers and administrators is that low teacher and administrator retention rates adversely
affect student achievement, especially for students attending urban schools and schools
with a high share of economically disadvantaged students (Béteille et  al., 2012; Branch
et al., 2012; Levy, Jablonski, & Fields, 2006; Miller, 2013; Rivkin, Hanushek, & Kain, 2005;
Ronfeldt et al., 2013; Vanderhaar, Muñoz, & Rodosky, 2006; Weinstein, Jacobowitz, Ely,
Landon, & Schwartz, 2009).
Various teacher and administrator characteristics tend to be associated with different
retention rates (Battle, 2010; Guin, 2004; Ingersoll, 2001; Ringel, Gates, Chung, Brown, &
Ghosh-Dastidar, 2004). Further, retention rates appear to differ based on the characteris­
tics of the school districts where teachers and administrators are employed (Hanushek &
Rivkin, 2007; Scafidi, Sjoquist, & Stinebrickner, 2007). Much of the literature describes
retention, attrition, and mobility rates among teachers and administrators in individual
schools (Guarino, Santibañez, & Daley, 2006). These analyses have, for example, described
the percentage of teachers and administrators who stay in the same school from one year
to the next, leave the public school system, or move to different schools within the system
(Goldring & Taie, 2014; Goldring, Taie, & Riddles, 2014). Researchers have used state-level
personnel data to examine retention, attrition, and mobility among teachers and adminis­
trators at the school level in other states, including Georgia, Illinois, North Carolina, and
Texas (Fuller & Young, 2009; Gates et al., 2006; Hanushek, Kain, & Rivkin, 2004; Ladd,
2011; Scafidi et al., 2007).
Few studies have focused on retention, attrition, and mobility at the school district level.
Those studies show that teacher retention at the district level is negatively associated with
district property wealth and the percentage of racial/ethnic minority students (Theobald,
1990). Similar to studies focused on school-level retention, attrition, and mobility (Inger­
soll, 2001), district-level retention appears lower among teachers with less than 3 years of
experience or more than 15 years of experience (Theobald, 1990). District expenditures on
professional development and central administration appear related to teacher retention.
Districts that spend more on central administration and less on teacher professional devel­
opment have lower retention rates than districts that spend less on central administration
and more on teacher professional development (Theobald & Gritz, 1996).
West Virginia stakeholders have expressed interest in identifying ways to improve teacher
and administrator retention in the state’s public school districts. Stakeholders also wanted
to know whether school districts located in rural settings were experiencing higher rates
of teacher and administrator attrition and mobility than districts in town, suburban, and
city locales. Likewise, stakeholders were concerned that newer teachers were leaving the
West Virginia public school system at a higher rate than teachers with more experience,
potentially contributing to staffing challenges in certain school districts.
The dearth of recent information about district-level retention, attrition, and mobili­
ty rates in West Virginia public school districts prompted members of the West Virginia
School Leadership Research Alliance to partner with Regional Educational Laboratory
(REL) Appalachia to conduct this descriptive study. The alliance consists of school dis­
tricts, regional education service agencies, and state education agencies, and its top priority
1

The few studies
that have focused
on retention,
attrition, and
mobility at the
school district
level show that
teacher retention
at the district
level is negatively
associated with
district property
wealth and the
percentage of
racial/ethnic
minority students

 is to support effective teachers and leaders. The purpose of this study is to provide alliance
members with information about average retention, attrition, and mobility rates among
teachers and administrators in West Virginia school districts to inform district policy and
programmatic decisions related to improving teacher and administrator retention. Districtlevel attrition and mobility data may be of particular interest to school district staff seeking
information about the costs of hiring new teachers and administrators to replace those
who are leaving the public school system or moving to different districts.
Because retention, attrition, and mobility among teachers and administrators are areas of
concern for other states and on a national level (Goldring & Taie, 2014; Goldring et al.,
2014), this study may also be of interest to policymakers beyond West Virginia. Thus, this
study’s findings serve as an important reference point for others working to understand
retention, attrition, and mobility among teachers and administrators.

What the study examined
The study addressed the following research questions:
•  What were the average retention and attrition rates among teachers and adminis­
trators in West Virginia public school districts between 2008/09 and 2012/13?
•  How did average retention and attrition rates vary by district, teacher or adminis­
trator characteristics, and district characteristics?
•  What were the average mobility rates among teachers and administrators in West
Virginia public school districts between 2008/09 and 2012/13?
•  How did average mobility rates vary by teacher or administrator characteristics
and district characteristics?
•  What were the annual and overall average attrition rates among beginning
teachers employed in West Virginia public school districts between 2008/09 and
2012/13?
Key terms used in this report are defined in box 1. The study’s data and methodology are
summarized in box 2 and explained in detail in appendix A.

Box 1. Key terms
Administrators. Superintendents, deputy superintendents, principals, assistant principals, or
vice principals, as defined by the West Virginia Department of Education.
Attrition. Refers to leaving the West Virginia public school system and not returning to the
public school system in any of the follow-up years. Average attrition rate refers to the average
percentage of teachers or administrators who left the system during those years. See table B1
in appendix B for average attrition data for teachers and administrators.
Baseline year. The first academic year the individual appears in the dataset. Possible baseline
years are 2008/09, 2009/10, 2010/11, or 2011/12. It is possible, but rare, for an individual
to appear in more than one baseline year for the purposes of the first and second research
questions. For example, an individual who appeared in the dataset in 2008/09 and left in
2009/10 may have returned to a West Virginia public school district in 2011/12.
(continued)

2

The purpose of this
study is to provide
alliance members
with information
about average
retention, attrition,
and mobility rates
among teachers
and administrators
in West Virginia
school districts
to inform district
policy and
programmatic
decisions related
to improving
teacher and
administrator
retention

 Box 1. Key terms (continued)
Beginning teachers. Teachers who had zero years of experience and were hired in a West Vir­
ginia public school district during one of the baseline years. If the baseline year is 2008/09 a
beginning teacher is a person who had zero years of experience and who was hired by a West
Virginia public school district in that year. (See table C1 in appendix C for cumulative attrition
data for beginning teachers.)
Follow-up year. The successive year after the individual’s baseline year; in this study follow-up
years are academic years 2009/10, 2010/11, 2011/12, and/or 2012/13.
Mobility. Refers to moving from one West Virginia public school district in one academic year
to a different West Virginia public school district in the next academic year. Average mobility
rate refers to the average percentage of teachers or administrators who moved to a different
school district in the West Virginia public school system from the baseline year to the follow-up
year across each of the four periods observed.
Retention. Refers to staying in the same West Virginia school district from one year to the
next. Average retention rate refers to the average percentage of teachers or administrators who
stayed in the same school district from the baseline year to the follow-up year across each of
the four periods observed.
Teachers. Staff who work in classroom instructional roles.

