Merging bulk data

[Pier Luigi Giardino]

Hello everyone,

I need to merge data about lobbying and PAC for all cycles from 1990 to this year, with the aim of assessing firms how much they contributed to political activities over the years.

Does anyone have idea about how to do it?

Thanks so much :)

[Kamron Eck]

Because there is no universal identifier that links the data perfectly, you will have to do some string matching. I used a python tool called SPlink, i’ve even left some code in another thread on this groups page. The official method for linking the data to firms would be considered probabilistic matching. I noticed that using a 80% probability to be the sweet spot for matching firm names from OpenSecrets and Capital IQ.