1. Geographical Ways of looking at segregation Rich Harris, University of Bristol, UK School of Geographical Sciences & Centre for Market and Public Organisation Rich Harris, University of Bristol, UK School of Geographical Sciences & Centre for Market and Public Organisation

2. Outline Focus on two opportunities Modelling micro data geographically –Mapping school catchment areas to identify polarization Building geographical models –Example of Geographically Weighted Regression Common framework for analysis –R Open source software for computing and statistics http://cran.r-project.org/

3. Outline Focus on two opportunities Modelling micro data geographically –Mapping school catchment areas to identify polarization Building geographical models –Example of Geographically Weighted Regression Common framework for analysis

4. School choice & Social segregation?

5. Ethnic polarization?

6. Geographical perspective Economic theory and government policy suggest schools operate within local markets to attract pupils and funding. However, there is a deficit of understanding about the scales and configurations of those admission spaces. Whilst competition for pupils and for school places is assumed to operate at some localised scale, the actual geographies of the markets, where they overlap and where they might be changing are generally unknown. Aim: To understand processes of polarization in the context of the local markets within which schools operate. Task: To use micro-data to model those markets

7. The data PLASC –Pupil Level Annual Census Returns Data on all pupils in primary (and secondary) schools in England 2005/6 data –Information on state educated primary school students (5- 11 years old) –'Self-identified' ethnic category collected from parents when students enrol –Also records postcode unit of pupils' homes –Which they school they attend –School type (selective? Faith school?) –Measure of deprivation (take a free school meal)?

8. Defining core catchments Imagine centring a polygon at (mid-x, mid-y) based on the residential postcodes of pupils attending a school –Let the polygon grow outwards

9. The direction of growth is determined as that which returns highest n1 / n2 –where n1 is number of pupils in area going to the school –n2 is all pupils in the area (go to any school) –Measuring prevalence

10. Continues until a certain proportion of all pupils who attend the school are enclosed… p = 0.30

11. Continues until a certain proportion of all pupils who attend the school are enclosed… p = 0.40

12. Continues until a certain proportion of all pupils who attend the school are enclosed… p = 0.50 Catchment is then defined as the convex hull for pupils of school within the search area.

13. London

14. Evidence of polarisation Are particular social (ethnic) groups travelling further to school than they need to? Are there (primary) schools with an intake not representative of the local community? Are there (primary) schools with shared admission spaces but where one has a very different intake to the other? Study region: London

15. Evidence of polarisation Are particular social (ethnic) groups travelling further to school than they need to? Are there (primary) schools with an intake not representative of the local community? Are there (primary) schools with shared admission spaces but where one has a very different intake to the other? Study region: London

16. Defining Near Define as being near to a pupil any primary school that has a core catchment that includes the pupils residential postcode Here the pupil has three near schools

17. Proportion attending any near school (target catchment p=0.50) LONDON

18. Evidence of polarisation Are particular social (ethnic) groups travelling further to school than they need to? Are there (primary) schools with an intake not representative of the local community? Are there (primary) schools with shared admission spaces but where one has a very different intake to the other? Study region: London

19. Pairwise Comparisons Look inside the catchments –Expected intake Vs Actual ethnic profile of each school Compare the profiles of locally competing schools –ones that overlap (strongly) in terms of their core catchment areas

20. Visual Summary (LONDON) Consider those schools with highest expected % Black Caribbean

21. Visual Summary (LONDON) Consider those schools with highest expected % Bangladeshi

22. Outline Focus on two opportunities Modelling micro data geographically –Mapping school catchment areas to identify polarization Building geographical models –Example of Geographically Weighted Regression Common framework for analysis

23. Example DataNumerator/DenominatorSource Y Higher education participation Successful entrants under 21 in UCAS data, for 2002–2005/ Census population 14–17 2007 Index of Multiple Deprivation X1X1 No qualifications Adults aged 25–54 in the area with no qualifications or with qualifications below NVQ Level 2, for 2001 /All adults aged 25– 54. 2007 Index of Multiple Deprivation X2X2 No post 16 qualifications Those aged 17 still receiving Child Benefit in 2006/ Those aged 15 receiving Child Benefit in 2004. 2007 Index of Multiple Deprivation X3X3 Average KS4 Points Total score of pupils taking KS4 in 2004 and 2005 in maintained schools from the NPD / All pupils in their final year of compulsory schooling in maintained schools for 2004 and 2005 from PLASC. 2007 Index of Multiple Deprivation X4X4 Four or more cars Four or more cars in household / total households2001 Census X5X5 AsianTotal Indian, Pakistani, Bangladeshi people / total people2001 Census

24. Example DataNumerator/DenominatorSource Y Higher education participation Successful entrants under 21 in UCAS data, for 2002–2005/ Census population 14–17 2007 Index of Multiple Deprivation X1X1 No qualifications Adults aged 25–54 in the area with no qualifications or with qualifications below NVQ Level 2, for 2001 /All adults aged 25– 54. 2007 Index of Multiple Deprivation X2X2 No post 16 qualifications Those aged 17 still receiving Child Benefit in 2006/ Those aged 15 receiving Child Benefit in 2004. 2007 Index of Multiple Deprivation X3X3 Average KS4 Points Total score of pupils taking KS4 in 2004 and 2005 in maintained schools from the NPD / All pupils in their final year of compulsory schooling in maintained schools for 2004 and 2005 from PLASC. 2007 Index of Multiple Deprivation X4X4 Four or more cars Four or more cars in household / total households2001 Census X5X5 AsianTotal Indian, Pakistani, Bangladeshi people / total people2001 Census

25. Global regression model (n = 31 378 ) βStandard errort value Significan t at α 0.01 ? (Intercept)3.6200.0213170.2Yes X 1 : No Qualifications-0.0270.0002-152.5Yes X 2 : No Post 16 Qualifications-0.0020.0001-15.1Yes X 3 : Average KS4 attainment0.0030.000252.6Yes X 4 : Four or more cars0.0180.000535.9Yes X 5 : Asian0.0120.000268.1Yes

26. But… Geographical variation in the Asian coefficient

27. What is it? –Extension of regression model –Allows model to vary over space How it works... Geographically Weighted Regression Regression Point Data Points

28. Summary of GWR model β (global value) β (u,v) Min β (u,v) 1 st decile β (u,v) 3 rd decile β (u,v) Median β (u,v) 7 th decile β (u,v) 9 th decile β (u,v) Max. β (u,v) IQR (Intercept)3.620 X 1 : No Qualifications -0.027-0.047-0.036-0.032-0.030-0.027-0.023-0.0140.006 X 2 : No Post 16 Qualifications -0.002-0.008-0.003-0.002-0.001 0.0000.0050.002 X 3 : Average KS4 attainment 0.0030.0000.0010.0020.003 0.0040.0060.001 X 4 : Four or more cars 0.018-0.0130.0110.0160.0210.0270.0400.1010.014 X 5 : Asian0.012-0.156-0.0060.0090.0120.0150.0200.2170.008

29. Geographical variation in the Asian coefficient

30. Outline Focus on two opportunities Modelling micro data geographically –Mapping school catchment areas to identify polarization Building geographical models –Example of Geographically Weighted Regression Common framework for analysis

31. Framework for analysis R –Open source software for statistical computing –Available at CRAN http://cran.r-project.org/ –WUN GIS Academy eSeminars about Spatial analysis in R http://www.wun.ac.uk/g gisa/

32. Thank you!