Box 2. Data and methods
This study used data provided by the West Virginia Department of Education. The data includ­
ed every administrator and teacher employed by the state of West Virginia in academic years
2008/09, 2009/10, 2010/11, 2011/12, and 2012/13. The dataset included information
about their gender, years of experience, highest degree earned, National Board for Professional
Teaching Standards certification status, annual salary, district/school level assignment, and
employment status. In addition, data on West Virginia school districts were obtained from the
National Center for Education Statistics Common Core of Data for 2008/09–2011/121 (U.S.
Department of Education, Institute of Education Sciences, National Center for Education Sta­
tistics, 2013). These data included district locales, student enrollment, percentage of students
who were racial/ethnic minority students, percentage of students who were eligible for the
school lunch program, and percentage of students who were limited English proficient students.
To determine the status of teachers and administrators for this study, each teacher or
administrator was assigned to a single professional role each year in order to compare his or
her employment from one year to the next. Teachers who assumed any type of administrative
position were included in the teacher retention rate if they were employed in the same school
district. Teachers who moved to another school district to become administrators were included
in the mobility rate. A similar approach was used for administrators who returned to classroom
teaching in the same school district or moved to a different school district to return to teaching.
For the study every teacher and administrator in West Virginia was assigned to a single
school district location in each year analyzed. (Those who worked in more than one district
were assigned to the one in which they worked the most.) This assignment then was compared
from one year to the next to determine whether the teacher or administrator was retained in
(continued)
3

 Box 2. Data and methods (continued)
the same school district, left the West Virginia public school system, or moved to a different
school district in West Virginia. Retention, attrition, and mobility rates were calculated for each
year-to-year comparison and then averaged across the four comparisons.
Beginning teachers in each of the baseline years were identified from the entire population
of teachers, and their employment status in the baseline year was compared with follow-up
years. That is, it was determined whether teachers who began teaching in 2008/09 were still
teaching in 2009/10, 2010/11, 2011/12, and 2012/13; whether teachers who began teach­
ing in 2009/10 were still teaching in 2010/11, 2011/12, and 2012/13; and so on. Attrition
rates were calculated by dividing the number of beginning teachers who left the system in each
academic year by the total number of beginning teachers identified (see appendix A for more
detail on this analysis).
Cross-tabulations were created to compare average retention, attrition, and mobility rates
among teachers and administrators across the different characteristics listed above. Similar
cross-tabulations compared average retention, attrition, and mobility rates based on the dis­
trict characteristics listed.
A 2  percentage point difference was used as the threshold for identifying (and discuss­
ing) substantive differences between comparison groups (for example, female and male; rural,
town, suburban, or city; and so on). This threshold is arbitrary, selected on the basis that it
would yield the most policy-relevant information. Differences of less than 2 percentage points
are not highlighted in the report narrative, but all comparisons are shown in the tables.
Note
1. Data from the Common Core of Data from 2012/13 were not used because the study did not calculate re­
tention and attrition rates for the 2012/13 academic year and because 2012/13 district characteristics were
not required for the analysis.

What the study found
This section summarizes average retention, attrition, and mobility rates among teachers
and administrators overall and by district, average retention and attrition rates’ relation­
ship with teacher and administrator characteristics and with district characteristics, and
annual and average attrition rates for beginning teachers.
Close to 90 percent of teachers and administrators stayed in their West Virginia school district
during the study period

On average 90.1  percent of teachers stayed in the same school district from one year to
the next over the study period, 9.0  percent left the West Virginia public school system,
and 0.9  percent moved to a different school district in the system. Among administrators
87.7  percent stayed in the same school district, 11.1  percent left the West Virginia public
school system, and 1.2 percent moved to a different school district in the system.
Average retention, attrition, and mobility rates varied by district, but rates among administrators
varied more than rates among teachers

Across the 55 West Virginia public school districts, average district-level retention rates
among teachers ranged from 83.0  percent to 93.6  percent, average district-level attrition
4

 rates among teachers ranged from 6.0  percent to 14.9  percent, and average district-level
mobility rates among teachers ranged from 0.2 percent to 3.9 percent (map 1; see also table
B1 in appendix B). Average district-level retention rates among administrators ranged from
66.2  percent to 96.7  percent, average district-level attrition rates among administrators
ranged from 3.3  percent to 31.2  percent, and average district-level mobility rates among
administrators were as high as 9.8 percent, though a few districts had rates at or close to
zero (map 2; see also table B2 in appendix B).
Teachers left the West Virginia public school system at different rates depending on their years of
experience, highest degree earned, national board certification status, and annual salary

The rate at which teachers left the West Virginia public school system varied by at least
2 percentage points depending on some of the teacher characteristics analyzed (table 1).
Years of experience. Teachers with fewer than 4 years of experience and teachers with 15
or more years of experience left the public school system at higher rates than did teachers
with 4–9 or 10–14 years of experience.
Highest degree earned and national board certification status. Teachers with a doctoral
degree left the public school system at a higher rate than did teachers with a master’s or
bachelor’s degree. Teachers without National Board for Professional Teaching Standards
Map 1. Average retention rates among teachers vary across the 55 West Virginia
public school districts, 2008/09–2012/13
Average retention rate (percent)

66–70
71–80
81–90
91–97

Note: See table B1 in appendix B for a complete list of average retention, attrition, and mobility rates among
teachers in each school district.
Source: Authors’ calculations based on school personnel data provided by the West Virginia Department of
Education.

5

Across the 55 West
Virginia public
school districts,
average districtlevel retention
rates among
teachers ranged
from 83.0 percent
to 93.6 percent

 Map 2. Average retention rates among administrators vary across the 55 West
Virginia public school districts, 2008/09–2012/13
Average retention rate (percent)

66–70
71–80
81–90
91–97

Note: See table B2 in appendix B for a complete list of average retention, attrition, and mobility rates among
administrators in each school district.
Source: Authors’ calculations based on school personnel data provided by the West Virginia Department of
Education.

certification left the public school system at a higher rate (9.0 percent) than did teachers
with the certification (6.9 percent).
Annual salary. Teachers earning the highest and lowest annual salaries left the public
school system at the highest rates.
Districts with a larger proportion of students eligible for the federal school lunch program had
higher teacher attrition rates than did districts with a smaller proportion

Average attrition rates among teachers did not differ across most district characteristics
—locale, student enrollment, percentage of students who are racial/ethnic minority stu­
dents, and percentage of students who are limited English proficient students. However,
teachers did leave the West Virginia public school system at different rates across districts
depending on the percentage of students eligible for the federal school lunch program. Dis­
tricts with a larger proportion of students eligible for the school lunch program had higher
teacher attrition rates than did districts with a smaller proportion (table 2).

6

Across the 55 West
Virginia public
school districts,
average districtlevel retention
rates among
administrators
ranged from
66.2 percent to
96.7 percent

 Table 1. Average retention and attrition rates among teachers in West Virginia
public school districts, by teacher characteristic, 2008/09–2012/13
Characteristic

Average number
of teachers

Average retention rate
(percent)

Average attrition rate
(percent)

Gender
Female
Male

15,004

90.5

8.6

4,443

88.6

10.1
10.4

Years of experience
Fewer than 4 years

3,312

86.9

4–9 years

4,035

93.3

5.6

10–14 years

2,387

95.1

4.3

15 or more years

9,713

88.6

11.1

8,995

90.7

8.2

Highest degree earned
Bachelor’s degree
Master’s degree

10,372

89.6

9.7

Doctoral degree

64

85.9

13.3

National Board for Professional Teaching Standards certification
Yes

289

92.6

6.9

18,188

90.0

9.0

Less than $37,082

4,749

88.7

9.1

No
Annual salary

a

$37,082–$44,722

5,368

93.9

5.1

$44,723–$51,861

4,413

93.3

6.4

More than $51,861

4,918

84.4

15.4

School level taught
Elementary school

8,062

91.0

8.4

Middle school

3,881

90.4

8.6

High school

5,027

89.2

9.6

Combination schoolb

2,081

89.4

9.6

Note: The average number of teachers is presented as a reference and may not sum to the same total in each
category of characteristics because of missing data, because of rounding, or because teachers taught at more
than one school level. Percentages do not sum to 100 because the table excludes teachers who moved to
another West Virginia public school district.
a. Reflects salary quartiles from the first year of data (2008/09).
b. Combination schools serve students in a variety of grade-level configurations and thus could not be classi­
fied as elementary, middle, or high school.
Source: Authors’ calculations based on teacher personnel data provided by the West Virginia Department of
Education.

Administrators left the West Virginia public school system at different rates depending on their
gender, years of experience, highest degree earned, national board certification status, annual
salary, and district- or school-level assignment

The rate at which administrators left the West Virginia public school system varied by at
least 2 percentage points for every administrator characteristic analyzed (table 3).
Gender. Female administrators left the public school system at a lower rate than male
administrators (8.8 percent versus 13.2 percent).

