Understanding Teacher Labor Markets: An Update Don BoydHamp Lankford University at Albany Susanna Loeb Jim Wyckoff Stanford UniversityUniversity at Albany.

Understanding Teacher Labor Markets:An Update

Don Boyd Hamp Lankford

University at Albany University at Albany

Susanna Loeb Jim Wyckoff

Stanford University University at Albany

Teacher Labor Markets and the Behavior of Teachers

How are the qualifications of teachers sorted?

How do teacher’s careers progress?

What is the geography of teacher labor markets?

What determines the initial match of teachers to jobs?

Who enters teaching?

What factors affect teacher quit and transfer decisions?

Several papers are available at www.teacherpolicyresearch.org .

Workforce Database

Universe TIME SOURCE:

Personnel data (BEDS)

All public school teachers, administrators and staff

1970-2002 NYSED

Certification and exam data

All those taking certification exams

1985-2002 NYSED

SUNY student data

All SUNY applicants 1990-2000 SUNY

School and district data

All public schools & districts 1970-2002 NYSED

Sorting of Teachers

• What is the distribution of teacher qualifications within and between districts?

• How do these qualifications vary with student attributes?

Teachers are sorted such that low-performing, non-white and poor students are most likely to be taught by the least qualified teachers.

What Explains These Between- and Within-District Differences in the Qualifications of Teachers?

What are the choices key in explaining the sorting of teachers?

• Initial job matches• Job transfers• Quits

What are the key determinants of each of these choices?

• Preferences of teachers • Factors considered by school officials • Institutional constraints

Teacher Qualifications by Proportion Failing Student Achievement Test, School Level, 2000

Percent of Students in

Level 1 4th Grade ELA

0 0 – 5% 5 - 20% >20%

Not Certified 0.03 0.04 0.09 0.22

Failed Gen. Exam 0.09 0.10 0.19 0.35

Most Comp. Coll. 0.11 0.11 0.09 0.08

Least Comp. Coll. 0.10 0.11 0.16 0.26

Factor 0.98 0.86 -0.30 -2.82

Select Teacher Qualifications by Student Attributes, 2000

Not certified in any subject taught

Failed LAST exam

Attended least comp. college

New York State

Non-White 0.17 0.21 0.21

White 0.04 0.07 0.10

Poor 0.21 0.28 0.25

Non-Poor 0.16 0.20 0.24

New York City

Non-White 0.21 0.26 0.25

White 0.15 0.16 0.25

Poor 0.22 0.30 0.27

Non-Poor 0.17 0.21 0.26

1995 Cohort of NYC Teachers Failing Certification Exam by Quartile of Student Performance

Initial match is 43% of gap after 6 years; transfers and quits make up 57% (Quartile 4 – Quartile 1)

15%

20%

25%

30%

35%

40%

1995 1996 1997 1998 1999 2000

% 1

995

Coh

ort

Fai

led

Exa

m

Quartile 1 Quartile 2 Quartile 3 Quartile 4

Percent of 1995 Cohort of NYC Teachers Failing Certification Exam by District Geography

Initial match is 75% of gap after 6 years , transfers and quits 25% (NYC Urban – NYC Suburban)

0%

5%

10%

15%

20%

25%

30%

1995 1996 1997 1998 1999 2000

% F

ail

ed

Ex

am

NYC - Urban NYC - SuburbanOther - Urban Other - Suburban

Proportion of NYC Teachers Who Transferred to Another District by Cohort

and Number of Years Teaching

0

0.02

0.04

0.06

0.08

0.1

0.12

0 2 4 6 8 10

number of years teaching

pro

po

rtio

n t

ran

sfe

rrin

g

(in

to

tal)

1990

1999

19931996

The Geography of Teacher Labor Markets

What defines the geography of the regions within which teachers look for jobs? How big are teacher labor markets?

How does geographical proximity affect the job choices of teachers?

Distance from Home to First Job, First Time Teachers, 1999-2002

85 percent of teachers take a first job within 40 miles of home

Region of First Job 0 to 15 miles

15 to 40 miles

40 to 100 miles

100 or more miles

Buffalo City 76.0 10.3 3.7 9.9Buffalo suburbs 71.0 18.3 4.3 6.3New York City 63.4 26.9 6.6 3.1New York City Sub 71.0 22.7 3.6 2.8Rochester City 54.2 11.4 18.7 15.8Rochester Suburbs 44.8 25.9 18.2 11.2All 60.8 23.9 8.6 6.7

Distance from Home to First Job

Distance from Home to First Job for Those Who Go Away to College, First Time Teachers, 1999-2002

Most who go away to college come home to take a first job.

Home to CollegeHome to Job 100 or more

miles0 to 15 miles 48.015 to 40 miles 24.040 to 100 miles 8.9100 or more miles 19.1All 23.9

Determinants of Employment Location of New Teachers

Multivariate analysis of location choice among 16 regions (separate urban and suburban areas of six MSAs and three rural areas).

• Distance from high school has largest effect on choice of region.

• Teachers are 4 times more likely to locate in a region that is five miles from home as one that is 40 miles from home.

• The effect of distance from college is smaller.

• Similarity of region is important, even holding distance constant.

• A teacher having a suburban home is 4.5 times as likely to teach in that suburban area as opposed to teaching in the adjacent urban district.

Urbanicity of Home and Urbanicity of First Job, First Time Teachers, 1999-2002

• Urban districts are net importers of teachers.

• The geographical distribution of teachers and their preferences for job proximity disadvantage urban schools.

