Did young people bother to vote in the EU referendum?
Despite claims that young people could not be bothered to vote, they did. In fact, turnout was 64% among 18-24 year olds and xx% among 25-39 year old, which is closer to the population average than could have been expected. 71% of people aged 18-24 voted to remain in the EU, and xx% of 25-39 year old’s. This confirms that the generational gap was really about how people voted, and not about whether they voted or not.
Much has been said of the generational gap between young and old voters in the referendum on EU membership, but whilst the contrast in terms of preference, with young people massively supporting a vote to remain in the EU, was unanimously accepted, the question of youth turnout was far more – and surprisingly – controversial.
Differences in turnout across age groups are rarely controversial for three main reasons. The first and most obvious one is that they usually do not matter. Most aspects of electoral behaviour do not tend to vary that much by age, and the relatively clear generational gap identified in this referendum is, in that sense, quite unusual. The second is that the fact that young people usually tend to vote very significantly less than older ones is well-known, so it was only questioned this time because some commentators claimed that the turnout was in fact even larger than usual (see below). The third and final reason is that in all general elections, we tend to have an exit poll and a British Election Study which both give clear insights into turnout so that there is relatively little uncertainty on turnout, while neither took place this time.
Perhaps because there was so much to comment on, and not exit poll to use, the media tried to find whatever information they could, and when it came to turnout, the first figures that hit the news were released by Sky Data in a Tweet and were understood to mean that only 36% of young people aged 18-24 voted in the referendum. Most articles referring to youth turnout used that figure, and yet, the only problem with it is: the figures that Sky Data released had effectively nothing to do with turnout in the referendum. As they explained in a further tweet, the figure came from a survey actually conducted for the 2015 General Election, more than a year ago, and looked at the proportion, within each generation who say that they always vote. Other figures floated around, but they seemed to be disconnected from any relevant public opinion data.
While the Sky Data figure has nothing to do with turnout in the referendum, it is in fact very interesting, because in a way, it tells us about what turnout could have been expected to be if young people had behaved the way they usually do on the day of the vote. In truth, however, they did not, and their turnout was much higher than that.
In a blog piece for the BBC, Prof. John Curtice points out that there seems to be some uncertainty about a lack of certainty about young voters’ actual turnout. He finds that a recent BMG survey estimates that 60% of 18-34 year old’s voted in the referendum against 80% of those aged 65 and over, whilst pointing out a key issue: all surveys get “too many voters” according to non-voters, in part because we know that people over-report participation in elections, and in part because many people who do not vote, equally do not wish to answer surveys. Comparing answers to a Survation poll for the Mail on Sunday that was conducted after the referendum in similar conditions to another poll that they conducted after the 2015 General Election, Prof.Curtice also finds that while there is no huge difference between the two cases, reported abstention was in fact 2% less for young people aged 18-34 whilst it was 2% more for the other two categories reported (35-54 and 55+). This would suggest that in fact, the turnout of young people was in fact closer to that of other age groups in the 2016 EU referendum than in the 2015 General Election although he also points out that differences are too small to exclude the possibility that the pattern was possibly quite similar in the two cases.
To make things worse, however, a crucial element to us is that most commercial surveys ask respondents whether they voted or not and use that figure to estimate turnout. The elephant in the room is that this is not, however, how turnout is calculated in real life. Legally, electoral participation is the proportion of people who vote amongst eligible (registered) voters, and not amongst the population as a whole. The distinction is critical because we know that young people are far more likely not to be registered electorally than any other age group. In fact, in July 2014, the Electoral Commission confirmed that “younger people (under 35) are considerably less likely to be registered” with only 70.2% of 20-24 year old on electoral registers against 95.5% of 65+ year olds.
While the Electoral Commission[1] has made tremendous efforts to reduce this gap, there are well-known structural reasons why younger people are always less likely to be correctly registered electorally in any country and this means that if we do not control for whether survey respondents are registered to vote or not, we will significantly under-estimate youth turnout compared to other categories.
The question of whether young people voted or not is politically important for two critical reasons:
1)because there continues to be a significant proportion of younger voters who say that they are unhappy with the result of the referendum and want to be heard, and one of the key arguments that have been made in answer to them is: “they should have bothered to vote if they cared that much”, and
2)because the Government chose not to give the right to vote to 16 and 17 year old’s in the referendum, and it is fair to ask whether allowing them to vote could have changed the result of the referendum or not.