7

Teachers with
fewer than 4 years
of experience and
teachers with 15
or more years of
experience left
the public school
system at higher
rates than did
teachers with 4–9
or 10–14 years
of experience

 Table 2. Average retention and attrition rates among teachers in West Virginia
public school districts, by district characteristic, 2008/09–2012/13
Characteristic

Average number
of teachers

Average retention rate
(percent)

Average attrition rate
(percent)

Locale
Rural

7,938

89.7

9.1

Town

5,289

90.0

9.0

Suburban

3,499

90.1

9.3

City

2,722

91.2

8.2

Fewer than 1,976 students

1,334

88.3

9.9

1,976–3,862 students

2,669

88.8

9.4

3,863–6,507 students

4,703

90.2

9.1

10,741

90.6

8.7

Student enrollment

More than 6,507 students

Percentage of students who are racial/ethnic minority students
Less than 1.5

1,348

89.5

9.3

1.5–2.2

3,301

90.5

8.3

2.3–5.1

4,459

89.9

9.1

10,339

90.1

9.1

More than 5.1

Percentage of students eligible for the federal school lunch program
Less than 47.7

4,692

91.9

7.5

47.7–52.5

7,082

90.5

8.7

52.6–57.5

3,664

89.3

9.6

More than 57.5

3,988

87.9

10.7

6,769

89.1

9.6

12,678

90.6

8.6

Percentage of students who are limited English proficient
Below median
(below 0.1 percent)
Median or higher
(0.1 percent or higher)

Note: Average number of teachers is presented as a reference and may not sum to the same total in each
category of characteristics because of missing data or because of rounding. Percentages do not sum to 100
because the table excludes teachers who moved to another West Virginia public school district. Ranges reflect
school district characteristics as of the 2008/09 academic year and represent quartiles in that year.
Source: Authors’ calculations based on teacher personnel data provided by the West Virginia Department of
Education and school district information obtained from the U.S. Department of Education, National Center for
Education Statistics, Common Core of Data.

Years of experience. Administrators with fewer than 4 years of experience and adminis­
trators with 15 or more years of experience left the public school system at about twice the
rate of administrators with 4–9 or 10–14 years of experience.
Highest degree earned and national board certification status. Administrators with
higher degrees had higher average attrition rates. Administrators without National Board
for Professional Teaching Standards certification left the public school system at a rate more
than double that of administrators with the certification (11.3 percent versus 4.7 percent).
Annual salary. Administrators earning an annual salary of more than $69,274 left the
public school system at a rate almost double that of administrators with an annual salary of
less than $69,275 (13.9 percent versus 8.1 percent).

8

Districts with a
larger proportion
of students eligible
for the school
lunch program had
higher teacher
attrition rates than
did districts with a
smaller proportion

 Table 3. Average retention and attrition rates among administrators in West
Virginia public school districts, by administrator characteristic, 2008/09–2012/13
Characteristic

Average number of
administrators

Average retention rate
(percent)

Average attrition rate
(percent)

Gender
Female

611

90.3

8.8

Male

639

85.2

13.2
13.9

Years of experiencea
Fewer than 4 years

8

80.6

4–9 years

115

90.0

7.6

10–14 years

149

92.6

6.1

15 or more years

978

86.8

12.2

23

92.2

7.8

Highest degree earned
Bachelor’s degree
Master’s degree

1,183

87.9

11.0

Doctoral degree

42

80.5

14.8

National Board for Professional Teaching Standards certification
Yes

21

94.1

4.7

1,172

87.6

11.3

Less than $61,482

314

90.0

8.1

No
Annual salaryb
$61,482–$69,274

305

90.8

8.3

$69,275–$77,346

301

85.1

13.9

More than $77,346

330

85.0

14.0

District- or school-level assignment
Central office

127

81.8

15.4

Elementary school

471

88.8

10.7

Middle school

238

90.0

9.0

High school

296

87.7

10.5

Combination schoolc

119

84.8

13.5

Note: The average number of administrators is presented as a reference and may not sum to the same total in
each category of characteristics because of missing data or because of rounding. Percentages may not sum to
100 because the table excludes teachers who moved to another West Virginia public school district.
a. Includes past experience as a teacher in West Virginia public schools, if applicable.
b. Reflects salary quartiles from 2008/09.
c. Combination schools serve students in a variety of grade-level configurations and thus could not be classi­
fied as elementary, middle, or high school.
Source: Authors’ calculations based on administrator personnel data provided by the West Virginia Depart­
ment of Education.

District/school level assignment. Administrators working in the central office left
the public school system at the highest rate (15.4  percent). Administrators working
in elementary, middle, and high schools left the system at a rate about a third lower
(9.0–10.7  percent). Administrators in combination schools (schools serving students
in a variety of grade-level configurations) left at a rate similar to that of administrators
working in the central office.

9

Female
administrators left
the public school
system at a lower
rate than male
administrators
(8.8 percent versus
13.2 percent)

 Administrators left the West Virginia public school system at different rates depending on district
locale, student enrollment, percentage of students who are racial/ethnic minority students,
percentage of students eligible for the federal school lunch program, and percentage of students
who are limited English proficient students

The rate at which administrators left the West Virginia public school system varied by at
least 2 percentage points for every district characteristic analyzed (table 4).
Locale. Administrators in school districts located in rural areas and towns left the public
school system at higher rates than did administrators in suburban and city school districts.
Student enrollment. Administrators in districts serving 1,976–3,862 students left the
public school system at the highest rate. Administrators in districts serving fewer than
Table 4. Average retention and attrition rates among administrators in West 

Virginia public school districts, by district characteristic, 2008/09–2012/13

Characteristic

Average number of
administrators

Average retention rate
(percent)

Average attrition rate
(percent)

Locale
Rural

531

85.7

12.4

Town

336

86.8

11.8

Suburban

220

90.9

8.9

City

163

91.6

7.9

98

84.2

12.8

Student enrollment
Fewer than 1,976 students
1,976–3,862 students

187

81.4

15.5

3,863–6,507 students

318

88.2

10.8

More than 6,507 students

647

89.8

9.6

79.7

17.5
11.3

Percentage of students who are racial/ethnic minority students
Less than 1.5

89

1.5–2.2

229

86.4

2.3–5.1

305

88.5

10.6

More than 5.1

627

88.9

10.3

Percentage of students eligible for the federal school lunch program
Less than 47.7

283

91.5

7.9

47.7–52.5

440

88.5

10.3

52.6–57.5

248

87.7

11.7

More than 57.5

277

82.6

14.9

Percentage of students who are limited English proficient students
Below median
(below 0.1 percent)

466

84.6

13.1

Median or higher
(0.1 percent or higher)

784

89.5

9.9

Note: Average number of administrators is presented as a reference and may not sum to the same total in
each category of characteristics because of rounding or because of missing data. Percentages in rows do not
sum to 100 percent because the table excludes teachers who moved to another West Virginia public school
district. Ranges reflect school district characteristics as of the 2008/09 academic year and represent quar­
tiles in that year.
Source: Authors’ calculations based on administrator personnel data provided by the West Virginia Depart­
ment of Education and school district information obtained from the U.S. Department of Education, National
Center for Education Statistics, Common Core of Data.

