dannyparsons
commented
7 years ago One potential issue I already found with the IDB data:
pr_min_ca_signdate is numeric. It looks like it could be number of days since 1960/1/1 (Stata format) but doesn't seem to line up exactly when converting in R. I tried calculating this date from contract data and merging back in. Most of the dates match up but some very different so not quite sure about this. Although it's not a hugely important variable.
I have now added two new datasets to the procurement library, from Europe Aid and IDB. I have cleaned them a bit just by correcting missing values and converting columns to sensible types. These can now be used for testing and let me know if there is any other tidying that could be done to make these easier to use from the library. I also duplicated the awarder and winner country columns and corrected them so they are recognised by the mapping. These are not the main procurement datasets but they are quite interesting. IDB has quite a good number of contracts, Europe Aid has much less but has much more columns.