banner image

Gender bias in the EU election

The UK is having elections for the European Parliament on 23 May 2019.

The data

Democracy Club are, as ever, collecting information about candidates.

They use it in, but anyone can download the CSV file and take a look for themselves.

I wrote some horrible code to generate Although in this case, a spreadsheet (ie: LibreOffice and Pilot Tables) would have been about as effective as the horrible code.

Gender split by party

I'm going to look at only the candidate in position 1 of each party's list of candidates. (The electorate votes for the party, not the candidate; the party's seats are given to candidates at the top of the list.)

I think this is accurate because candidates low down the list are very unlikely to be elected. Besides, the dataset does not have gender set for candidates (yet).

Candidates in position 1, by party

(Full list at

The gender balance of candidates in position 1 of the list is:

         M  F
UKIP    12  0
Con      9  3
Brexit   8  3
TIG      5  6
Lab      5  6
LD       5  6
Green    3  8

I would assume the best candidate is equally likely to be a woman or a man. If that's true, how likely are these distributions to be down to chance?

Bad behaviour among candidates

Some candidates have been deleting their social media profiles:

I understand the desire not to have foolish statements splashed in the press. But this deletion of the historical record makes me angry: how can I trust what you say, when you are destroying all your past statements? How will you represent me, now that you've closed communication?