10

Administrators in
school districts
located in rural
areas and towns
left the public
school system at
higher rates than
did administrators
in suburban and
city school districts

 1,976 students left at a somewhat lower rate, and administrators in districts serving more
than 3,862 students left at the lowest rate.
Percentage of students who are racial/ethnic minority students. Administrators in
schools with the lowest percentage of students who are racial/ethnic minority students
(less than 1.5 percent) left the public school system at the highest rate. This pattern differs
from previous research, which has shown that administrators tend to leave schools and
districts serving a larger proportion of racial/ethnic minority students (Gates et al., 2006;
Ringel et al., 2004).
Percentage of students eligible for the federal school lunch program. The percentage of
administrators who left the public school system was nearly twice as high in districts where
more than 57.5 percent of students were eligible for the federal school lunch program as in
districts where less than 47.7 percent were eligible.
Percentage of students who are limited English proficient students. Administrators in
districts serving a smaller percentage of limited English proficient students left the public
school system at a higher rate than did administrators in districts serving a larger percent­
age of limited English proficient students.
The average mobility rate among teachers in West Virginia school districts was less than 3 percent
across all personal characteristics and varied depending on teachers’ years of experience and
annual salary

For all personal characteristics the average mobility rate was lower than the average attri­
tion rate and was more than 80 percentage points lower than the average retention rate
(see box 1 for definitions of mobility and attrition). This suggests that teachers in the West
Virginia public school system who leave their school district are leaving the system and
not simply moving to other West Virginia school districts (table 5).
Years of experience. Teachers with fewer than 4 years of experience moved to a different
West Virginia school district at a higher rate than did teachers with 10–14 years of experi­
ence and teachers with 15 or more years of experience.
Annual salary. Teachers earning an annual salary of less than $37,082 moved to a dif­
ferent district at a higher rate than did teachers earning an annual salary of more than
$51,861.
The average mobility rate among administrators in West Virginia school districts was less than
6 percent across all personal characteristics and varied depending on administrators’ years of
experience and highest degree earned

Across all personal characteristics the average mobility rate among administrators was
lower than the average attrition rate, except for national board certification status, and was
more than 75 percentage points lower than any average retention rate. This suggests that
administrators in the West Virginia public school system are leaving the system and not
simply moving to other West Virginia school districts (table 6).

11

The percentage
of administrators
who left the public
school system was
nearly twice as
high in districts
where more than
57.5 percent of
students were
eligible for the
federal school
lunch program
as in districts
where less than
47.7 percent
were eligible

 Table 5. Average mobility rates among teachers in West Virginia public school
districts, by personal characteristic, 2008/09–2012/13
Characteristic

Average number of
teachers

Average mobility rate
(percent)

Gender
Female
Male

15,004

0.9

4,443

1.3

Years of experience
Fewer than 4 years

3,312

2.7

4–9 years

4,035

1.1

10–14 years

2,387

0.6

15 or more years

9,713

0.3

8,995

1.1

Highest degree earned
Bachelor’s degree
Master’s degree

10,372

0.7

Doctoral degree

64

0.8

National Board for Professional Teaching Standards certification
Yes

289

0.5

18,188

1.0

Less than $37,082

4,749

2.2

$37,082–$44,722

5,368

1.0

$44,723–$51,861

4,413

0.3

More than $51,861

4,918

0.2

No
Annual salary

a

School level taught
Elementary school

8,062

0.6

Middle school

3,881

1.0

High school

5,027

1.2

Combination schoolb

2,081

1.0

Note: Average number of teachers is presented as a reference and may not sum to the same total in each
category of characteristics because of missing data, because of rounding, or because teachers taught at more
than one school level.
a. Reflects salary quartiles from 2008/09.
b. Combination schools serve students in a variety of grade-level configurations and thus could not be classi­
fied as elementary, middle, or high school.
Source: Authors’ calculations based on personnel data provided by the West Virginia Department of
Education.

Years of experience. Administrators with fewer than four years of experience moved to a
different West Virginia school district at more than twice the rate that administrators with
four or more years of experience did.
Highest degree earned. Administrators with a doctoral degree moved to a different
school district at four times the rate that administrators with a bachelor’s or master’s
degree did.
Level assigned. Administrators working in the central office moved to a different school
district at a higher rate than administrators working in elementary schools did but at about
the same rate that administrators in middle, high, and combination schools did.

12

Teachers with
fewer than 4 years
of experience
moved to a
different West
Virginia school
district at a
higher rate than
did teachers with
10–14 years of
experience and
teachers with 15
or more years
of experience

 Table 6. Average mobility rates among administrators in West Virginia public
school districts, by personal characteristic, 2008/09–2012/13
Characteristic

Average number of
administrators

Average mobility rate
(percent)

Gender
Female

611

0.9

Male

639

1.6

8

5.5

Years of experiencea
Fewer than 4 years
4–9 years

115

2.4

10–14 years

149

1.3

15 or more years

978

1.0

23

0.0

Master’s degree

1,183

1.1

Doctoral degree

42

4.7

Highest degree earned
Bachelor’s degree

National Board for Professional Teaching Standards certification
Yes

21

1.2

1,172

1.1

Less than $61,482

314

1.9

$61,482–$69,274

305

0.9

$69,275–$77,346

301

1.0

More than $77,346

330

1.0

Central office

127

2.8

Elementary school

471

0.5

No
Annual salary

b

District level or school level assigned

Middle school

238

1.0

High school

296

1.8

Combination schoolc

119

1.7

Note: Average number of administrators is presented as a reference and may not sum to the same total in
each category of characteristics because of missing data or because of rounding.
a. Includes past experience as a teacher in West Virginia public schools, if applicable.
b. Reflects salary quartiles from 2008/09.
c. Combination schools serve students in a variety of grade-level configurations and thus could not be classi­
fied as elementary, middle, or high school.
Source: Authors’ calculations based on personnel data provided by the West Virginia Department of
Education.

Teachers moved to a different district at about the same rate regardless of the characteristics of
their baseline district

Regardless of the district characteristic compared, average teacher mobility rates were
about the same, ranging from 0.6 percent to 1.8 percent (table 7).

13

Administrators
with fewer than
four years of
experience moved
to a different
West Virginia
school district at
more than twice
the rate that
administrators with
four or more years
of experience did

 Table 7. Average mobility rates among teachers in West Virginia public school
districts, by district characteristic, 2008/09–2012/13
Characteristic

Average number of
teachers

Average mobility rate
(percent)

Locale
Rural

7,938

1.2

Town

5,289

1.0

Suburban

3,499

0.6

City

2,722

0.6

Fewer than 1,976 students

1,334

1.8

1,976–3,862 students

2,669

1.8

3,863–6,507 students

4,703

0.7

10,741

0.7

Student enrollment

More than 6,507 students

Percentage of students who are racial/ethnic minority students
Less than 1.5

1,348

1.2

1.5–2.2

3,301

1.2

2.3–5.1

4,459

1.0

10,339

0.8

More than 5.1

Percentage of students eligible for the federal school lunch program
Less than 47.7

4,692

0.6

47.7–52.5

7,082

0.8

52.6–57.5

3,664

1.1

More than 57.5

3,988

1.4

Percentage of students who are limited English proficient students
Below median
(below 0.1 percent)
Median or higher
(0.1 percent or higher)

6,769

1.3

12,678

0.8

Note: Average number of teachers is presented as a reference and may not sum to the same total in each
category of characteristics because of missing data or because of rounding. Ranges reflect school district
characteristics as of the 2008/09 academic year and represent quartiles in that year.
Source: Authors’ calculations based on teacher and administrator personnel data provided by the West Vir­
ginia Department of Education and school district information obtained from the U.S. Department of Educa­
tion, National Center for Education Statistics, Common Core of Data.

Administrators moved to a different district at varying rates based on the characteristics of the
students the district served

Administrators moved to a different school district at different rates depending on district
characteristics, including student enrollment and percentage of students who are racial/
ethnic minority students (table 8).
Student enrollment. Administrators in West Virginia school districts enrolling fewer stu­
dents changed districts at a higher rate than did administrators in districts enrolling a
larger number of students.
Percentage of students who are racial/ethnic minority students. Administrators in dis­
tricts serving a smaller proportion of racial/ethnic minority students had higher average
mobility rates than did administrators in districts serving a larger proportion of racial/
ethnic minority students.
14

Administrators
in West Virginia
school districts
enrolling fewer
students changed
districts at a
higher rate than
did administrators
in districts
enrolling a larger
number of students

 Table 8. Average mobility rates among administrators in West Virginia public
school districts, by district characteristic, 2008/09–2012/13
Characteristic

Average number of
administrators

Average mobility rate
(percent)
1.9

Locale
Rural

531

Town

336

1.4

Suburban

220

0.2

City

163

0.5

98

3.0

Student enrollment
Fewer than 1,976 students
1,976–3,862 students

187

3.1

3,863–6,507 students

318

1.0

More than 6,507 students

647

0.6

Percentage of students who are racial/ethnic minority students
Less than 1.5

89

2.8

1.5–2.2

229

2.3

2.3–5.1

305

0.9

More than 5.1

627

0.8

Percentage of students eligible for the federal school lunch program
Less than 47.7

283

0.6

47.7–52.5

440

1.2

52.6–57.5

248

0.6

More than 57.5

277

2.5

Percentage of students who are limited English proficient students
Below median
(below 0.1 percent)

466

2.3

Median or higher
(0.1 percent or higher)

784

0.6

Note: Average number of administrators is presented as a reference and may not sum to the same total in
each category of characteristics because of missing data or because of rounding. Ranges reflect school dis­
trict characteristics as of the 2008/09 academic year and represent quartiles in that year.
Source: Authors’ calculations based on teacher and administrator personnel data provided by the West Vir­
ginia Department of Education and school district information obtained from the U.S. Department of Educa­
tion, National Center for Education Statistics, Common Core of Data.

