Closed MansMeg closed 2 years ago
I'm currently joining together the two lists (i.e., the one from the Git repository and the member meta-data one). One thing that bothers me is that party names change over the years. For instance, from 2006--2010 it's "Miljöpartiet de Gröna" and thereafter it's "Miljöpartiet de gröna" (lowercase g), 2014 it becomes Miljöpartiet. For joining the tables together, this is tedious. I can imagine that it also bears potential for errors when filtering (since you would have to do something like
filter(party %in% c("Miljöpartiet de Gröna", "Miljöpartiet de gröna", "Miljöpartiet"))
). Would it make sense to add another column with just the party abbreviation for
filter(party == "MP")
(just an example)? If so, I can fix that right away for the "current" parties from which I know the abbreviation. I have to do it anyways to join the tables properly.
I think it sounds like a good idea. What do @ninpnin say?
We have already done the same thing for other parties. I think the work should be finalized
Though we use harmonized versions of the full names (Miljöpartiet de gröna, Miljöpartiet -> Miljöpartiet) now?
? Sorry, I dont understand.
I fixed it now with the abbreviations @Stubbendorff suggested. See discussion in #47 .
Done.
Cool! Is this done? Where is that data?
At least we merged this at some point?
? But where is the twitter data then?
There is a column for the twitter handles and apparently 1 value ?
In what file? The MOP file?
Ah yes. This remind me that we should formulate the mop data into at least the third normal form for databases. Ill file this as an issue.
http://makthavare.se/2018/09/19/kartlaggning-har-ar-riksdagsledamoterna-pa-twitter-2/