Croydon and the referendum: how we voted

By - Wednesday 15th February, 2017

Tracking voting data across the borough’s wards

Photo public domain.

More has emerged on how people voted in the EU referendum through an investigation by the BBC. Reporter Martin Rosenbaum used freedom of information requests to access more detail of how people voted from a cross-section of councils.

This data was not previously available; indeed, it is not data that councils were required to collect at all, because this was a referendum and not an election. All that mattered was the total number of votes cast for and against, totalled across the UK.

With no requirement for constituency tallies, votes were counted for administrative convenience, in our case by Croydon Council. Without a ward or parliamentary constituency-level breakdown, up until now we could only use models to estimate (read ‘guess’) how the voters of our three parliamentary constituencies split for Remain or Leave.

Fortunately Croydon and our neighbours Lambeth, Bromley and Merton all counted the votes down to ward level. Martin Rosenbaum gathered this data and kindly supplied me with his analysis. The only slight wrinkle is that some postal votes were mixed in prior to counting, so ward-level voting data cannot be considered exact. Not perfect, perhaps, but since postal votes divided similarly (although slightly more Remain-y) to the votes cast at polling stations, then the data is not too bad at all.

The primary indicator was level of education

Mr. Rosenbaum’s conclusions are along the lines of those I previously published in the Citizen. The primary indicator of voting was level of education, with a much weaker correlation with age. People with higher qualifications, especially younger ones, were more likely to vote Remain. Ethnic minorities were also more likely to vote Remain, although some sub groups leaned towards Leave.

What more can this new information tell us? Most obviously it is clearer that Gavin Barwell’s Croydon Central constituency voted to Leave. My estimate from the cruder data had been that this was by a narrow margin, but the ward-level data shows Croydon Central went 50.3%/49.7% in favour of Leave, slightly less than was the case for the whole of the UK. Steve Reed’s Croydon North and Chris Philp’s Croydon South went approximately 59% and 54% for Remain.

All the factors indicate a high Remain vote

Looking at the ward-level data for ourselves and our neighbours illustrates the correlations. Lambeth, with its relatively younger, better-qualified, high ethnic minority population combines all the factors that would indicate a high Remain vote. Indeed in the event the highest Leave voting ward in Lambeth at just less than 30% is lower than the lowest Leave voting ward in Croydon at just shy of 34%.

At the other extreme, Fieldway and New Addington are up there with the highest Leave votes in our area. They combine the main factor (lower education level) with a secondary factor (low ethnic minority) that would indicate a high Leave vote. They also show the weakness of the age correlation, given that they are in the middle to lower end of the average age range.

An additional factor that had an impact on the result was the high turnout. Nationally it was 72.2%, compared to the 2015 general election’s 66.1%. In Croydon, the referendum turnout was 69.8%, compared to the 2015 general election’s 66.6%. What is interesting here is to ask ourselves who were the people who made the effort to turn out for the referendum who did not do so in the last general election?

Turnout was over 70%

The new information confirms that these newly energised voters were disproportionately from groups more likely to vote Leave. Comparing referendum turnout by ward, Fieldway and New Addington turnout were both over 70%, whereas in the general election in 2010 (the most recent general election for which I could find ward-level voting figures), the turnout was only 56%.

Other Croydon wards typically increased turnout by a factor of 1.1 compared to general elections. These two wards increased theirs by a factor of more than 1.3 for the referendum. This goes a long way to explaining why the opinion polls underestimated the Leave vote.

For Croydon – and more particularly Croydon’s politicians – there are some lessons for the future. Gavin Barwell, an enthusiastic Remainer whose constituents voted Leave, can with justification claim the backing of his constituents for his approach to Brexit.

Interesting times ahead

Both Labour, Tories and UKIP will look to the newly energised voters in Fieldway and New Addington as opportunities (or in Labour’s case, threats). New Addington residents who complain that Labour has forgotten them may well find they will be increasingly love bombed as the May 2018 council election approaches. Croydon UKIP’s purple colours will also – I suspect – become increasingly evident on New Addington’s Central Parade. Interesting times behind us – and interesting times ahead.

This article was updated on 15th February to correct an inaccurate voting figure for Croydon Central.