Region

of Home Urban Suburban Rural All

Urban % Row 87.8 10.4 1.8 100.0%Col 59.9 6.4 2.9 27.1

Suburban % Row 25.7 66.9 7.4 100.0%Col 36.1 85.7 24.6 55.8

Rural % Row 9.2 20.0 70.9 100.0%Col 4.0 7.9 72.4 17.1

All PctN 39.7 43.6 16.7 100.0

Region of First Job

Initial Matches of Teachers and Jobs

• What preferences by teachers and employers determines the initial match of teachers to jobs?

• What additional monetary and non-monetary compensation would be needed to attract sufficient numbers of highly qualified teachers to traditionally hard to staff schools?

Estimating Compensating Differentials

(salary premiums needed to attract more qualified teachers to hard-to-staff schools)

The traditional approach has been to employ hedonic wage models.

Problems with the hedonic approach:

• Conceptual (theoretical) problems.

• Empirical results typically are inconsistent.

These difficulties have lead us to develop an alternative approach that accounts for important features of teacher labor markets.

Two-sided matching model

Initial results are very encouraging.

• Disentangle the factors affecting the separate choices of teachers and school officials responsible for hiring.

• Preliminary estimates of some compensating differentials

• Inferences regarding what attributes of teachers are important to school officials in their hiring decisions?

Some Major Findings and Implications

• There is extensive sorting of teachers such that the lowest-achieving, minority and poorest students typically are taught by the least qualified teachers.

• Initial job matches, transfers and quits each significantly contribute to this sorting.

• The unfolding retirement of large numbers of current teachers is likely to exacerbate current difficulties in attracting highly able teachers to hard-to-staff schools.

• Teachers' preferences for job proximity and the geographical distribution of teachers disadvantage urban schools and underscore the importance of teacher preparation.

• Many questions remain regarding how best to attract more qualified teachers to hard-to-staff schools.

Identifying the Next Generation of School Leaders     Hamp Lankford, Ray O’Connell and Jim Wyckoff

University at Albany, SUNY

We appreciate the advice and support of Dr. Kevin McGuire, Leadership Project, NYSED

Age Distribution of New York State Principals, 1990-2000

  Age Distribution of Principals 1990, 1995, 2000

0

100

200

300

400

500

35 40 45 50 55 60 65

Age

Pri

ncip

als

1990

1995

2000

Cumulative Number of Individuals Certified to be Principals by Age, NYS 2000

0

4000

8000

12000

20 25 30 35 40 45 50

Age

Num

be

r o

f In

div

iduals

# Certified Principals

Research Questions 

Why did they seek administrative certification?

What were their experiences in searching for administrative positions?

What are their career aspirations?

Are they interested in pursuing school leadership positions?

Do they feel prepared for an administrative position?

What conditions would they assume an administrative position?

Data 

Mail survey of all 7809 individuals with administrative certification, not older than 50 who are working NYS public schools and not currently serving as principals or superintendents.

Useable response rate of 39 percent.

Results reported here are for 1170 individuals who are currently non-administrators; separate results for those currently serving in administrative positions are presented in the paper.

What Was Your Primary Reason for SecuringAdministrative Certification?

(By Gender and Age) 

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

male female < 36 36 to 45 46 to 52

Pro

po

rtio

n Serve as admin

Teaching cert

Encouraged

Salary

Have You Ever Applied for anAdministrative Position?

(By Gender and Age, Proportion Yes) 

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

male female < 36 36 to 45 46 to 52

For How Many Years Following Certification Did You Apply for Administrative Positions?

(by Gender and Age) 

0

0.1

0.2

0.3

0.4

male female < 36 36 to 45 46 to 52

Less than 1

1 to 2

3 to 5

6 or more

Were You Offered an Administrative Position You Turned Down?

(by Gender and Age, Proportion Yes) 

0

0.1

0.2

0.3

0.4

0.5

male female < 36 36 to 45 46 to 52

What Factors Affected Your Decision to Turn Down an Administrative Position?

(by Gender) 

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Relocation Salary Tenure Personal Workingconditions

Pro

po

rtio

n

male

female

If Ever Applied for Administrative Positions, Have You Stopped?

(by Gender and Age, Proportion Yes) 

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

male female < 36 36 to 45 46 to 52

If Stopped Applying for Administrative Positions, Why?

(by Gender and Age) 

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

male female < 36 36 to 45 46 to 52

Pro

po

rtio

n Tenured

Compensation

Working conditions

Stress

If Stopped Applying for Administrative Positions, Why (con’t)?

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

male female < 36 36 to 45 46 to 52

Pro

po

rtio

n Unsuccessful

Not prepared

Competition

Few positions

What Are Your Feelings About Pursuing First Administrative Position

(by Gender and Age) 

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

male female < 36 36 to 45 46 to 52

Interested now

No longer interested

Never interested

Alw ays interested

Are There Conditions Under Which You Would be Willing to Assume First Administrative Position?

(by Gender and Age, Proportion Yes) 

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

male female < 36 36 to 45 46 to 52

Factors that Would Encourage You to Consider First Administrative Position

(by Gender and Age) 

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

male female < 36 36 to 45 46 to 52

Approached

Compensation

Prepared

Mentoring

Minimum Additional Compensation Required to Induce You to Take First Administrative Position

(by Gender and Age) 

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

male female < 36 36 to 45 46 to 52

Up to $5000

$5001-$10,000

More than $10,000

No amount

Implications

Although there a large number of individuals certified to be school leaders, only 27 percent of non-administrators continue to apply for administrative positions.

Compensation, working conditions and stress are factors frequently cited for reduced interest in administrative positions.

However, most individuals indicate that they would be interested in administrative positions under the right conditions.

Compensation, being encouraged by an administrator, and being mentored are factors that would induce individuals to consider administrative positions.