The Massachusetts Teacher Workforce:Status and Challenges
By Antoniya Owens (MPP 2010) Harvard Kennedy School
May 2010
EDUCATION
RAPPAPORT INSTITUTEfor Greater Boston
About This Paper
This working paper, which was originally prepared as a Policy Analysis Exercise, Harvard Kennedy School’s equivalent of a master’s thesis, was named one of the best PAEs written by an HKS student in the 2009-2010 academic year. Antoniya Owens, the PAE’s author, was a Rappaport Public Policy Fellow in the summer of 2009 and graduated from the Kennedy School in May 2010.
The Rappaport Institute for Greater Boston
The Rappaport Institute for Greater Boston aims to improve the governance of Greater Boston by fostering better connections between scholars, policy-makers, and civic leaders. The Rappaport Institute was founded and funded by the Phyllis and Jerome Lyle Charitable Foundation, which promotes emerging leaders in Greater Boston. More information about the Institute is available at http://www.hks.harvard.edu/rappaport.
The Taubman Center for State and Local Government
The Taubman Center and its affi liated institutes and programs are the Kennedy School of Government’s focal point for activities that address urban policy, state and local governance and intergovernmental relations.
The Policy Analysis Exercise
The Policy Analysis Exercise (PAE), which is the capstone of the Kennedy School’s Master in Public Policy (MPP) curriculum, is a professional product, meant to clarify and address a practical policy or management problem for a real-world client. The clients for this PAE were two senior offi cials in the Massachusetts Department of Elementary and Secondary Education: Carrie Conaway, Director of Planning, Research, and Evaluation and David Haselkorn Associate Commissioner for Educator Policy, Preparation, Licensure and Leadership. The faculty advisors were Suzanne Cooper, Senior Lecturer in Public Policy and Associate Academic Dean at Harvard’s Kennedy School of Government and Professor Timothy Nelson, Lecturer in Public Policy at Harvard’s Kennedy School of Government. More information about PAEs is available at http://www.hks.harvard.edu/degrees/oca/students-alumni/pae.
© 2010 by Antoniya Owens. The contents of this paper refl ect the views of the author (who is responsible for the facts and accuracy of the research herein) and do not represent the offi cial views or policies of the Rappaport Institute for Greater Boston, the Taubman Center for State and Local Government, the Harvard Kennedy School, or Harvard University.
The Massachusetts Teacher Workforce: Status and Challenges
by Antoniya Owens John F. Kennedy School of Government
Harvard University
A Report to the Massachusetts Department of Elementary
and Secondary Education, Malden, MA
Advisors:
Professor Suzanne Cooper Professor Timothy Nelson
March 2010
Page i Page i
T��� !" C!#$ #$%
Executive Summary………………………………………………………………… 1
Introduction…………………………………………………………………………5
The Current Status of the Teacher Workforce in Massachusetts……………………. 6
Demographic Characteristics…………………………..……………………. 6
Mobility and Turnover in the Teacher Workforce……..……………………. 8
Effective Teachers and Their Distribution across the State………………….. 11
Academic Subjects and the Shortage of Effective Teachers…………………. 16
Future Supply and Demand Dynamics: Hiring Needs and the Role of Retirements… 20
Methodology………………………………………………………………... 20
Results: Projections of Student Enrollment, Total Teacher Demand,
Teacher Hiring Needs, and Retirements…………..………………………… 21
Strengthening the Teacher Workforce: Policy Options and Current Initiatives ………26
Recruiting New Teachers…………………………………………………… 26
Improving Teacher Retention………………………………………………. 34
Recommendations……………………………………………….………………….. 42
End notes....……………………………………………….……………………….... 46
Bibliography……………………………………………….………………………... 50
Appendix: Data Tables……………………………………………….……………... 55
A&'#!(� )* + #$%
I am truly grateful to Carrie Conaway and David Haselkorn at the Massachusetts Department of Elementary and Secondary Education for making this analysis possible. Their guidance on the parameters of the project and their thoughtful comments on various drafts were instrumental in shaping the current version. I am also thankful to Matthew Borek and Craig Weller at the Department for patiently answering my questions and handling my data requests. At the Harvard Kennedy School, I am highly indebted to my faculty advisor, Professor Suzanne Cooper. Her expertise, patient guidance, detailed feedback, and calming presence during the difficult stages of this project were absolutely critical for its successful completion. Special thanks also go to Professor Timothy Nelson for leading the Human Services, Labor, and Education PAC Seminar, and to my classmates in that seminar, for providing me with valuable feedback and encouragement throughout the entire year.
Page ii
Page iii
Page 1
A ny plan to improve student learning and decrease the lingering socioeconomic and racial achievement gaps
must include a concerted effort to staff every classroom in Massachusetts with an effective teacher. To that end, the Massachusetts Department of Elementary and Secondary Education is developing a coherent strategic approach to cultivating a teacher workforce that is adequate in size, equitable in distribution, demonstrably effective, and more robustly diverse. To inform this process, this report seeks to answer a number of questions about the state’s teacher workforce, including its demographic characteristics, current patterns of turnover, the equity of the distribution of effective teachers across geographic areas and academic subjects, and the future teacher hiring needs, including potential shortages and the impact of teacher retirements. The report begins with two data analysis sections—one evaluating the current status of the Massachusetts teacher workforce, and the other projecting future supply and demand dynamics to identify potential shortages and the role of retirements in creating them. The findings reveal a series of challenges plaguing the Commonwealth’s teacher workforce. The workforce is less racially and ethnically diverse than the student population. The distribution of effective teachers is far from equitable across the state: Low-income, high-minority, and urban districts find it harder to recruit and retain talented teachers, and teacher turnover in those districts is high. Subjects like mathematics, natural sciences, special education, bilingual education, and foreign languages are facing shortages of qualified teachers. Finally, although Massachusetts is unlikely to experience a statewide shortage of teachers over the next decade, several large urban districts with rising student enrollments and high teacher turnover will likely face greater teacher hiring needs and more potential shortages—a development that may exacerbate the already inequitable distribution of qualified teachers across the state. In response to these challenges, the Commonwealth needs to devise strategies to attract diverse and talented teachers into high-demand subjects and difficult-to-staff areas, and then retain them there. To that end, the report also includes a review of existing program and policy approaches that have either shown promise or imparted lessons in recruiting or retaining talented teachers. It concludes with a set of recommendations that the state should consider when devising a comprehensive educator workforce strategy. The key findings, policy review insights, and recommendations are summarized below.
The Current Status of the Teacher Workforce Demographic characteristics • Teachers in Massachusetts are less likely to be very young
and more likely to be over the age of 49, compared with all workers in the state. Less than a third of all workers but more than 37 percent of teachers are aged 49 and older, signaling that the impact of the baby boomer retirements may be more strongly felt in the teaching profession than in other employment sectors.
• The teacher workforce in Massachusetts is substantially less diverse than its student population. Ninety-three percent of the state’s teachers are non-Hispanic whites, compared with 70 percent of all students.
• The racial breakdown of teachers varies across the state. Districts with large enrollments of minority and low-income students have more diverse teacher workforces, especially if they are located in urban areas. However, because students in such districts are more likely to be nonwhite themselves, a racial mismatch still exists between them and their teachers. In fact, such districts have fewer minority teachers per 1000 minority students than do low-minority, non-poor, or suburban districts.
Turnover and mobility • Between 2007-08 and 2008-09, 5,117 teachers left the
teaching profession or moved out of the state, for a total attrition rate of 7 percent. Another 1,883 teachers, or 2.6 percent, moved across school districts within the state.
• Mobility among teachers is not uniform, but rather varies by age, length of teaching experience, and district and school characteristics. Relatively old teachers with lengthy work experience leave teaching at very high rates, presumably to retire. Very young teachers with few years of experience have a high propensity both to leave teaching and to move across school districts.
• Teacher turnover in urban, low-income, or high-minority districts is considerably higher than in less diverse, wealthier, or non-urban areas. High turnover contributes to teacher shortages across the state and the nation, has negative implications for student learning, and costs thousands of dollars in teacher replacement costs.
Distribution of effective teachers across districts and academic subjects • Given the lack of a singular and proven measure of
teacher quality, this report evaluates the distribution of effective teachers using a number of observable teacher characteristics as proxies for teacher effectiveness, including licensure status, educational attainment, and teaching experience.
E�������� S���� �
Page 2
• Each of these indicators points to a systematic sorting of less effective teachers to urban districts with large populations of minority and low-income students. Such districts have lower shares of fully licensed teachers, teachers deemed highly qualified, and teachers with master’s degrees. Schools in these districts also report more annual teaching vacancies and more difficulties in staffing these vacancies with fully qualified teachers.
• The distribution of effective teachers across academic subjects is also uneven, with certain subjects consi-derably more difficult to staff than others. A series of indicators, such as the share of licensed teachers, the number of waivers granted, and difficulties reported in staffing teaching positions, reveal five academic fields with persistent teacher shortages: mathematics, science, foreign languages, special education, and bilingual/ESL education.
Future Supply and Demand Dynamics and the Role of Retirements • To project future dynamics in the teacher workforce,
this report employs a supply and demand model from the literature. The projections span academic years 2010-11 through 2019-20 and are undertaken for the state as a whole and for its ten largest school districts.
• The model first projects annual total demand for teachers using forecasts of student enrollment and assumptions about student-teacher ratios. It then matches total demand to the expected supply of teachers retained from the previous year. Each year, the gap between the projected total demand and the returning supply is the demand for new teachers for that year. The role of retirement is evaluated by the share of the workforce that retires each year, the fraction of total attrition that retirees constitute, and the portion of new hiring needs that they necessitate.
• Over the next decade, the state will need to hire about 45,500 new teachers to meet total teacher demand—with annual hiring needs ranging from 4,600 in 2010-11 to 4,300 in 2019-20 (see ES Figure 1). Though these numbers appear high, they are not large enough to cause concern about teacher shortages at the state level: If the supply of new teachers remains at current levels, it will be sufficient to meet and even exceed the state’s hiring needs.
• The projections of district-level hiring needs, however, vary widely across the state, with some districts more likely to experience teacher shortages than others. For example, due to enrollment declines and high teacher retention rates, hiring needs in Lawrence and Worcester are likely to be relatively low. In contrast, districts like Springfield and Fall River will likely face
significantly higher demand for new hires and may encounter difficulties in finding sufficient numbers of fully qualified teachers to meet this demand.
• About 19,000 teachers are expected to retire in Massachusetts over the next decade. Annual retirements will represent between 2.4 and 2.9 percent of each year’s teaching workforce. They will account for between 36 and 40 percent of annual teacher attrition, and will create between 38 and 45 percent of annual demand for new teachers (see ES Figure 1). As with hiring needs, retirement trends and their impact differ substantially across districts.
Policy Options and Current Initiatives The challenges of uneven hiring needs, inequitable distribution of effective teachers across regions and subjects, high turnover, and lack of diversity in the teacher workforce call for strategic and targeted policy responses. To develop a teaching workforce that is more equitably distributed, more stable, and more diverse, the Commonwealth needs to attract sufficient numbers of diverse and talented teachers into difficult-to-staff subjects and areas, and retain them over an extended period of time. This goal would be best served by targeted recruitment and retention approaches that create incentives for prospective teachers to train in shortage subjects, teach in high-need areas, and do it as a long-term career rather than a temporary stint. This section of the report discusses four types of recruitment and retention policies: financial incentives and alternative certification programs to expand the supply of traditional and non-traditional recruits willing to teach in high-need areas and subjects; and salary incentives and work condition policies to help retain them there.
ES Figure 1. Projected number of new teachers needed to
meet demand each year, 2010-2019
Source: Massachusetts Department of Elementary and Secondary Education
Hiring needs
20
30
40
50
60
70
80
90
100
4,000
4,100
4,200
4,300
4,400
4,500
4,600
4,700
4,800
2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 2019-20
Mid-point estimate
of hiring needs
Percent
% of hiring needs created
by retirements (right axis)
Page 3
Financial incentives for new teachers • The financial incentive programs reviewed here—
the MINT/Signing Bonus program, the Attracting Excellence to Teaching program, and North Carolina’s Teaching Fellows and Math/Science/Special Education Teacher Bonus programs—reveal two key lessons.
• First, if an important program goal is to increase the number of teachers in shortage subjects or difficult-to-staff areas, programs whose eligibility criteria explicitly target candidates with such commitment will likely attain this goal more effectively than programs with universal eligibility.
• Second, implementation flaws like lack of publicity, clarity, and good communication between admi-nistrators and participants may doom the success of even the most thoughtfully designed programs.
Alternative certification programs for nontraditional candidates
• To recruit more mid-career professionals, a traditionally underused pool of teaching talent, most states have created accelerated alternative teacher certification programs that reduce time and cost barriers to teaching. Review of two such programs in Massachusetts—the Boston Teacher Residency and TEACH! Southcoast—reveals that such programs hold promise for increasing the number of teachers in shortage subject and high-need areas as long as they are carefully and strategically designed with local needs in mind.
Teacher salaries
• Because pre-retirement turnover contributes signi-ficantly to teacher shortages in low-income urban areas and high-need subjects, increasing the supply of new teachers alone is unlikely to resolve these shortages. Compensation reform and improved working conditions are two retention strategies that could slow down “the revolving door” of turnover, alleviating teacher shortages and improving the equity of their distribution across the state.
• Teacher salaries lag behind those in comparable occupations, and have grown more slowly over the past 15 years. Not surprisingly, salary increases have been popular strategies in other states for attracting and retaining teachers in the profession.
• Given the localized nature of teacher shortages in Massachusetts, an across-the-board salary increase is likely to be prohibitively expensive as well as an inefficient allocation of resources. If salary increases are to be used to retain teachers in shortage subjects and difficult-to-staff districts, they may be more effective if they are explicitly targeted to teachers in such positions. However, the political feasibility of
differentiated compensation may be an issue due to qualified union support. Differentiated compen-sation may also be difficult to introduce because of potential adverse impact on collegiality, which is of greater importance in the teaching profession than elsewhere.
Working conditions
• Working conditions for teachers vary more widely across the state than does compensation. Teachers in poorer, high-minority, and urban communities face less favorable working conditions, including more behavioral problems, larger class sizes, and little empowerment or influence over their school’s decision-making processes.
• Interventions to improve working conditions may be an important strategy to reduce teacher attrition, particularly in difficult-to-staff areas. As local contexts and needs differ, working condition policies are best created “from the ground up,” with schools and districts taking a leading role in devising and implementing interventions that best suit their needs. However, the state has an important role to play in guiding, informing, and supporting local initiatives.
Recommendations Based on insights from the data analysis and the policy review, this section distills three recommendations about steps the state can take to respond to the challenges presented by its teacher workforce.
1. Target programs and initiatives to candidates committed to teaching in high-need areas and in difficult-to-staff subjects.
To tackle the inequitable distribution of teachers across districts and subjects, and alleviate potential regional shortages, the Commonwealth would be best served by targeted approaches that create incentives for prospective teachers to train in shortage subjects, teach in high-need areas, and do it as a long-term career. The state could improve the targeting of current and future programs in several ways:
• Restrict eligibility for financial incentives, such as tuition assistance, loan reimbursement, and signing bonuses, to candidates committed to teaching in shortage areas and subjects. Programs with eligibility requirements re-stricted to candidates committed to teaching in less desirable areas or shortage fields produce more teachers in such positions than programs with universal eligibility.
Page 4
• Encourage districts to develop alternative teacher training programs, or modify existing ones, to target candidates willing to teach in schools or subjects with shortages in those locations. Explicitly targeted district-level programs like the Boston Teacher Residency and TEACH! Southcoast have shown significant promise in alleviating local teacher shortages. However, because such programs work best when designed in response to local needs, their undertaking is best left to school districts.
• Minimize implementation flaws like lack of publicity, clarity, or poor communication—and advise districts to do the same for their locally run programs. Potential channels through which program implementation can be improved include intensive advertising in teacher training programs and district hiring offices, ensuring the clarity and consistency of program features, and providing reliable communication channels between program administrators and applicants/participants.
2. Assist districts with policies that reduce turnover but are better handled at the local level, such as reforming compensation structures and improving working conditions. Differentiated pay and improved working conditions may reduce teacher turnover, alleviate local shortages, and improve the distribution of qualified teachers across the state. Though these two approaches are best handled at the local level, the state can play an important role in guiding, informing, and supporting local initiatives:
• Create standards, guidelines, and a bank of knowledge to store and disseminate “best practices” that have been effective in improving teaching conditions or implementing differentiated pay structures in Massachusetts and elsewhere. School and district leaders can thus learn more easily about promising approaches and modify them to suit local needs.
• Provide financial incentives for schools and districts to experiment with differentiated pay and to create teaching condition improvement plans. Such incentives should ideally be targeted to low-income high-need districts, which stand to benefit the most both from enhanced teaching conditions and from higher pay for teachers choosing to work there.
• Engage the teachers’ unions. Because political feasibility is a constraint at least as important as fiscal viability, the state should seek the collaboration of union officials in all stages of the process—and encourage district leaders to do the same.
3. Promote diversity in the teacher workforce.
Recent evidence suggests that own-race teachers are associated with improved student achievement. Thus, making the teacher workforce more diverse may be an important way to reduce the racial achievement gap. Furthermore, since teachers, on average, prefer to teach near where they grew up, increasing the popularity of teaching among minority students may help alleviate teacher shortages in high-minority districts. There are several steps the state can take to enhance the diversity of its teaching workforce:
• Increase awareness of teaching programs and incentives in minority communities. This can be done through advertising in minority community media, holding information sessions in neighborhoods, high schools, and community colleges with large minority populations, and recruiting teacher trainees from historically black colleges and universities outside New England.
• Provide incentives for high schools to develop initiatives that encourage minority students to consider teaching as a career. Extracurricular activities designed to expose middle and high school minority students to teaching may generate their interest in pursuing the profession.
• Continue state and local efforts to increase high school graduation, college enrollment, and college completion among minority students. Because low educational attainment among minority residents is one reason for the relative lack of minority teachers, improving their college completion rates may indirectly advance the objective of enhancing teacher diversity. Increasing minority students’ educational attainment would also generate significant and direct social and economic benefits for the students themselves as well as for the state as a whole.
� �
Page 5
I�� ��������
S tudent achievement in Massachusetts is impressive. Our students’ scores on National Assessment of Educational Progress (NAEP) tests and on the SAT are among the highest in the nation; their dropout rates are among the lowest.
And their performance on the Massachusetts Comprehensive Assessment System (MCAS) exams continues to improve: The share passing the MCAS on their first try nearly doubled from 47 percent in 2000 to 87 percent in 2009.1 Despite this success, the state’s public education system faces a number of challenges. Stellar academic achievement is far from uniform across the Commonwealth. The academic performance of low-income and minority students continues to trail that of their more privileged peers, indicating a lingering achievement gap. For example, 91 percent of white students passed the 2007 MCAS exams in both math and English, compared with three quarters of African American students and only two-thirds of Hispanic students.2 Voluminous research has identified teacher quality as having a strong influence on student achievement—an influence that can last for years after the teacher and the student have parted ways. Unfortunately, poor and minority students are substantially less likely to have a highly qualified and talented teacher for every subject. As a result, any plan to improve student learning in Massachusetts must include a concerted effort to staff every classroom in the state with an effective teacher. To that end, the Massachusetts Department of Elementary and Secondary Education is developing a coherent strategic approach to cultivating a teacher workforce that is adequate in size, equitable in distribution, demonstrably effective, and more robustly diverse. To inform this process and to aid the Department in accomplishing its goals, this report seeks to answer the following questions:
• What does the Massachusetts teacher workforce look like in terms of size and demographic characteristics? • What are the current patterns of teacher turnover and how do they vary by district location and students’
ethnicity and poverty? • Which districts are finding it difficult to recruit and retain sufficient numbers of talented teachers? What
academic subjects are facing critical shortages of effective teachers? • What does the future of the teacher workforce look like in terms of supply and demand? Are the state and its
districts likely to experience teacher shortages, particularly in light of the impending baby boomer retirements? To answer these questions, the report begins with two detailed data analysis sections—the first evaluating the current status of teachers in the Commonwealth, and the second projecting future supply and demand dynamics to identify potential teacher shortages and the role of retirements in creating them. These analyses rely primarily on teacher data from the Department’s Education Personnel Information Management System (EPIMS) and student data from its Student Information Management System (SIMS). Whenever possible, these have been supplemented by data from other sources, such as the U.S. Department of Education’s Schools and Staffing Survey, and the state’s Massachusetts Teaching, Learning and Leading Survey. The findings reveal a series of challenges plaguing the state’s teacher workforce. Teachers as a whole are less racially and ethnically diverse than their students, as the state’s minority residents are less likely to graduate from college and less likely to opt for a teaching career. In addition, the distribution of effective teachers, as measured by a number of indicators, is far from equitable across the state. Low-income, high-minority, and urban districts find it harder to recruit and retain talented teachers, and teacher turnover in those districts is considerably higher. Furthermore, subject areas like mathematics, natural sciences, special education, bilingual education, and foreign languages are experiencing shortages of fully qualified teachers, as the candidates to teach those fields often face more favorable labor market opportunities outside of teaching. Finally, the Commonwealth is unlikely to experience a statewide shortage of teachers over the next decade, as teacher supply may meet or even exceed demand in the aggregate. However, several large urban districts with rising student enrollments and high teacher turnover will likely experience greater teacher hiring needs and more potential shortages—a development that may exacerbate the already inequitable distribution of qualified teachers across the state. To tackle these challenges, Commonwealth needs to attract sufficient numbers of diverse and talented teachers into difficult-to-staff subjects and areas, and retain them over an extended period of time. The report continues with a discussion of existing program and policy approaches that have shown promise or imparted lessons in recruiting or retaining talented teachers. It concludes with a set of recommendations the state should consider when devising a comprehensive educator workforce strategy.
Page 6
D uring academic year 2008-09, public schools in Massachusetts employed 72,433 teachers. (Because
some teachers work part-time, the total number of full-time equivalent (FTE) teachers was slightly lower at 70,396.) One third of the state’s teachers work in early childhood and elementary education. One in five teaches middle school, and just over a quarter teach at the high school level. The remaining fifth teach classes that span multiple or all grades. The vast majority of teachers—over 85 percent of all FTEs—are employed in general education; 14.5 percent work in special education (see Figure 1). Of the teachers in general education, a third work as generalists, followed by English and mathematics instructors, who respectively account for 15 percent and 10 percent of the general education teachers. One in ten teaches science.
Demographic Characteristics
Age and gender
Compared with the state’s total workforce, teachers in Massachusetts are less likely to be very young and more likely to be over the age of 49 (see Figure 2). Young teachers, and very young teachers in particular, are disproportionately less represented: Only a quarter are under the age of 32, and a mere six percent are 25 years old or younger. In contrast, 30 percent of all workers in the state are 32 years old or younger, and 16 percent are 25 or younger. Teachers between the ages of 33 and 48 account for 37 percent of the teacher workforce—a share similar to that of all workers in this age range. Finally, less than a third of all workers in the Commonwealth but more than 37 percent of its teachers are aged 49 and older, signaling that the impact of the looming baby boomer retirements may be more strongly felt in the teaching profession than in other employment sectors. As in the nation as a whole, teachers in Massachusetts are disproportionately more likely to be female than male. In academic year 2008-09, three out of four teachers in the state were women (see Figure 3). The share of female teachers is particularly high at the elementary level, and tends to decline in higher grades. In 2003-04, for example, 91 percent of all elementary school teachers in the state were female. At the same time, women comprised three quarters of middle school teachers and just over half of all high school teachers.3
T�� C� ��� S����� �� ��� M������������ T����� W� ��� ��
Source: Massachusetts Department of Elementary and Secondary Education
Special education,
14.5%
General
education, 85.5%
Generalists, 34%
English, 15%Math, 12%
Science, 10%
Arts, 8%
History, 8%
Languages, 6%
Figure 1. Massachusetts teachers by academic subject, 2008-09
Other, 8%
5.9
19.4 19.8
17.6
23.0
13.3
0.9
16.1
13.7
17.6
19.6
17.3
11.3
4.4
0
5
10
15
20
25
25 or younger 26 to 32 33 to 40 41 to 48 49 to 56 57 to 64 65 and over
Teachers
All workers
Figure 2. Age distribution of teachers and of all workers in
Massachusetts, 2008-09
Source: Mass Department of Elementary and Secondary Education, 2008 American Community Survey.
Percent
75.9
24.1
93.2
2.6 1.1 2.5 0.6
48.651.4
69.9
8.25.1
14.3
2.5
0
20
40
60
80
100
Female Male White African American Asian Hispanic Other
Teachers
Students
Figure 3. Massachusetts teachers and students,
by gender and race/ethnicity, 2008-09
Source: Massachusetts Department of Elementary and Secondary Education
Note: Other includes Native American, Native Hawaiian, Pacific Islander, and non-Hispanic multi-race.
