It’s hard to measure who took more of a battering in the wake of the 2015 General Election. Some might say the Labour Party, others would point unabashedly to the Liberal Democrats, whilst the more holistic observer may take the view that it was the polling companies who never appeared to even hint at the idea of a Conservative majority.
So far was the pollsters’ fall from grace that the governing bods stepped in, with the British Polling Council and the Market Research Council raising an independent inquiry into just how this level of inaccuracy occurred, and what changes could be made to processes in the future to stop something of this ilk occurring again.
Last year the recommendations were revealed, including: having questions in surveys to determine whether respondents had already voted by post; reviewing existing methods for determining turnout probabilities; reviewing allocation methods for respondents who say they don’t know; obtaining more representative samples of the electorate; and investigating the extent to which those who are naturally interested in politics are more likely to involve themselves in the survey.
These of course are all solid recommendations, but unfortunately there is very little that surveys and polls can do to mitigate the impact of human nature on survey results. There is a certain desirability bias that comes into play with a voting intention poll – notably, a social desirability bias that affects respondents’ behaviour. Often the way questionnaires are answered are skewed, as respondents often want to display the best version of themselves (much to the chagrin of researchers, who are far happier with the imperfect, and most importantly valid, truth). In particular here, voting is a desirable social trait. More people thus report that they will vote than actually will, and particularly in terms of the “shy Tory factor”, as in certain circles it isn’t seen as desirable to vote for the “nasty party”.
Fortunately, we can get a better view into the actual behaviour of citizens, although it is to be stressed in advance that this is in no way statistically reliable data. Consider it an online personal diary of sorts, but in fact, Google searches and trend data can provide a strong insight into the enquiring minds of the UK.
Firstly, here’s how popular Google searches were for the two political parties before, during and after the airing of May vs Corbyn: The Battle for Number 10 on Sunday night between 20:30 and 22:00. These metrics work in a way that track popularity in searches, where 100 equates to the most searches for either group during the entire timeframe, and then sets the benchmark for comparison as to how often the parties were searched throughout the four hours. What’s noticeable here is that even when Jeremy Corbyn was taking questions and being interviewed, the Conservative Party was still more frequently being looked up – in terms of what they stand for, who they are, and even in some cases, what they are.
However, when it comes to the leaders themselves – arguably, the whole point of Sunday’s programme was to pit one against the other – a very different story plays out. Apart from the point where attention turned to Theresa May for the first time, Jeremy Corbyn was a far more popular search amongst British Googlers. In particular, searches for Corbyn peaked at the end of the programme, despite him featuring at the start. The resounding idea left in the heads of those looking to explore what they’d seen further was resolutely Jeremy.
This may not be surprising, we are after all in an era of politics where the party does not matter as much as the personal, where the traditional definitions of a party and what it stands for are beginning to blur as the traditional definitions of “left” and “right” become increasingly redundant.
So, with that in mind, let’s take a bigger overview of what people in the UK are searching for in the past month, in terms of both the leaders and the parties in a side-by-side comparison.
The first most notable thing here is that Theresa May is by far and away more the result of internet inquisition than her party. Undeniably, May matters to this election campaign; it’s not as simple as Conservatives vs Labour.
This wasn’t so much the case for Corbyn, for whom the Labour Party was predominantly the dominant search up until 26 May – but more about that shortly. Particularly here, Labour’s peak on 16 May proved to be far more researched on the day of the party’s Manifesto launch (scoring 74) than Corbyn (43). In comparison, on 18 May, the day of the Conservative Manifesto launch, the eponymous party scored 37, with its leader May on 27. On this day, Labour searches scored 45 and Corbyn scored 28. Even on a day which should have belonged to the Tories, people in the UK were more likely to want to know about the opposition.
May’s peak on 22 May was, unsurprisingly, the day of her notorious U-turn on the so-called Dementia Tax, as well as her difficult interview with Andrew Neil. Interestingly, on the same day, the Labour Party was an ever-so-slightly more popular search term than May herself. Perhaps a sign of a rising search for an alternative. Opposingly, Corbyn piqued interest strongly on 26 May, when he made comments that the Manchester attacks were in part an unfortunate ramification of the UK’s foreign policy over a number of years.
What it does tell us is how individuals are reacting to the General Election story as it plays out. Anyone in the spin department of Conservative HQ will be disheartened to hear that it is only a very small portion indeed of Corbyn searches that include the phrase “IRA”, no matter how hard that angle was peddled. What we can see is that people want to know more about Corbyn and the Labour party. If Google is the centre of exploration for those looking for information – potentially, the floating voters – people are looking to gather their own information on the Labour leader in particular. Whether this translates to taking these lines of enquiry into the polling booth can only really be known on 9 June.
Main image by Carlos Luna