cluelessresearch.com

political methodology, brazilian politics, etc.

Linking IBGE to TSE databases

As promised, I have been working on linking TSE (electoral) data to IBGE (maps, census, etc) data by using municipalities names. They are very far from consistent, so I created some adhoc programs in Stata that seem to work quite well. The strategy was to strip all accents and “extra” words like “de”. “da”, state names, and merge using the processed names state by state.

The problematic municipalities remaining are the following:

In IBGE data, but not in TSE:

  • augusto severo RIO GRANDE DO NORTE
  • januario cico RIO GRANDE DO NORTE
  • lagoa mirim RIO GRANDE DO SUL
  • presidente juscelino RIO GRANDE DO NORTE
  • In TSE data, but not in IBGE:

  • aroeiras itaim PIAUI
  • boa esperanca norte MATO GROSSO
  • boa saude RIO GRANDE DO NORTE
  • campo grande RIO GRANDE DO NORTE
  • ipiranga norte MATO GROSSO
  • I suspect most of them are newly created municipalities, since IBGE data is from 2001 and TSE is from 2002. If you know about any of these cases, drop me an eline . I don’t have access to GIS tools here in Rochester, but I will probably be in NYC next week so I can work on it and post the data (GIS shape files and TSE data with IBGE geocodes).

    2 Comments so far

    1. Danny Hidalgo January 22nd, 2006 1:41 am

      Isn’t this already available on the IPEA website? What’s the additional data you’re trying to match? We’re working on something similar and haven’t had these problems.

    2. Eduardo January 22nd, 2006 3:14 am

      Danny -

      do you care to provide a specific link in IPEA? I would be interested to check it out.

      I would also like to point out that the problem comes simply from not having the tse data and the ibge maps in sync, that is both coming from the same configuration of municipalities. Thus, even if you do not have the problem for a single year, you _will_ have it if you compare across time.