About a fifth of beginning teachers left the West Virginia public school system after their first year of
teaching, and a third left by the end of the fourth year

On average 19.5 percent of beginning teachers who were initially employed in the West
Virginia public school system during one of the four baseline years left the system after
their first year of teaching. Among those newly hired in the 2008/09, 2009/10, or 2010/11
academic years, 26.8 percent left after two years of teaching. Among those first hired in
2008/09 or 2009/10, 30.8  percent left after three years, and of those hired in 2008/09,
32.0 percent left after four years (were no longer employed in 2012/13; table 9). Cumulative
attrition rates among beginning teachers in each district are provided in appendix C.

15

Administrators in
districts serving a
smaller proportion
of racial/ethnic
minority students
had higher average
mobility rates than
did administrators
in districts serving
a larger proportion
of racial/ethnic
minority students

 Table 9. Percentage of beginning teachers who left the West Virginia public school
system after each year of teaching, 2008/09–2012/13
Baseline
year

Number of
beginning
teachers

Cumulative attrition rate
(percent)
After one year

After two years

After three years

After four years
32.0

2008/09

807

20.1

23.9

28.5

2009/10

811

21.6

28.5

33.1

na

2010/11

832

20.7

28.0

na

na

2011/12

776

15.7

na

na

na

Average

807

19.5

26.8

30.8

32.0

na is not applicable because beginning teachers had not yet taught that many years by 2012/13.
Source: Authors’ calculations based on data provided by the West Virginia Department of Education.

Implications of the study findings
The study’s findings are important for two reasons. First, the findings contribute to the
sparse literature on retention, attrition, and mobility among teachers and administrators
at the school district level in West Virginia. The findings indicate that the teacher and
administrator workforce is largely stable. On average West Virginia public school districts
retain 90 percent of teachers and 88 percent of administrators from one year to the next;
however, a third of beginning teachers who started teaching in the West Virginia public
school system in 2008/09 left by 2012/13. Furthermore, administrators with fewer than
4 years of experience and administrators with 15 or more years of experience left the system
at double the rate that other administrators did (about 13 percent versus about 7 percent).
Second, the study findings can be used to inform state and district policy initiatives that
aim to improve teacher and administrator retention in school districts serving specific
student populations. For example, the average attrition rate among teachers and admin­
istrators was highest in school districts with the highest proportion of students eligible for
the federal school lunch program. Policies providing resources for retaining personnel in
these districts might lower the attrition rates. Likewise, upgraded hiring practices, better
recruitment efforts, more competitive salary structures, and adequate investment in pro­
fessional development and mentoring might increase teacher and administrator retention
(Brill & McCartney, 2008; Odden, 2011; Smith & Ingersoll, 2004). At the state level the
information could be used to inform policies aimed at improving retention of newer teach­
ers (teachers who were identified as beginning teachers in one of the years covered in this
study)—who left the West Virginia public school system at a higher rate. About 32 percent
of beginning teachers employed by the West Virginia public school system in 2008/09 left
the system by 2012/13, compared with 9 percent of teachers overall.1
The data in this study can inform actions at the district and state levels, which might
relate to estimating the cost of teacher and administrator attrition. Similarly, the findings
can be used to identify differences in attrition and mobility rates that might be linked
to district-level initiatives aimed at recruiting, retaining, and improving human resource
management practices.

16

On average
19.5 percent of
beginning teachers
who were initially
employed in the
West Virginia
public school
system during
one of the four
baseline years
left the system
after their first
year of teaching

 Limitations of the study
The dataset did not include information describing why individual teachers or adminis­
trators moved to a different school district or left the West Virginia public school system.
Without such information, the reason these teachers or administrators made the decisions
can be inferred only from patterns in rates across personal or district characteristics. Addi­
tional information about the reasons teachers and administrators left the West Virgin­
ia public school system would allow policy and programmatic responses to target those
working conditions, increasing retention.
The dataset also lacked information describing whether individual teachers and admin­
istrators left the public school system permanently, moved to a different state, or left the
public school system to work in a private school. This information would provide import­
ant insights about the condition of the teacher and administrator labor market in West
Virginia.
Further, the dataset did not contain information on the effectiveness of teachers and school
administrators who stayed in the same school district, moved to a different school district,
or left the West Virginia public school system. Thus, the study findings cannot address the
comparative effectiveness of the teachers and administrators who were retained by districts
throughout West Virginia and those who were not.
Finally, the study findings are descriptive and thus cannot assess the impact that higher
attrition or mobility rates may have on school district performance.

17

The findings
contribute to the
sparse literature
on retention,
attrition, and
mobility among
teachers and
administrators
at the school
district level in
West Virginia
and indicate that
the teacher and
administrator
workforce is
largely stable

 Appendix A. Data and methodology
This appendix describes the data sources and details the study methodology used to com­
plete this analysis.
Variables and sources

Teacher, administrator, and school district variables and their data sources are presented in
table A1. All variables were assigned to teachers and administrators in each baseline year
and follow-up year (2008/09, 2009/10, 2010/11, 2011/12, and 2012/13). Full-time equivalent
status was used to determine whether a teacher or administrator was employed within the
West Virginia public school system.
Description of the study population

The study included the entire population of teachers and administrators employed in West
Virginia public school districts in 2008/09, 2009/10, 2010/11, 2011/12, and 2012/13. The
dataset contained 77,787 teachers and 4,998 administrators in 55 school districts.

Table A1. Variables used in the retention, attrition, and mobility analyses
Variable

Source

Teacher and administrator variables
Teacher or administrator ID

West Virginia Department of Education

National Board for Professional Teaching Standards certification status

West Virginia Department of Education

Gender

West Virginia Department of Education

Highest degree earned

West Virginia Department of Education

Years of experiencea
Teacher or administrator position code

West Virginia Department of Education
b

West Virginia Department of Education

Annual salary

West Virginia Department of Education

Full-time equivalent statusc

West Virginia Department of Education

District or school level (central office, elementary, middle, high, combination)

West Virginia Department of Education

School district variables
District ID

West Virginia Department of Education/
National Center for Education Statistics

Locale (rural, town, suburban, or city)

National Center for Education Statistics

Total student enrollment

National Center for Education Statistics

Percentage of students who are racial/ethnic minority students

National Center for Education Statistics

Percentage of students eligible for the federal school lunch program

National Center for Education Statistics

Percentage of students who are limited English proficient students

National Center for Education Statistics

Note: Variable names presented in the table have been revised for clarity. Actual variable names contained in the dataset provided by
the West Virginia Department of Education were abbreviated.
a. Includes past experience as a teacher in West Virginia public schools, if applicable, for administrators.
b. Refers to a teacher’s or administrator’s primary area of responsibility. This code was used to filter from the dataset employees who
were not working as teachers or administrators.
c. Approximates the amount of time a teacher or administrator spends in a particular role, school faculty, or school district. Teachers
and administrators were assigned to the district in which the majority of their full-time equivalent assignment was located.
Source: Variables were provided by the West Virginia Department of Education and obtained from the National Center for Education Sta­
tistics Common Core of Data (U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics,
2013).

