Eu Referendum Demographics: A Different Take.

I've been interested to find out how many people from each age group actually voted in the Eu referendum.

From my findings below I reckon the votes cast by the 18 - 24 age group amounted to barely 6% of the total vote cast. So for every young voter who turned out there were four pensioners who turned out to vote.

We've all seen the standard Eu voting demographic which looks like this:



This is from a quite detailed set of statistics surrounding the referendum by Lord Ashcroft.
The full Lord Ashcroft data is (HERE) and the BBC version is  (HERE).

Both though miss one very important (or vital) statistic. What was the turnout demographic?

Look as you might, there is very little about how many people by each age group voted rather than just which way they cast their vote.

There is this from The Financial Times (HERE) that shows a trend that the older a towns population, the higher the Brexit vote. But while it is qualitatively interesting it gives no real indication of how many people from each age band voted.



But there was one source based on a poll and reported by Sky data on Twitter. Polls, as you know are far from definitive but I think it would be fair to say that it could be regarded as reliable to +/- 5%.

Here is the tweet.


(The caveat that was here relating to the Sky Data has been removed as it proved to be a red herring) 

From this we can see that while 73% 18-24 year olds  voted to remain, only 36% of that age band in total actually turned up to vote. The turn out for this age band was dismal.

In order to translate this into an approximate number of voters, we need to know the population density within that age group.

Here is a demographic table for the UK from Wikipedia (HERE)


We get (approximately)

18 - 24   5,900,000
25 - 34   8,400,000
35 - 44   8,800,000
45 - 54   8,700,000
55 - 64   7,400,000
65+       10,500,000

Clearly, not only is the 18-25 age group the least likely to vote but they are also (by far) the smallest group of voters.

If you now use the the Sky data you can get a rough idea of how many people voted in total from each age band

18 - 24 (5,900,000 * 0.36) 2,125,000
25 - 34 (8,400,000 * 0.58) 4,875,000
35 - 44 (8,800,000 * 0.72) 6,335,000
45 - 54 (8,700,000 * 0.75) 6,525,000
55 - 64 (7,400,000 * 0.81) 5,995,000
65+    (10,500,000 * 0.83) 8,715,000

This above table totals to 34,570,000 whereas the actual total who voted was 33,577,342 which is less than 3% different (2.95%)

I would suggest that within a 5% tolerance the figures above are fairly good.

So from this we can see that for every young vote there were four pensioner votes. The young vote (18- 24) was only 6% of the total vote cast.

If the young had voted with the same dedication as anyone above the age of 35 then the vote would have been with remain.


4 comments:

Anonymous said...

We need to verify those sky stats are from the referendum and not the 2015 GE

chris scott said...

We need the stats verifying that they were from the referendum and not the 2015 GE

BilloTheWisp said...
This comment has been removed by the author.
BilloTheWisp said...

Hi Chris Scott,

I deleted my initial reply to you after doing a some more research into the Sky Data which now looks solid. In my initial reply to you I think I fell into the same trap regarding GE 2015 as you have and pulled the post.

I think the confusion with GE 2015 is caused by their reply to Frederieke Hegger‏ @FrederiekeH
Her query was:

Frederieke Hegger ‏@FrederiekeH Jun 25
@SkyData Can you explain this tweet? How did you establish that people didn't vote?

Their reply was:

Sky Data ‏@SkyData Jun 25
Based on 9+/10 certainty to vote, usually/always votes, voted/ineligible at GE2015

So all they are saying is that those included in this poll were high potential voters who had either voted in GE 2015 or were at that time ineligible to vote.

Of course these criteria therefore apply across all age bands.

In fact you could construe that it was unduly "kind" to young voters as it included those previously too young and ineligible. They therefore had no history of reliable voting.

So it looks like the Sky Data does relate to the Eu Referendum and so (as far as any poll can) is reliable.