Percent
Page 7
Race/Ethnicity
The teacher workforce in Massachusetts is substantially less diverse than its student population. The vast majority—93 percent—of the state’s teachers are non-Hispanic whites, compared with 70 percent of all students (see Figure 3 on previous page). Furthermore, while Hispanics and African Americans account for 14 percent and 8 percent of students, respectively, each of these two categories describes less than three percent of teachers. The racial breakdown of teachers varies widely across the state. Districts with high enrollments of minority students and low-income students have more diverse teacher workforces, especially if they are located in urban areas. For example, in 2008-09, the share of minority teachers in districts with large minority student enrollments was three times as high as their share in low-minority districts: 4.3 percent and 1.4 percent, respectively (see Figure 4).i Similarly, teachers in low-income districts, particularly in urban locations, were more likely to be nonwhite than their peers elsewhere.ii More than one in ten teachers in urban districts was nonwhite, compared with 2.5 percent of those in suburban or rural ones. These differences are even more pronounced by quartile: More than six percent of teachers in the most diverse quartile of districts and five percent of those in the poorest quartile are minority, compared with 1.3 percent in the least diverse and 2.7 in the least poor quartiles (see Appendix Table 1). At the same time, because students in low-income, urban, and by definition, high-minority districts are far more likely to be nonwhite themselves, a racial mismatch still exists between them and their teachers. In fact, such districts have fewer minority teachers per 1000 minority students than do low-minority, non-poor, or suburban districts (see Figure 5). In 2008-09, both high-minority districts and low-income districts employed roughly 14 minority teachers per 1000 minority students, compared with 19 to 20 minority teachers in low-minority and in high-income districts. Quartile analysis reveals similar findings: districts in the poorest and most diverse quartiles had only half as many minority teachers per 1000 minority students as districts in the least diverse and least poor quartiles (see Appendix Table 1). While the relative lack of minority teachers may not be a problem in and of itself, it should be a cause for concern to the extent that it adversely affects student learning. The limited empirical literature on this question offers somewhat mixed findings. Older studies examining the relationship between teachers’ race and student outcomes have found no significant impact.4 For example, a 1994 study using national
i High-minority districts are those in which the share of minority students is higher than the states' median share. Low-minority districts are those below the state median share. For a full list, see Appendix Table 14. ii Low-income districts are those in which the share of students eligible for free or reduced-price lunch is higher than the state’s median share. High-income districts are those below the state median share. For a full list, see Appendix Table 14.
4.3
1.4
3.4
2.4
10.8
2.5
0
1
2
3
4
5
6
7
8
9
10
11
High-minority districts
Low-minority districts
Low-income districts
High-income districts
Urban districts Suburban/rural districts
Figure 4. Minority teachers as share of all teachers, 2008-09
Source: Massachusetts Department of Elementary and Secondary Education
Percent
14.0
19.7
14.5
19.2
15.816.9
0
4
8
12
16
20
24
High-minority districts
Low-minority districts
Low-income districts
High-income districts
Urban districts Suburban/rural districts
Figure 5. Number of minority teachers
for every 1000 minority students, 2008-09
Source: Massachusetts Department of Elementary and Secondary Education
Percent
5.87.1
4.13.4 3.2
21.8
44.8
7.0
4.4 4.32.9 2.5
1.5 1.2 1.32.6
0
5
10
15
20
25
25 or younger
26 to 32 younger
33 to 40 years
41 to 48 years
49 to 56 years
57 to 64 years
65 or older All teachers
Exiting teachers
Moving teachers
Source: Massachusetts Department of Elementary and Secondary Education
Percent
Figure 6. Share of teachers in each age group who left teaching or moved across districts between 2007-08 and 2008-09
Page 8
longitudinal data on high school students found little association between teachers’ race, ethnicity, and gender, and the standardized test scores of students of different races.5 In contrast, a more recent analysis evaluates the relationship between teachers’ race and student learning in a randomized experiment setting, and finds that an own-race teacher has a significant positive impact on the math and reading achievement of black and white students alike.6 Moreover, a 1998 literature review on the subject concludes that racial dynamics between teachers and students may be influencing learning and contributing to the black-white student achievement gap.7 If the latter findings are valid, policies that enhance diversity in the teacher workforce may have important implications for improving the quality of learning for minority students—especially in racially diverse districts, which host many of the state’s low-achieving schools.
Mobility and Turnover in the Teacher Workforce
The most recent national estimates place the share of teachers who exit the profession each year at about 8 percent.8 The annual exit rate for Massachusetts public school teachers is slightly lower, though generally in this ballpark. Between 2007-08 and 2008-09—the only two academic years for which such data are available—5,117 teachers left the teaching profession or moved out of the state, for a total attrition rate of 7 percent. Another 1,883 teachers, or 2.6 percent, moved across school districts within the state. Similarly to the nation, mobility among teachers in Massachusetts is not uniform, but rather varies by age, length of teaching experience, and district and school characteristics. Generally, relatively old teachers with lengthy work experience tend to leave teaching at very high rates, presumably to retire. On the other hand, very young teachers with few years of experience have a high propensity both to leave teaching and to move across school districts. Finally, urban low-income districts with high shares of minority students experience considerably higher teacher turnover than districts that are less diverse, wealthier, or located in non-urban areas.
Turnover and age
Between 2007-08 and 2008-09, age-specific exit rates in Massachusetts followed a U-shaped distribution, with relatively young and relatively old teachers more likely to leave the profession compared to their middle-aged colleagues (see Figure 6 on previous page). However, this distribution was still heavily skewed toward older age groups, which include large numbers of retirees. Close to 45 percent of teachers aged 65 and older left teaching during this period, as did more than a fifth of teachers between the ages of 57 and 64. Overall, these two age groups account for 45 percent of all teachers leaving the profession but for only 11 percent of those staying put (see Appendix Table 2A). Retirement was the most frequent reason for leaving, cited by nearly 44 percent of all exiting teachers. Furthermore, looming retirements among older teachers, many of whom are members of the baby boom generation, are likely to disproportionately increase future turnover in the profession, as retirees are typically replaced with younger and relatively inexperienced colleagues who themselves have high attrition and mover rates. Younger teachers were more likely to move across districts than any other age groups—more than 4 percent of those under the age of 32 did, compared with less than three percent of middle-aged teachers and just over one percent of teachers
I know that I cannot continue working at
such a fevered pitch forever, and high teacher turnover is emotionally difficult for students and logistically difficult for departments.
- Massachusetts public school teacher *
“
“ * Gerson-Nieder, J., Tamerat, J., Gelinas, A. “Teachers from Somerville, Boston, and Longmeadow talk about being asked to do too much with too little.” (2008). CommonWealth Magazine, 13 (3): 61-66, p.63.
Page 9
older than 57 (see Figure 6 on page 7). As a result, despite the fact that teachers aged 32 or younger formed a quarter of the workforce, they accounted for 42 percent of teachers who moved across districts (see Appendix Table 2A). The largest share of moving teachers took similar positions in other districts. Nearly a quarter moved as a result of district actions, such as layoffs or refusal to renew contracts. Another 19 percent cited personal reasons for their move. (Some teachers in the latter group may have found similar positions in their new districts, though they did not report these as the primary reason for their move.)
Turnover and length of service in district
The unevenness of teacher mobility is even more pronounced by teachers’ length of service in their current district. Teachers with little experience and those with the longest service in their district had the highest turnover (see Figure 7). The high degree of “churning” among relatively new teachers is particularly evident from these comparisons. Between 2007-08 and 2008-09, a quarter of all teachers with one to two years of service in their current district either left teaching or moved to another district. Overall, teachers who have taught in their current district for five or fewer years accounted for 42 percent of stayers but for nearly half of all leavers and a whopping 80 percent of all movers (see Appendix Table 2B). These numbers generally mirror the national trends. For example, Ingersoll (2003) finds that between 40 and 50 percent of all beginning teachers leave the profession within five years.9 In large urban districts like the Boston Public Schools, turnover among young teachers is often even higher. A recent study found that 47 percent of Boston public school teachers left the system within their first three years of service.10
Turnover and district characteristics
Research at the national level and in many metropolitan areas has consistently found that urban schools and districts with high enrollments of low-income and minority students have higher teacher turnover rates. A district-level analysis in Massachusetts reveals similar trends. Between 2007-08 and 2008-09, teachers in districts with large enrollments of minority students were substantially more likely to leave teaching or move elsewhere, compared with teachers in low-minority areas (see Figure 8). Twelve percent of teachers in high-minority districts left the profession or moved, compared with 10 percent of their colleagues in districts with lower shares of nonwhite students.iii The discrepancy was driven primarily by the fact that the districts with the most diverse students—those in the top quartile in terms of minority student enrollment—had substantially higher
2.6
1.6
2.3
1.9
4.2
2.02.1
1.2
1.9
1.3
3.3
1.5
0
1
2
3
4
5
High-minority districts
Low-minority districts
Low-income districts
High-income districts
Urban districts Suburban/rural districts
As share of all teachers (%)
Per 1000 students
Figure 9. Massachusetts teachers with waivers,
by district characteristics, 2008-09
Source: Massachusetts Department of Elementary and Secondary Education
iii The district-level turnover rate is defined as the share of teachers who left teaching or who moved to another district within the state. Since teachers in the latter group continue to teach in Massachusetts, they are not included in the state attrition rate. As a result, the state attrition rate is lower than most district attrition rates.
iii
0.3
14.7
6.1
2.8 3.3
8.2
27.3
7.0
0.0
9.0
4.0
1.50.7 0.3 0.3
2.6
0
5
10
15
20
25
30
Less than 1 year
1 to 2 years 3 to 5 years 6 to 10 years 11 to 20 years
21 to 30 years
30 years or more
All teachers
Exiting teachers
Moving teachers
Source: Massachusetts Department of Elementary and Secondary Education
Percent
Figure 7. Share of teachers in each experience group who left teaching or moved across districts between 2007-08 and 2008-09
12.1
10.3
11.8
10.7
13.5
11.1
6
8
10
12
14
High-minority districts
Low-minority districts
Low-income districts
High-income districts
Urban districts Suburban/rural districts
Figure 8. Turnover rates of Massachusetts teachers
between 2007-08 and 2008-09, by district characteristics
Source: Massachusetts Department of Elementary and Secondary Education
Percent
Page 10
teacher turnover than districts in the three lower quartiles, whose turnover rates were similar to each other (see Appendix Table 3). Similarly, low-income districts were more likely to lose teachers than high-income ones. Nearly 12 percent of teachers in low-income districts left between 2007-08 and 2008-09, compared with 10.7 percent of those in high-income districts (see Figure 8 on previous page). Teachers in the poorest quartile of districts were the most likely to leave: At 13 percent, their turnover rate was more than a quarter higher than the rates in the lower three quartiles (see Appendix Table 3). Finally, between 2007-08 and 2008-09, 13.5 percent of teachers in urban districts left their district, compared with 11 percent of teachers in suburban or rural locations.
Why does turnover matter?
High teacher turnover should cause concern for several reasons. First, and perhaps most importantly, it has troubling implications for student learning. Persistent turnover means that experienced teachers are regularly replaced with less experienced colleagues who also tend to be less effective. Research has also found that it is the higher-ability teachers who are more likely to leave the profession. For example, Henke et al. (2000) finds that four years after college graduation, nearly a third of teachers who scored in the top quartile on their college entrance exams had left teaching, compared with one fifth of teachers in the middle quartiles, and only 16 percent of their lowest-scoring colleagues.11 These dynamics often translate into lower student achievement. Guin (2004) finds that students in schools with high teacher turnover were less likely to meet standards on statewide math and reading assessments.12 High teacher turnover also means that experienced teachers are frequently stretched thin by the need to mentor and support a constantly changing group of novice teachers. At the same time, without adequate support, new recruits can feel overwhelmed by the novelty and stress of teaching, leading them to exit the profession and contributing to a downward spiral of never-ending teacher mobility that takes a further toll on student learning.
Photograph: Erica Harrison
Page 11
Second, high turnover is among the key factors contributing to the teacher shortages occurring in certain academic subjects and in various regions of the country. For example, teachers trained in high-demand disciplines, such as math and science, face more favorable employment opportunities in private-sector technology, science, and engineering occupations than in teaching. As a result, studies have found that math and science teachers are more likely to exit the teaching profession, compared with their colleagues who teach English or social studies.13 Furthermore, while the rising demand for new teachers due to enrollment growth and retirements is cited as the major cause of regional teacher shortages, researchers like Ingersoll (2003) argue that such staffing problems are mostly due to high teacher turnover—the proverbial “revolving door” of the profession—rather than to an insufficient supply of new teachers.14 The analysis of hiring needs and teacher attrition in Massachusetts, presented in the next section, reveals that this is indeed the case in several large districts in the Commonwealth. Finally, persistent turnover is costly, as schools and districts are forced to spend scarce financial resources on never-ending cycles of recruiting, hiring, processing, and training new teachers. In 2005, the Alliance for Excellent Education estimated that replacing teachers who leave the profession or transfer elsewhere costs the nation $4.9 billion annually.15 The fiscal drain created by high turnover can be considerable for schools in high-poverty and urban districts, which often have the highest teacher attrition rates and the most modest coffers. For example, in 2004-05, the Boston Public Schools spent more than $3 million on replacing 194 exiting teachers, for an average cost of $17,000 per teacher.16 Barnes, Crowe, and Shaeffer (2007) arrive at similar estimates for the Chicago Public Schools and the Milwaukee Public Schools, which spent, on average, $17,872 and $15,325, respectively, to replace a leaving teacher.iv,17
Effective Teachers and Their Distribution across the State
The relationship between teacher quality and student achievement has been a subject of a large volume of empirical research. Not surprisingly, findings consistently show that teachers have an important impact on student learning.18 At the same time, there is far more disagreement on which teacher characteristics drive this impact. Studies have found that some characteristics, like subject-matter training and overall academic proficiency, make a difference, especially at the secondary level, while others—such as degree level—matter less.19 Given the lack of a singular and proven measure of teacher quality, this report evaluates the distribution of effective teachers in Massachusetts using a series of observable teacher characteristics as proxies for teacher effectiveness. Though each indicator is by itself an imperfect measure of teacher quality, their collective use helps to construct a more multi-faceted and hopefully more accurate portrait of effective teachers in the state and the equity of their distribution across districts and academic subjects. Unfortunately but not unexpectedly, each of these indicators points to a systematic sorting of less effective teachers to urban districts with large populations of minority and low-income students. This trend that has been extensively documented both at the national level and in many large metropolitan areas; it appears that the Commonwealth is no exception.20
iv Because of data limitations, Barnes, et al. do not adjust these cost estimates for salary differentials between departing and replacement teachers. However, they argue that district savings from such differentials are likely to be small for two reasons: most exiting teachers leave before they reach higher salary levels, and exiting teachers who are veterans are not always replaced by inexperienced and hence less costly teachers. (Barnes, et al., p.84)
HIGH TURNOVER IS AMONG THE KEY FACTORS CONTRIBUTING TO THE TEACHER SHORTAGES IN CERTAIN ACADEMIC SUBJECTS AND IN VARIOUS REGIONS OF THE COUNTRY.
EACH INDICATOR POINTS TO A SYSTEMATIC SORTING OF LESS EFFECTIVE TEACHERS TO URBAN DISTRICTS WITH LARGE POPULATIONS OF MINORITY AND LOW-INCOME STUDENTS.
Page 12
Teacher licensure/certification
Historically research findings on the effect of teacher licensure on student outcomes have been mixed. However, recent and more methodologically sound studies have found that proper certification does make a difference. For instance, several studies using data from New York City and North Carolina find evidence that students with fully certified teachers outperform in both mathematics and reading students whose instructors lack certification.21 Goldhaber and Brewer (1997, 2000) also find that having a high school teacher certified in math has stronger positive effects on student math achievement than does having a teacher certified in another subject.22 In a similar vein, Darling-Hammond et al. (2005) compare Teach for America teachers with other teachers and find that uncertified TFA teachers perform less well than fully certified teachers and about the same as non-TFA uncertified teachers; TFA teachers who attain certification, however, are found to be as effective as regular certified teachers.23 Given these findings, it is encouraging that in 2008-09—the most recent academic year for which data are available—96.5 percent of teachers in Massachusetts were licensed in their teaching assignment (see Table 1).v Moreover, this share has been consistently on the rise over the past several years—and so have the shares of teachers identified as highly qualified under the No Child Left Behind definition and the shares of core academic classes taught by such teachers.vi Unfortunately, the number of waivers—granted to teachers who have not met the requirements for licensure—has also skyrocketed.vii Between academic years 2003-04 and 2007-08, the number of waivers granted in the state more than doubled, rising from about 1,500 to more than 3,500 in just five years. In 2008-09, that number declined to just over 2,500 but teachers with waivers still accounted for nearly four percent of the teacher workforce in the state. Furthermore, the distribution of effective teachers, as proxied by their licensure status, is less than equitable across school districts. Although in 2008-09 districts with high enrollments of minority students employed similar shares of licensed teachers compared with low-minority districts, the former were less likely to offer core classes taught by highly
v Teachers are considered licensed if they have attained a Preliminary, Initial, or Professional license in the area in which they teach. Charter schools are exempt from the requirement to hire licensed teachers. vi To meet the NCLB definition of highly qualified, teachers must possess a valid Massachusetts teaching license at the Preliminary, Initial, or Professional level and demonstrate subject matter competency in the areas they teach. vii Waivers are granted for the duration of one year but allow unlimited renewals.
Table 1. Effectiveness indicators of Massachusetts teachers
% of teachers
licensed in teaching
assignment
% of core academic
teachers identified as
highly qualified
% of core classes
taught by highly
qualified teachers
Number of
waivers
% of teachers
with waivers
2003-04 93.9 93.9 1,489 2.1
2004-05 93.9 93.0 1,643 2.5
2005-06 94.4 93.7 2,398 3.6
2006-07 95.4 95.1 2,946 4.0
2007-08 95.8 95.7 3,521 5.0
2008-09 96.5 96.5 2,578 3.7
Source: Massachusetts Department of Elementary and Secondary Education
Page 13
-qualified teachers. High-minority districts also employed disproportionately more teachers with waivers: 2.6 percent of their teachers were uncertified, compared with 1.6 percent in low-minority districts (see Figure 9 on page 9). The analysis of teacher certification by students’ socioeconomic status reveals similar trends. Low-income districts had lower shares of teachers licensed in their field and lower shares of core subjects taught by highly qualified teachers, compared with wealthier districts (see Appendix Table 4). In addition, low-income districts were more likely to hire teachers with waivers—they employed nearly 2 such teachers per 1000 students, compared with 1.3 in high-income areas (see Figure 9 on page 9). A quartile analysis reveals similar findings: Districts in the poorest quartile employed lower shares of licensed or highly qualified teachers and higher shares of teachers with waivers than districts in the wealthiest quartile (see Appendix Table 4). For example, between eight and nine percent of all teachers in the low-income districts of Springfield, Holyoke, and Fall River were granted waivers. In contrast, high-income suburban districts like Wellesley and Wayland employed no teachers with waivers. Finally, urban districts—all of which have large enrollments of low-income and minority students—also tend to hire fewer qualified teachers, as proxied by licensure status. For example, the waiver indicators reveal a far more prevalent employment of uncertified teachers in urban districts: at 4.2 percent, their share was more than twice as high as in suburban and rural areas. Urban districts employed, on average, more than 3 teachers with waivers per 1000 students, compared with 1.5 teachers in non-urban districts (see Figure 9).
Teacher education
The research evidence linking teachers’ education and student achievement is more limited than the evidence on licensure status, especially when teacher education is defined simply as degree level. In a review of more than 100 studies on predictors of teacher effectiveness, Hanushek (1986) notes that fewer than five percent found a positive statistically significant relationship between a teacher’s degree level and the academic outcomes of his or her students.24 However, there is some evidence that degree level may be important in certain contexts, especially at the secondary level. Using the 1988 National Educational Longi-tudinal Study, which links individual students with specific teachers and tracks them over time, Goldhaber and Brewer (1997) find that high-school math and science teachers with advanced degrees in their subjects have a more positive impact on student achievement than do comparably trained English and history teachers.25 According to data from the 2003-04 Schools and Staffing Survey—the only source of data on educational credentials of Massachusetts teachers—98 percent of all teachers in the Commonwealth have at least a bachelor’s degree (see Table 2 on page 15). (This finding should not be surprising given that the state’s licensure requirements mandate a bachelor’s degree.) Moreover, there is little variation in the share of teachers with bachelors’ degrees between schools in urban and suburban locations or with different enrollments of minority students and low-income students. In addition, more than 60 percent of teachers in the Commonwealth have at least a master’s degree. Unlike undergraduate degrees, teachers’ attainment of advanced degrees differs across schools with different characteristics. On average, 63 percent of teachers have a master’s degree in schools in which the majority of students are non-Hispanic whites, but only 52 percent of their colleagues in high-minority schools do. Similarly, 64 percent of teachers in schools with small shares of low-income students have a master’s degree, compared with just over one half of teachers in schools in which low-income students are the majority. Finally, one half of teachers in urban schools but nearly two thirds of teachers in suburban or rural schools have attained a master’s degree.
URBAN DISTRICTS, WHICH HAVE LARGE ENROLLMENTS OF LOW-INCOME AND MINORITY STUDENTS, TEND TO HIRE FEWER QUALIFIED TEACHERS.
Page 14
Length of teaching experience
Teaching experience is among the more extensively studied proxies of teacher quality. For example, nearly 40 percent of the studies reported in Hanushek’s 1986 review article find evidence that experience is a statistically significant predictor of teacher effectiveness.26 However, this does not nece-ssarily reveal a causal relationship, as seniority transfer policies often match teachers with longer tenures to classes with higher-achieving students. More recent studies that correct for such non-random matching find that experience tends to increase teacher effectiveness, though the payoff is not linear across years of experience. The positive impact of experience on teacher quality appears to be highest in the first 3 to 5 years and tapers off somewhat after that.27 Since state data on total teaching experience are unavailable, this report uses length of service in a teacher’s current school district as a proxy for experience—albeit an imperfect one that understates the share of experienced teachers because it ignores their experience in previous districts. According to this indicator, as of 2008-09, a signi-ficant plurality of teachers in Massachusetts (43 percent) were relatively inexperienced, having taught for five or fewer years in their current district (see Figure 10). Twenty-two percent had between 6 and 10 years of experience in their district, while another fifth have taught between 10 and 20 years. Only five percent of teachers in the state have taught for more than 30 years in their current district. Interestingly, there is not a particularly strong connection between teachers’ tenure in their district and characteristics like urbanicity or students’ race (see Appendix Table 5). The shares of teachers in each experience group are similar both across urban and suburban districts and across districts with low and high concentrations of minority students. A more detailed quartile analysis of teachers’ experience by students’ racial breakdown reveals no particular trends either (see Appendix Table 5). District-level differences are more pronounced by student socioeconomic status. Low-income districts have lower shares of teachers with medium length of service—between 3 and 10 years—but higher shares of teachers with longer tenures (see Appendix Table 5). Low- and high-income districts have comparable shares of very inexperienced teachers. A quartile analysis confirms these trends. In particular, districts in the poorest quartile are two-thirds more likely than the least poor districts to have teachers with more than 20 years of experience. Whether these trends are beneficial to poor students depends on whether the teachers who stay longer are more or less effective than those who leave within the first few years—a question we cannot answer given the current nature of data availability. However, the higher shares of teachers with long tenures
9.6
17.316.3
22.1
19.9
9.5
5.3
0
5
10
15
20
25
Less than 1 year
1 to 2 years 3 to 5 years 6 to 10 years 11 to 20 years
21 to 30 years
30 years or more
Source: Massachusetts Department of Elementary and Secondary Education
Percent
Figure 10. Massachusetts teachers by length of service
in their current district, 2008-09
RECENT STUDIES FIND THAT EXPERIENCE TENDS TO INCREASE TEACHER EFFECTIVENESS, THOUGH THE PAYOFF IS NOT LINEAR ACROSS YEARS. THE POSITIVE IMPACT OF EXPERIENCE ON TEACHER QUALITY APPEARS TO BE HIGHEST IN THE FIRST 3 TO 5 YEARS AND TAPERS OFF SOMEWHAT AFTER THAT.
Page 15
Table 2. Educational attainment of Massachusetts teachers
Percentage distribution of school teachers by degree earned, 2003-04
Less than
bachelor's
Bachelor's
degreeAll schools 2.0 98 60.8
Share of minority students in schoolBelow 50% 2.1 97.9 63.3 *50% or higher 1.9 98.1 52.3
Share of low-income students in schoolBelow 50% 2.4 97.6 63.5 *50% or higher 0.8 99.2 51.2
Urbanicity of the school
Urban 2.2 97.8 50.4 **
Suburban/rural 2.0 98 65.3
Source: National Center for Education Statistics, 2003-04 Schools and Staffing Survey (SASS), Public School File.