A-1

 Methods for descriptive analyses

Before analysis began the quality of the data received from the West Virginia Department
of Education and the extent to which there were missing or invalid values were assessed.
SAS® was used to create frequency counts of values for each of the key variables in the
dataset provided by the West Virginia Department of Education. The frequencies included
the number of nonmissing values, the number of values outside the valid range, and the
number and percentage of missing values. The information on missing data was used to
identify incomplete records for removal. Less than 5 percent of values were missing within
and across all variables. Teacher and administrator records containing missing data were
excluded from the analyses only when applicable data were missing (that is, pairwise dele­
tion was used).
Next, individual teachers and administrators were identified. The dataset included dupli­
cate personnel records reflecting teachers and administrators with multiple assignments in
a school district. Duplicate records posed a challenge for these analyses, as teachers and
administrators who appear in more than one record can inflate retention, attrition, and
mobility estimates. To eliminate duplicate records and ensure unduplicated counts, teach­
ers and administrators were assigned to a single school district based on their full-time
equivalent assignment during that baseline year.
The school district where the majority of the teacher’s or administrator’s full-time equiv­
alent assignment was located was considered his or her primary district assignment and
was used to determine whether the teacher or administrator stayed in the district, left
the school system, or moved to another district in the follow-up year. In the event that a
teacher’s or administrator’s assignment was equally divided between two school districts,
the district to which the largest proportion of salary could be attributed was designated as
the primary assignment. A similar approach has been used elsewhere in statewide analyses
of teacher retention, which have focused primarily on retention, attrition, and mobility
at the school level (Lochmiller, Sugimoto, & Muller, 2016; Plecki, Elfers, Loeb, Zahir, &
Knapp, 2005).
To further prevent duplicate entries, the position code field was used to exclude employees
for whom a portion of their assignments did not involve instruction or school adminis­
tration. The position code field was used not in the descriptive analyses but as a filtering
mechanism to identify teachers and administrators as of the baseline year. This helped
ensure that the analyses focused on retention, attrition, and mobility among teachers and
administrators and did not include staff working in other roles.
Next, the dataset was merged with district-level information obtained from the National
Center for Education Statistics, Common Core of Data. Unique identification numbers for
school districts were used to link individual teacher and administrator information with
district information. The merge resulted in a single dataset that included teacher, adminis­
trator, and district characteristics.
Calculating retention, attrition, and mobility rates. After data were merged, the analy­
ses involved calculating frequencies and percentages among teachers and administrators
who stayed in the same school district, left the West Virginia public school system, or
moved to another West Virginia school district—from one baseline year to the follow-up
A-2

 year. Retention, attrition, and mobility rates were calculated for four baseline-to-follow-up
periods: from 2008/09 to 2009/10, from 2009/10 to 2010/11, from 2010/11 to 2011/12, and
from 2011/12 to 2012/13. Average rates represent the mean of these four separate one-year
rates.
Calculating attrition rates among beginning teachers. Beginning teachers were defined
as having zero years of teaching experience in their baseline year and were identified from
the entire population of teachers in each academic year 2008/09, 2009/10, 2010/11, and
2011/12. The number of beginning teachers identified ranged from 776 to 832, depend­
ing on the year. Each beginning teacher’s initial employment status was compared with
each subsequent academic year. For example, teachers who began teaching in 2008/09
were identified in 2009/10, 2010/11, 2011/12, and 2012/13 to determine whether they were
still teachers or had left the West Virginia public school system; beginning teachers in
2009/10 were compared with 2010/11, 2011/12, and 2012/13; and so on. Teachers who were
not found in the dataset as teachers in any follow-up year were considered to have left the
system. These annual attrition rates were calculated by dividing the number of beginning
teachers who left the public school system in each follow-up year by the total number
of beginning teachers identified as of the baseline year (2008/09, 2009/10, 2010/11, or
2011/12).

A-3

 Appendix B. Average retention, attrition, and mobility rates among
teachers and administrators by West Virginia public school district
This appendix provides information about district-level retention, attrition, and mobility
among teachers and administrators in West Virginia public schools. The data may be of
interest to school district administrators seeking information about their districts, as well
as to other stakeholders who are interested in knowing more specifically how retention,
attrition, and mobility rates vary across the 55 West Virginia public school districts.
Table B1. Average retention, attrition, and mobility rates among teachers in West Virginia public
school districts, 2008/09–2012/13
District

Average number
of teachers

Average retention rate
(percent)

Average attrition rate
(percent)

Average mobility rate
(percent)

Barbour

653

88.7

9.0

Berkeley

4,981

92.3

7.1

0.6

Boone

1,495

90.4

8.7

0.9

Braxton

706

88.4

8.8

2.8

Brooke

980

90.7

8.8

0.5

Cabell

2.3

3,353

90.8

8.6

0.6

Calhoun

340

85.6

12.6

1.8

Clay

569

92.8

6.5

0.7

Doddridge
Fayette
Gilmer
Grant
Greenbrier
Hampshire
Hancock
Hardy

355

89.9

8.7

1.4

2,010

86.2

12.3

1.5

261

84.3

14.6

1.1

514

85.4

10.7

3.9

1,461

88.3

10.7

1.0

931

87.4

10.4

2.2

1,182

91.8

8.0

0.2

543

92.6

7.2

0.2
0.8

Harrison

2,865

91.4

7.8

Jackson

1,335

92.1

7.3

0.6

Jefferson

2,281

91.3

7.9

0.8

Kanawha

7,824

88.4

10.9

0.7

696

89.7

7.6

2.7

Lewis
Lincoln

1,080

85.4

12.7

1.9

Logan

1,709

88.1

10.7

1.2

Marion

2,307

91.8

7.5

0.7

Marshall

1,319

90.8

8.8

0.4

Mason

1,251

88.6

10.3

1.1

McDowell

1,034

83.0

14.3

2.7

Mercer

2,583

89.1

10.6

0.3

Mineral

1,196

92.3

7.4

0.3

Mingo

1,291

83.7

14.9

1.4

Monongalia

2,832

90.9

8.3

0.8

549

88.2

10.6

1.2
1.3

Monroe
Morgan

661

90.5

8.2

Nicholas

1,173

91.0

8.2

0.8

Ohio

1,428

92.6

7.1

0.3

(continued)
B-1

 Table B1. Average retention, attrition, and mobility rates among teachers in West Virginia public
school districts, 2008/09–2012/13 (continued)
District
Pendleton

Average number
of teachers

Average retention rate
(percent)

Average attrition rate
(percent)

Average mobility rate
(percent)

90.4

8.1

1.5

333

Pleasants

409

87.3

11.0

1.7

Pocahontas

340

93.2

6.2

0.6

1,275

90.0

8.9

1.1
0.4

Preston
Putnam

2,552

93.6

6.0

Raleigh

3,233

91.2

8.1

0.7

Randolph

1,283

92.0

6.9

1.1

Ritchie

443

88.3

9.5

2.2

Roane

691

88.9

9.6

1.5

Summers

421

87.9

10.7

1.4

Taylor

612

91.0

7.5

1.5

Tucker

320

89.1

9.4

1.5

426

89.0

9.6

1.4

Upshur

Tyler

1,104

90.6

8.1

1.3

Wayne

2,053

91.2

8.3

0.5

Webster

452

88.7

9.3

2.0

Wetzel

833

90.3

7.8

1.9
1.6

309

91.3

7.1

Wood

Wirt

3,653

91.1

8.2

0.7

Wyoming

1,297

90.8

8.2

1.0

Note: Average number of teachers is presented as a reference. Percentages may not sum to 100 percent because of rounding.
Source: Authors’ analysis based on teacher personnel data provided by the West Virginia Department of Education.