Robert Ward

Robert Ward

Engineer and project manager specialised in helping businesses make better strategic decisions and improve safety, quality and effectiveness. Conservative Party Councillor representing Selsdon and Addington Village on Croydon Council. He tweets as @moguloilman.

More Posts

  • Michael Swadling

    Robert great write up. With more recent generations so much more likely to attend college and university I’m not sure age and education level can be readily separated.

    We have had the Croydon data from about a week after the referendum. I make it 50.3% Leave in Croydon Central but I’m happier to take your figure!

    Part of the reason for the high turnout in New Addington and Fieldway was great work done by the Vote Leave team canvasing the area.

    The other part is of course the way they have been let down by successive Labour and Conservative governments. These governments and the whole media establishment have been far readier to accuse them of xenophobia than listen to their concerns or even accept it acceptable to believe British laws should be made in Britain.

    The most interesting result of course was in Ashburton ward which had an exact 50/50 split in votes.

    • Robert Ward

      Thanks Michael. I take your point that age and education level are to some extent proxies of one another.

      I was surprised at the statistics on the level of higher education of people now in their seventies and eighties. These show in our area as being much higher than I had expected given that the chance of a working class child going to university at that time was tiny. My guess is that over time London job opportunities attracted degree level people from across the country and indeed from abroad which has pushed the number up.

      • baldfeller

        Or better educated people don’t end up doing jobs that shorten life expectancy, so more actually make it to old age. Apologies for replying a year late.

    • KeepRightcroydOn

      Any idea what the turnout in New Addington and Fieldway was in GE2017?
      I wonder if those disillusioned people who rarely vote, but decided to vote Leave, came out once more in 2017.

      • Robert Ward

        Hard to say exactly because postal votes are all lumped together. However, making an estimate suggests the turnout in those two wards was well up on GE2015, but not as high as for the referendum.

        • KeepRightcroydOn

          Thanks for the figures. If replicated nationally it would imply that a significant minority who voted Leave, do not generally vote in GEs. Enough to give Leave that tiny majority.

          Such a shame that these people, who presumably feel dienfranchised and hence don’t generally exercise their democratic right to vote, did so last year in the misconception that the EU is to blame for their word. Farage and Ukip did a good job in selling lies to these people.

      • Michael Swadling

        It looks like the number of people voting on the day in the referendum in the two wards in New Addington was 10,361. For GE2017 this was 7190.

        The interesting number to have (which I can’t find) is the the number of voters on the day in 2015 to see if 2017 was a big change,.

        • Robert Ward

          Voting on the day at GE2015 was 6,730.

          • Michael Swadling

            Thanks Robert. This means turn out on the day was 6% higher across New Addington compared to 3.6% higher total turnout (postal and polling station) across the constituency as a whole.

            Suggests to me quite likely there was some bump from people who voted for the first time or first time for years in the referendum and voted again in the General Election.

  • Peter Staveley

    Vote Leave (as indeed the Remain side) did a lot of sampling at the count and I can confirm what you have stated. The postal vote generally reduced the Leave votes at the polling stations by around 1%.

    The (pseudo) ward data was made available by Croydon Council immediately after the Referendum and was on their website for a while. All that the BBC have done is contact each Council for that information and publish it.

    Regarding who turned out. We have to remember that the Referendum nationally was the highest turnout for decades and that was because people knew that their vote would actually matter. The problem with First Past The Post is that for the vast majority of constituencies and wards we know before the election which Party and Candidate will win. The Government and Croydon Council Administration is decided in only a few wards and so what is the point of voting if you do not want to vote for that winning Party?

    • Robert Ward

      Thanks Peter, I had not realised the ward level data had been available from Croydon Council for a while. The extra work done by the bbc reporter was mainly slogging through the 2011 Census data to get the age, education, ethnicity data for each ward and then doing the analysis which previously had only been done at a much cruder level.

  • Robert Ward

    Apologies to people reading the article but somehow I put the wrong number in for the Leave/Remain margin for Croydon Central. As Michael Swadling has already pointed out below the correct number is 50.3%/49.7% in favour of Leave. Once again my sincere apologies for this mistake.