School Autonomy in the UK / Damon Clark

School Autonomy in the UK

Damon Clark

Cornell University

Presentation at Workshop on Evaluation of Education Policies, October 7 2011

Background: How to Improve Schools?

Three approaches:

(3) Structural reforms

(1) Resources

Spending, class size reductions

Autonomy, choice, competition

(2) Standards & Accountability

Incentives for schools, teachers and students

Outline of the talk

(A) UK school system

(B) Autonomy reforms in the UK

(C) Evaluation of a radical UK autonomy reform

(D) Outlook for school autonomy in the UK

(E) Summary, conclusions and questions


Brief History

State schooling gradually more important in 19th & early 20th century

By 1950, most schooling was state-provided (private school ~7%)

1950-1975: moved to comprehensive system of state schooling

In 1950, state system had two tiers: elite & non-elite schools


1975

National Government

Local Schools Authority

Neighbourhood Enrollment

Needs-Based Funding

S1 S2 S3

(A) UK school system1975 1988

National Government


Neighbourhood Enrollment

Needs-Based Funding

Open Enrollment

Money follows Pupils

League Tables

‘Virtual Voucher’ system=

S1 S2 S3 S1 S3S2


National Government

(B) UK Autonomy reforms

Reform #1: Local Management of Schools (1988)

S3S2


National Government

S1

More decisions given to schools (especially on budgets)


Reform #2: Grant Maintained Schools (1988)

S3S2


National Government

S1

‘Opt-Out’ = Quasi-independent

(must win vote among parents)


GM schools like private schools:

Owned the school grounds and buildings

Employed all teaching and non-teaching staff

GM schools like (UK) public schools:

Had to follow National Curriculum

Same current funding as non-GM schools (but from central gvt)

Admitted students (but no fees or tests)

GM schools like non-profits:

Established a board (governing body), principal like CEO

Complex formula for capital funding


Around one-third of high schools voted on GM status…

050

100

150

200

1988 1989 1990 1991 1992 1993 1994 1995 1996 19971 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3

Lose Vote Win Vote

Around one-quarter of high schools became GM

1988 1990 1992 1994 1996

0.3

0.2

0.1

Why not hold a GM vote? Why lose a GM vote?

Local resistance to GM reform

GM Voting: Typically

Principal For

Against

District Neutral

District Oppose

Teachers support Win ?

Teachers oppose ? Lose

(Q)

(A)


rb

r

r

bb b

b

b

bb

br

r

bb

bb

r

brr

b

bb

b rbr

b

r

b

b

b

r

bb

b

b

b

rb b

bb

bb

r br

r

bb

bb r

b

bb bb

b

bbb

b

b

b r

bb

b b bbb

b

b

b

b

b

b

b

b

b b

r

b

r

b

bb

b

r

b

b

r

b

bb

r

br r

bb

r

r rbb

b

r

bb

rbbb b

r

b

bb

b

b

r b

b

b

bb bb

rbr

b

r

b

b bbb bb

bb

bb

Labour

Conservative


GM votes in London

No – GM and never-GM schools may already be different

(2) Compare narrow vote winners and narrow vote losers

Yes – pre-GM vote, these schools should be same on avg

Possible methods

(1) Compare GM schools with never-GM schools?

(C) Evaluation of UK Autonomy Reform

-> narrow losers = good counterfactual for outcomes of narrow winners without GM status

= “regression discontinuity” design

Outcome = school pass rate

Sample = all secondary schools that voted

Age Grade

5 1

6 2

7 3

8 4

9 5

10 6

11 7

12 8

13 9

14 10

15 11

18

19

20

16 12

17 13‘O’ Level Exams

‘A’ Level Exams

Primary School

High School

College

University


11 7

12 8

13 9

14 10

15 11

‘O’ Level Exams (now GCSEs)

High School

Outcome = % students in school that pass >=5 exams

Sample


“School pass rate”


020

4060

8010

0

0 20 40 60 80 100

Vote counts by vote share (Figure 1)

GM status by vote share (Figure 2)

00.

250.

50.

751

0 20 40 60 80 100


School performance before the vote (Figure 3a)

School performance two years after the vote (Figure 3b)

3540

4550

0 20 40 60 80 100

-50

5

0 20 40 60 80 100


Estimates of the effects of becoming GM:

02

46

810

0 1 2 3 4 5 6 7 8

Percentage points

Year after vote

-> effects large: 10% of bass pass rate, 0.25 st devin in base pass rate s

-> effects are stable for 8 years after the vote


How did schools improve?

(1) No evidence of student cream-skimming, at least in short-run

(2) Evidence for reduced student absences

(3) Evidence for increased teacher hiring

(4) Evidence for increase in teacher turnover following vote

(5) Evidence for more flexible teacher pay (e.g., bonuses)

(D) Outlook for Autonomy in the UK

GM policy made by Conservative gvt, in power until 1997

Labor government that came to power in 1997 froze policy

No new GM schools but existing GM schools kept freedoms

Non-GM schools given more flexibility -> became more like GM

“Academy schools” created

New magnet-like schools in poor areas

Given a lot of resources, given GM-like freedoms


Coalition government elected in May 2010

Flagship policy = “free schools”

New schools (like start-up charter schools in the US)

Can be set up by parents, teachers, charities but not for-profits

Only a handful of free school applications/approvals


Coalition government elected in May 2010

More important policy = let schools rated “outstanding” become academies

200 new Academies created between January 2002 and election

68 new Academies created since election

194 schools converted to Academy status since election

-> another (quieter) revolution

(E) Summary & Conclusions

In 1975, UK school system like that in most countries

1988-1997: first autonomy revolution; appears to be have been successful

2010-?: second autonomy revolution; results as yet unknown

Interesting questions:

What do successful autonomour schools need? E.g., Great leaders?

Why can autonomy help? E.g., flexibility re staff decisions?

School Autonomy in the UK / Damon Clark

Education

new gm schools

gm academy schools

existing gm schools

tests gm schools

free schools new schools

school autonomy

private schools

schools ratedoutstanding