Table B2. Average retention, attrition, and mobility rates among administrators in West Virginia public
school districts, 2008/09–2012/13
District

Average number of
administrators

Average retention rate
(percent)

Average attrition rate
(percent)

Average mobility rate
(percent)

Barbour

51

86.3

3.9

9.8

Berkeley

259

91.1

8.1

0.8

Boone

99

86.9

10.1

3.0

Braxton

53

77.4

15.1

7.5

Brooke

60

90.0

10.0

0.0

Cabell

0.9

213

89.7

9.4

Calhoun

28

82.1

10.7

7.2

Clay

44

86.4

13.6

0.0

Doddridge
Fayette
Gilmer

26

88.5

11.5

0.0

135

83.7

14.8

1.5

28

71.4

28.6

0.0

Grant

36

80.6

16.7

2.7

Greenbrier

94

85.1

14.9

0.0

Hampshire

56

83.9

14.3

1.8

(continued)
B-2

 Table B2. Average retention, attrition, and mobility rates among administrators in West Virginia public
school districts, 2008/09–2012/13 (continued)
District
Hancock
Hardy

Average number of
administrators

Average retention rate
(percent)

Average attrition rate
(percent)

Average mobility rate
(percent)

92.5

6.3

1.2

80
42

85.7

14.3

0.0

Harrison

181

88.4

10.5

1.1

Jackson

93

88.2

11.8

0.0

Jefferson

120

96.7

3.3

0.0

Kanawha

503

89.9

9.7

0.4

44

90.9

6.8

2.3

Lewis

69

69.6

27.5

2.9

Logan

Lincoln

138

88.4

11.6

0.0

Marion

0.8

134

85.8

13.4

Marshall

75

92.0

8.0

0.0

Mason

74

78.4

20.3

1.3

77

66.2

31.2

2.6

Mercer

McDowell

148

85.1

13.5

1.4

Mineral

89

88.8

11.2

0.0

Mingo

110

90.0

9.1

0.9

Monongalia

149

91.9

8.1

0.0
6.5

Monroe

31

67.7

25.8

Morgan

60

90.0

10.0

0.0

Nicholas

85

91.8

7.1

1.1

Ohio

87

90.8

9.2

0.0

Pendleton

23

91.3

4.3

4.4

Pleasants

39

87.2

5.1

7.7

Pocahontas

28

82.1

14.3

3.6

Preston

72

76.4

19.4

4.2

Putnam

169

94.1

5.9

0.0

Raleigh

183

89.1

10.4

0.5

88

87.5

11.4

1.1

Ritchie

36

88.9

11.1

0.0

Roane

45

71.1

24.4

4.5

Summers

26

96.2

3.8

0.0
9.3

Randolph

Taylor

43

81.4

9.3

Tucker

27

77.8

14.8

7.4

Tyler

24

91.7

8.3

0.0

Upshur

76

94.7

3.9

1.4

Wayne

148

87.2

12.2

0.6
3.5

Webster

29

79.3

17.2

Wetzel

56

80.4

17.9

1.7

Wirt

20

95.0

5.0

0.0

211

92.9

7.1

0.0

84

94.0

6.0

0.0

Wood
Wyoming

Note: Average number of administrators is presented as a reference. Percentages may not sum to 100 percent because of rounding.
Source: Authors’ analysis based on administrator personnel data provided by the West Virginia Department of Education.

B-3

 Appendix C. Cumulative attrition rates among 

beginning teachers in West Virginia public school districts

Table C1 provides cumulative attrition rates among beginning teachers in West Virginia
public school districts for 2008/09–2012/13.

Table C1. Cumulative attrition rates among beginning teachers who left the West Virginia public
school system, by district, 2008/09–2012/13

District

Number of
beginning teachers

Cumulative attrition rate (percent) as of 2012/13 among beginning
teachers first employed in a West Virginia public school district in:
2008/09

2009/10

2010/11

2011/12

Barbour

25

12.5

66.7

42.9

28.6

Berkeley

238

24.0

26.3

25.0

10.0

Boone

82

35.3

26.3

31.6

11.1

Braxton

34

20.0

40.0

35.7

40.0

Brooke

29

57.1

11.1

20.0

25.0

Cabell

157

41.9

40.7

38.3

15.4

Calhoun

27

66.7

22.2

50.0

20.0

Clay

22

40.0

0.0

37.5

25.0

a

Doddridge
Fayette

84

40.0

a

Gilmer
Grant

a

33

a

37.5

a

a

40.7

a

50.0

14.3

a

32.3

a

35.7

a

25.0

Greenbrier

69

27.3

20.0

13.6

0.0

Hampshire

54

25.0

20.0

18.8

31.3

Hancock

43

0.0

0.0

6.7

15.4

Hardy

12

50.0

20.0

33.3

0.0

Harrison

83

50.0

26.7

16.0

8.7

Jackson

36

25.0

25.0

55.6

14.3

Jefferson

105

26.1

21.1

35.7

30.8

Kanawha

278

50.0

59.0

43.8

9.5

30

20.0

33.3

33.3

0.0

Lincoln

64

33.3

25.0

19.1

16.7

Logan

101

30.8

45.5

37.5

3.9
13.6

Lewis

Marion

81

30.8

8.3

22.7

Marshall

25

60.0

33.3

7.1

0.0

Mason

65

34.8

68.8

40.0

50.0

McDowell

80

23.5

47.8

10.0

26.7

Mercer

97

62.1

20.0

27.8

10.0

Mineral

18

0.0

0.0

100.0

33.3

Mingo

108

25.0

33.3

53.9

22.2

Monongalia

107

38.5

40.0

13.3

7.7

Monroe

15

25.0

60.0

0.0

25.0

Morgan

30

40.0

12.5

0.0

50.0

Nicholas

39

35.7

25.0

20.0

14.3

Ohio

43

0.0

10.0

18.8

22.2

Pendleton

17

0.0

20.0

50.0

0.0

(continued)
C-1

 Table C1. Cumulative attrition rates among beginning teachers who left the West Virginia public
school system, by district, 2008/09–2012/13 (continued)

District
Pleasants

Number of
beginning teachers

Cumulative attrition rate (percent) as of 2012/13 among beginning
teachers first employed in a West Virginia public school district in:
2008/09

a

2009/10

a

2010/11

a

2011/12

a

a

Pocahontas

12

33.3

0.0

16.7

0.0

Preston

48

21.4

33.3

20.0

25.0

Putnam

71

0.0

20.0

20.0

4.6

Raleigh

121

29.4

46.4

23.5

8.0

Randolph

42

36.4

20.0

0.0

18.2

Ritchie

19

100.0

37.5

14.3

0.0

Roane

37

50.0

14.3

38.5

40.0

Summers

20

0.0

0.0

40.0

27.3

Taylor

22

0.0

60.0

25.0

16.7

Tucker

10

0.0

100.0

0.0

0.0

Tyler

19

50.0

25.0

50.0

11.1

Upshur

40

30.0

16.7

33.3

16.7

Wayne

112

19.2

38.5

17.2

5.6

Webster

25

0.0

33.3

33.3

28.6

Wetzel

32

0.0

30.0

11.1

0.0

Wirt
Wood
Wyoming

16

20.0

0.0

12.5

0.0

162

31.7

33.3

32.7

10.0

62

35.7

15.4

19.1

7.1

Note: The number of beginning teachers is presented as a reference and includes all beginning teachers employed within the district.
The cumulative attrition rate was calculated by dividing the total number of beginning teachers who left the school district by 2012/13
by the total number of beginning teachers employed in the district as of the 2008/09, 2009/10, 2010/11, and 2011/12 academic
years. When an attrition rate is listed as 0.0 percent, it means that none of the beginning teachers in that year had left by 2012/13.
a. Value suppressed to comply with privacy requirements.

Source: Authors’ calculations based on teacher personnel data provided by the West Virginia Department of Education.