Table 3. Methods used by schools to fill teacher vacancies in Massachusetts, 2003-04
All schools Urban
% of schools with vacancies 77.2 86.0
Percent of schools using each method to fill vacancies:
Hired substitutes 35.4 53.3
Hired less than fully licensed
teacher17.9 27
Expanded class size 12.6 12.9
Assigned a teacher of another
subject or grade level to teach
those classes
8.5 9.7
Added sections to other
teachers' normal teaching loads6.1 5.6
Cancelled planned course
offerings3.8 0
Assigned administrator or
counselor0.9 1.4
Source: National Center for Education Statistics, 2003-04 Schools and Staffing Survey (SASS), Public School File.
27.6
73.9
Suburban/
rural
Master's
degree
Note: Asterisks indicate that the two numbers are statistically significantly different: * at the 10% level, ** at the
5% level, and *** at the 1% level.
0.7
5.5
6.3
7.9
12.5
13.9
Page 16
and the more prevalent shortages of talented teachers (as revealed by other indicators) may make poor districts disproportionately vulnerable to the upcoming wave of teacher retirements and less able to weather them without further deterioration in teacher quality.
Teacher vacancies Another variable often used as an indicator of the availability of qualified teachers is the number of vacancies in a given district or subject that remain unfilled or get cancelled.28 However, this indicator is somewhat unreliable as vacancies may be unfilled or cancelled due to factors unrelated to teacher supply, such as budget cuts. And even if a legitimate shortage of qualified teachers exists, schools rarely leave teaching positions vacant. Instead, vacancies are filled by other methods, such as hiring uncertified teachers, using teachers licensed in other subjects, or expanding class sizes. Methods of filling vacancies are thus a more telling indicator of shortages of qualified teachers than the number of positions left unfilled.29 Data from the Schools and Staffing Survey allow us to investigate how schools in Massachusetts tended to fill vacant positions in academic year 2003-04. Due to the relatively small number of schools sampled from the state, comparisons by students’ race and socioeconomic background are unreliable. However, it is still possible to evaluate the methods used by urban and suburban schools to fill vacant teaching positions and thus to compare the difficulties each type of school faces in recruiting highly qualified teachers. Overall, more than three quarters of schools in Massachusetts had teacher vacancies in 2003-04 (see Table 3 on previous page). Given their higher teacher turnover, it is not surprising that urban schools were more likely to post vacancies. In 2003-04, 86 percent had at least one vacancy, compared with 74 percent of schools in suburban or rural areas. In addition, urban schools were much more likely to fill vacant positions by hiring less qualified teachers. Over half of all urban schools with vacancies hired a substitute teacher for at least one position, compared with 28 percent of suburban and rural schools. Twenty-eight percent of urban schools filled at least one vacancy with a less than fully licensed teacher, while only 14 percent of non-urban schools did. In contrast, schools in non-urban areas were more likely to cancel the course or add it to the normal teaching load of another already licensed teacher. Schools in urban and suburban locations were equally likely to expand class size as a method of eliminating teaching vacancies.
Academic Subjects and the Shortage of Effective Teachers
As with geography, the distribution of effective teachers across academic subjects is far from even, with certain subjects appearing considerably more difficult to staff than others. Nationally, the American Association for Employment in Education has documented persistent teacher shortfalls in special education, mathematics, physical science, and bilingual education, along with surpluses in subjects like elementary education, English, and social studies, among others.30 A series of indicators, such as the share of teachers licensed in their field, the number of waivers granted, and the difficulties reported in staffing teacher positions, reveal similar supply and demand imbalances by academic subject in Massachusetts. Analysis of data from the past two school years reveals mathematics, science, and foreign languages as the three core academic subjects most difficult to staff with qualified teachers. In both years, the shares of teachers licensed in their teaching
URBAN SCHOOLS ARE MORE LIKELY TO FILL VACANT POSITIONS BY HIRING LESS QUALIFIED TEACHERS. OVER HALF OF ALL URBAN SCHOOLS WITH VACANCIES HIRED A SUBSTITUTE TEACHER FOR AT LEAST ONE POSITION, COMPARED WITH 28 PERCENT OF SUBURBAN AND RURAL SCHOOLS.
Page 17
assignment and the shares of core classes taught by teachers deemed highly qualified were lowest in these three subject areas (see Table 4 on next page). For example, in 2008-09, only 91.9 percent of foreign language teachers were licensed in their teaching assignment. For both math and science teachers, the respective shares were around 95 percent—still considerably lower than the state average of 98 percent for all core academic subjects. Similarly, only 92 percent of foreign language classes were staffed with teachers deemed highly qualified, along with 94 and 95 percent of science and math classes, respectively. In contrast, nearly 98 percent of all core classes, on average, were taught by highly qualified teachers. Analysis of waivers granted to teachers of various subjects confirms these findings. Between 2003-04 and 2007-08, the number of waivers grew dramatically not only overall but for each of the subjects identified as experiencing shortages at the national level: special education and bilingual/ESL education at all grade levels, and science, mathematics, and foreign languages at the secondary level. The rise in waivers was particularly dramatic for special education teachers, nearly quadrupling from 404 to 1,429 in this five-year period (see Appendix Table 6). The numbers of waivers granted to math, science, bilingual/ESL, and foreign language teachers grew as well, though they remained considerably lower than the number of special education waivers. For instance, in 2007-08, the subject with the second highest number of waivers was high-school science, but at 233, its number of waivers was still dwarfed by special education. As a result, over the past six years, special education has accounted for the largest, and an ever growing, share of all waivers in the state. Fewer than 30 percent of all waivers in 2003-04 were in special education, but by 2008-09, this field accounted for more than half (see Figure 11 on page 19). Despite the fact that the number of waivers granted to the other shortage fields also increased, their relative shares of waivers declined as a result of the unprecedented growth in special education waivers. As of 2008-09, science, math, bilingual/ESL education, and foreign languages each accounted for less than seven percent of all waivers in the state.
Photograph: iStockphoto
Page 18
Table 4. Licensed and highly qualified Massachusetts teachers by academic subject
2007-08 2008-09 2007-08 2008-09
All Core Subjects 97.7 98.1 97.5 97.8
Civics/Government 98 97.7 96.6 96.3
Geography 97.8 97.7 95.4 97.6
Reading 97.2 98 94.1 95.1
Economics 97 98.7 93 95.8
Social Studies 97 97.4 94.3 95.3
History 96.8 97.6 95.9 96.8
Arts 96.6 97.6 97 97.5
English Language Arts 96.2 97.1 95.5 96.6
Mathematics 94.4 95.5 93.5 94.9
Science 94 95.4 92.2 94
Foreign Languages 90.1 91.9 90.5 92
Source: Massachusetts Department of Elementary and Secondary Education
% of schools
reporting vacancies
% of schools with
significant difficulties
in filling vacancies
Foreign languages (secondary) 64.3 40.7
Mathematics (secondary) 81.0 39.0
Special education (all levels) 86.1 26.0
Biology (secondary) 58.2 24.7
Physical sciences (secondary) 55.0 21.1
Bilingual/ESL (all levels) 43.4 18.9
English (secondary) 69.6 8.6
Music/Art 41.8 3.8
Social Studies (secondary) 67.6 2.4
General elementary education 94.1 0.0
Source: National Center for Education Statistics, 2003-04 Schools and Staffing Survey (SASS), Public School File.
Highly-qualified teachers are those that meet the NCLB definition of highly-qualified. To meet the definition, teachers must possess
a valid Massachusetts teaching license at the Preliminary, Initial, or Professional level and demonstrate subject matter competency in
the areas they teach.
Notes: Percent of teachers licensed in teaching assignment is the percentage of teachers with Preliminary, Initial, or Professional
licensure in the area in which they are teaching. Charter schools are not required to hire licensed teachers.
Table 5. Schools reporting teacher vacancies and significant difficulties filling vacancies by
subject in Massachusetts, 2003-04
% of teachers licensed in teaching
assignment
% of core classes taught by highly
qualified teachers
Note: Schools with significant difficulties are defined as those reporting it very difficult or impossible to fill vacancies in a given
subject.
Page 19
The growing shortage of qualified teachers in special education, bilingual/ESL education, math, science, and foreign languages is also evident from the fact that an ever rising fraction of teachers in each subject have waivers (see Figure 12). In 2007-08, 17 percent of bilingual/ESL teachers and 14 percent of special education teachers had waivers—nearly four times their respective shares five years earlier. In addition, between six and eight percent of math, science, and foreign language teachers had waivers at the secondary level—still higher than the state average of five percent. Another indicator of the shortfall of qualified teachers is the difficulty schools experience in filling vacant positions in each subject. Data on this indicator, available from the 2003-04 Schools and Staffing Survey, are presented in Table 5. It appears that although many schools reported having vacancies in all subjects, disproportionate shares faced significant difficulties in filling these vacancies in the previously identified shortage subject areas: foreign languages, math, and science at the secondary level, and special education and bilingual/ESL classes at all grade levels. According to this measure, foreign languages and mathematics are the subject areas most difficult to staff, with 40 percent of all schools reporting that they found it very difficult or impossible to fill their vacancies with fully qualified teachers. About a quarter of the schools reported similar difficulties for special education and biology, and approximately one fifth—for physical sciences and bilingual education. In contrast, while 94 percent of schools had teaching vacancies in elementary education, none reported significant difficulties in filling them. Finally, the length of teachers’ work experience by subject may also serve as an indirect sign of teacher shortfalls. In particular, high shares of teachers with relatively little experience in certain subjects may indicate a greater “churning” of the teacher workforces in those subjects—with more teachers leaving early in their career and generating vacancies typically filled by (equally) inexperienced teachers. Appendix Table 7 reports these data for various subject areas from the 2003-04 Schools and Staffing Survey. Though the patterns here are not as definitive as those that arise from the waiver or teacher vacancy analyses, they provide a tentative confirmation of teacher shortages in at least mathematics and natural sciences. Forty percent of math teachers and 30 percent of science teachers had no more than 5 years of teaching experience, compared with a state average of 27 percent. About a quarter of special education and foreign language teachers were new teachers, while the sample of bilingual/ESL teachers was too small to generate reliable results.
Source: Massachusetts Department of Elementary and Secondary Education
Percent
0
10
20
30
40
50
60
0
2
4
6
8
10
12
14
16
18
2003-04 2004-05 2005-06 2006-07 2007-08
Bilingual/ESL (all levels)
Mathematics (secondary)
Science (secondary)
Foreign language (secondary)
Special education (all levels)
Figure 11. Waivers by subject area, as share of all waivers,
2003–2009Percent
(Special Education)
Figure 12. Percentage of teachers in each subject
who have waivers, 2003–2008
Source: Massachusetts Department of Elementary and Secondary Education
Percent
0
2
4
6
8
10
12
14
16
18
2003-04 2004-05 2005-06 2006-07 2007-08
All teachers
Special education (all levels)
Bilingual/ESL (all levels)
Mathematics (secondary)
Science (secondary)
Foreign language (secondary)
Page 20
N umerous studies have warned about the prospects of impending teacher shortages in many parts of the country, as more and more teachers of the
baby boom generation approach retirement and student enrollment keeps growing.31 Indeed, given that the national teacher workforce has a larger share of older workers than the college educated workforce in general, the prospect of “an unprecedented loss in work experience” due to teacher retirements is daunting.32 Similar trends hold in Massachusetts, where 35 percent of teachers in the Commonwealth are aged 50 and older, compared with 31 percent of all college-educated workers.33 To project future demand and supply dynamics in the teaching profession in Massachusetts, this report employs a teacher supply and demand model similar to that used by Hussar (1998) and Aaronson & Meckel (2009). The analysis spans academic years 2010-11 through 2019-20 and is undertaken for the state as a whole and for the ten largest school districts in terms of student enrollment.34 The model first projects annual total demand for teachers based on forecasts of student enrollment and assumptions about student-teacher ratio levels. It then matches total demand to the expected supply of teachers retained from the previous school year, estimated using age-specific retention rates. Each year, the gap between the projected total demand and the returning supply is the demand for new teachers for that year—that is, the number of new teachers the state or district will need to hire to fully meet the demand created by student enrollment levels and teacher attrition from the previous year. The role of retirement is evaluated by the share of the workforce that retires each year, the fraction of total attrition that retirees constitute, and the portion of new hiring needs that they necessitate. This section describes in more detail the methodology behind each part of the model, and then discusses its findings, focusing particularly on the teacher hiring needs and retirement projections at both the state and the district level.
Methodology
Student enrollment and total demand for teachers Demand projections are based on two key pieces of data: student enrollment projections and student-teacher ratios. For the state as a whole, the model uses student enrollment projections constructed by the Department of Elementary and Secondary Education. For the ten largest districts, student enrollment over the next ten years is estimated using projections of their cohorts of five-year olds, the average propensities of their 5-year old children to attend public school kindergarten, and the average grade progression rates of their students from grades 1 through 12. To set up a range of plausible projections for total teacher demand, the resulting enrollment estimates are then divided by three different values for each district’s student-teacher ratio—its average, its lowest, and its highest ratio from over the past six years.
F��� � S����� ��� D����� D�������:
H� ��! N���� ��� ��� R��� �� R��� ������
SINCE THE NATIONAL TEACHER WORKFORCE HAS A LARGER SHARE OF OLDER WORKERS THAN THE COLLEGE EDUCATED WORKFORCE IN GENERAL, THE PROSPECT OF “AN UNPRECEDENTED LOSS IN WORK EXPERIENCE” DUE TO TEACHER RETIREMENTS IS DAUNTING.
Page 21
Returning supply for teachers, new hiring needs, and retirements Calculations of the supply of teachers returning each year begin with the actual number of teachers in academic year 2008-09, broken down by single years of age. To calculate the number of teachers retained the following year, I apply actual age-specific retention rates from between 2007-08 and 2008-09, which I assume will remain constant over the next decade.viii Not surprisingly, retention rates are lower for younger teachers at the beginning of their careers and especially low for older teachers who are retiring in large numbers. Each year the resulting supply of retained teachers is lower than the total enrollment-driven demand for teachers. This gap between them represents the number of additional teachers, or “new hires,” that the state or district needs to hire each year in order to staff all classrooms. In the model, the deficit each year is filled by adding the number of new hires necessary to exactly equate total teacher demand with teacher supply. The new hires are assumed to replicate the actual age distribution of new teachers hired between 2007-08 and 2008-09—a distribution assumed to remain constant. All teachers retained from the previous year are then made a year older, and the retention calculations are rerun on the resulting group of new hires and retained teachers, again using the age-appropriate retention rates. This algorithm is repeated through 2019-20, continuously filling the gap between total demand and returning supply with new teachers and updating the age of the retained teachers. Finally, to ascertain the impact of retirements on teacher hiring needs, I assume that all teachers aged 58 and older—the national median retirement age for teachers—who leave the workforce do so to retire.35 After calculating the number of retirees each year, I evaluate the importance of retirements for the state and for each district by looking at the share of the workforce that retires each year, the fraction of total attrition due to retirements, and the portion of new hires created to replace retiring teachers.
Results: Projections of Student Enrollment, Total Teacher Demand, Teacher Hiring Needs, and Retirements
Student enrollment and total demand for teachers
According to the Department’s state projections, public school enrollment in Massachusetts is expected to decline slowly but steadily through 2019-20, at an average rate of about half a percent a year, from nearly 924,000 students in 2010-11 to 885,000 ten years later (see Figure 13 on next page). This trend is not uniform across the state. Some of the districts evaluated here face declining enrollments, while others may see the number of their students remain steady or even rise. Projections for Boston, for example, indicate that in 2019-20 there will be 16 percent fewer enrolled students than in 2008-09 (see Appendix Table 8). Enrollments in Lawrence and Newton are projected to decline by 9 percent and 7 percent, respectively. In contrast, enrollments in Brockton, Fall River, and New Bedford are expected to increase substantially, particularly in the latter, which will likely enroll 15 percent more in 2019-20 than it did last year. Applying the three different values of the state’s student-teacher ratio yields total demand for teachers in Massachusetts in the high 60,000s and low 70,000s each year over the next decade. For example, using the highest and the lowest state ratios of 13.6 and 13.2 students per teacher from the past six years delivers a range for the total demand for teachers in 2010-11 of between 70,200 and 72,300; using the state’s average ratio of 13.4 yields a mid-point estimate of teacher demand of about 71,100 for that year. Detailed results for total teacher demand in the state as a whole and in the ten school districts are presented in Appendix Table 9.
viii Undoubtedly, using average retention rates from several years would notably improve the accuracy of these projections. Unfortunately, these data are have only been tracked since 2007.
Page 22
Hiring needs
Table 6 presents the hiring need projections for Massachusetts and the state’s ten largest school districts. According to these estimates, the state will need to hire about 45,500 new teachers over the next ten years in order to fully meet teacher demand—with annual hiring needs of just over 4,600 in 2010-11 and declining to 4,300 by 2019-20 (see also Figure 14). These projections represent mid-point estimates that assume the statewide student-teacher ratio will remain constant at 13.4, its average level over the past six years. Appendix Table 10 shows lower-bound and upper-bound estimates for hiring needs, calculated using the highest and the lowest student-teacher ratios from the past six years. As the three assumed ratios are not very different from one another, the resulting range of hiring need projections is fairly small: Total hiring needs over the next decade range from 45,000 to just over 46,000. Though these numbers are sizeable, they are not high enough to cause concerns about overall teacher shortages at the state level. Because student enrollment is estimated to decrease steadily, thereby lowering the total demand for teachers, the projected annual demand for new teacher hires actually accounts for a slightly smaller share of the previous year’s teaching workforce—6.6 percent, on average—than the seven percent that new hires constituted between 2007-08 and 2008-09 (see Table 6B). Furthermore, a total of 5,125 new teachers—the equivalent of about 5,000 full-time teachers—were hired between 2007-08 and 2008-09. And in each of the past three years, at least 5,000 students are estimated to have completed teacher training programs leading to initial licenses in the state.36 If the supply of new teachers, including both new graduates and re-entrants into the profession, remains at these levels, it will be sufficient to meet and even exceed the state’s overall hiring needs. Regional results, however, paint a very diverse picture of teacher hiring needs across the state, with some districts more likely to experience teacher shortages than others. Due to enrollment declines and high teacher retention rates, hiring needs in Lawrence and Worcester, for example, are likely to be relatively low: In these two districts, annual demand for new teachers is projected to be between six and eight percent of the previous year’s workforce (see Table 6B). Meanwhile, districts like Springfield and Fall River—and to a lesser degree, New Bedford and Newton—are likely to face significantly higher demand for new hires and may encounter difficulties in finding sufficient numbers of fully qualified teachers to meet this demand. Fall River and Springfield, for example, are projected to replace about a fifth of their teachers each year; New Bedford, Newton, and Boston are slightly behind, with annual hiring needs over the next decade ranging from 12 to 14 percent of the previous year’s teachers.
600,000
650,000
700,000
750,000
800,000
850,000
900,000
950,000
1,000,000
Figure 13. Public school enrollment in Massachusetts, 2010-2019
Source: Massachusetts Department of Elementary and Secondary Education
Students
projectedactual
Figure 14. Projected number of new teachers needed to
meet demand each year, 2010-2019
Source: Massachusetts Department of Elementary and Secondary Education
Hiring needs
20
30
40
50
60
70
80
90
100
4,000
4,100
4,200
4,300
4,400
4,500
4,600
4,700
4,800
4,900
2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 2019-20
Mid-point estimate
of hiring needs
Percent
% of hiring needs created
by retirements (right axis)
Page 23
Table 6. Projected new teacher hiring needs, 2010-11 through 2019-20
Massachusetts and the 10 largest school districts
A. Number of new teacher hires needed to meet total demand
Massachusetts
Boston
Springfield
WorcesterBrockton
Lowell
Lynn
New
Bedford
Law
rence
New
ton
Fall R
iver
2010-11
4,676
445
396
9490
90111
123
50143
162
2011-12
4,624
461
425
103
103
104
113
135
52147
168
2012-13
4,685
471
425
114
94112
119
130
52146
175
2013-14
4,684
472
431
121
110
120
125
130
61144
172
2014-15
4,621
480
444
136
107
116
126
135
65141
175
2015-16
4,611
492
446
136
110
118
134
139
66140
177
2016-17
4,559
465
450
132
112
127
129
137
68132
177
2017-18
4,397
457
459
140
111
126
134
138
66129
179
2018-19
4,388
455
465
129
116
122
132
144
63124
180
2019-20
4,297
446
469
137
114
128
135
143
64123
182
Total 2010-2019
45,543
4,642
4,41
11,24
21,06
71,164
1,259
1,353
606
1,370
1,746
B. New teacher hires as a percentage of previous year's workforce
Massachusetts
Boston
Springfield
WorcesterBrockton
Lowell
Lynn
New
Bedford
Law
rence
New
ton
Fall R
iver
2010-11
6.6
11.2
18.5
6.0
8.4
9.2
11.5
13.8
6.1
14.2
20.7
2011-12
6.6
11.8
19.8
6.6
9.6
10.6
11.8
14.7
6.4
14.6
21.3
2012-13
6.7
12.1
19.7
7.4
8.8
11.3
12.3
14.0
6.5
14.5
22.0
2013-14
6.7
12.3
19.8
7.9
10.2
12.1
12.8
13.8
7.7
14.3
21.5
2014-15
6.7
12.6
20.2
8.7
9.8
11.5
12.7
14.1
8.1
14.0
21.5
2015-16
6.7
13.0
20.1
8.7
9.9
11.6
13.3
14.2
8.2
14.0
21.6
2016-17
6.6
12.3
20.1
8.4
9.9
12.2
12.7
13.7
8.4
13.2
21.4
2017-18
6.4
12.2
20.2
8.8
9.6
11.9
12.9
13.7
8.1
13.0
21.4
2018-19
6.4
12.3
20.2
8.1
9.9
11.5
12.6
14.0
7.8
12.6
21.4
2019-20
6.3
12.2
20.1
8.5
9.6
11.9
12.7
13.7
7.8
12.6
21.4
Note: F
or lower and upper bound estim
ates of new
teacher hiring needs, see Appendix Table 10.
Source: Author's calculations.
Page 24
The supply and demand dynamics in the districts with the highest hiring needs typically result from growing student enrollments, high teacher turnover, or both. For instance, the substantial teacher hiring needs in Fall River are due both to the fact that its student enrollment is estimated to grow over the next ten years and to its substantial annual teacher attrition rate (more than 19 percent). Similar trends hold for nearby New Bedford, whose student population is projected to grow by 15 percent by 2020—the fastest among the ten largest districts. (For enrollment projections, see Appendix Table 8). In Springfield, by contrast, the demand for new teachers is almost entirely due to high attrition. While its student population is projected to grow by only 2 percent over the next decade, nearly 19 percent of its teachers left the district between 2007-08 and 2008-09. The varying influence of enrollment growth and teacher attrition often results in vastly different hiring needs even in districts with otherwise similar enrollments. For example, Springfield and Worcester have similar numbers of students—about 23,000 and 25,000 in academic year 2008-09, respectively. However, Springfield’s teacher attrition rate is more than twice as high, and its student population is projected to grow while Worcester’s is expected to shrink by five percent. As a result, annual hiring needs are projected to equal about 20 percent of the previous years’ teacher workforce in Springfield, but only about eight percent in Worcester. In fact, given its stably growing student population and its higher teacher turnover, Springfield is expected to need roughly the same numbers of new teachers as Boston, whose student enrollment is more than twice as large.
The role of teacher retirements
Projections indicate that about 19,000 teachers in Massachusetts will retire over the next decade (see Table 7 on page 25). Annual retirements between 2010-11 and 2013-14 are estimated to be just over 2,000 and to decline to about 1,600 by the end of the decade. In relative terms, these numbers represent between 2.4 and 2.9 percent of each year’s total teaching workforce—shares similar to the three percent that retired between 2007-08 and 2008-09. Statewide, retirements will account for a significant portion of total teacher attrition—between 36 and 40 percent each year—and will create between 38 and 45 percent of the demand for new teachers (see Figure 14 on page 22, and Appendix Table 11). As with hiring needs, retirement trends and their impact differ substantially across districts, primarily due to differences in the age distribution of each district’s teachers and in their propensity to leave teaching at different ages. Over the next decade, the annual number of teacher retirees is estimated to decline gradually in districts like Boston, Springfield, Lawrence, Newton, and Fall River, and to remain constant or increase in Worcester, Lowell, Lynn, and New Bedford (see Table 7). The district most significantly affected by retirements is Worcester, where 40 percent of all teachers are aged 50 or older. In that district, more than three percent of all teachers are projected to retire each year. The role of retirements is slightly smaller in Boston, Springfield, Lowell, Lynn, New Bedford, and Fall River, where between 2.5 and 3 percent of teachers are expected to retire each year. In Brockton, Lawrence, and Newton, these shares are lower still. At the same time, due to differences in teachers’ age-specific attrition rates and in projected student enrollments, the impact of retirements in creating teacher hiring needs, and thus potential teacher shortages, differs even for districts with similar shares of retirees. For example, turnover among young teachers in Fall River and Springfield is particularly high, and enrollment in both districts is projected to grow. As a result, retirees are expected to account for a relatively small share of annual exits and to create only about 12 to 15 percent of new teacher openings (see Appendix Table 11). Instead, the majority of hiring needs in these districts will be necessitated by growing student enrollment and high turnover among younger teachers. In contrast, although slightly higher shares of teachers in Worcester are projected to retire each year, the impact of these retirements will likely be much more noticeable. Because of shrinking student enrollment and high retention of younger teachers, Worcester retirees will account for about 40 percent of annual teacher attrition and create two out of every five new teaching positions.
