search

rpradosiqueira / sidrar

A R interface to IBGE's SIDRA API
87 stars 11 forks source link

are neighborhood codes incorrect? #8

Open Rodrigo-Fontoura opened 4 years ago

Rodrigo-Fontoura commented 4 years ago

https://github.com/ipeaGIT/geobr/issues/182

Hello everyone,

I'm opening here the same issue that i post at geobr github package. I wish that you could help me to solve this problem.

Thanks a lot!

rafapereirabr commented 4 years ago

Hi all. I've looked into the original shape files of the census tracts and neighborhoods published by IBGE and they haven't published any update on the data. This is to say that the data sets in geobr are exactly the same as the ones available on IBGE´s ftp: ftp://geoftp.ibge.gov.br/organizacao_do_territorio/malhas_territoriais/malhas_de_setores_censitarios__divisoes_intramunicipais/censo_2010/setores_censitarios_shp

So I'm not entirely sure what could be causing this incompatibility between neighborhood codes between geobr and sidrar, I'm looking forward to this collaboration so we can sort this out.

rafapereirabr commented 3 years ago

Hi all. Is there any update on this issue?

cassianord commented 3 years ago

A bit late:

The problem might be that neighborhood is not the same as census tracts. Sidra's table 1378 lists 14.216 neighborhoods, while Brazil has 449.880 census tracts.

The neighborhoods on Sidra have 10 digits, 7 for state+municipality, and the last 3 for the neighborhood.

IBGE's census tracts, on the other hand, have 12 to 14 digits, 7 for state+municipality, 2 to 4 for the district+subdistrict, and then the last 3 for the tract.