C-2

 Note
1.  The national rate of beginning-teacher mobility has been estimated at 13 percent, and
the national rate of beginning-teacher attrition has been estimated at 7 percent (Gol­
dring et al., 2014). However, a direct comparison between the current study and Gol­
dring et al. (2014) is not possible because beginning teachers in that study were defined
as those with up to three years of experience, and the dataset for the current study did
not account for teachers who left West Virginia public schools but may have gone on
to teach in another state or in West Virginia private schools.

Notes-1

 References
Battle, D. (2010). Principal attrition and mobility: Results from the 2008–09 Principal Fol­
low-up Survey (NCES No. 2010–337). Washington, DC: U.S. Department of Educa­
tion, National Center for Education Statistics. http://eric.ed.gov/?id=ED511937
Béteille, T., Kalogrides, D., & Loeb, S. (2012). Stepping stones: Principal career paths
and school outcomes. Social Science Research, 41(4), 904–919. http://eric.ed.gov/
?id=ED522078
Branch, G. F., Hanushek, E. A., & Rivkin, S. G. (2012). Estimating the effect of leaders on
public sector productivity: The case of school principals (CALDER Working Paper No.
66). Washington, DC: National Center for Analysis of Longitudinal Data in Educa­
tion Research. http://eric.ed.gov/?id=ED529175
Brill, S., & McCartney, A. (2008). Stopping the revolving door: Increasing teacher reten­
tion. Politics & Policy, 36(5), 750–774.
Fuller, E., & Young, M. D. (2009). Tenure and retention of newly hired principals in Texas.
Austin, TX: University Council for Educational Administration. Retrieved February
19, 2014, from http://acy.hs-cluster-1.net/upimages/Texas_Principal_Turnover.pdf.
Gates, S. M., Ringel, J. S., Santibañez, L., Guarino, C., Ghosh-Dastidar, B., & Brown,
A. (2006). Mobility and turnover among school principals. Economics of Education
Review, 25(3), 289–302. http://eric.ed.gov/?id=EJ739025
Goldring, R., & Taie, S. (2014). Principal attrition and mobility: Results from the 2012–13
Principal Follow-up Survey. First look. (NCES No. 2014–064). Washington, DC:
U.S. Department of Education, National Center for Education Statistics. http://eric.
ed.gov/?id=ED545366
Goldring, R., Taie, S., & Riddles, M. (2014). Teacher attrition and mobility: Results from
the 2012–13 Teacher Follow-up Survey (NCES No. 2014–077). Washington, DC:
U.S. Department of Education, National Center for Education Statistics. http://eric.
ed.gov/?id=ED546773
Guarino, C. M., Santibañez, L., & Daley, G. A. (2006). Teacher recruitment and retention:
A review of recent empirical literature. Review of Educational Research, 76(2), 173–208.
http://eric.ed.gov/?id=EJ751152
Guin, K. (2004). Chronic teacher turnover in urban elementary schools. Education Policy
Analysis Archives, 12(42), 1–30. http://eric.ed.gov/?id=EJ853508
Hanushek, E. A., Kain, J. F., & Rivkin, S. G. (2004). Why public schools lose teachers.
Journal of Human Resources, 39(2), 326–354. http://eric.ed.gov/?id=EJ746489
Hanushek, E. A., & Rivkin, S. G. (2007). Pay, working conditions, and teacher quality.
Future of Children, 1(1), 69–86. http://eric.ed.gov/?id=EJ795875

Ref-1

 Ingersoll, R. M. (2001). Teacher turnover and teacher shortages: An organizational analy­
sis. American Educational Research Journal, 38(3), 499–534.
Kane, T. J., & Staiger, D. O. (2008). Estimating teacher impacts on student achievement: An
experimental evaluation (NBER Working Paper No. 14607). Cambridge, MA: National
Bureau of Economic Research. Retrieved November 19, 2014, from http://www.nber.
org/papers/w14607.
Ladd, H. (2011). Teachers’ perceptions of their working conditions: How predictive of
planned and actual teacher movement? Educational Evaluation and Policy Analysis,
33(2), 235–261. http://eric.ed.gov/?id=EJ927621
Levy, A. J., Jablonski, E., & Fields, E. (2006, October). Overview of research: What we know
and don’t know about the consequences of science and math teacher turnover. Paper pre­
sented at the National Commission on Teaching and America’s Future Symposium on
the Scope and Consequences of K–12 Science and Mathematics Teacher Turnover,
Racine, WI.
Lochmiller, C. R., Sugimoto, T. J., & Muller, P. A. (2016). Teacher retention, mobility, and
attrition in Kentucky public schools from 2008 to 2012 (REL 2016–116). Washington,
DC: U.S. Department of Education, Institute of Education Sciences, National Center
for Education Evaluation and Regional Assistance, Regional Educational Laboratory
Appalachia. http://eric.ed.gov/?id=ED562734
Miller, A. (2013). Principal turnover and student achievement. Economics of Education
Review, 36(C), 60–72.
Odden, A. (2011). Strategic Management of Human Capital in Education. New York, NY:
Routledge Press.
Plecki, M. L., Elfers, A. M., Loeb, H., Zahir, A., & Knapp, M. S. (2005). Teacher retention
and mobility: A look inside and across districts and schools in Washington State. Seattle,
WA: University of Washington. http://eric.ed.gov/?id=ED485567
Ringel, J., Gates, S., Chung, C., Brown, A., & Ghosh-Dastidar, B. (2004). Career paths of
administrators in Illinois: Insights from an analysis of state data. Technical report pre­
pared for the Wallace Foundation. Santa Monica, CA: RAND Corporation.
Rivkin, S. G., Hanushek, E. A., & Kain, J. F. (2005). Teachers, schools, and academic
achievement. Econometrica, 73(2), 417–458.
Ronfeldt, M., Lankford, J., Loeb, S., & Wyckoff, J. (2013). How teacher turnover harms
student achievement. American Educational Research Journal, 50(1), 4–36. http://eric.
ed.gov/?id=EJ995828
Scafidi, B., Sjoquist, D. L., & Stinebrickner, T. R. (2007). Race, poverty, and teacher mobil­
ity. Economics of Education Review, 26(2), 145–159. http://eric.ed.gov/?id=EJ756857

Ref-2

 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.
http://eric.ed.gov/?id=EJ727495
Theobald, N. D. (1990). An examination of the influence of personal, professional, and
school district characteristics on public school teacher retention. Economics of Educa­
tion Review, 9(3), 241–250. http://eric.ed.gov/?id=EJ416483
Theobald, N. D., & Gritz, R. M. (1996). The effects of district spending priorities on the
exit paths of beginning teachers leaving the district. Economics of Education Review,
15(1), 11–22. http://eric.ed.gov/?id=EJ522740
U.S. Department of Education, Institute of Education Sciences, National Center for Edu­
cation Statistics. (2013). Common Core of Data. Local Education Agency (School
District) Universe Survey Data. Retrieved December 11, 2014, from https://nces.ed.gov/
ccd/pubagency.asp.
Vanderhaar, J. E., Muñoz, M. A., & Rodosky, R. J. (2006). Leadership as accountability for
learning: The effects of school poverty, teacher experience, previous achievement, and
principal preparation programs on student achievement. Journal of Personnel Evalua­
tion in Education, 19(1–2), 17–33. http://eric.ed.gov/?id=EJ784979
Weinstein, M., Jacobowitz, R., Ely, T., Landon, K., & Schwartz, A. E. (2009). New schools,
new leaders: A study of principal turnover and academic achievement at new high schools
in New York City (NYU Wagner Research Paper No. 2011–09). New York, NY: Insti­
tute for Education and Social Policy, New York University.

Ref-3

 The Regional Educational Laboratory Program produces 7 types of reports

Making Connections
Studies of correlational relationships

Making an Impact
Studies of cause and effect

What’s Happening
Descriptions of policies, programs, implementation status, or data trends

What’s Known
Summaries of previous research

Stated Briefly
Summaries of research findings for specific audiences

Applied Research Methods
Research methods for educational settings

Tools
Help for planning, gathering, analyzing, or reporting data or research