Page 25
Table 7. Projected teacher retirements, 2010-11 through 2019-20
Massachusetts and the 10 largest school districts
A. Number of teacher retiring each year
Massachusetts
Boston
Springfield
WorcesterBrockton
Lowell
Lynn
New
Bed
ford
Law
rence
New
ton
Fall R
iver
2010-11
2,03
9122
6942
1526
2419
1729
29
2011-12
2,08
0122
6745
1430
2819
1626
24
2012-13
2,04
9120
6652
1328
2426
1522
26
2013-14
2,02
5118
6748
1832
2823
1520
24
2014-15
1,98
2116
6554
1530
2424
1419
23
2015-16
1,89
3107
6255
1527
2823
1317
23
2016-17
1,82
1106
6150
1632
2823
1316
23
2017-18
1,74
5105
6156
1328
2724
1314
23
2018-19
1,67
798
6150
1426
2823
1214
22
2019-20
1,61
996
6151
1325
2624
1113
23
Total 2010-20
1918,930
1,110
641
504
145
284
265
229
139
189
240
B. Retirees as a percentage of previous year's teaching workforce
Massachusetts
Boston
Springfield
WorcesterBrockton
Lowell
Lynn
New
Bed
ford
Law
rence
New
ton
Fall R
iver
2010-11
2.9
3.1
3.2
2.7
1.4
2.7
2.4
2.1
2.1
2.8
3.7
2011-12
2.9
3.1
3.1
2.9
1.3
3.0
2.9
2.1
2.0
2.6
3.1
2012-13
2.9
3.1
3.1
3.4
1.2
2.8
2.5
2.8
1.9
2.2
3.3
2013-14
2.9
3.1
3.1
3.1
1.6
3.2
2.9
2.5
1.9
1.9
3.0
2014-15
2.9
3.0
3.0
3.5
1.4
2.9
2.4
2.5
1.7
1.9
2.8
2015-16
2.7
2.8
2.8
3.5
1.3
2.7
2.8
2.4
1.6
1.7
2.8
2016-17
2.6
2.8
2.7
3.2
1.4
3.0
2.7
2.4
1.6
1.6
2.8
2017-18
2.5
2.8
2.7
3.6
1.1
2.7
2.6
2.4
1.6
1.5
2.8
2018-19
2.5
2.7
2.6
3.2
1.2
2.5
2.7
2.2
1.4
1.4
2.7
2019-20
2.4
2.6
2.6
3.2
1.1
2.3
2.4
2.3
1.4
1.4
2.7
Note: F
or additional data on projected teacher retirem
ents, see Appendix Table 11.
Source: Author's calculations.
Page 26
G iven the projected enrollment declines and the sufficient number of new teaching graduates each year, the Commonwealth is unlikely to face an overall
shortage of teachers over the next decade. However, while teacher supply may meet or even exceed demand in the aggregate, the state will continue to face major challenges in strengthening its teacher workforce. Several large urban districts with rising low-income and minority student enrollments and high teacher turnover will likely experience greater teacher hiring needs and more potential shortages than others—a development that may exacerbate the already inequitable distribution of qualified teachers across the state. Furthermore, subject areas like mathematics, natural sciences, special education, and foreign languages are already experiencing shortages of fully qualified teachers, as the candidates to teach those fields often face more favorable labor market opportunities outside of teaching. Finally, the teaching workforce as a whole is far less racially and ethnically diverse than the student population, as the state’s minority residents are both less likely to graduate from college and less likely to opt for a teaching career.37 The challenges of uneven hiring needs, inequitable distribution of teachers across regions and subjects, high teacher turnover, and lack of diversity in the workforce call for strategic and targeted policy responses. To develop a teaching workforce that is more equitably distributed, more stable, and more diverse, the Commonwealth needs to devise strategies to attract sufficient numbers of diverse and talented teachers into high-demand subjects and difficult-to-staff areas, and then retain them over an extended period of time. This goal would be best served not by a general policy of increasing the number of college graduates going into teaching, but by targeted recruitment and retention approaches that create incentives for prospective teachers to train in shortage subjects, teach in high-need areas, and do it as a long-term career rather than a temporary stint. This section discusses four types of recruitment and retention policies, with specific examples from programs that have been implemented in Massachusetts or in other states. These programs have either shown promise or imparted design or implementation lessons that help inform recommendations on what else the state can do to attract and retain the best talent into the teaching profession. The four policy approaches include financial incentives and alternative certification programs to expand the supply of traditional and non-traditional recruits willing to teach in high-need areas and subjects; and salary incentives and work condition policies to help retain them.
Recruiting New Teachers
Financial Incentives for New Teachers
College education, including teacher preparation programs, is expensive and requires sizeable investments on the part of the students. Yet teaching pays less than most other highly-skilled occupations, forcing students who may have chosen teaching to opt for more lucrative professions. This is especially true for mathematics and science students, whose training is in high demand in the knowledge economy and who command much greater compensation outside of teaching. In an attempt to alleviate the financial burdens that may discourage graduates from teaching—especially in areas with high cost of living like our
S� ��!������! ��� T����� W� ��� ��:
P����� O������ ��� C� ��� I����������
SEVERAL LARGE URBAN DISTRICTS WITH RISING STUDENT ENROLLMENTS AND HIGH TEACHER TURNOVER WILL EXPERIENCE GREATER TEACHER HIRING NEEDS AND MORE POTENTIAL SHORTAGES THAN OTHER AREAS.
THE COMMONWEALTH NEEDS STRATEGIES TO ATTRACT SUFFICIENT NUMBERS OF DIVERSE AND TALENTED TEACHERS INTO HIGH-DEMAND SUBJECTS AND DIFFICULT-TO-STAFF AREAS, AND RETAIN THEM OVER AN EXTENDED PERIOD OF TIME.
Page 27
state—most states have undertaken at least one fiscal incentive program, such as tuition support, loan forgiveness or repayment assistance, or signing bonuses.38 These programs vary widely in design, financial magnitude, and eligibility require-ments. Some finance significant portions of a teacher’s education; others provide a one-time signing bonus only. Some are available to all students interested in teaching, while others target those who choose shortage subjects or prefer to teach in hard-to-staff schools and districts. Given the large variation in program design and implementation, little is known about the overall effectiveness of financial incentives in increasing the supply of talented teachers. However, individual programs that have proven successful (or ineffective) can offer important lessons on which features bolstered (or compromised) their effectiveness. This section highlights several such examples from Massachusetts and elsewhere, and draws insights about how to enhance the effectiveness of current and future initiatives in the Commonwealth. The MINT/Signing Bonus Program for New Teachers The Massachusetts Signing Bonus Program, now defunct, was among the programs established by the 1998 Teacher Quality Enhancement Act (also known as the “12 to 62 Plan”) in a major state effort to improve the educator workforce and encourage “the best and brightest candidates to teach in the public schools.” 39 Throughout its existence, the signing bonus program operated in conjunction with the Massachusetts Institute for New Teachers (MINT), an accelerated teacher training program, as all bonus recipients were required to go through the MINT program. Given how closely the two programs are inter-related, they are discussed jointly in this section, despite the fact that the MINT program is an alternative certification program of the type reviewed in the next section. The primary goals of the two programs were to reduce teacher shortages in high-demand subjects and in high-need schools by attracting high-quality recruits into the profession from among both recent college graduates and mid-career professionals.40 In addition, a plan was in place to market the program intensively to candidates from underrepresented populations (though the eligibility criteria did not specifically target such candidates). Participants in the signing bonus program received a total of $20,000, paid out over four years and targeted to prospective teachers in the highest-need core subject areas. Their tuition for the MINT program was also covered. As part of the MINT program, bonus recipients—along with a roughly equal number of
Photograph: iStockphoto
Page 28
non-bonus participants—undertook seven weeks of summer training and support seminars during their first year of teaching in order to move from provisional certification to provisional certification with advanced standing.ix Overall, in 2001, the MINT program trained 220 participants, of whom 115 were bonus recipients. The Signing Bonus program was by far the more expensive of the two, costing a total of $2.3 million for the 115 recipients, excluding MINT scholarship expenses. The costs of the MINT program were substantially lower, at approximately $600,000 in 2001.41 Finally, the Bonus program was discontinued after three years of operation. The MINT program continued to operate until 2008, when its funding from the federal Transition to Teaching grant ran out.42 A 2003 evaluation of the 12-to-62 initiatives uses data from interviews, focus groups, and surveys, and reveals mixed results about effectiveness of the two programs, though generally more positive for the MINT than the Signing Bonus program. Selection seems to have successfully targeted high-quality recruits, as the vast majority of interviewed principals were satisfied with MINT participants, and 89 percent stated that they would hire them again without reservations. The MINT program was also effective at attracting non-traditional candidates, as two thirds of its participants were mid-career professionals. One clear finding is that the accelerated MINT program was of higher importance to new recruits than the signing bonuses. Though the bonus feature may have increased the publicity of the initiative, more than three quarters of surveyed participants stated that they would have considered enrolling even without the bonuses. In addition, 92 percent of participants regarded the expedited nature of the training program as a significant factor in their decision to enroll, while just over half felt the same way about the bonuses. Liu, Johnson, and Peske (2004) confirm these findings. Their analysis followed 13 MINT participants who were also bonus recipients and found that the alternative certification program was far more influential than the signing bonus in attracting them.43 Perhaps most importantly, the MINT/Signing Bonus program was effective at producing more teachers in shortage subjects but fell short of its objectives to induce them to teach in high-need schools and to attract minority candidates. Overall, nearly 60 percent of all MINT graduates taught subjects with demonstrated teacher shortages: A third became science teachers, 17 percent taught math, and another 10 percent taught special education or foreign languages. At the same time, only one third of both MINT participants and of the subset of bonus recipients took teaching jobs in high-need urban districts. Finally, success at attracting minority candidates was negligible at best: the share of minority participants was in the low single digits.44 Two characteristics of the MINT/Signing Bonus program appear to have hampered its effectiveness. Many interview respondents expressed doubts that the bonus segment was the wisest use of limited financial resources. They felt that while it was too low to make a difference, it still generated resentment among non-bonus receiving teachers and attracted people with no long-term attachment to the profession.45 Second, despite the fact that this program was among the most visible initiative in the 12 to 62 Plan, a number of key stakeholders like school principals, union leaders, and higher education officials still did not know much about its specifics. Poor communication with the Department of Education, especially with regards to program requirements and various deadlines, was a frequent complaint among program participants.
ix Provisional certification and provisional certification with advanced standing have now been replaced with preliminary licenses and initial licenses, respectively. While both require completion of a bachelor’s degree and passage of the state’s MTEL test, applicants for initial license must also complete an approved teacher preparation program, either through a higher education institution or an alternative certification track.
THE MINT/SIGNING BONUS PROGRAM WAS EFFECTIVE AT PRODUCING MORE TEACHERS IN SHORTAGE SUBJECTS BUT FELL SHORT OF ITS OBJECTIVE TO INDUCE THEM TO TEACH IN HIGH-NEED SCHOOLS.
Page 29
Attracting Excellence to Teaching Program (AET) The Attracting Excellence to Teaching Program was established in 1995 as a major component of the 1993 Massachusetts Education Reform Act. Its objective was to increase the number of high-ability graduates into the teaching profession by attracting talented candidates from both Massachusetts and other states. The program provided loan reimbursement of up to $1,800 per year for four years to graduates who were currently teaching in public schools, had graduated either with honors or in the top 15 percent of their class, and were actively repaying student loans.46 Though eligibility was not restricted to teachers in high-need districts, they were given priority whenever the number of applicants exceeded the funds available for loan assistance. Between 1997 and 2001, nearly 2,500 teachers received loan assistance through the program. In 2000, for example, the program served 700 teachers, at a total cost of $1.08 million in loan reimbursements.47 However, in an attempt to close a large budget deficit in fiscal year 2004, the state legislature eliminated funding for the program and effectively ended it.48 According to the 2003 evaluation of the 12 to 62 Plan, the AET program attracted participants who, on average, replicated the demographic characteristics of the teaching workforce—80 percent were women, and 90 percent were non-Hispanic whites. In other words, the program does not appear to have been effective at attracting a more diverse body of teachers. Three quarters of program participants taught in subject areas like elementary education, English, history, and social sciences, none of which have experienced significant shortages of qualified teachers. Admittedly, however, neither the creation of a diverse teacher workforce nor the alleviation of teacher shortages by subject were among the program’s main objectives.49 Effectiveness with respect to explicitly stated program goals—to increase the number of high-achieving individuals in teaching in general and in high-need areas in particular—was also limited. The AET program does not appear to have been a significant factor in graduates’ decision to enter teaching. Three quarters of survey respondents considered the program an insignificant factor in their career choice, and 98 percent claimed that they would have entered teaching even in its absence. And although two thirds taught in high-need schools at the time of the evaluation, 87 percent had no knowledge of the program when they initially decided where to teach. In any case, 78 percent claimed that the program would not have influenced their choice of location, even if loan reimbursements were restricted to teachers in high-need areas. Furthermore, 90 percent of those surveyed claimed that the reimbursement levels would need to be significantly increased in order to truly impact their decision about whether and where to teach. For example, nearly two thirds of respondents considered an annual reimbursement level of $4,000—more than twice as high as the actual level—to be an more effective incentive; 89 percent thought that $5,000 would have made the difference.50 Interestingly, the AET program appears more effective at retaining existing teachers than at attracting new members into the profession. The program seems to have boosted morale among teachers: 45 percent claimed that the loan payments increased their satisfaction with their career, and 16 percent claimed that they were more likely to remain teachers for longer as a result of the program. In contrast to the MINT/Signing Bonus program, the vast majority of AET participants were pleased with the way the Department of Education administered the program and communicated its features and requirements. Instead, what may have limited its success at attracting high-quality candidates is the lack of publicity it received during its existence. According to the evaluation, districts and teacher training programs do not seem to have used the program much as a recruiting tool.
WHAT MAY HAVE LIMITED THE AET PROGRAM’S SUCCESS AT ATTRACTING HIGH-QUALITY CANDIDATES IS THE LACK OF PUBLICITY IT RECEIVED DURING ITS EXISTENCE.
Page 30
Eighty-six percent of program participants learned about it when they were already teaching, often by accident. Initiatives in other states Despite their popularity, many of the programs to recruit and retain teachers into the profession in general and into shortage subjects and hard-to-staff schools in particular are relatively new and most have not been evaluated.51 The North Carolina Teaching Fellows (NCTF) program, one of the few exceptions, has been cited as one of the most successful state initiatives of this type. Enacted in 1986, the NCTF program fully funds the college tuition of hundred of high-achieving students who commit to teach in the state’s public schools for at least four years. Since its inception, the program has trained more than 8,000 teachers. It has been evaluated as effective both at increasing the supply of teachers in shortage academic fields and at attracting higher shares of both male and minority graduates into the profession. Each year at least 20 percent of program participants are minority and 30 percent are male.52 Finally, a 1998 evaluation that followed recipients for several years found that three quarters of them were still teaching in the state’s public schools seven years after graduation.53 Another North Carolina program that has proven moderately effective is the North Carolina Math/Science/Special Education (MSSE) Teacher Bonus program. The program awards annual bonuses of up to $1,800 to mathematics, science, and special education teachers in schools with high enrolments of poor and low-performing students. An analysis comparing the mobility of teachers in schools eligible for the program and in schools that barely missed the eligibility threshold found that the bonus reduced teacher turnover by about 12 percent. However, the authors also found that implementation issues, such as lack of clear communication about eligibility, may have reduced the impact of the program.54
Photograph: Erica Harrison
Page 31
Key lessons Despite the relatively sparse research on what makes financial incentive programs successful, several important, albeit unsurprising, insights emerge. To be effective at attaining their stated objectives, programs need to be both strategically designed and smartly implemented. The examples reviewed here reveal that, at the very least, a program’s features should be designed with its primary goals explicitly in mind. For example, if an important policy goal is to increase the number of teachers in shortage academic fields, a program whose eligibility expressly targets candidates with such commitment would likely perform better at attaining this goal than programs with broader eligibility criteria. For instance, that the MINT/Signing Bonus program was successful at increasing the supply of teachers in shortage subjects but not in hard-to-staff areas should not be surprising, given that program features did not include any location-based incentives. Furthermore, financial incentives need to be set at generous enough levels—though what levels are adequate would certainly be a very difficult parameter to determine. An annual bonus of $1,800 appears to have been high enough to make the North Carolina program effective, while the identical amount in loan reimbursement provided by the AET program here was widely viewed as insufficient to have an impact. Differences between teacher salaries in the two states may explain some, but likely not all, of the difference in program effectiveness: At the time the AET program was in operation, $1,800 equaled just over three percent of the average teacher salary in Massachusetts; in North Carolina, this fraction was about four percent. Finally, implementation errors may doom the success of even the most thoughtfully designed programs. In particular, clarity and consistency of program eligibility and high levels of publicity are key to enabling program effectiveness. The lack of public awareness and the confusion about eligibility and deadlines hampered the success of both the MINT/ Signing Bonus program and the AET loan reimbursement program; similar issues may have also reduced the impact of the North Carolina bonus program.
Expanding the Non-Traditional Pipeline
While financial incentives like tuition assistance and signing bonuses may attract more students onto the teaching track in college, they are unlikely to entice large numbers of mid-career professionals to invest the time and money needed to complete a traditional four-year teacher training program. Through previous training and work experience, these professionals have accumulated valuable subject-matter knowledge, often in fields like mathematics and science, and may consider teaching if entry into the profession were eased. To tap into this underused pool of talent, most states have created accelerated alternative teacher certification programs that do not require the completion of another bachelor’s degree, thereby reducing the barriers to teaching for mid-career professionals. Such programs have skyrocketed in popularity over the past decade: Between 2000 and 2004, the number of alternative program completers nationwide increased by 40 percent. Currently, nearly all states have implemented at least one such program, and alternative routes produce a fifth of all teachers in the nation.55 In Massachusetts, these programs still account for a relatively small share of teachers: In 2007-08, only eight percent of teaching candidates who passed the state’s teacher licensing test came from alternative training routes.56
IF A PROGRAM’S GOAL IS TO INCREASE THE NUMBER OF TEACHERS IN SHORTAGE SUBJECTS, PROGRAMS WHOSE ELIGIBILITY CRITERIA EXPLICITLY TARGET SUCH CANDIDATES WILL LIKELY ATTAIN THIS GOAL MORE EFFECTIVELY THAN PROGRAMS WITH UNIVERSAL ELIGIBILITY.
Page 32
The structure of alternative certification programs varies widely, though most offer a combination of classroom experience and at least some academic coursework. They are hosted by a variety of institutions, including school districts, colleges and universities, and non-profit organizations.57 Despite their rising popularity, there is limited and inconclusive research on what features make these programs more (or less) effective at producing high-quality teachers.58 However, evaluations of individual programs across the country are beginning to yield insights into how alternative program graduates differ from traditionally trained teachers in terms of initial employment, retention, and diversity. A longitudinal study in Texas followed graduates of alternative and of traditional certification programs for five years after graduation in the mid-1990s, and found significant differences in their initial employment as teachers and in their subsequent attrition from the profession. Alternative program graduates were significantly more likely to enter teaching: 91 percent took teaching jobs within their first year, compared with 71 percent of traditional program trainees. Alternative program completers, however, also tended to leave the profession at higher rates. By the end of year 5, 19 percent had exited teaching, compared with 12 percent of their traditionally trained peers. Darling-Hammond (2000) finds similar employment rates, though much greater attrition. Summarizing estimates from programs in several states, she concludes that more alternative program graduates than traditional graduates enter teaching—80 percent and 70 percent, respectively—but that three years after graduation only a third of alternative program completers are still teaching, compared with half of traditional program graduates.59 Finally, alternative programs have been found to enlist significantly more minority teaching candidates, making them a potentially attractive conduit to a more diverse teaching workforce.60 Given the wide variety in alternative program design and their mixed impact on teacher recruitment and retention, a mere proliferation of such programs would not be a panacea to the state’s teacher workforce issues. Nonetheless, smart and strategic design and thoughtful implementation may make such programs an important mechanism for increasing the number of non-traditional teachers in high-need areas and high-demand subjects. For example, the MINT program, discussed in the previous section, was particularly effective at producing teachers in subjects with demonstrated teacher shortages, with nearly 60 percent of program graduates teaching math, science, special education, or foreign languages. The rest of this section profiles two other initiatives from Massachusetts that have shown promise. The Boston Teacher Residency The Boston Teacher Residency (BTR) is a form of Urban Teacher Residency—a relatively new type of alternative teacher training program operated by school districts. UTRs incorporate best practices in teacher recruiting, preparation, placement, and induction in order to attract a diverse and talented pool of trainees, prepare them to be successful teachers in shortage subjects and high-need urban schools, and retain them in teaching. While design, duration, and costs vary across programs, participants are typically chosen through a rigorous selection process. Similarly to the medical residency model, UTRs combine master’s level course-work at a partnering educational institution with a year-long classroom residency alongside an experienced “master teacher.” Participants then become teachers with their own classrooms but continue to receive mentoring and support.61 The Boston Teacher Residency itself is a highly-selective 13-month long program, first established in 2003. In 2007-08, the program enrolled 84 residents out of a pool of 450 applicants. While in residency, BTR participants work alongside a
SMART DESIGN AND IMPLEMENTATION MAY MAKE ALTERNATIVE TRAINING PROGRAMS AN IMPORTANT MECHANISM FOR INCREASING THE NUMBER OF NON-TRADITIONAL TEACHERS IN HIGH-NEED AREAS AND SHORTAGE SUBJECTS.
Page 33
mentor teacher and take graduate-level coursework at the University of Massachusetts. They graduate with a master’s degree in education and receive dual licensure in special education. Annual program costs amount to about $2.5 million, as residents receive a first-year stipend of $11,100, $3,700 for their master’s degree tuition (paid for with AmeriCorps funds), and $10,000 to cover the program’s tuition. The latter amount is typically in the form of loans, forgiven over a three-year period if the resident remains a teacher in the Boston Public Schools.62 An important advantage of UTR programs is that districts can customize their design to meet specific district needs, as has been the case with the BTR. In launching the program, the Boston Public Schools placed a high priority on recruiting teachers in high-need academic subjects and special education, and from more diverse racial and ethnic backgrounds. As a result, nearly 60 percent of program participants teach in high-need subjects and virtually all teach in high-need schools.63 Finally, as an important source of African-American and Hispanic teachers, the BTR has become the district’s main strategy for strengthening the diversity of its teacher workforce. Fifty-five percent of BTR recruits in 2007-08 were nonwhite, making their cohort more diverse than teachers in the district, and substantially more so than teachers in the state.64 Since UTRs have been in existence for only a few years, there is insufficient data to determine how effective their graduates are at promoting student learning. The BTR has commissioned outside research to determine the impact of the program on student outcomes, though results from this research are unavailable yet. However, school administrators’ assessments of BTR graduates appear very positive. More than 88 percent of residents have been rated by their principals as just as effective or more effective than their peer teachers. In addition, 94 percent of principals were willing to hire more BTR graduates in the future.65 Finally, UTRs like the Boston Teacher Residency have been significantly more successful at retaining their graduates than both traditional and alternative certification programs. After three years, an average of 90 percent of BTR graduates still teach. In other UTRs like Chicago’s Academy for Urban School
Photograph: Kim Martin
Page 34
Leadership, the three-year retention rate is even higher, at 95 percent. According to researchers, the extremely high retention rates of UTR graduates are the product of a confluence of factors, such as the high-quality preparation and the continuing mentoring and support provided by the programs, the up-front requirement to commit to teaching in shortage subjects and in high-need schools, and the financial penalties for those failing to meet these commitments.66 The promise of UTR programs like the Boston Teacher Residency has attracted the attention of both federal and state education policy makers. The Higher Education Opportunity Act passed by Congress in 2008 authorized millions of dollars for the development of UTRs across the country. In addition, an Urban Teacher Residency Institute has been established to provide technical assistance to districts interested in developing their own UTRs.67 Finally, as part of the Readiness Project, Massachusetts Governor Deval Patrick announced a plan to partner with teacher colleges to establish a statewide teacher residency program modeled after the BTR by middle of this decade.68 TEACH! SouthCoast TEACH! SouthCoast is another alternative teacher preparation program intended to attract high-quality nontraditional candidates who already have a college degree to teach shortage subjects in high-need urban schools. It is run in partnership by the University of Massachusetts Dartmouth, the Fall River and New Bedford school districts, and the SouthCoast Education Compact. With only 20 participants each year, the program is smaller than the Boston Teacher Residency.69 Though no evidence of its effectiveness is yet available, it shares many of the BTR characteristics which may contribute to its success. TEACH! SouthCoast is specifically intended for prospective middle school and high school mathematics and science teachers. Like the BTR, it integrates academic coursework and teaching practice in a year-long program in two large urban school districts with documented shortages of high-quality teachers—Fall River and New Bedford. Prior to enrolling, participants are required to have passed the MTEL—the state’s teacher licensing test. Upon graduation, they receive an initial teaching license. Throughout the program, they receive intensive mentoring from veteran teachers.70 After the first year, they are typically hired by one of eight participating schools in the two districts in small groups, providing them with an in-school support system composed of their fellow graduates.71 Program participants receive a 50-percent reduction in tuition at the UMASS Dartmouth campus as well as a stipend of $5,000 during their first year. In return, they commit to teaching in the two districts for at least three years.72 Because of acute teacher shortages in some schools, several participants have been hired as full-time working teachers during their first year, instead of spending it in apprenticeship alongside an experienced teacher. Though this was not the original intent of the program and it is unclear whether these teachers are truly ready for the classroom after only a summer of training, this opportunity enables them to begin earning full pay immediately—an incentive which may prove important for mid-career switchers unwilling to take too much time off.73
Improving Teacher Retention
Pre-retirement turnover is among the most important factors contributing to teacher shortages in low-income urban areas and in high-need academic subjects. As a result, simply increasing the supply of new teachers is unlikely to resolve these shortages. As Ingersoll (2003) notes, increasing supply without addressing the turnover challenge is like constantly refilling “a bucket rapidly losing water because of holes in the bottom.”74 Effective retention strategies are needed to slow down “the revolving door” of teacher turnover, thereby minimizing the loss of talented teachers, reducing its negative impact on
Page 35
student achievement, and saving schools and districts thousands in teacher replacement costs.75 There is voluminous research on the factors that drive teachers to leave the profession or influence their preferences about where to teach. Compensation and working conditions are two of the most powerful predictors of teacher choices.76 That teachers are sensitive to compensation is hardly surprising. Teaching remains a relatively low-paid occupation and implies a high opportunity cost for teachers with stellar academic credentials, especially math and science teachers, who face more favorable employment opportunities elsewhere. At the same time, given that teacher salary schedules vary little within and across districts, recent research has shown that working conditions may matter even more than compensation.77 Teachers appear particularly drawn to schools with low concentrations of poor, minority, and low-achieving students, as these characteristics are typically proxies for schools with scarce resources, inferior facilities, more disciplinary problems and safety concerns, and high student turnover.78 This section discusses teacher compensation and working conditions in more detail and draws insights about how the state can use them to reduce teacher turnover.
