Standard

Posted by

Posted on

2 Jul 2014

Posted under

Democracy

Comments

7 Comments

Any psephologists out there? Or, failing that, good mathematicians?

Here is the data about the missing 6% of voters. They appear to form an almost perfect bell curve with the largest losses concentrated in the middle-size areas in percentage terms. Page 1 is a frequency chart, page 2 is the raw data by area and page 3 is an analysis of the raw data.

EDDC electors

7 thoughts on “Any psephologists out there? Or, failing that, good mathematicians?”

  1. Nice analysis and graph, but as a (30+ year lapsed) mathematician I am not sure that there is any conclusion that can be drawn from this.

    I would be interested to compare East Devon with other Devon districts and with Somerset, Dorset and Cornwall districts surrounding Devon to see how localised this is.

    Has anyone got the time to look into this?

    Like

      • I have tried to find statistics on neighbouring district council web sites, but surprisingly the EDDC site is the only one which gives voter numbers. Not even the Electoral Commission site can provide statistics. I have raised FoI requests for some numbers from the neighbouring councils, but until they get answered further analysis is impossible.

        Like

      • That’s odd because EVERY council was supposed to collate this information and publish it in February each year. Definitely curioser and curioser and one can already see the pitfalls between now and the General Election.

        Like

      • It may be something to do with the very recent change in voter registration – it was previously the local council’s responsibility to register voters but as of last month (?) it became a centralised responsibility.

        So the obligation to publish voter numbers may have now gone away. But I couldn’t find voter numbers on any of the other local council web sites that I looked at and had to raise FoIs as a way to get the information.

        Like

  2. Who is responsible for the electoral roll? My guess: Mark Williams in his capacity as Returning Officer for East Devon. Some searching questions to him perhaps (in writing, in triplicate or whatever the rules are this week) and at a public meeting. How do you loose 6,000 people from an increasing population in one year?

    Like

  3. As far as the electoral numbers are concerned, a bell curve is exactly what you would expect. The author of the blog, I think, has got it wrong. The bell curve does not have the size of parishes on one side of the graph. It is the number of parishes within a certain band of percentage change that is represented in the graph. These are ‘always’ bell curves. If you sample the speeds of cars on an ordinary road, there will always be more doing between 30 and 40mph, than there are those doing 70-80mph, or 10-20mph.

    So nothing sinister or surprising there.

    Also you always get more ‘eccentric outliers’ when the sample is from a small community, so the biggest changes tend to be in the tiniest parishes. All the big parishes are in a fairly narrow band. This again is to be expected.

    The only exception to all this is Broadclyst, but of course that is due to Cranbrook.

    So nothing odd about the data.

    However, the decline in electoral numbers is highly mysterious and must represent a change in the way EDDC are doing things.

    Like

Comments are closed.