Rebecca%20Allen%20and%20Anna%20Vignoles

1
Can school competition improve standards?The
case of faith schools in England

• Rebecca Allen and Anna Vignoles
• Institute of Education, University of London
• r.allen_at_ioe.ac.uk and a.vignoles_at_ioe.ac.uk
• Presentation to Bristol Choice Conference, 9th
  June 2009

2
Motivation for paper

• Current policy trends to encourage school
  autonomy and parental choice
• Over 100 years of school choice in England due to
  faith schools

• Research Questions
• Is there evidence that choice and competition for
  pupils takes place between faith and secular
  schools?
• Does competition from faith schools raise pupil
  achievement across all schools in an area?
• Does the presence of faith schools lead to pupil
  sorting in the local schooling market?

3
English Schools Data
Primary schooling Secondary schooling
Age 5-11 Age 11-16
National test data at ages 5, 7 and 11 Ability on entry to secondary school Total attainment in age 11 tests GCSE exams at age 16 (Z-score of best 8 subjects) English Maths Science

• Annual administrative records on pupils in state
  sector tell us
• Key individual characteristics e.g. sex,
  ethnicity, mother tongue, special needs status,
  free school meals status
• Pupil home postcode (nearest c.11 houses)
  allows matching of census data to provide
  estimates of socio-economic background (IMD and
  ACORN)

4
Faith schools in the state sector

• State-funded
• Religious bodies control governing bodies
• Control pupil admissions

5
Who chooses faith schools?
CHOICE

• Typical admissions policy at a faith school
  prioritises
• own denomination
• related denominations
• other religions
• non-religious families based on proximity
• Cannot clearly identify religious families who
  choose
• Religious self-identification by families tends
  not to be strong
• Church-going is uncommon in England
• Demonstration of religious adherence may not be
  onerous
• Families can adjust church-going behaviour to
  satisfy requirements

6
Models of school choice
CHOICE
Household non-schooling consumption depends on
take-home pay less housing interest payments
Household with childs ability and income
Childs final educational attainment depends on
peer group and initial ability
7
Model of religious school choice
CHOICE
Utility from schools religious characteristics
Home-school travel
Idiosyncratic preference for a location
Household religious characteristics
Disutility from demonstrating schools religious
admissions requirements
8
Choice between sectors is active
CHOICE

1. Faith schools recruit from wide geographical
   areas
2. Transitions between faith and secular schools
   take place at age 11
3. Not all faith school attendees are from religious
   families

9
How schools compete
COMPETITION

• Effort directed at improving test scores
• Encouraging academic ethos
• Monitoring teachers class test scores
• Altering teacher recruitment strategy
• Directing effort at certain high reward pupils
• Improving quality of peer intake
• Improving quality of applicants to school
• Altering published admissions criteria
• Altering the application of admissions criteria

10
Estimation problem
COMPETITION
Secular school quality
Proportion of faith school places in the area
Pupil achievement at GCSE
Historical religious population
Religious families
11
Pupil-level achievement models
COMPETITION
GCSE exam z-scores
Area controls, including religious composition
Pupil controls, including prior attainment
12
Additional specifications
COMPETITION
Within-area differences
Non-symmetrical responses
Attending a religious school
Other school controls
13
Instrumenting Catholic school supply
COMPETITION
14
Main results faith schools
COMPETITION
15
Main results Catholic schools
COMPETITION
16
Sorting in local competition spaces
SORTING
17
Place in local schooling hierarchy
SORTING
Faith schools
Secular schools
Local ranking of school by proportion of top
ability pupils
18
Levels of school stratification
SORTING
19
Conclusions

• Faith schools are associated with more stratified
  local schooling markets
• Cream-skimming
• Parental choice strategies
• No evidence that faith schools improve (or
  damage) area-wide academic achievement
• Is choice and competition genuine?
• Schools unable to respond to threat
• Competition muted by sorting