Teacher Salaries
In 2007, the average Massachusetts teacher earned $58,257—over a third more than in 1997 in nominal terms and 5 percent more when adjusted for inflation.x Absolute salary levels are, however, less useful in informing policy than the analyses of how teacher salaries here compare with those in neighboring states, in comparable occupations, and across communities within the Commonwealth. Inter-state comparisons In 2007, the latest year for which inter-state data are available, the average teacher salary in Massachusetts ranked 7th in the nation but near the middle among neighboring states (see Appendix Table 12). Teachers in Connecticut, Rhode Island, and New York earned more than in the Commonwealth, while those in Maine, New Hampshire and Vermont earned considerably less. (The cost of living in the three northern New England states is considerably lower, however, suggesting that Connecticut, Rhode Island, and New York form a more appropriate comparison group for our state.) Similar findings hold when average beginning salaries are compared across the region.79 In 2005, the average beginning teacher in Massachusetts earned $35,421, ranking 11th in the nation but still behind Connecticut and New York.80 At the same time, teacher salaries in Massachusetts grew faster than in all neighboring states, except Rhode Island. According to the American Federation of Teachers, between 1997 and 2007, the average inflation-adjusted teacher salary in the state increased by 4.7 percent, just below Rhode Island’s growth rate. The average teacher salaries in Vermont and New Hampshire grew at a much slower pace, while in Maine, Connecticut, and New York, they actually declined.81 Occupation comparisons National comparisons indicate that teachers earn considerably less than workers in comparable occupations. (State-level data on such comparisons are unavailable.) For example, in 2007, teachers earned an average of $51,009 nationwide—a quarter less than the median worker in other occupations that require a college degree,
x In 2008, the average teacher salary in Massachusetts increased to $64,166. However, in order to maintain comparability with another salary data source (The American Federation of Teachers), the time period used in the state-level analysis here is from 1997 to 2007. * Gerson-Nieder, J., Tamerat, J., Gelinas, A. “Teachers from Somerville, Boston, and Longmeadow talk about being asked to do too much with too little.” (2008). CommonWealth Magazine, 13 (3): 61-66, p.62.
Unless class size, planning time, and
pay are addressed, we will continue to see the least experienced teachers funneled into the highest-needs schools and then, after a few years, depart for better paid and easier w o r k— e v e n t h o s e teachers who under different circumstances would have happily stayed in an urban setting.
- Massachusetts public school teacher *
“
“
Page 36
such as accountants, engineers, and computer pro-grammers.82 At the same time, Census estimates for 2008 show that teachers in Massachusetts earned salaries on par with those of other public sector workers in the state.83 Furthermore, in comparison to the average worker in the private sector (not just those employed in comparable occupations), teachers in Massachusetts earned about eight percent more in 2005. Still, this represents a considerable decline from ten years earlier, when the average teacher salary was more than a quarter higher than that in the private sector.84 The relative decline is due to the fact that teacher salaries grew at a considerably slower pace than did salaries in other occupations. For example, a nationwide analysis of teachers’ weekly wages shows that between 1996 and 2004 they rose by less than 1 percent in real terms, compared with a 12-percent growth in the weekly wages of all workers.85 Beginning teacher salaries fared even worse in comparison with the average private sector earnings. In 2005, the average beginning teacher in our state earned only 70 percent of what the average private worker did—down from 76 percent a decade earlier.86 Salaries within the Commonwealth In the years since the passage of the 1993 Education Reform Act—one of whose objectives was to equalize educational resources across the state—regional differences in teacher salaries have declined somewhat (see Figure 15). Nonetheless, teachers in communities with high concentrations of low-income students still earn less than their colleagues in wealthier districts. For example, teachers in districts with shares of low-income students above the state’s median earned, on average, more than $2,000 less than teachers in the remaining districts (see Figure 16). Teachers in the poorest districts—those in the top quartile in terms of their share of students eligible for free or reduced price lunch—earned over $3,500 less than their peers in the least poor quartile. In other words, teachers in the poorest districts were paid only 94 cents for every dollar earned by teachers in the wealthiest districts (see Appendix Table 13). Interestingly, these findings do not hold when teacher salaries are compared across districts with different concentrations of minority students or in different locations. Teachers in the more diverse half of districts earned an average salary of $63,433, compared with $60,812 for those in the remaining districts (see Figure 16). Quartile analysis reveals that, in fact, only teachers in the most diverse quartile of districts earned more than their colleagues; average salaries in the remaining three quartiles were similar (see Appendix Table 13). Lastly, teachers in urban districts earned over $5,000 more, on average, than their colleagues in suburban and rural locations. The salary differentials across high- and low-minority districts and urban and suburban locations are perhaps indicative of the fact that
Figure 15. Inflation-adjusted average teacher salaries,
by districts’ share of low-income students
Source: Massachusetts Department of Elementary and Secondary Education
1997 dollars
$35,000
$37,000
$39,000
$41,000
$43,000
$45,000
$47,000
$49,000
$51,000
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Least poor quartile Second quartile
Third quartile Poorest quartile
$4,322
$5,538
63,433
60,812 60,943
63,302
66,966
61,874
50,000
55,000
60,000
65,000
70,000
High-minority districts
Low-minority districts
Low-income districts
High-income districts
Urban districts Suburban/rural districts
Figure 16. Average teacher salaries,
by district characteristics, 2007-08
Source: Massachusetts Department of Elementary and Secondary Education
Dollars
Page 37
higher compensation is needed to attract and retain teachers to these districts, as their schools are more frequently plagued by problems, such as lack of supplies, inadequate facilities, and poor student discipline. Finally, it is important to note that average salaries are a rather crude indicator for analyzing teacher compensation differences across the state. The aggregate data make it impossible to control for teacher characteristics that directly affect compensation, such as educational credentials and years of experience. As a result, the district variation is not only due to differing pay scales across districts but also to differences in their teachers’ education and experience. For example, if low-income districts have more inexperienced teachers on staff than wealthier locations, the average salaries in those districts would be lower even if the underlying pay scales are identical. Policy options Since teachers’ preferences of whether and where to teach are at least partially sensitive to compensation, salary increases have been among the most popular strategies for attracting and retaining teachers in the profession.87 Raising salaries across the board has been frequently suggested given the substantial slip in teachers’ pay over the past decade.88 Proponents of overall salary increases point to Connecticut’s experience over the past two decades. As part of its 1986 Education Enhancement Act, Connecticut raised its minimum beginning teacher salaries and equalized salary levels across the state. The average teacher salary in Connecticut is now the second highest nationwide, behind only California’s.89 Within three years, teacher shortages were virtually eliminated across the state, even in urban locations. Furthermore, the salary increases, along with policies to strengthen teacher training and licensing processes, have catapulted the state’s student achievement in reading and mathematics to the top of the national distribution.90 Whether this approach can be replicated in Massachusetts with equal success is uncertain. To be effective, an overall salary increase would need to bring teachers’ pay considerably closer to that in competing occupations. As teacher salaries lag substantially behind those of comparably qualified workers, the adequate across-the-board increase would be sizeable and therefore very expensive. (The Connecticut salary initiative, for instance, cost the state more than $300 million in 1986—an equivalent to $587 million in today’s dollars.91) Furthermore, while the average teacher salary in Massachusetts is lower than in New York, Connecticut, and Rhode Island, the differences are small—in the magnitude of a couple of
Photograph: iStockphoto
Page 38
thousands of dollars. As a result, to attract teachers from those states despite the associated moving costs, salaries in the Commonwealth would need to increase by considerably more.
In addition, an overall salary increase is likely to lead to an inefficient allocation of scarce financial resources.92 An across-the-board policy would necessarily benefit teachers who never intended to leave or who teach in subjects experiencing teacher surpluses. At the same time, such an increase is unlikely to be high enough to reduce the attrition of teachers with stellar educational credentials and those in shortage subjects like math and science, whose employment opportunities outside of teaching would still be more lucrative, especially in a knowledge-based economy like Massachusetts. As a result—and similarly to the financial incentives for prospective teachers—if salary policies are to be used as a strategy to retain teachers in shortage subjects and difficult-to-staff districts, they may be more effective if they are explicitly targeted.93 Thus, policy makers should consider alternative pay options, such as higher base pay or bonuses for teachers of subjects with documented teacher shortages and for teachers in districts or schools proving particularly hard to staff (i.e., “combat pay”). In fact, the popularity of differentiated pay options among policy makers at both the state and the federal level has increased considerably over the past few years. In the Readiness Project, which summarizes Governor Patrick’s education agenda, this strategy figures prominently. To reduce the pay gap between teaching and the other sectors competing for talented workers, the Governor has voiced support for a system of differentiated pay that will initially focus on teachers opting for high-need low-income schools and districts, math and science teachers, and teachers with significant experience with special education students or English language learners.94 The federal government in turn has created a grant program—the Teacher Incentive Fund—to assist districts that have proposed innovative differentiated pay alternatives to attract and retain talented teachers in hard-to-staff schools.95 Nonetheless, few districts have used differentiated pay to recruit and retain teachers in shortage subjects or less desirable locations. In 2003-04, only 4.6 percent of districts nationwide used such incentives to reduce teacher turnover; in Massachusetts, this share was negligible, at 1.9 percent. The use of differentiated pay was slightly more popular for retaining teachers in hard-to-staff locations, used by 9 percent and 12 percent of districts in the state and the nation, respectively. Given the lack of experience with such pay options, little hard evidence is available as to their effectiveness in education. Research on other sectors has shown promise, though such results may not be readily applicable to teaching, where factors like collegiality are usually of greater importance. Nonetheless, a recent study by Kowal, Hassel, and Hassel (2008) evaluates financial incentives for difficult-to-staff positions in sectors like the military, medicine, and private industries. Their evidence shows that differentiated salaries and various retention bonuses have been a common and effective strategy to recruit and retain workers in hard-to-staff positions. They also note that, in order to be more successful in education, such strategies would need to be specifically tailored to the needs of both teachers and their districts.96 Finally, differentiated compensation approaches are less controversial among teachers than options like performance-based pay, for instance. According to a 2000 national survey of teachers, 44 percent considered it a good idea to pay
I work more than 65 hours a week,
but I cannot imagine being able to buy a house given my salary, or figure out how I would raise a family while working these hours.
- Massachusetts public school teacher *
“
“
* Gerson-Nieder, J., Tamerat, J., Gelinas, A. “Teachers from Somerville, Boston, and Longmeadow talk about being asked to do too much with too little.” (2008). CommonWealth Magazine, 13 (3): 61-66, p.62.
Page 39
higher salaries to math and science teachers.97 A later survey revealed that an even higher share—70 percent—favor pay incentives for teachers to work in less desirable locations.98 In addition, 88 percent of teachers affiliated with the American Federation of Teachers and surveyed in 2008 expressed support for additional pay for teachers in hard-to-staff schools. In contrast, only 35 percent favored higher compensation for teachers whose students’ test scores improved.99 Nonetheless, differential pay policies are likely to encounter political difficulties. The fixed salary schedules negotiated in collective bargaining agreements considerably reduce districts’ flexibility to experiment with alternative pay options. And teachers’ unions have been less than enthusiastic about such pay reforms. For example, Massachusetts Teachers Association—the larger of the two teacher unions in the state—has voiced support for higher pay in difficult-to-staff schools but opposes salary differentiation by subject on the argument that it harms teacher collegiality within the school.100 In addition, the American Federation of Teachers, with which the other state union AFT Massachusetts is affiliated, supports differentiated pay for teachers in hard-to-staff schools only as long as it is locally negotiated between district officials and local union leaders, and is voluntary, schoolwide, available to all teachers, and fostering a collaborative environment.101 As a result, any strategies to introduce differentiated compensation would need to include negotiations with union leaders and would likely be more feasible if undertaken at the district level rather than as a statewide initiative.
Working Conditions
Although differences in teacher salaries across districts do exists, they are often not substantial in practical terms. The variation in working conditions across the state, however, is more significant and may very well drive teacher attrition more than compensation does. Research on the importance of working conditions in determining teacher choices is ample. Numerous studies have documented that teachers tend to disproportionately leave schools and districts with poor working conditions, such as student discipline problems, lower readiness to learn, fewer resources, and less teacher empowerment—conditions that are frequently correlated with student and district characteristics. In addition, teachers have been consistently found to be more sensitive to characteristics like student poverty or race—frequent proxies for working conditions—than to salary differentials.102 Teachers in poorer, high-minority, and urban communities typically face less favorable working conditions, including more behavioral problems, larger class sizes, and less influence over the school’s decision-making process.103 Therefore policies that improve conditions in hard-to-staff schools in less desirable areas may hold significant promise for plugging the holes in the “leaking bucket” of teacher attrition. Working conditions across Massachusetts As a recent state-sponsored survey of more than 40,000 educators in Massachusetts shows, the Commonwealth is no exception to these trends. The Massachusetts Teaching, Learning and Leading Survey, or Mass TeLLS, assessed educators’ views on the status and impor-tance of teaching and learning conditions in schools across the state, and found that teaching conditions strongly impact teachers’ decisions about where to teach.104 Teachers with positive impressions of their working conditions were much more likely to continue to teach in their current schools. For example, only one in five teachers who switched schools agreed that their previous schools had a positive atmosphere of trust and respect, compared with two thirds of those who stayed. Overall, teachers in the state were strongly committed to their work and to their current schools. More than three quarters viewed their schools as good places to work, and 83 percent planned to continue teaching there.105 Nonetheless, they voiced a series of concerns about the quality of various teaching conditions in their schools. The majority of
POLICIES THAT IMPROVE CONDITIONS IN HARD-TO-STAFF SCHOOLS IN LESS DESIRABLE AREAS MAY HOLD SIGNIFICANT PROMISE FOR PLUGGING THE HOLES IN THE “LEAKING BUCKET” OF TEACHER ATTRITION.
Page 40
teachers felt excluded from decision-making processes in their schools, and complained about their school administration’s lack of responsiveness to their concerns. About half expressed unease about the quality, safety, and cleanliness of their schools; an even greater number noted the lack of sufficient instructional time to meet their students’ needs.106 Not surprisingly, teachers in areas with large enrollments of low-income students and in urban locations had less positive views about their working conditions. Only two thirds of teachers in high-poverty schools thought their schools were good places to work and learn, compared with 85 percent of teachers in schools with low concentrations of poor students. Teachers in high-poverty schools felt far less support from the communities where they taught. They were also less likely to agree that they were provided with adequate resources or instructional materials, and more likely to report student behavioral problems, inadequate physical safety, and inferior environmental conditions in their schools.107 A comparison of two very different districts—Springfield and Wellesley—illustrates these disparities more vividly. Springfield and Wellesley occupy virtually opposite sides of the spectrum in terms of student characteristics. Nearly 85 percent of students in urban Springfield are nonwhite, and more than three quarters are from low-income families. In contrast, minority students comprise less than a fifth of all students in suburban Wellesley, only 4 percent of whom are low-income. More than half of the teachers surveyed in Springfield viewed their large class sizes as unreasonably large, and two thirds noted that managing student behavior significantly adds to their workload. In comparison, similar views were held by only about a quarter of teachers in Wellesley. Half of all teachers in Springfield viewed their access to instructional materials and technologies as insufficient, but only one in ten Wellesley teachers did. Finally, nearly 40 percent of Springfield teachers believed that their school environments are neither well maintained nor physically safe, compared with only 17 and 24 percent, respectively, of teachers in Wellesley. Given such disparities in perceptions, it is hardly surprising that only a third of teachers in Springfield but three quarters of their colleagues in Wellesley strongly agreed that their school is a good place to work and learn.108 These findings are broadly confirmed by the 2000 Schools and Staffing Survey. According to that survey, more than 40 percent of Massachusetts teachers viewed student discipline as a hindrance to their teaching; a third felt the same way about student tardiness and class cutting. Nearly two thirds felt that paperwork and routine non-teaching duties interfered with their teaching work. Physical safety was also a significant concern, as an alarming 15 percent reported being either threatened with injury or physically attacked in the past 12 months.109 Policy options Given that teachers are at least as responsive to working conditions as they are to salary incentives, interventions to improve conditions may be an important strategy to reduce the attrition of teachers and improve the equity of their distribution across the state. Unfortunately, not all working conditions are easily changed or equally easy to manipulate. For example, while it may be challenging to improve factors like teacher workload or school cleanliness, increasing collegiality among teachers or enhancing their sense of autonomy and empowerment is an infinitely more difficult and time-consuming undertaking. Furthermore, research on which conditions are more easily influenced and which interventions are most effective is still limited.110
Urban teachers need smaller classes,
more planning time, and compensation that reflect the tasks that are being asked of them.
- Massachusetts public school teacher *
“ “ * Gerson-Nieder, J., Tamerat, J., Gelinas, A. “Teachers from Somerville, Boston, and Longmeadow talk about being asked to do too much with too little.” (2008). CommonWealth Magazine, 13 (3): 61-66, p.62.
Page 41
Nonetheless, teachers’ views in the Commonwealth reveal that working condition interventions hold significant promise. More than half of the teachers in the Mass TeLLS survey strongly agreed that their career choices are influenced by school conditions, such as staff collegiality, leadership support, and teacher empower-ment; less than 40 percent ascribed similar importance to salary. When asked about the effectiveness of different incentives to attract and retain teachers in difficult-to-staff schools, similar proportions—about half of all teachers—viewed both improved working conditions and financial incentives as “extremely effective.” In particular, between 46 and 48 percent considered more planning time and reduced teaching load as extremely important, while 51 percent felt the same way about loan forgiveness and other forms of tuition support.111 Teachers in low-income communities appear even more sensitive to teaching conditions than their peers in wealthier districts. For example, nearly two thirds of teachers in Lawrence and 56 percent in Fall River consider additional planning time as an extremely effective incentive, compared with only 42 and 43 percent, respectively, in the affluent communities of Weston and Wellesley.112 While it is unclear whether policies to provide more planning time, reduce class sizes, and secure better supplies are less costly than salary increases (across-the-board or targeted), they appear to be a more politically feasible alternative. The two teacher unions in the Commonwealth—the Massachusetts Teachers Association and AFT Massachusetts—have explicitly called on state and local policy makers to improve working conditions as a strategy to attract highly effective teachers to high-poverty schools.113 As local contexts and needs differ greatly, working condition policies are probably best created “from the ground up,” with schools and districts taking a leading role in devising and implementing interventions that best suit their needs. However, the state has an important role to play, primarily in guiding, informing, and supporting local initiatives. In particular, state policy makers can create guidelines and a bank of knowledge that stores and disseminates “best practices” that have repeatedly proven effective at enhancing teaching conditions and that schools and districts can adapt to their specific contexts. The state can also provide schools with financial incentives to create improvement plans—with resources preferably targeted toward low-income high-need schools, which would benefit the most from enhanced teaching conditions. Finally, given how valuable the Mass TeLLS survey has proven in understanding teachers’ perceptions, state policy makers should consider commissioning similar surveys periodically. They should also encourage schools and districts to review their findings carefully in order to assess their needs, measure the effectiveness of their current initiatives, and improve future policies.
Photograph: iStockphoto
Page 42
M assachusetts continues to grapple with a series of challenges related to its teacher workforce. Low-income, high-minority, and urban districts have more difficulties in staffing teaching positions with fully qualified teachers,
indicating a less than equitable distribution of effective teachers across the state. Turnover among teachers in those districts is higher, with adverse implications for student learning, teacher collegiality, and district finances. Subject areas like mathematics, sciences, special education, ESL, and foreign languages are experiencing teacher shortages, as candidates trained in these disciplines—particularly in math and science—often face better-paid job opportunities elsewhere. Over the next decade, several large urban districts with rising student enrollments and high teacher attrition are expected to face significant teacher hiring needs and more potential shortages. Finally, teachers in Massachusetts are less racially and ethnically diverse than their students, and the racial mismatch is greater in low-income, high-minority, and urban districts. To develop a teaching workforce that is more equitably distributed, more stable, more adequate in size, and more diverse, the Commonwealth needs to attract sufficient numbers of diverse and talented teachers into high-demand subjects and difficult-to-staff areas, and retain them over an extended period of time. Using insights from the data analysis and the policy review, this section distills three broad recommendations about steps the state can take to respond to the challenges presented by its teacher workforce:
• Target programs and initiatives to candidates committed to teaching in high-need areas and in difficult-to-staff subjects.
• Assist districts with policies that reduce turnover but are better handled at the local level, such as modifying compensation structures and improving working conditions.
• Promote diversity in the teacher workforce.
1. Target programs and initiatives to candidates committed to teaching in high-need areas and in difficult-to-staff subjects. The uneven distribution of qualified teachers across the state and across academic subjects is a larger problem than their overall supply. Future supply and demand imbalances are likely to be greater for large urban low-income districts than for the state as a whole. Teacher turnover in hard-to-staff subjects and less desirable areas is also high, contributing to current and potential regional shortages. As a result, the Commonwealth would be best served not by a general policy of increasing the overall number of college graduates going into teaching, but by targeted approaches that create incentives for prospective teachers to train in shortage subjects, teach in high-need areas, and do it as a long-term career. The state could improve the targeting of current and future programs in several ways:
• Restrict eligibility for financial incentives, such as tuition assistance, loan reimbursement, and signing bonuses, to candidates committed to teaching in high-need areas or difficult-to-staff subjects. Since the Commonwealth faces regional and subject imbalances but not a general teacher shortage, restricting program eligibility to candidates with the desired commitments is likely to be both more effective and more cost-effective at addressing the imbalances than implementing programs with universal eligibility. The lessons reviewed in this report reveal that programs with eligibility requirements restricted to candidates committed to teaching in shortage areas or fields tend to generate more teachers in such positions than programs with universal eligibility. For instance, the state’s MINT/Signing Bonus program aimed to reduce teacher shortages in high-demand subjects and in high-need schools but its requirements only targeted prospective teachers in the shortage subject areas; they did not include any location-based incentives. As a result, the MINT/Signing Bonus program was effective at producing more teachers in shortage subjects but fell short of its objective to induce them to teach in high-need schools. Similarly, whenever funds were insufficient, the Attracting Excellence to Teaching program gave priority to candidates willing to teach in high-need districts, but it did not explicitly restrict eligibility to those candidates. Its success in increasing the number of high-achieving teachers in high-need areas was limited.
