Descrepancies in District and Assembly Constituencies boundaries

[NiranjanaPrasad]

Hi,

I am hoping to superimpose the district shape file on the Assembly Constituencies (ACs) shapefiles. While I understand that ACs are almost never a hundred percent in one district or for that matter have the same boundaries, there seem to be large shift across most AC boundaries. This is especially the case when I merge it with other datasets that use GPS coordinates. I have attached a screenshot (Red is AC and black is district boundaries).

I know you had mentioned a shift in data, could you elaborate on that?

Also, could you let me know which AC names are entered wrong?

Thanks a lot!

[cvipul]

This is actually a reality. Assembly constituencies don't always follow district boundaries as they are different classifications.

Districts -> Mandals/Talukas/Tehsil -> Panchayat - are administrative divisions
Lok Sabha constituencies -> Assembly constituencies - are electoral divisions

Their boundaries will match within their system, not necessarily with each other.

[DavidSorge]

Yes, it's an issue that folks in the social sciences have been trying to figure out how to deal with for years; there's been a long history of different solutions discussed in Economic and Political Weekly. So far the best I've seen is the solution Subramanian et al used in their paper. Here's one from their team detailing their method:
https://cga-download.hmdc.harvard.edu/publish_web/CGA_PDF_Maps/DI_LIV_19_110519_Jeffrey_C_Blossom.pdf