1. Click to add side text Click to add copy her Global Mapping Resources: Insights from Spatial Analysis & Exploration of Data Deborah Balk Baruch College, School of Public Affairs & CUNY Institute for Demographic Research 25 March 2008 2 nd Annual Census Workshop Series, Baruch College

2. Click to add side text Click to add copy her Population Distribution 15 years of progress Counts  Models More than just population distribution –Urbanization –Mortality –Other

3. Click to add side text Click to add copy her Evolution in global collection of population and poverty data 192019301940195019601970198019902000 Population count Population projection Population location Economic Output Poverty count Poverty location Urban Population Urban locations More attention to global scope More attention to comparability More attention to problem-oriented science More attention to spatial frameworks

4. Click to add side text Click to add copy her http://sedac.ciesin.columbia.edu/gpw Global Population Distribution

5. Click to add side text Click to add copy her Two views of Population Distribution (2000): Density estimates at the National level vs. 2.5 ’ grid Spatial data: Drilling down to finer resolution

6. Click to add side text Click to add copy her GPW v1 (1995)GPW v2 (2000)GPW v3 (2003) Benchmark 1994Benchmark 1990, 1995Benchmark 1990, 1995, 2000 19,000 input units127,000 input units400,000+ input units globally 102,000 units in Africa Population Counts (gridded) http://sedac.ciesin.columbia.edu/gpw

7. Click to add side text Click to add copy her http://sedac.ciesin.columbia.edu/gpw

8. Click to add side text Click to add copy her http://sedac.ciesin.columbia.edu/gpw

9. Click to add side text Click to add copy her http://sedac.ciesin.columbia.edu/gpw

10. Click to add side text Click to add copy her Score card on global data Global Extent Data Quality Data Availability Institutional Cooperation Population  Boundaries  Urban Areas Roads  Poverty 

11. Click to add side text Click to add copy her The US Census in International Perspective

12. Click to add side text Click to add copy her

13. Click to add side text Click to add copy her MEASURE DHS GPS Data Availability October 20, 2005 Slide courtesy of Livia Montana, Harvard University data available from http://www.measuredhs.com/

14. Click to add side text Click to add copy her What makes a GIS special? Data Visualization Data Exploration Data Integration Data Analysis –Service provision, public & constituency participation

15. Click to add side text Click to add copy her Visualization The organization of spatial information is different than that of tabular data. –That organization is often intrinsically visual Identification of neighbors –Construction of neighborhoods Identification of factors that share characteristics –Cites that are situated on a coast, along a river, etc

16. Click to add side text Click to add copy her Brazil 33 Cambodia 97 Cameroon 95 Australia 5 Afghanistan 168 China 30 IMR … Zimbabwe 78 In some scholarly traditions, the world is not only flat but also alphabetized.

17. Click to add side text Click to add copy her www.ciesin.columbia.edu/povmap Subnational underweight database also available (sparser coverage)

18. Click to add side text Click to add copy her Data exploration Example courtesy of Professor Juliana Maantay, Lehman College, CUNY

19. Click to add side text Click to add copy her “Layers” of GIS Information Municipalities

20. Click to add side text Click to add copy her “Layers” of GIS Information Census Tracts

21. Click to add side text Click to add copy her “Layers” of GIS Information Lakes and Rivers

22. Click to add side text Click to add copy her “Layers” of GIS Information Polluting Companies

23. Click to add side text Click to add copy her “Layers” of GIS Information Schools

24. Click to add side text Click to add copy her Exploration  Identification

25. Click to add side text Click to add copy her Human Settlements: Rendered as Points

26. Click to add side text Click to add copy her Human Settlements: Render with spatial form or “Polygons” Note the variety of shape Spatial location of large and small cities Form conveys much more than points

27. Click to add side text Click to add copy her Integration Overlay or combine units in a spatial framework to produce estimates or analysis –School buffers (in above example) –Cities and coastal flooding

28. Click to add side text Click to add copy her Administrative Boundaries Urban population and coastal flooding Calculations based on spatial overlays –All data are gridded + urban extent boundaries+ low elevation coastal buffer Cambodia Vietnam Ho Chi Minh City

29. Click to add side text Click to add copy her Which country has the greatest number of persons living at risk of coastal flooding? CountryPopulation in LECZ % of Population in LECZ China143,879,60011% India63,188,2086% Bangladesh62,524,04846% Vietnam43,050,59355% Indonesia41,609,75420% Japan30,477,10624% Egypt25,655,48138% USA22,859,3598% Thailand16,478,44826% Philippines13,329,19118% But, countries with the highest % of their populations in the zone include the populous deltaic countries and islands.

30. Click to add side text Click to add copy her Countries with greatest share of its population at risk of coastal flooding % urban % rural Broken down by urban and rural population

31. Click to add side text Click to add copy her Integration  Analysis Poverty (derived from econometric model for subnational units) + Elevation (derived from satellites, measured on a contiguous grid)

32. Click to add side text Click to add copy her Ecuador: Poverty Rate Urban areas are centers of population & more affluent High-poverty parroquias:  are numerous  more spatially distributed  of much lower population densities

33. Click to add side text Click to add copy her Ecuador: + Elevation Not all of Ecuador's poorest parroquias are found at high elevations, but there is a strong association: –Of the low-poverty parroquias, no non-urban ones are found at elevations above 2000 meters –In contrast, of the high-poverty parroquias almost half are found at elevations above 2000 meters, and nearly two-thirds are above 1000 meters. In reaching the poor, account for access associated with elevation. www.ciesin.columbia.edu/povmap

34. Click to add side text Click to add copy her Analysis Spatial –Characteristics & patterns associated with Distance Spatial relationships (e.g., neighbors) Spatial correspondence (i.e., to other factors) Non-spatial based on spatial integration –Analysis of omitted variables May result in maps or tables, or both May be “descriptive” or “analytic”

35. Click to add side text Click to add copy her Same pattern in Asia Largest cities tend to be near coasts Elevation overlay show that they also tend to be in low lying areas

36. Click to add side text Click to add copy her Final remark: Confidentiality Already a concern with information is collected from survey or census respondents Investigators and practitioners are ethically obligated to maintain respondent confidentiality –Geocoding may increase the difficulty in so doing

37. Click to add side text Click to add copy her Frequency of cluster size (ordered by cluster ID number below) Ranges from 2 to 36 persons per cluster

38. Click to add side text Click to add copy her

39. Click to add side text Click to add copy her US Census data is an excellent model There are increasingly diverse and high quality data being produced & distributed throughout the world In rich and poor countries alike Though coverage and consistency remain barriers to global coverage for many variables of interest Using international data does not alter responsibility to standards, such as maintaining confidentiality