1 / 21

Causality issues in attributing the role of place – Setting the Scene

Causality issues in attributing the role of place – Setting the Scene. Simon Burgess July 2006. Attributing the role of place . Setting the scene: Greater availability of richer data New theoretical approaches Persistent policy questions

morag
Download Presentation

Causality issues in attributing the role of place – Setting the Scene

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Causality issues in attributing the role of place – Setting the Scene Simon Burgess July 2006

  2. Attributing the role of place • Setting the scene: • Greater availability of richer data • New theoretical approaches • Persistent policy questions • The importance of getting methods of analysis right, and understanding what can and cannot be learnt from a dataset. www.bris.ac.uk/CMPO

  3. Data • New data, or matching of different datasets: • administrative datasets • geographical data. • Four examples: • Children in a local area, and friendships • Area deprivation and life chances • Physical aspects of places • Patterns of criminality in areas www.bris.ac.uk/CMPO

  4. Example 1: • Avon Longitudinal Study of Parents and Children (ALSPAC) • Census birth cohort from one area (City of Bristol and surrounding areas, “CUBA”); spatial information included. • Look at peer and friendship links between children – mapping links • Effects of peers on behaviour?? www.bris.ac.uk/CMPO

  5. Example 2: • Match large and rich longitudinal dataset – British Household Panel Survey (BHPS) with (very) local area data • Use bespoke neighbourhoods and Census data to characterise these neighbourhoods • Does living in a poor neighbourhoods influence your future life chances? www.bris.ac.uk/CMPO

  6. Income level and growth www.bris.ac.uk/CMPO

  7. Example 3: • Physical aspects of places • Traditionally pretty small scale studies, because of the cost of looking at many places. • But new technologies … www.bris.ac.uk/CMPO

  8. www.bris.ac.uk/CMPO

  9. www.bris.ac.uk/CMPO

  10. www.bris.ac.uk/CMPO

  11. www.bris.ac.uk/CMPO

  12. www.bris.ac.uk/CMPO

  13. Example 4: • Patterns of criminality in local areas • Investigate aspects of place as measured by patterns of crime • Bringing together administrative data plus google map technology: • www.chicagocrime.org www.bris.ac.uk/CMPO

  14. www.bris.ac.uk/CMPO

  15. www.bris.ac.uk/CMPO

  16. www.bris.ac.uk/CMPO

  17. www.bris.ac.uk/CMPO

  18. www.bris.ac.uk/CMPO

  19. Theoretical Approaches • Economic analysis of social networks is popular at the moment. • Mapping networks • Who joins networks? • What is the impact of networks on individual outcomes? • Networks are certainly not necessarily spatial, but same fundamental set of issues. www.bris.ac.uk/CMPO

  20. Policy Questions • Economists thought of as obsessed with causality. • Causality is certainly not the only interesting empirical question to answer. • But in terms of policy advice, establishing causality is the key. • A correlation between poverty and place (for example) could arise in many different ways: the appropriate policy response depends on what is the causal factor acting on poverty. www.bris.ac.uk/CMPO

  21. All this points to the huge importance of getting the techniques right: • Getting the most out of the rich new data resources, • Giving the right answers for policy. •  “Empirical Analysis of Social Interactions” by Charles Manski www.bris.ac.uk/CMPO

More Related