R��������������
Page 43
It is encouraging that at least one of the two tuition assistance programs currently in existence in Massachusetts explicitly targets students interested in teaching academic subjects with demonstrated teacher shortages. The state’s Incentive Program for Aspiring Teachers provides tuition assistance to candidates enrolled in a state-approved teacher education program in a shortage field. Eligibility for the other program, Tomorrow’s Teachers Scholarship, does not require students to commit to teaching in districts or subjects with teacher shortages, though such students are given priority in granting scholarships.
• Encourage districts to develop new teacher training programs, or modify existing ones, to specifically target candidates willing to teach in schools or subjects with documented teacher shortages in those districts. Because such programs work best when designed to respond to local needs, their design and undertaking is best left to school districts. And although Governor Patrick has expressed support for establishing a statewide teacher residency program, designing and implementing such programs statewide may be severely hampered by limited funds, especially in the current fiscal environment.114 Explicitly targeted and well designed district-level programs like the Boston Teacher Residency and TEACH! Southcoast, meanwhile, have shown significant promise in alleviating local teacher shortages. For example, the Boston Teacher Residency requires candidates to teach in the Boston Public Schools for at least three years and all participants graduate with a dual license in their subject and in special education—a critical shortage subject area. As a result, nearly 60 percent of program graduates now teach in high-need academic fields and virtually all teach in high-need schools.115 TEACH! Southcoast, on the other hand, only trains mathematics and science teachers for employment in Fall River and New Bedford, two districts with documented shortages of teachers in general and in those two subjects in particular. Program graduates are already helping to alleviate acute shortages in some local schools.116
• Minimize implementation flaws like lack of publicity, clarity, or poor communication that can significantly impede a program’s success—and advise districts to do the same for their locally run programs. High levels of publicity, clarity, and consistency of program eligibility are key to enhancing program effectiveness.
An important lesson from prior experiences is that even the most strategically designed and well targeted programs can fail because of implementation errors. For example, the lack of public awareness and the confusion about program eligibility and deadlines impeded the success of both the MINT/Signing Bonus program and the AET loan reimbursement program. Similar issues appear to have reduced the effectiveness of the North Carolina bonus program as well. Potential channels through which implementation can be improved include more intensive advertising in teacher training programs and district hiring offices, ensuring the clarity and consistency of program features (eligibility, reward levels, and conditions), and providing reliable communication channels between program administrators and program applicants/participants. In short, programs cannot work (or work well) if prospective teachers never hear about them, do not know how to apply, and cannot get their questions answered.
2. Assist districts with policies that reduce turnover but are better handled at the local level, such as
modifying compensation structures and improving working conditions. Teachers’ decisions about whether, where, and for how long to teach are influenced at least in part by compensation and teaching conditions. Differentiated pay, in which teachers in hard-to-staff schools and/or shortage subjects receive higher compensation, appears to be a more promising and more cost-effective way to attract and retain teachers in the high-demand schools and subjects than across-the-board salary increases. Improving working conditions, particularly in low-income high-minority areas, holds additional promise for retaining teachers. As a result, if well designed and properly executed, differentiated pay and improved working conditions may reduce turnover among qualified teachers, equalize their distribution across the state, and alleviate local shortages.
Page 44
These two approaches, however, are best handled at the local level. Because working conditions vary widely across the state, improvement policies are likely to be more effective if designed and implemented “from the ground up”—with schools and districts taking a leading role in devising and implementing interventions that best suit their needs. And local implementation of differentiated pay is likely to be more politically feasible. Teachers’ unions in general support higher pay for teachers in challenging schools, but are adamant that such reforms should be negotiated locally between district leaders and union officials. Finally, limited financial resources currently prevent the state from initiating large-scale statewide programs to improve working conditions or reform pay structures. Nonetheless, the Commonwealth can play an important role in guiding, informing, and supporting local initiatives:
• Create standards and guidelines as well as a bank of knowledge that stores and disseminates “best practices” that have repeatedly proven effective in improving teaching conditions or implementing differentiated pay structures in Massachusetts and in other states. School and district leaders can use this information to better understand what makes condition improvement policies or compensation reforms effective. They can also learn about promising approaches implemented elsewhere that can be easily modified to suit their local needs without having to reinvent the wheel.
• Provide financial incentives for schools and districts to experiment with differentiated pay and to create teaching condition improvement plans. Such incentives ideally should be targeted to low-income high-need districts, which stand to benefit the most both from enhanced teaching conditions and from higher pay for teachers choosing to work there. Restricting funds to a select number of districts would also reduce the cost of such incentives, which is particularly important given the current paucity of state funds.
• Engage the teachers’ unions. Because political feasibility is a constraint at least as important as fiscal
viability, the state should seek the collaboration of union officials in all stages of the process—and encourage district leaders to do the same. Political feasibility is likely to be a bigger concern for compensation reforms than for condition improvement policies. Both of the Commonwealth’s teacher unions have explicitly called on state and local policy makers to improve working conditions as a strategy to attract effective teachers to high-poverty schools. Their support for differentiated pay, however, has been more qualified: Union leaders express support for higher pay in challenging schools, as long as the compensation structures are implemented schoolwide and, in the interest of collegiality, incentives are available to all teachers—and not to teachers of specific subjects only.
3. Promote diversity in the teacher workforce. Because recent evidence suggests that own-race teachers are associated with improved student achievement, making the teacher workforce more diverse can be an indirect but important way to reduce the racial achievement gap. In addition, research suggests that teachers, on average, prefer to teach near where they grew up or in schools with similar demographics to those they attended as students. Thus, increasing the popularity of teaching among minority students may also help to alleviate teacher shortages in high-minority districts.117 One reason for the lack of diversity among teachers is the fact that the state’s teacher licensing criteria require a bachelor’s degree but minorities are disproportionately less represented among college graduates than in the general population. In 2008, for example, African Americans and Hispanics accounted for six and nine percent of the state’s population, respectively, but each group made up only three percent of college graduates in Massachusetts.118 This suggest that improving diversity among teachers is partly rooted in solutions to the larger problem of unequal educational attainment. Though increasing higher education pursuit among minority students is a long-term and large-scale endeavor beyond the scope of teacher workforce policies, there are several steps that the state can take to promote diversity among its teachers:
• Increase awareness of teaching programs and incentives in minority communities. This can be done through means like advertising in minority community media, holding information sessions in neighborhoods, high schools, and community colleges with large minority populations, and recruiting teacher trainees from historically black colleges and universities outside New England. The latter option,
Page 45
however, needs to be approached carefully. Because many teachers prefer to teach in or near their home locations, it is uncertain whether this approach could entice many minority students from outside New England to relocate to Massachusetts, especially given the higher costs of living here.
• Provide incentives for high schools to develop initiatives that encourage minority students to consider teaching as a career. In a similar vein, the 62 to 12 Plan provided funding for multiple Tomorrow’s Teachers Clubs across the state—extracurricular activities designed to expose middle and high school students to teaching as a profession. The success of these clubs has not been fully assessed, but according to preliminary findings, 70 percent of club advisors predicted that two fifths or more of club members would eventually become teachers.119 Replicating such clubs in schools with high concentrations of minority students may generate similar interest in the teaching profession.
• Continue state efforts and promote district- and school-based efforts to increase the rates of high
school graduation, college enrollment, and college completion among minority students. If successful, such efforts will generate significant social and economic benefits—for the students themselves as well as for the state as a whole. Importantly, they may also indirectly advance the objective of enhancing teacher diversity: Increasing the number of nonwhite residents with at least a bachelor’s degree would expand the pool of potential teaching candidates, at which the state and districts can then target minority teacher recruiting efforts.
� �
Page 46
1 “Ready for 21st Century Success: The New Promise of Public Education.” (June 2008). The Patrick Administration Education Action Agenda, pp.2-4 2 Ibid, p.7 3 National Center for Education Statistics, 2003-04 Schools and Staffing Surveys (Public School Teacher File). 4 Loeb, S. and Reininger, M. (April 2004). Public Policy and Teacher Labor Markets: What We Know and Why It Matters. Education Policy Center. Michigan State University. p. 15. 5 Ehrenberg, R., Goldhaber, D., and Brewer, D. (1994). “Do Teacher’s Race, Gender, and Ethnicity Matter? Evidence from the National Education Longitudinal Study of 1988.” Industrial and Labor Relations Review, 48: 547-61. 6 Dee, T. S. (February 2004). “Teachers, Race and Student Achievement in a Randomized Experiment.” The Review of Economics and Statistics, 86 (1): 195-210. 7 Ferguson, R.F. (1998). “Teachers’ Perceptions and Expectations and the Black-White Test Score Gap,” in C. Jencks and M. Phillips (Eds.), The Black-White Test Score Gap. Washington, DC: Brookings Institution Press. 8 Marvel, J., Lyter, D.M., Peltola, P., Strizek, G.A., and Morton, B.A. (2006). Teacher attrition and mobility: Results from the 2004–05 Teacher Follow-up Survey. NCES 2007–307. Washington, DC: U.S. Department of Education, National Center for Education Statistics., p.7. 9 Ingersoll, R. M. (2003). Is There Really a Teacher Shortage? Center for the Study of Teaching and Policy. University of Washington., p.13. 10 Birkeland, S. E., & Curtis, R. (2006). Ensuring the Support and Development of New Teachers in Boston Public Schools. Boston: Boston Public Schools. 11 Henke, R. R., Chen, X., and Geis, S. (2000). Progress Through the Teacher Pipeline: 1992–93 College Graduates and Elementary/Secondary School Teaching as of 1997. Washington, DC: National Center for Education Statistics. 12 Guin, K. (2004). “Chronic teacher turnover in urban elementary schools.” Education Policy Analysis Archives, 12(4). 13 For example, see Ingersoll (2003), Murnane and Olsen (1990), and Murnane and Olsen (1989). 14 Ingersoll, R. M. (2003). Is There Really a Teacher Shortage? Center for the Study of Teaching and Policy. University of Washington. 15 Alliance for Excellent Education. (2005). Teacher Attrition: A Costly Loss to the Nation and to the States (Issue Brief). Washington, DC: Alliance for Excellent Education. 16 Birkeland and Curtis (2006). 17 Barnes, G., Crowe, E., and Schaefer, B. (2007). The Cost of Teacher Turnover in Five School Districts: A Pilot Study. Washington, DC: National Commission on Teaching and America’s Future. 18 For example, see Hanushek (1992), Goldhaber (2007), Rockoff (2004), Rivkin, Hanushek, and Kain (2005), Sanders and Horn (1998). For a summary of related literature, see Goldhaber (2008) in Ladd and Fiske (ed.) (2008), p.146. 19 For example, see Hanushek (1986, 1997), Goldhaber, Brewer, and Anderson (1999), Rivkin, Hanushek, & Kain (2005). For a summary of related literature, see Goldhaber (2008) in Ladd and Fiske (ed.) (2008), p.147. 20 For example, see Loeb, S. and Reininger, M. (2005), pp. 29-30. 21 Goldhaber (2007), Clotfelter, Ladd, and Vigdor (2007), Boyd, Grossman, et al. (2006). 22 Goldhaber and Brewer (1997a) and Goldhaber and Brewer (2000). 23 Darling-Hammond, L., Holtzman, DJ, Gatlin, S.J., and Heilig, J.V. (2005). “Does Teacher Preparation Matter? Evidence about Teacher Certification, Teach for America, and Teacher Effectiveness.” Education Policy Analysis Archives, 13 (42). 24 Hanushek, E.A. (1986) “The Economics of Schooling: Production and Efficiency in Public Schools.” Journal of Economic Literature, 24 (3): 1141-1177. 25 Goldhaber , D. and Brewer, D.J. (1997b). “Evaluating the Effect of Teacher Degree Level on Educational Performance.” In J.W. Flower (ed.), Developments in School Finance 1996 (pp. 197-210). Washington, DC: National Center for Education Statistics. 26 Hanushek, E.A. (1986). 27 For example, see Goldhaber (2007), Rockoff (2004), Rivkin, Hanushek, and Kain (2005), Clotfelter, Ladd, and Vigdor (2006), Kane, Rockoff, and Staiger (2008). For a summary of related literature, see Goldhaber (2008) in Ladd and Fiske (ed.) (2008), p.148. 28 For example, see Arnold, C.L., et al. (1993), and Lindsay, J., et al. (2009) 29 Arnold, C.L. et al. (1993), pp. 16-20. 30 For example, see Darling-Hammond, L. (2000), and various Educator Supply & Demand Research reports from the American Association for Employment in Education. 31 Darling-Hammond, L. (2000), p.11., and Ingersoll (2003), p.5.
E�� N����
Page 47
32 Aaronson, Daniel and Katherine Meckel. (2009) “The impact of baby boomer retirements on teacher labor markets.” Chicago Fed Letter No. 254. Federal Reserve Bank of Chicago, p.1 33 Author’s calculations using data from the Massachusetts Department of Elementary and Secondary Education and the 2008 American Community Survey. 34 Aaronson, D. and Meckel, K. (2009). “How Will Baby Boomer Retirements Affect Teacher Labor Markets?” Economic Perspectives, 4Q: 2-15. Chicago, IL: Federal Reserve Bank of Chicago. 35 Podgursky, M. and Ehlert, M. (2007). Teacher Pensions and Retirement Behavior: How Teacher Pension Rules Affect Behavior, Mobility, and Retirement. Washington, DC: National Center for Analysis of Longitudinal in Education Research (CALDER). The Urban Institute, p.8. 36 Massachusetts Department of Elementary and Secondary Education. 37 Author’s calculation s using data from the 2008 American Community Survey. 38 Loeb, S. and Miller . (2006). L.C. A Review of State Teacher Policies. Institute for Research on Education Policy & Practice (IREPP). School of Education, Stanford University, p. 46 39 Center for Education Policy. (2003). An Analysis and Evaluation of the 12 to 62 Plan for Recruiting and Retaining Teachers in Massachusetts. University of Massachusetts-Amherst. Commissioned by the Massachusetts Department of Education, p.5 40 Ibid, p.1. 41 Ibid, p. ii and iii. 42 Massachusetts Department of Elementary and Secondary Education: The Massachusetts Initiative for New Teachers (MINT) Home Page. URL: https://www.doemass.org/mint/ 43 Liu, E., Jonson, S.M. and Peske, H.G. (2004). “New Teachers and the Massachusetts Signing Bonus: The Limits of Inducements.” Educational Evaluation and Policy Analysis, 26 (3): 217-236. 44 Center for Education Policy. (2003), p.8 45 Ibid, and Abeille, A., Hurley, N., and Nesbitt, J. (2002). Teacher Supply and Career Development: Positive Pathways for Massachusetts. Stoneham, MA: Learning Innovations. 46 Center for Education Policy. (2003), p.8. 47 Abeille, A., Hurley, N., and Nesbitt, J. (2002), p.10. 48 Na’im, A. and Legg Greenberg, C. (January 2004). Cuts That Hurt: An Examination of Some of the Painful Cuts in the FY2004 State Budget. Boston, MA: Massachusetts Budget and Policy Center. 49 Center for Education Policy. (2003), p.55-79. 50 Center for Education Policy. (2003), p.viii. 51 Boyd, D., Lankford, H., and Wyckoff, J. (2008). “Increasing the Effectiveness of Teachers in Low-Performing Schools.” Chapter 30 in H. F. Ladd and E.B. Fiske (ed.), Handbook of Research in Education Finance and Policy (pp. 535-550). New York, NY: Routledge, p. 545. 52 North Carolina Teaching Fellows Program Home Page. URL: http://www.teachingfellows.org/prospectivestudents/ 53 Darling-Hammond, L. Recruiting and Retaining Teachers: What Matters Most and What Can Government Do?. A Testimony, p.3. 54 Clotfelter, C.T., Glennie, E., Ladd, H. and Vigdor , J.(2006). “Would Higher Salaries Keep Teachers in High-Poverty Schools.” NBER Working Paper No. 12285. 55 U.S. Department of Education. (2006). The Secretary’s Fifth Annual Report On Teacher Quality. Washington, DC: Office of Postsecondary Education, U.S. Department of Education., p.iii. 56 U.S. Department of Education. (2009). Title II Higher Education Act: Massachusetts State Report 2009. Washington, DC: U.S. Department of Education. URL: https://title2.ed.gov/Title2DR/StateHome.asp 57 Berry, B., Montgomery, D., Curtis, R., Hernandez, M., Wurtzel, J., and Snyder, J. (2008). Creating and Sustaining Urban Teacher Residencies. The Aspen Institute and the Center for Teaching Quality, p.9 58 Loeb, S. and Reininger, M. (2004), p.59. 59 Darling-Hammond, L. (2000), pp.22-24. 60 Education Commission of the States. (2003). Eight Questions on Teacher Preparation: What Does the Research Say? A Summary of the Findings. Denver, CO: Education Commission of the States., p.4. 61 Berry, B., et al. (2008), p.5. 62 Ibid. 63 Ibid, p.21 64 Ibid, p.17. 65 Ibid, p.20. 66 Ibid, p.21. 67 Ibid, p.7. 68 “Ready for 21st Century Success: The New Promise of Public Education.” (June 2008). The Patrick Administration Education Action Agenda, p.18.
Page 48
69 Compendium of Strategies to Reduce Teacher Turnover in the Northeast and Islands Region. (2008). Washington, DC: REL Northeast & Islands. National Center for Education Evaluation and Regional Assistance. U.S. Department of Education. URL: http://www2.edc.org/relnei/teacherdb/ViewProgram.aspx?ProgramID=5 70 TEACH! SouthCoast Home Page. URL: http://www.umassd.edu/cusp/teach/, and Transition to Teaching: Massachusetts 2006 Grant Abstract. Washington, DC: U. S. Department of Education. URL: http://www.ed.gov/programs/transitionteach/2006abstracts/ma.html 71 Vaznis, J. (September 25, 2008) “Fast-track Program Combats Math, Science Teacher Shortage.” The Boston Globe. 72 Compendium of Strategies to Reduce Teacher Turnover in the Northeast and Islands Region. (2008). 73 Vaznis, J. (2008). 74 Ingersoll (2003), p.17. 75 Ibid. 76 Boyd, D., Lankford, H., and Wyckoff, J. (2008), p. 537. 77 Boyd (2008), p.537. 78 Hanushek, E.A., Kain, J.F., and Rivkin, S.G. (2004). “Why Public Schools Lose Teachers.” Journal of Human Resources, 39(2): 326-354. 79 American Federation of Teachers. (2008). Survey and Analysis of Teacher Salary Trends 2007. Washington, DC: American Federation of Teachers. 80 American Federation of Teachers. (2007). Survey and Analysis of Teacher Salary Trends 2005. Washington, DC: American Federation of Teachers. 81 American Federation of Teachers. (2008), p.19. 82 American Federation of Teachers. (2008), p.12. 83 Authors calculations using data from the 2008 American Community Survey. 84 American Federation of Teachers. (2007), p.27. 85 Allegretto, S.A., Corcoran, S.P., and Mishel, L. (2004). How Does Teacher Pay Compare? Methodological Challenges and Answers. Washington, DC: Economic Policy Institute, p.2 86 American Federation of Teachers. (2007), p.32. 87 Loeb, S. and Reininger, M. (2004), p.53. 88 Hanushek, E.A., and Rivkin, S.G. (Spring 2007). “Pay, Working Conditions, and Teacher Quality.” The Future of Children, 17(1): 69-86, p.13 89 American Federation of Teachers. (2008), p.16. 90 Darling-Hammond, L. (2000). Solving the Dilemma of Teacher Supply, Demand, and Standards. New York, NY: National Commission on Teaching & America’s Future, pp.18-19. 91 Bradley, A. (1989) “In Connecticut, Moving Past Pencil and Paper: Teachers Evaluated On Class Behavior.” Education Week. http://www.edweek.org/ew/articles/1989/09/13/09040014.h09.html 92 Loeb, S. and Reininger, M. (2004), p.54. 93 Ibid. 94 “Ready for 21st Century Success: The New Promise of Public Education.” (June 2008). The Patrick Administration Education Action Agenda, p.18. 95 Goldhaber, D. (November 2008). Addressing the Teacher Qualification Gap. Center for Reinventing Public Education. University of Washington, p.17 96 Kowal, J., Hassel, B.C., and Hassel, E.A. (2008). Financial Incentives for Hard-To-Staff Positions. Washington, DC: Center for American Progress. 97 Public Agenda. (2000). A Sense of Calling: Who Teaches and Why? New York, NY: Public Agenda. 98 Public Agenda. (2003). Stand by Me: What Teachers Really Think about Unions, Merit Pay and Other Professional Matters. New York, NY: Public Agenda. 99 American Federation of Teachers. Differentiated Pay Plans: What AFT Members Think. Washington, DC: American Federation of Teachers, p.1. URL: http://archive.aft.org/topics/teacher-quality/compensation/pollresults.htm 100 Riley, D. (June 25, 2008). “Patrick wants to recruit, retain more math, science teachers.” The MetroWest Daily News. 101American Federation of Teachers. Differentiated Pay. URL: http://archive.aft.org/topics/teacher-quality/compensation/performance-pay.htm 102 For example, see Allen (2005), Guarino, Santibanez, and Daley (2006), Hanushek, Kain, and Rivkin (2004), Ingersoll (2001), Stinebrickner, Scafidi, and Sjoquist (2007), and Boyd, Lankford, Loeb, and Wyckoff (2003). 103 National Center for Education Statistics (NCES). (1997). America's Teachers: Profile of a Profession, 1993-94. Washington, DC: U.S. Department of Education. Table A 4.15 104 Hirsch, E., Freitas, C., Church, K. and Villar, A.(2009). Massachusetts Teaching, Learning and Leading Survey (Mass TeLLS).Santa Cruz, CA: New Teacher Center. University of California Santa Cruz.
Page 49
105 Ibid, p.v. 106 Ibid, p.vi. 107 Ibid, p.viii. 108 Massachusetts Teaching, Learning and Leading Survey (Mass TeLLS) Home Page. URL: http://masstells.org/ 109 Gruber, K.J., Wiley, S.D., Broughman, S.P., Strizek, G.A., and Burian-Fitzgerald, M. (2002). Schools and Staffing Survey, 1999-2000: Overview of the Data for Public, Private, Public Charter, and Bureau of Indian Affairs Elementary and Secondary Schools. NCES 2002-313. Washington, DC: National Center for Education Statistics. U.S. Department of Education., pp.39, 41. 110 Goldhaber, D. (November 2008), p.15-16. 111 Hirsch, E., Freitas, C., Church, K. and Villar, A.(2009), pp.27-28. 112 Massachusetts Teaching, Learning and Leading Survey (Mass TeLLS) Home Page. URL: http://masstells.org/ 113 Massachusetts Teachers Association and AFT Massachusetts. (2006). Fulfilling the Promise of Education Reform. Boston, MA: Massachusetts Teachers Association and AFT Massachusetts, pp.23-24. 114 “Ready for 21st Century Success: The New Promise of Public Education.” (June 2008). The Patrick Administration Education Action Agenda, p.18. 115 Berry, B., et al. (2008), p.21. 116 Vaznis, J. (2008). 117 Loeb, S. and Reininger, M. (2004), p.48. 118 Author’s calculations using the 2008 American Community Survey. 119 Center for Education Policy. (2003). p.vii.
Page 50
Aaronson, D. and Meckel, K. (2008). “The impact of baby boomer retirements on teacher labor markets.” Chicago Fed Letter No. 254. Chicago, IL: Federal Reserve Bank of Chicago. Aaronson, D. and Meckel, K. (2009). “How Will Baby Boomer Retirements Affect Teacher Labor Markets?” Economic Perspectives, 4Q: 2-15. Chicago, IL: Federal Reserve Bank of Chicago. Abeille, A., Hurley, N., and Nesbitt, J. (2002). Teacher Supply and Career Development: Positive Pathways for Massachusetts. Stone-ham, MA: Learning Innovations. Study commissioned by the Massachusetts Education Reform Review Commission (MERRC). Allen, M.B. (2005). Eight Questions on Teacher Recruitment and Retention: What does the Research Say. Denver, CO: Education Commission of the States Allegretto, S.A., Corcoran, S.P., and Mishel, L. (2004). How Does Teacher Pay Compare? Methodological Challenges and Answers. Washington, DC: Economic Policy Institute. Alliance for Excellent Education. (2005). Teacher Attrition: A Costly Loss to the Nation and to the States (Issue Brief). Wash-ington, DC: Alliance for Excellent Education. American Association for Employment in Education. AAEE Educator Supply and Demand Research Annual Reports. http://www.aaee.org/cwt/external/wcpages/resource/ American Federation of Teachers. Differentiated Pay Plans: What AFT Members Think. Washington, DC: American Federa-tion of Teachers. URL: http://archive.aft.org/topics/teacher-quality/compensation/pollresults.htm American Federation of Teachers. Differentiated Pay. Washington, DC: American Federation of Teachers. URL: http://archive.aft.org/topics/teacher-quality/compensation/performance-pay.htm American Federation of Teachers. (2007). Survey and Analysis of Teacher Salary Trends 2005. Washington, DC: American Federation of Teachers American Federation of Teachers. (2008). Survey and Analysis of Teacher Salary Trends 2007. Washington, DC: American Federation of Teachers Arnold, C.L., Choy, S.P., and Bobbit, S.A. (1993). Modeling teacher supply and demand, with commentary (NCES 93461). Wash-ington, DC: National Center for Education Statistics. Barnes, G., Crowe, E., and Schaefer, B. (2007). The Cost of Teacher Turnover in Five School Districts: A Pilot Study. Washing-ton, DC: National Commission on Teaching and America’s Future. Berry, B., Montgomery, D., Curtis, R., Hernandez, M., Wurtzel, J., and Snyder, J. (2008). Creating and Sustaining Urban Teacher Residencies: A New Way to Recruit, Prepare, and Retain Effective Teachers in High-Needs Districts. The Aspen Institute and the Center for Teaching Quality. Birkeland, S. E., & Curtis, R. (2006). Ensuring the Support and Development of New Teachers in Boston Public Schools. Boston, MA: Boston Public Schools. Boyd, D., Lankford, H., and Wyckoff, J. (2008). “Increasing the Effectiveness of Teachers in Low-Performing Schools.” Chapter 30 in H. F. Ladd and E.B. Fiske (ed.), Handbook of Research in Education Finance and Policy (pp. 535-550). New York, NY: Routledge.
