Talk:2016 Oxford City Council election

Percentage changes

Percentage changes here are calculated with respect to 2012, which seems methodologically unsound as well as not in keeping with past practice (the 2014 results page has percentages with respect to 2012 as well). Will edit over the next few days to put them in respect to 2014, as is standard practice, unless there are objections. — Preceding unsigned comment added by 192.76.8.9 (talk) 21:20, 6 May 2016 (UTC)

It is in line with 2014 election page practice. Although it says at the top that the plus/minuses are from 2012, that's actually only true of the summary table section. If you go and check some of the numbers in the results by ward section, you will see that the plus/minuses come from a comparison with the 2010 election, not 2012. The 2012 results by ward section also follow the same practice, comparing with 2008 not 2010.

It seemed like a grey area to me as to what would be best, so I went with previous practice. 2012 was the last time that these particular seats were elected, as 24 of the 48 seats are elected each time, and lots of the councillors have been re-elected, so comparing with 2012 compares them with the last time they ran. On the other hand, it doesn't show change in voting by party since the most recent 2014 election.

Anyway, that was my reasoning. Kastrel (talk) 15:00, 8 May 2016 (UTC)

Sorry, I thought I'd replied when I made the initial reversion. If the precedent on the Oxford City pages is different from how it was explained, then I'd argue to set aside the precedent. While technically you are right that these seats were last contested four years ago, the wards were not, and the voters voted last in the wards two years ago. For example, if one were to plot a line graph of party vote share in a particular ward over time, they wouldn't use two lines per party, they'd use one, with points at each two years.

I think as it stands, the use of four years ago as a reference point is somewhat misleading (a party might on this metric have a negative change of the vote based on last time, which makes it look like they've declined, when actually, compared to two years ago, their share of the vote has increased) and doesn't fit with the general, ordinary way that people talk about election results - which is in reference to the last election in the ward, not the last election for the seat. Certainly I had leaflets through my door where bar charts were used by more than one party which referred to the last election (2014), and not the last time this particular iteration of the seat was up.

So, as I say, I think this is misleading, and think there's a much stronger case to show vote share change with respect to 2014 than to 2012. 192.76.8.12 (talk) 20:47, 8 May 2016 (UTC)

Sounds good. In that case I think I might suggest changing the previous elections' changes in line with that. Sorry for the revert boomeranging. Kastrel (talk) 09:52, 9 May 2016 (UTC)