In that context, the ECREP electoral psychology team at the LSE and Opinium have collaborated to try and get a clearer sense of the turnout of young people using two surveys where we controlled – among other things – for respondents’ electoral registration.
The first is the panel study (meaning that the exact same people were interviewed three times in late April, late May, and late June) that Opinium fielded on our behalf with 3,008 respondents from April to June as part of our comparative study of the psychology of voters in 25 countries as well as a joint attempt to understand how British people’s social, political, and consumer behaviour would likely be affected by the referendum result. However, whilst panel studies are in many ways the golden standard for election studies when it comes to try and understand election effects and “what happened” throughout a campaign, they are not ideal to infer such elements as electoral choice and turnout from the population at the end because we know that they tend to disproportionately lose abstentionists across waves and also because the very fact that people are asked three times about the referendum can make them more interested in it than average voters. In other words, those surveys are ideal to understand processes and effects, but not to take a snapshot of the population. That study suggested a turnout of young people of about 70% for 18-24 year old’s and 67% for 25-39 year old’s after weighing for the over-reporting of participation, but again, had to be taken with caution.
The second study that we used, simply used a fresh sample of 2,008 voters, but unlike most commercial surveys, specifically asked respondents to tell if they were registered to vote or not in order to be able to evaluate actual turnout rather than the gross proportion of a sample that reported voting. After weighing the results to match the actual overall turnout of 72%, we then found that turnout was 64% for 18-24 year old’s and 65% for 25-39 year old’s. It was almost identical for 40-54 year old (66%) but then increases to 74% for 55-64, and 90% for 65+ year old’s.
What this suggests is that whilst young people voted a little bit less than average, they were probably quite close to the national average (only 8 points below according to our survey). This is of course notwithstanding the methodological issue of the difficulty to get abstentionnists to answer surveys rightly mentioned by Prof.Curtice, but of course, this affects all types of voters and not only young ones, so while it imposes that we consider those figures with caution, there is no particular reason to believe that they would overestimate (as oppose to underestimate) youth turnout specifically, especially as it undoubtedly also affect the ability of survey companies to capture non-registered voters which higher proportion amongst young people is, as mentioned, one of the reasons why turnout among that age group is likely often underestimated.
This figure suggests, at the very least, that whilst 18-24 year old voted less than voters above 55 in particular, the differential was certainly not more than in general elections, and quite probably less, suggesting that young people made at least as much and in fact probably more of an effort to vote in the EU referendum than they did in recent General Elections. In that sense, it is certainly unfair to criticise their right to have an opinion on the outcome on the ground that they “did not bother to vote”.
The question of what it tells us about what might have happened had 16-17 year old’s been allowed to vote is of course far more complex. Our electoral psychology team has done a lot of work on both first time voters and the electoral behaviour of 16-17 year old’s with recent 7 country and 4 country comparative studies. Invariably, we find that 16-17 year old’s vote less than older voters but also significantly more than the “18-30” age bracket. We had confirmation of this recently in both Austrian elections and the Scottish independence referendum, and other studies such as the ICM post-referendum survey in Scotland confirmed the same, with suggested participation rates of 75% among 16-17 year old’s as opposed to only 54% of 28-24 year old’s[2].
In fact, we know that 16-17 year old’s (as well as 18-19 year old’s) are also significantly more likely to register than 20-24 year old’s. In both cases, there are structural reasons why early 20s are “the wrong age” to register to vote or vote, whilst 16-17 is not: many young people in their early 20s study or move away from home, are often less aware of local political stakes (something which the political science literature confirms to be an extremely important predictor of voting), are more likely than any other group to travel or live abroad and are often not even aware that this does not affect their right to vote as well as present serious practical complications if they have registered to vote in a place where they are not physically present on the date of the vote (something which was particularly relevant with a referendum that took place after the end of examination periods in many UK universities). By contrast, 16-17 year old’s often live at home with their parents, are more likely to be present both to register and on the day of the vote, are in a context where they are more likely to be connected with political debates and local networks.
Allowing 16-17 year old’s would have added nearly 1.6 million potential citizens to the electorate, but it is of course extraordinarily difficult to know if it might have affected the outcome of the referendum. On balance, the results of our surveys on the turnout of 18-24 year old’s would suggest that it would not have been enough to overturn the result of the referendum unless 16-17 year old’s were almost unanimously voting in favour of remaining, but it would have almost certainly reduced the advantage of leave to such a point (likely less than 500,000 votes) that the very concept of a majority would have been highly controversial.
[1]
[2]