B�'���! ����
Page 51
Boyd, D., Grossman, P., Lankford, H., Loeb, S., and Wyckoff, J. (2006). “How Changes in Entry Requirements Alter the Teacher Workforce and Affect Student Achievement.” Education Finance and Policy, 1 (2): 176-216. Boyd, D., Lankford, H., Loeb, S., and Wyckoff, J. (2003). “Analyzing the Determinants of the Matching of Public School Teachers to Jobs: Estimating Compensating Differentials in Imperfect Labor Markets.” National Bureau of Economic Research: NBER Working Paper No. 9878. Bradley, A. (1989) “In Connecticut, Moving Past Pencil and Paper: Teachers Evaluated On Class Behavior.” Education Week. http://www.edweek.org/ew/articles/1989/09/13/09040014.h09.html Center for Education Policy. (2003). An Analysis and Evaluation of the 12-to-62 Plan for Recruiting and Retaining Teachers in Mas-sachusetts. University of Massachusetts-Amherst. Commissioned by the Massachusetts Department of Education. Clotfelter, C.T., Ladd, H., and Vigdor, J. (2006). “Teacher-Student Matching and the Assessment of Teacher Effective-ness.” Journal of Human Resources, 41 (4): 778-820. Clotfelter, C.T., Ladd, H., and Vigdor, J. (March 2007). “How and Why Do Teacher Credentials Matter for Student Achievement.” Working Paper #2. Washington, DC: National Center for the Analysis of Longitudinal Data in Education Research. The Urban Institute. Clotfelter, C.T., Glennie, E., Ladd, H. and Vigdor , J.(2006). “Would Higher Salaries Keep Teachers in High-Poverty Schools.” National Bureau of Economic Research: NBER Working No. 12285. Compendium of Strategies to Reduce Teacher Turnover in the Northeast and Islands Region. (2008). Washington, DC: REL Northeast & Islands. National Center for Education Evaluation and Regional Assistance. U.S. Department of Education. http://www2.edc.org/relnei/teacherdb/ViewProgram.aspx?ProgramID=5 Darling-Hammond, L. Recruiting and Retaining Teachers: What Matters Most and What Can Government Do?. A Testimony. http://www.ewa.org/docs/linda_darlinghammond_testimony.doc Darling-Hammond, L. (2000). Solving the Dilemma of Teacher Supply, Demand, and Standards. New York, NY: National Com-mission on Teaching & America’s Future. Darling-Hammond, L., Holtzman, D.J., Gatlin, S.J., and Heilig, J.V. (2005). “Does Teacher Preparation Matter? Evidence about Teacher Certification, Teach for America, and Teacher Effectiveness.” Education Policy Analysis Archives, 13 (42). Dee, T. S. (February 2004). “Teachers, Race and Student Achievement in a Randomized Experiment.” The Review of Eco-nomics and Statistics, 86 (1): 195-210. Education Commission of the States. (2003). Eight Questions on Teacher Preparation: What Does the Research Say? A Summary of the Findings. Denver, CO: Education Commission of the States. Ehrenberg, R., Goldhaber, D., and Brewer, D. (1994). “Do Teacher’s Race, Gender, and Ethnicity Matter? Evidence from the National Education Longitudinal Study of 1988.” Industrial and Labor Relations Review, 48: 547-61. Ellis, P., Grogan, M., Jurist Levy, A., Tucker-Seeley, K. (2008). Developing the “Compendium of Strategies to Reduce Teacher Turnover in the Northeast and Islands Region. Washington, DC: REL Northeast & Islands. National Center for Education Evaluation and Regional Assistance. U.S. Department of Education. Ferguson, R.F. (1998). “Teachers’ Perceptions and Expectations and the Black-White Test Score Gap,” in C. Jencks and M. Phillips (Eds.), The Black-White Test Score Gap. Washington, DC: Brookings Institution Press. Gerson-Nieder, J., Tamerat, J., and Gelinas, A.“Teachers from Somerville, Boston, and Longmeadow talk about being asked to do too much with too little.” (2008). CommonWealth Magazine, 13 (3): 61-66.
Page 52
Goldhaber, D. (2007). “Everyone’s Doing It, but What Does Teacher Testing Tell Us about Teacher Effectiveness?” Journal of Human Resources, 42 (4): 765-794. Goldhaber, D. (2008). “Teachers Matter; Effective Teacher Quality Policies are Elusive.” Chapter 9 in H. F. Ladd and E.B. Fiske (ed.), Handbook of Research in Education Finance and Policy (pp. 146-165). New York, NY: Routledge. Goldhaber, D. (November 2008). Addressing the Teacher Qualification Gap: Exploring the Use and Efficacy of Incentives to Reward Teachers for Tough Assignments. Center for Reinventing Public Education. University of Washington. Goldhaber, D. and Brewer, D.J. (1997a). “Why Don’t Schools and Teachers Seem to Matter? Assessing the Impact of Unob-servables on Educational Productivity.” Journal of Human Resources, 32 (3): 505-23. Goldhaber, D. and Brewer, D.J. (1997b). “Evaluating the Effect of Teacher Degree Level on Educational Performance.” In J.W. Flower (ed.), Developments in School Finance 1996 (pp. 197-210). Washington, DC: National Center for Education Statistics. U.S. Department of Education. Goldhaber, D. and Brewer, D.J. (2000). “Does Teacher Certification Matter? High School Teacher Certification Status and Student Achievement.” Educational Evaluation and Policy Analysis, 22 (2): 129-145. Goldhaber, D., Brewer, D.J., and Anderson, D. (1999) “A Three-way Error Components Analysis of Educational Productivi-ty.” Education Economics, 7 (3): 199-208. Gruber, K.J., Wiley, S.D., Broughman, S.P., Strizek, G.A., and Burian-Fitzgerald, M. (2002). Schools and Staffing Survey, 1999-2000: Overview of the Data for Public, Private, Public Charter, and Bureau of Indian Affairs Elementary and Secondary Schools. NCES 2002-313. Washington, DC: National Center for Education Statistics. U.S. Department of Education. Guarino, C.M., Santibanez, L. and Daley, D.A. (2006). “Teacher Recruitment and Retention: A Review of the Recent Em-pirical Literature.” Review of Educational Research, 76 (2): 173-208. Guin, K. (2004). “Chronic teacher turnover in urban elementary schools.” Education Policy Analysis Archives, 12(4). Hanushek, E.A. (1986) “The Economics of Schooling: Production and Efficiency in Public Schools.” Journal of Economic Liter-ature, 24 (3): 1141-1177. Hanushek, E.A. (1992) “The Tradeoff between Child Quantity and Quality.” Journal of Political Economy, 100 (1): 84-117. Hanushek, E.A. (1997) “Assessing the Effects of School Resources on Student Performance: An Update.” Educational Evalua-tion and Policy Analysis, 19 (2): 141-164. Hanushek, E.A., Kain, J.F., and Rivkin, S.G. (2004). “Why Public Schools Lose Teachers.” Journal of Human Resources, 39(2): 326-354. Hanushek, E.A., and Rivkin, S.G. (Spring 2007). “Pay, Working Conditions, and Teacher Quality.” The Future of Children, 17(1): 69-86. Henke, R. R., Chen, X., and Geis, S. (2000). Progress Through the Teacher Pipeline: 1992–93 College Graduates and Elementary/Secondary School Teaching as of 1997. Washington, DC: National Center for Education Statistics. U.S. Department of Education. Hirsch, E., Freitas, C., Church, K. and Villar, A. (2009) Massachusetts Teaching, Learning and Leading Survey (Mass TeLLS).Santa Cruz, CA: New Teacher Center. University of California Santa Cruz. Hussar, W. (1998). Predicting the need for newly hired teachers in the United States to 2008-09. Washington, DC: U.S. Department of Education, National Center for Education Statistics. Ingersoll, R. M. (2001). “Teacher Turnover and Teacher Shortages: An Organizational Analysis.” Is American Educational Re-search Journal, 38 (3): 499-534.
Page 53
Ingersoll, R. M. (2003). Is There Really a Teacher Shortage? Center for the Study of Teaching and Policy. Seattle, WA: University of Washington. Ingersoll, R.M. and Kralik, J.M. (2004). “The Impact of Mentoring on Teacher Retention: What the Research Says.” Research Review. Denver, CO: Education Commission of the States. Kane, T.J., Rockoff, J.E., and Staiger, D.O. (2008). “What Does Certification Tell Us About Teacher Effectiveness? Evi-dence from New York City.” Economics of Education Review, 27 (6): 615-631. Kowal, J., Hassel, B.C., and Hassel, E.A. (2008). Financial Incentives for Hard-To-Staff Positions. Washington, DC: Center for American Progress. Lindsay, J., Wan, Y., and Gossin-Wilson, W. (September 2009). “Methodologies Used by Midwest Region States for Studying Teacher Supply and Demand. Washington, DC: National Center for Education Evaluation and Regional Assistance. U.S. Department of Education. Loeb, S. and Miller , L.C. (2006). A Review of State Teacher Policies. Institute for Research on Education Policy & Practice (IREPP). School of Education, Stanford University. Loeb, S. and Reininger, M. (April 2004). Public Policy and Teacher Labor Markets: What We Know and Why It Matters. Education Policy Center. Michigan State University. Liu, E., Jonson, S.M. and Peske, H.G. (2004). “New Teachers and the Massachusetts Signing Bonus: The Limits of Induce-ments.” Educational Evaluation and Policy Analysis, 26 (3): 217-236. Marvel, J., Lyter, D.M., Peltola, P., Strizek, G.A., and Morton, B.A. (2006). Teacher attrition and mobility: Results from the 2004–05 Teacher Follow-up Survey. NCES 2007–307. Washington, DC: U.S. Department of Education, National Center for Education Statistics. Massachusetts Teachers Association and AFT Massachusetts. (2006). Fulfilling the Promise of Education Reform. Boston, MA: Massachusetts Teachers Association and AFT Massachusetts. Massachusetts Teaching, Learning and Leading Survey (Mass TeLLS) Home Page. URL: http://masstells.org/ Murnane, R. J., and Olsen, R. J. (1989). “The Effect of Salaries and Opportunity Costs on Duration in Teaching: Evidence from Michigan.” The Review of Economics and Statistics, 71(2): 347-352. Murnane, R. J., and Olsen, R. J. (1990). “The Effects of Salaries and Opportunity Costs on Length of Stay in Teaching: Evi-dence from North Carolina.” The Journal of Human Resources, 25(1): 106-124. Na’im, A. and Legg Greenberg, C. (January 2004). Cuts That Hurt: An Examination of Some of the Painful Cuts in the FY2004 State Budget. Boston, MA: Massachusetts Budget and Policy Center. National Center for Education Statistics (NCES). (1997). America's Teachers: Profile of a Profession, 1993-94. Washington, DC: U.S. Department of Education. Table A 4.15 Podgursky, M. and Ehlert, M. (2007). Teacher Pensions and Retirement Behavior: How Teacher Pension Rules Affect Behavior, Mobility, and Retirement. Washington, DC: National Center for Analysis of Longitudinal in Education Research (CALDER). The Urban Institute. Public Agenda. (2000). A Sense of Calling: Who Teaches and Why? New York, NY: Public Agenda. Public Agenda. (2003). Stand by Me: What Teachers Really Think about Unions, Merit Pay and Other Professional Matters. New York, NY: Public Agenda.
Page 54
“Ready for 21st Century Success: The New Promise of Public Education.” (June 2008). The Patrick Administration Education Action Agenda, p.18. Rivkin, S., Hanushek, E.A., and Kain, J.F. (2005). “Teachers, Schools and Academic Achievement.” Econometrica, 73 (2): 417-458. Rockoff, J.E. (2004). “The Impact of Individual Teachers on Student Achievement: Evidence from Panel Data.” American Economic Review, 94 (2): 247-252. Riley, D. (June 25, 2008). “Patrick wants to recruit, retain more math, science teachers.” The MetroWest Daily News. Sanders, W. and Horn, S. (1998). “Research Findings from the Tennessee Value-added Assessment System Database: Impli-cations for Educational Evaluation and Research.” Journal of Personnel Evaluation in Education, 12 (3): 247-256. Stinebrickner, Todd R., Benjamin Scafidi, and David L. Sjoquist. (2007). “Race, Poverty, and Teacher Mobility.” Economics of Education Review, 26 (2): 145–159. U.S. Department of Education. (2006). The Secretary’s Fifth Annual Report On Teacher Quality. Washington, DC: Office of Postsecondary Education, U.S. Department of Education. U.S. Department of Education. (2009). Title II Higher Education Act: Massachusetts State Report 2009. Washington, DC: U.S. De-partment of Education.
Page 55
Appendix Table 1. Race/ethnicity of Massachusetts teachers, 2008-09
By racial/ethnic breakdown of districts' students
High-minority
districts
Low-m
inority
districts
Bottom quartile
(least diverse)
Second
quartile
Third
quartile
Top quartile
(most diverse)
Minority teachers (%
)4.3
1.4
1.3
1.6
2.5
6.2
Minority students (%)
25.2
6.1
4.4
7.9
13.1
37.4
Minority teachers per 100
0 minority
students
14.0
19.7
24.6
14.7
14.9
13.0
By socioeconomic status of districts' students
Low-income
districts
High-inco
me
districts
Bottom quartile
(least poor)
Second
quartile
Third
quartile
Top quartile
(poorest)
Minority teachers (%
)3.4
2.4
2.7
2.0
1.9
4.9
Minority students (%)
20.9
10.3
11.3
9.2
11.3
30.7
Minority teachers per 100
0 minority
students
14.5
19.2
22.3
15.7
16.7
12.4
By urbanicity of district
Urban
districts
Suburban/rura
l districts
Minority teachers (%
)10.8
2.5
Minority students (%)
45.4
14.3
Minority teachers per 100
0 minority
students
15.8
16.9
Source: M
assachusetts Departm
ent of Elementary and Secondary Education
Urban district are those coded as being located in
a large or medium-sized city in the National Center for Education Statistics' Common Core of Data.
Low-income districts are those in
which the share of students eligible for free or reduced priced lu
nch is above the state’s median share. H
igh-income districts are
those below the median share.
Notes: H
igh-m
inority districts are those in
which the share of minority students is higher than the states' m
edian share. Low-m
inority districts are those below the
A�������: D��� T�'���
Page 56
Appendix Table 2. Stayers, leavers, and movers, 2007-08 to 2008-09
Stayers Leavers Movers
Number of teachers 65,846 5,117 1,883
Percent of teachers 90.4 7.0 2.6
A. Stayers, leavers, and movers by age
Stayers Leavers Movers
Percent of teachers in each age group
25 or younger 89.8 5.8 4.4
26 to 32 younger 88.6 7.1 4.3
33 to 40 years 92.9 4.1 2.9
41 to 48 years 94.2 3.4 2.5
49 to 56 years 95.3 3.2 1.5
57 to 64 years 77.0 21.8 1.2
65 or older 53.9 44.8 1.3
All teachers 90.4 7.0 2.6
Percent of teachers in each mobility group
25 or younger 6.4 5.3 10.9
26 to 32 younger 18.3 19.0 30.9
33 to 40 years 19.8 11.4 21.8
41 to 48 years 18.0 8.3 16.4
49 to 56 years 26.1 11.2 14.0
57 to 64 years 10.9 39.8 5.7
65 or older 0.5 5.1 0.4
All teachers 100 100 100
B. Stayers, leavers, and movers by length of service in district
Stayers Leavers Movers
Percent of teachers in each age group
Less than 1 year 99.7 0.3 0.0
1 to 2 years 76.3 14.7 9.0
3 to 5 years 90.0 6.1 4.0
6 to 10 years 95.7 2.8 1.5
11 to 20 years 96.0 3.3 0.7
21 to 30 years 91.5 8.2 0.3
30 or more years 72.4 27.3 0.3
All teachers 90.4 7.0 2.6
Percent of teachers in each mobility group
Less than 1 year 13.0 0.5 0.0
1 to 2 years 13.7 33.9 56.6
3 to 5 years 15.1 13.1 23.4
6 to 10 years 23.7 8.8 13.2
11 to 20 years 20.4 9.0 5.2
21 to 30 years 9.2 10.6 1.0
30 or more years 5.0 24.1 0.7
All teachers 100.0 100 100
Source: Massachusetts Department of Elementary and Secondary Education
Page 57
Appendix Table 3. Turnover rates of Massachusetts teachers, 2007-08 to 2008-09
Percentage of teachers in each district who moved to another district or left teaching completely
By racial/ethnic breakdown of districts' students
Turnover rate
High-minority districts 12.1
Low-minority districts 10.3
Bottom quartile (least diverse) 11.0
Second quartile 9.7
Third quartile 11.2
Top quartile (most diverse) 13.1
By socioeconomic status of districts' students
Turnover rate
Low-income districts 11.8
High-income districts 10.7
Bottom quartile (least poor) 11.0
Second quartile 10.4
Third quartile 10.3
Top quartile (poorest) 13.3
By urbanicity of district
Turnover rate
Urban districts 13.5
Suburban and rural districts 11.1
Source: Massachusetts Department of Elementary and Secondary Education
Urban district are those coded as being located in a large or medium-sized city in the National Center for
Education Statistics' Common Core of Data.
Low-income districts are those in which the share of students eligible for free or reduced priced lunch is
above the state’s median share. High-income districts are those below the median share.
Notes: High-minority districts are those in which the share of minority students is higher than the states'
median share. Low-minority districts are those below the median.
Page 58
Appendix Table 4. District Analysis of Effectiveness Indicators, 2008-09
By racial/ethnic breakdown of districts' students
High-minority
districts
Low-minority
districts
Bottom quartile
(least diverse)
Second
quartile
Third
quartile
Top quartile
(most diverse)
Teachers licensed in teaching
assignment (%)98.1 98.2 98.4 98.0 98.7 97.6
Core classes taught by highly
qualified teachers (%)97.0 97.8 97.9 97.7 97.5 96.4
Waivers as share of all teachers (%) 2.6 1.6 1.7 1.5 2.3 2.9
Waivers per 1000 students 2.1 1.2 1.2 1.2 1.8 2.3
By socioeconomic status of districts' students
Low-income
districts
High-income
districts
Bottom quartile
(least poor)
Second
quartile
Third
quartile
Top quartile
(poorest)
Teachers licensed in teaching
assignment (%)97.8 98.6 98.6 98.6 98.1 97.4
Core classes taught by highly
qualified teachers (%)97.0 97.8 97.7 97.8 97.7 96.3
Waivers as share of all teachers (%) 2.3 1.9 1.7 2.0 1.9 2.8
Waivers per 1000 students 1.9 1.3 1.2 1.5 1.5 2.2
By urbanicity of district
Urban
districts
Suburban/rura
l districts
Teachers licensed in teaching
assignment (%)96.4 98.2
Core classes taught by highly
qualified teachers (%)95.7 97.5
Waivers as share of all teachers (%) 4.2 2.0
Waivers per 1000 students 3.3 1.5
Source: Massachusetts Department of Elementary and Secondary Education
Percent of teachers licensed in teaching assignment is the percentage of teachers with Preliminary, Initial, or Professional licensure in the area in which
they are teaching. Charter schools are not required to hire licensed teachers.
Urban district are those coded as being located in a large or medium-sized city in the National Center for Education Statistics' Common Core of Data.
Low-income districts are those in which the share of students eligible for free or reduced priced lunch is above the state’s median share. High-income
districts are those below the median share.
Notes: High-minority districts are those in which the share of minority students is higher than the states' median share. Low-minority districts are
those below the median.
Page 59
Appendix Table 5. Massachusetts teachers' length of service in current district, 2008-09
By racial/ethnic breakdown of districts' students
High-minority
districts
Low-minority
districts
Bottom quartile
(least diverse)
Second
quartile
Third
quartile
Top quartile
(most diverse)
Less than 1 year 9.5 9.0 9.6 8.4 8.8 10.2
1 to 2 years 16.4 15.2 16.1 14.2 15.8 17.1
3 to 5 years 15.7 16.8 17.5 16.0 15.9 15.6
6 to 10 years 22.2 22.4 21.9 22.8 22.2 22.3
11 to 20 years 20.8 20.7 19.5 21.8 21.5 20.2
21 to 30 years 10.1 10.9 10.3 11.4 10.9 9.2
More than 30 years 5.1 5.2 4.9 5.3 4.9 5.5
By socioeconomic status of districts' students
Low-income
districts
High-income
districts
Bottom quartile
(least poor)
Second
quartile
Third
quartile
Top quartile
(poorest)
Less than 1 year 9.2 9.3 10.1 8.5 8.7 9.6
1 to 2 years 15.4 16.2 16.5 15.9 14.6 16.2
3 to 5 years 15.2 17.4 18.6 16.1 15.4 14.9
6 to 10 years 20.9 23.6 24.4 22.8 21.9 20.0
11 to 20 years 21.7 19.8 18.8 20.9 22.8 20.6
21 to 30 years 11.7 9.3 7.8 10.8 11.0 12.4
More than 30 years 5.9 4.4 3.8 5.1 5.5 6.3
By urbanicity of district
Urban
districts
Suburban/rura
l districts
Less than 1 year 8.8 9.3
1 to 2 years 17.6 15.7
3 to 5 years 16.6 16.3
6 to 10 years 20.7 22.4
11 to 20 years 20.8 20.8
21 to 30 years 10.5 10.5
More than 30 years 5.1 5.2
Source: Massachusetts Department of Elementary and Secondary Education
Urban district are those coded as being located in a large or medium-sized city in the National Center for Education Statistics' Common
Core of Data.
Low-income districts are those in which the share of students eligible for free or reduced priced lunch is above the state’s median share. High-
income districts are those below the median.
Notes: High-minority districts are those in which the share of minority students is higher than the states' median share. Low-minority
districts are those below the median.
Page 60
Appendix Table 6. Massachusetts teachers with waivers, by academic subject
2003-04 2004-05 2005-06 2006-07 2007-08 2008-09
Total number of waivers 1,489 1,643 2,398 2,946 3,521 2,578
Special education (all levels) 404 530 936 1,367 1,429 1,305
Science (secondary) 133 129 179 184 233 167
Mathematics (secondary) 179 182 276 246 219 146
Bilingual/ESL (all levels) 95 93 176 171 187 113
Foreign language (secondary) 108 110 139 132 131 121
English (secondary) 88 99 129 115 117 66
Arts (all levels) 72 79 99 105 103
History (secondary) 67 71 63 66 88
Latin and Classical Humanities 10 12 8 6 10
Civics and government (secondary) - - - 9 7
Middle School Generalist 12 8 2 3 5
Economics (secondary) - - - - -
Geography (secondary) - - - - -
Career/technical education (secondary) 67 65 91 - -
Source: Massachusetts Department of Elementary and Secondary Education
Appendix Table 7. Teaching experience by subject area, 2003-04
Share of teachers by years of teaching experience
5 years or
fewer
6 to 10
years
11 to 20
years
21 to 30
years
More than
30 years
All teachers 27.2 19.6 21.6 17.7 14
Mathematics 39.6 15.9 18.9 17 8.5
Social Sciences 32.9 20.2 17.3 13.4 16.1
Natural Science 30.0 6.7 28 16 19.3
Early Childhood/General Elementary 26.2 19.2 22.3 15.4 17
Foreign Languages 25.7 13.9 22.4 24.5 13.5
Special Education 25.6 28.3 21.1 24.4 0.6
Vocational/Technical Education 25.6 22.5 28.7 18.1 5.2
Health/Physical Education 24.8 17.5 24.7 19.9 13.1
English/Language Arts 23.6 26.6 17.2 13.5 19
Arts/Music 18.7 14.4 18.3 25.5 23.1
ESL/Bilingual Education ‡ ‡ ‡ ‡ ‡
Source: National Center for Education Statistics, 2003-04 Schools and Staffing Survey (SASS), Public School Teacher File.
Note: ‡ indicates that sample was too small to report results.
Page 61
Appendix Table 8. Projected student enrollment, 2010-11 through 2019-20
Massachusetts and the 10 largest school districts
2008-09
(actual)
2010-11
2011-12
2012-13
2013-14
2014-15
2015-16
2016-17
2017-18
2018
-19
2019
-20
% change
2008
-201
9
Massachusetts
958,91
092
3,66
791
5,68
590
9,69
8904,73
0900,15
3896,87
5894,129
890,513
887,803
884,783
-7.7
Boston
55,923
50,819
49,999
49,396
48,896
48,554
48,470
48,035
47,570
47,190
46,760
-16.4
Springfield
25,360
23,827
23,972
24,091
24,256
24,535
24,777
25,025
25,340
25,661
25,972
2.4
Worcester
23,109
21,516
21,300
21,165
21,203
21,353
21,482
21,624
21,769
21,832
22,007
-4.8
Brockton
15,312
14,795
14,907
14,925
15,117
15,304
15,533
15,780
16,030
16,301
16,517
7.9
Lowell
13,400
12,559
12,551
12,641
12,767
12,888
13,075
13,271
13,451
13,588
13,806
3.0
Lynn
13,273
12,677
12,617
12,697
12,833
13,019
13,237
13,385
13,593
13,750
13,963
5.2
New
Bed
ford
12,609
12,443
12,771
12,931
13,115
13,365
13,657
13,869
14,056
14,306
14,511
15.1
Law
rence
12,221
11,127
10,959
10,808
10,815
10,844
10,900
10,978
11,029
11,078
11,122
-9.0
New
ton
11,700
11,292
11,248
11,266
11,279
11,265
11,254
11,183
11,119
11,004
10,918
-6.7
Fall R
iver
9,98
59,81
69,90
810,024
10,094
10,208
10,311
10,399
10,507
10,609
10,707
7.2
Source: State-level projections: M
assachusetts Departm
ent of Elementary and Secondary Education
District-level p
rojections: Author's calculations using data from the Massachusetts Departm
ent of Elementary and Secondary Education and the Bureau of Health
Inform
ation, Statistics, Research, and Evaluation, D
ivision of Research and Epidem
iology at the Massachusetts Departm
ent of Public H
ealth.
Page 62
Appendix Table 9. Projected total demand for teachers, 2010-11 through 2019-20
Massachusetts and the 10 largest school districts
"Mid-point" estimate using each district's average student-teacher ratio between 2003-04 and 2008-09
2010-11
2011
-12
2012-13
2013-14
2014-15
2015
-16
2016-17
2017-18
2018-19
2019
-20
Massachusetts
71,144
70,529
70,068
69,685
69,333
69,080
68,869
68,590
68,382
68,149
Boston
3,98
53,921
3,873
3,834
3,80
73,801
3,767
3,730
3,70
03,667
Spring
field
2,13
82,151
2,162
2,177
2,20
22,223
2,246
2,274
2,30
32,331
Worcester
1,56
71,551
1,541
1,544
1,55
51,564
1,575
1,585
1,59
01,603
Brockton
1,06
21,070
1,072
1,085
1,09
91,115
1,133
1,151
1,170
1,186
Lowell
983
982
989
999
1,00
81,023
1,038
1,052
1,06
31,080
Lynn
964
960
966
976
990
1,007
1,018
1,034
1,04
61,062
New
Bed
ford
893
916
928
941
959
980
995
1,009
1,02
71,041
Law
rence
818
806
795
795
797
801
807
811
815
818
New
ton
1,00
91,005
1,007
1,008
1,00
71,006
1,000
994
984
976
Fall R
iver
780
788
797
802
811
820
827
835
843
851
Table continues on next page.
Page 63
Appendix Table 9 (continued).
Lower-bound estimate using each district's highest student-teacher ratio between 2003-04 and 2008-09
2010-11
2011
-12
2012-13
2013-14
2014-15
2015
-16
2016-17
2017-18
2018-19
2019
-20
Massachusetts
70,185
69,578
69,123
68,746
68,398
68,149
67,940
67,666
67,460
67,230
Boston
3,39
53,340
3,299
3,266
3,24
33,238
3,209
3,178
3,15
23,123
Spring
field
1,85
81,870
1,879
1,892
1,91
31,932
1,952
1,976
2,00
12,026
Worcester
1,51
01,495
1,486
1,488
1,49
91,508
1,518
1,528
1,53
21,545
Brockton
1,04
11,049
1,051
1,064
1,07
71,093
1,111
1,128
1,147
1,163
Lowell
925
924
931
940
949
963
977
990
1,00
11,017
Lynn
920
916
922
932
945
961
972
987
998
1,014
New
Bed
ford
832
854
865
877
894
914
928
940
957
971
Law
rence
768
757
746
747
749
753
758
762
765
768
New
ton
932
929
930
931
930
929
923
918
908
901
Fall R
iver
731
738
747
752
760
768
775
783
790
798
Upper-bound estimate using each district's lowest student-teacher ratio between 2003-04 and 2008-09
2010-11
2011
-12
2012-13
2013-14
2014-15
2015
-16
2016-17
2017-18
2018-19
2019
-20
Massachusetts
72,312
71,687
71,218
70,829
70,471
70,214
69,999
69,716
69,504
69,268
Boston
4,43
94,367
4,315
4,271
4,24
14,234
4,196
4,155
4,12
24,084
Spring
field
2,30
22,316
2,327
2,343
2,37
02,393
2,417
2,448
2,47
92,509
Worcester
1,63
61,620
1,609
1,612
1,62
41,634
1,644
1,655
1,66
01,673
Brockton
1,10
31,111
1,112
1,127
1,14
11,158
1,176
1,195
1,215
1,231
Lowell
1,03
11,030
1,038
1,048
1,05
81,073
1,089
1,104
1,11
51,133
Lynn
999
995
1,001
1,012
1,02
61,044
1,055
1,072
1,08
41,101
New
Bed
ford
932
956
968
982
1,00
11,023
1,039
1,052
1,07
11,087
Law
rence
850
838
826
827
829
833
839
843
847
850
New
ton
1,05
11,047
1,048
1,049
1,04
81,047
1,041
1,035
1,02
41,016
Fall R
iver
820
828
838
844
853
862
869
878
887
895
Source: Author's calculations.
Page 64
Appendix Table 10. Number of new teacher hires needed to m
eet total demand, 2010-11 through 2019-20
Massachusetts and the 10 largest school districts
Lower-bound estimate M
assachusetts
Boston
Springfield
Worcester
Brockton
Lowell
Lynn
New
Bedford
Law
rence
New
ton
Fall R
iver
2010-11
4,627
377
342
9188
85104
114
48131
149
2011-12
4,574
395
369
99101
98107
126
49135
156
2012-13
4,632
405
369
110
92104
113
121
49135
163
2013-14
4,630
405
375
118
108
114
120
121
58134
161
2014-15
4,569
412
386
131
105
109
120
126
61131
164
2015-16
4,559
421
388
131
108
111
128
130
62130
165
2016-17
4,508
398
392
128
110
120
123
128
65122
165
2017-18
4,347
390
400
136
109
118
128
129
62119
168
2018-19
4,339
388
404
125
114
115
126
135
60114
169
2019-20
4,249
380
408
132
112
121
129
133
60114
171
Total 201
0-2019
45,033
3,970
3,833
1,20
21,046
1,096
1,197
1,262
574
1,265
1,632
Upper-bound estimate M
assachusetts
Boston
Springfield
Worcester
Brockton
Lowell
Lynn
New
Bedford
Law
rence
New
ton
Fall R
iver
2010-11
4,736
499
427
9893
95117
129
52150
172
2011-12
4,686
516
458
107
106
109
118
141
54154
178
2012-13
4,750
525
458
118
97118
123
135
54151
185
2013-14
4,748
526
464
126
115
126
130
136
63150
181
2014-15
4,685
534
478
141
111
122
131
141
67147
184
2015-16
4,674
547
480
141
114
124
140
145
68146
186
2016-17
4,622
515
485
138
116
133
135
143
70137
186
2017-18
4,457
506
494
145
115
132
139
144
68134
188
2018-19
4,448
503
500
134
121
128
137
150
66129
190
2019-20
4,356
493
504
143
118
134
140
149
66128
191
Total 201
0-2019
46,163
5,163
4,749
1,29
01,106
1,221
1,309
1,412
628
1,427
1,840
Source: Author's calculations.
Page 65
Appendix Table 11. Projected teacher retirements as share of annual attrition and annual hiring needs, 2010-11 through 2019-20
Massachusetts and the 10 largest school districts
Retirees as a percentage of each year's attrition
Massachusetts
Boston
Springfield
Worcester
Brockton
Lowell
Lynn
New
Bedford
Law
rence
New
ton
Fall R
iver
2010-11
38.7
22.6
16.7
35.9
16.1
25.9
20.0
16.8
26.1
18.6
17.7
2011-12
39.7
23.1
16.2
38.1
14.9
28.5
23.9
17.2
24.6
17.2
15.2
2012-13
39.8
23.1
16.0
42.0
14.1
26.5
21.3
22.3
24.2
15.3
15.7
2013-14
40.0
23.0
16.1
40.6
18.3
29.2
24.3
19.8
24.7
13.7
14.6
2014-15
39.9
22.9
15.5
43.2
16.2
27.7
21.6
20.4
22.3
13.1
13.7
2015-16
38.9
21.5
14.8
43.7
15.6
26.4
23.8
19.6
20.8
12.2
13.7
2016-17
38.2
21.2
14.3
40.9
16.5
28.2
23.3
19.3
21.0
11.4
13.5
2017-18
37.3
21.3
14.1
43.5
13.8
25.3
22.9
19.6
21.4
10.7
13.5
2018-19
36.5
20.3
14.0
40.5
14.6
23.6
23.7
18.3
19.3
10.2
13.0
2019-20
35.7
20.1
13.7
41.1
13.1
22.5
21.7
18.8
18.4
10.1
13.1
Percentage of each year's hiring needs due to retirements
Massachusetts
Boston
Springfield
Worcester
Brockton
Lowell
Lynn
New
Bedford
Law
rence
New
ton
Fall R
iver
2010-11
43.6
27.3
17.5
44.7
17.1
29.3
21.2
15.3
34.5
20.0
17.7
2011-12
45.0
26.4
15.7
43.9
13.7
28.7
24.9
14.2
30.5
17.6
14.5
2012-13
43.7
25.4
15.6
45.6
13.9
24.8
20.3
20.3
29.4
15.1
14.9
2013-14
43.2
24.9
15.6
39.7
16.0
26.8
22.3
17.8
24.5
13.5
14.2
2014-15
42.9
24.2
14.6
39.8
14.2
25.4
19.2
17.7
21.5
13.2
13.0
2015-16
41.1
21.8
14.0
40.7
13.2
23.1
20.9
16.7
19.4
12.3
13.1
2016-17
39.9
22.8
13.6
37.7
13.9
24.8
21.2
17.1
19.3
11.9
12.9
2017-18
39.7
23.0
13.3
40.2
11.6
22.5
20.2
17.7
20.2
11.2
12.9
2018-19
38.2
21.6
13.1
39.1
12.1
21.6
21.6
16.0
18.2
11.0
12.4
2019-20
37.7
21.6
12.9
37.2
11.3
19.5
19.1
16.8
17.5
10.7
12.6
Source: Author's calculations.
Page 66
Appendix Table 12. Average and beginning teacher salaries by state
Average salaries, 2006-07 Beginning salaries, 2004-05
State Average salary 10-year change State Beginning salary 10-year change
1 California $63,640 14.6% 1 Connecticut $39,259 7.3%
2 Connecticut $61,039 -7.7% 2 Alaska $38,657 -13.3%
3 New Jersey $59,730 -7.1% 3 New Jersey $38,408 -0.8%
4 New York $59,557 -4.0% 4 Illinois $37,500 14.0%
5 Rhode Island $58,420 5.0% 5 New York $37,321 2.9%
6 Illinois $58,275 6.5% 6 Maryland $37,125 8.0%
7 Massachusetts $58,178 4.7% 7 Delaware $35,854 17.8%
8 Maryland $56,927 6.8% 8 Hawaii $35,816 9.9%
9 Michigan $55,541 -8.9% 9 California $35,760 11.2%
10 Pennsylvania $54,977 -9.7% 10 Michigan $35,557 10.4%
11 Alaska $54,678 -13.9% 11 Massachusetts $35,421 12.4%
12 Delaware $54,537 1.9% 12 Colorado $35,086 31.1%
13 Ohio $53,536 6.8% 13 Pennsylvania $34,976 -7.2%
14 Hawaii $51,916 13.1% 14 Georgia $34,442 14.4%
15 Oregon $51,080 -3.8% 15 Rhode Island $33,815 9.5%
16 Wyoming $50,771 23.9% 16 Texas $33,775 18.1%
17 Georgia $49,836 8.1% 17 New Mexico $33,730 17.6%
18 Minnesota $49,719 1.0% 18 Oregon $33,699 7.9%
19 Nevada $49,426 -6.3% 19 Ohio $33,671 30.9%
20 Virginia $49,130 6.7% 20 Florida $33,427 11.0%
21 Washington $47,880 -2.0% 21 Virginia $33,200 3.6%
22 Indiana $47,832 -4.7% 22 Tennessee $32,369 19.6%
23 Vermont $47,645 2.3% 23 Minnesota $31,632 5.4%
24 Florida $47,219 8.0% 24 Wyoming $31,481 13.2%
25 New Hampshire $46,797 0.5% 25 Alabama $31,368 -0.4%
26 Wisconsin $46,707 -4.5% 26 Louisiana $31,298 26.0%27 North Carolina $46,137 14.6% 27 Washington $30,974 3.2%
28 Colorado $45,832 -2.2% 28 Indiana $30,844 1.8%29 Texas $45,392 8.4% 29 Kentucky $30,619 10.4%
30 Idaho $45,094 11.6% 30 Arizona $30,404 -0.5%
31 Arizona $44,700 4.2% 31 Nebraska $29,303 9.1%
32 Arkansas $44,493 11.0% 32 Missouri $29,281 7.6%
33 South Carolina $44,335 4.5% 33 Oklahoma $29,174 -4.8%
34 Tennessee $43,815 -0.9% 34 Arkansas $28,784 7.6%
35 Kentucky $43,787 0.3% 35 South Carolina $28,568 5.5%
36 Alabama $43,389 3.2% 36 New Hampshire $28,279 -3.2%
37 Kansas $43,318 1.3% 37 Mississippi $28,200 12.0%
38 Iowa $42,922 -0.1% 38 Nevada $27,957 -11.5%
39 Louisiana $42,816 14.2% 39 North Carolina $27,944 5.8%
40 New Mexico $42,780 11.4% 40 Kansas $27,840 1.9%
41 Oklahoma $42,379 8.7% 41 Idaho $27,500 10.6%
42 Maine $42,103 -3.2% 42 Iowa $27,284 1.0%
43 Nebraska $42,044 2.4% 43 West Virginia $26,704 -5.1%
44 Montana $41,146 6.3% 44 Maine $26,643 3.9%
45 West Virginia $40,534 -5.7% 45 Utah $26,521 5.6%
46 Missouri $40,384 -5.7% 46 Vermont $26,461 -13.0%
47 Mississippi $40,182 11.6% 47 South Dakota $26,111 5.4%
48 North Dakota $38,586 7.8% 48 Montana $25,318 0.8%
49 Utah $37,775 -8.2% 49 Wisconsin $25,222 -19.0%
50 South Dakota $35,378 1.2% 50 North Dakota $24,872 8.2%
U.S. Average $51,009 2.8% U.S. Average $31,753 3.3%
Sources: Beginning salaries: Survey and Analysis of Teacher Salary Trends 2005. (2007). Washington, D.C.: American Federation of Teachers
Average salaries: Survey and Analysis of Teacher Salary Trends 2007. (2008). Washington, D.C.: American Federation of Teachers
Page 67
Quartile analysis by socioeconomic status of districts' students
Average salary
Bottom quartile (least poor) 65,176
Second quartile 61,334
Third quartile 60,395
Top quartile (poorest) 61,491
Quartile analysis by race/ethnicity of districts' students
Average salary
Bottom quartile (least diverse) 60,811
Second quartile 60,813
Third quartile 61,807
Top quartile (most diverse) 65,058
Source: Massachusetts Department of Elementary and Secondary
Education
Appendix Table 13. Quartile analysis of average teacher
salaries, by various district characteristics, 2007-08
Appendix Table 14. Massachusetts school districts by students' characteristics, 2007-08
Low-income districts: High-income districts: High-minority districts: Low-minority districts: Urban districts:
Acushnet Abington Acton Abington Barnstable
Adams-Cheshire Acton Acton-Boxborough Acushnet Boston
Agawam Acton-Boxborough Amherst Adams-Cheshire Cambridge
Amesbury Andover Amherst-Pelham Agawam Fall River
Amherst Arlington Andover Amesbury Framingham
Amherst-Pelham Ashburnham-Westminster Arlington Ashburnham-Westminster New Bedford
Athol-Royalston Ashland Ashland Auburn Newton
Attleboro Bedford Athol-Royalston Belchertown Peabody
Auburn Belmont Attleboro Bellingham Pittsfield
Avon Berkley Avon Berkley Quincy
Ayer Berlin Ayer Berkshire Hills Springfield
Barnstable Berlin-Boylston Barnstable Berlin Waltham
Belchertown Billerica Bedford Berlin-Boylston Worcester
Bellingham Boxborough Belmont Billerica
Berkshire Hills Boxford Beverly Blackstone-Millville
Beverly Boylston Boston Boxford
Blackstone-Millville Braintree Bourne Boylston
Boston Brewster Boxborough Brewster
Bourne Bridgewater-Raynham Braintree Bridgewater-Raynham
Brockton Brimfield Brockton Brimfield
Brookfield Brookline Brookline Brookfield
Cambridge Burlington Burlington Carver
Carver Canton Cambridge Central Berkshire
Central Berkshire Carlisle Canton Chesterfield-Goshen
Chatham Chelmsford Carlisle Clarksburg
Chelsea Cohasset Chatham Cohasset
Chesterfield-Goshen Concord Chelmsford Danvers
Chicopee Concord-Carlisle Chelsea Dartmouth
Clarksburg Conway Chicopee Deerfield
Clinton Danvers Clinton Dighton-Rehoboth
Dartmouth Dighton-Rehoboth Concord Douglas
Dedham Douglas Concord-Carlisle Dover-Sherborn
Deerfield Dover Conway Dudley-Charlton Reg
Dennis-Yarmouth Dover-Sherborn Dedham Duxbury
Dracut Dudley-Charlton Reg Dennis-Yarmouth East Bridgewater
Eastham Duxbury Dover East Longmeadow
Easthampton East Bridgewater Dracut Eastham
Edgartown East Longmeadow Easthampton Fairhaven
Erving Easton Easton Florida
Everett Foxborough Edgartown Foxborough
Fairhaven Franklin Erving Franklin
Fall River Freetown Everett Freetown
Falmouth Freetown-Lakeville Fall River Freetown-Lakeville
Farmington River Reg Georgetown Falmouth Frontier
Fitchburg Gosnold Farmington River Reg Gateway
Florida Grafton Fitchburg Georgetown
Framingham Granville Framingham Gloucester
Frontier Groton-Dunstable Gardner Gosnold
Gardner Halifax Gill-Montague Granby
Gateway Hamilton-Wenham Grafton Granville
Gill-Montague Hampden-Wilbraham Greenfield Groton-Dunstable
Appendix Table 14 (continued)
Low-income districts: High-income districts: High-minority districts: Low-minority districts: Urban districts:
Gloucester Hampshire Hadley Halifax
Granby Hanover Hampden-Wilbraham Hamilton-Wenham
Greenfield Harvard Harwich Hampshire
Hadley Hatfield Haverhill Hancock
Hancock Hingham Hawlemont Hanover
Harwich Holliston Holbrook Harvard
Haverhill Hopedale Holyoke Hatfield
Hawlemont Hopkinton Hudson Hingham
Holbrook Ipswich Lawrence Holland
Holland King Philip Lee Holliston
Holyoke Kingston Leominster Hopedale
Hudson Lakeville Leverett Hopkinton
Hull Lenox Lexington Hull
Lanesborough Lexington Lincoln Ipswich
Lawrence Lincoln Lincoln-Sudbury King Philip
Lee Lincoln-Sudbury Longmeadow Kingston
Leicester Littleton Lowell Lakeville
Leominster Longmeadow Lynn Lanesborough
Leverett Lunenburg Malden Leicester
Lowell Lynnfield Marlborough Lenox
Ludlow Manchester Essex Regional Marthas Vineyard Littleton
Lynn Mansfield Mashpee Ludlow
Malden Marblehead Maynard Lunenburg
Marlborough Marion Medford Lynnfield
Mashpee Marshfield Melrose Manchester Essex Regional
Medford Marthas Vineyard Methuen Mansfield
Methuen Masconomet Milford Marblehead
Middleborough Mattapoisett Millbury Marion
Milford Maynard Milton Marshfield
Millbury Medfield Mount Greylock Masconomet
Milton Medway Nahant Mattapoisett
Mohawk Trail Melrose Nantucket Medfield
Monson Mendon-Upton Nashoba Medway
Mount Greylock Middleton Natick Mendon-Upton
Narragansett Millis Needham Middleborough
New Bedford Nahant New Bedford Middleton
New Salem-Wendell Nantucket Newton Millis
North Adams Nashoba North Adams Mohawk Trail
North Brookfield Natick North Andover Monson
Northampton Nauset Northampton Narragansett
Northbridge Needham Northborough Nauset
Norton Newburyport Norwood New Salem-Wendell
Norwood Newton Oak Bluffs Newburyport
Oak Bluffs Norfolk Orleans Norfolk
Orange North Andover Oxford North Attleborough
Orleans North Attleborough Peabody North Brookfield
Oxford North Middlesex Pelham North Middlesex
Palmer North Reading Petersham North Reading
Peabody Northboro-Southboro Pittsfield Northboro-Southboro
Petersham Northborough Plainville Northbridge
Pioneer Valley Norwell Provincetown Norton
Pittsfield Old Rochester Quincy Norwell
Plymouth Pelham Randolph Old Rochester
Appendix Table 14 (continued)
Low-income districts: High-income districts: High-minority districts: Low-minority districts: Urban districts:
Provincetown Pembroke Revere Orange
Quabbin Pentucket Rockland Palmer
Quaboag Regional Plainville Salem Pembroke
Quincy Plympton Saugus Pentucket
Ralph C Mahar Reading Savoy Pioneer Valley
Randolph Richmond Sharon Plymouth
Revere Rochester Sherborn Plympton
Rockland Sandwich Shirley Quabbin
Rockport Scituate Shrewsbury Quaboag Regional
Rowe Seekonk Shutesbury Ralph C Mahar
Salem Sharon Somerville Reading
Saugus Sherborn South Hadley Richmond
Savoy Shrewsbury Southborough Rochester
Shirley Shutesbury Southbridge Rockport
Somerville Silver Lake Springfield Rowe
South Hadley Somerset Stoneham Sandwich
Southbridge Southampton Stoughton Scituate
Southern Berkshire Southborough Sturbridge Seekonk
Southwick-Tolland Stoneham Sudbury Silver Lake
Spencer-E Brookfield Sturbridge Sunderland Somerset
Springfield Sudbury Taunton Southampton
Stoughton Sutton Tisbury Southern Berkshire
Sunderland Swampscott Truro Southwick-Tolland
Taunton Swansea Up-Island Regional Spencer-E Brookfield
Tisbury Tantasqua Waltham Sutton
Triton Tewksbury Ware Swampscott
Truro Topsfield Wareham Swansea
Uxbridge Tyngsborough Watertown Tantasqua
Wales Up-Island Regional Wayland Tewksbury
Waltham Wachusett Webster Topsfield
Ware Wakefield Wellesley Triton
Wareham Walpole West Springfield Tyngsborough
Watertown Wayland Westborough Uxbridge
Webster Wellesley Westfield Wachusett
Wellfleet West Boylston Westford Wakefield
West Springfield West Bridgewater Weston Wales
Westfield Westborough Westwood Walpole
Westhampton Westford Weymouth Wellfleet
Westport Weston Whately West Boylston
Weymouth Westwood Williamstown West Bridgewater
Williamsburg Whately Winchendon Westhampton
Williamstown Whitman-Hanson Winchester Westport
Winchendon Wilmington Winthrop Whitman-Hanson
Winthrop Winchester Woburn Williamsburg
Woburn Wrentham Worcester Wilmington
Worcester Wrentham
High-minority districts are those in which the share of minority students is higher than the states' median share. Low-minority districts
are those below the median.
Notes: Low-income districts are those in which the share of students eligible for free or reduced-price lunch is higher than the state’s
median share. High-income districts are those below the median share.