Last week, two Cambridge student newspapers, TCS and Varsity, published surveys of political opinion amongst University of Cambridge students. These both suggested that opinion amongst students is a lot less pro-Lib-Dem than indicated by the Ashcroft poll of Cambridge voters in March. TCS have kindly sent me the full tables from their survey, so here’s a more detailed look at their data.
First of all, the data includes a timestamp for each survey response. This means that we can look at how the vote shares shifted as each survey response came in. This makes for quite an intriguing graph:
This graph shows the share of the vote as each of the 732 responses came in, between March 3rd and April 19th – though of course not evenly spread out over that time. At first there is a good deal of wobbling about due to random variation. After a while, the Greens briefly take the lead, but Labour soon recover pole position; by the midway point Labour are at around 33%, and settle down there for the remainder of the survey. After their strong start, the Greens sag a bit to finish on 21%, while the Lib Dems drift upwards during the second half of the responses, just taking second place. The Conservatives have a little jump towards the end, but still finish fourth.
It’s notable that most of the parties have periods where there’s a sustained rise in their vote share; it’s likely that these represent times when they were contacting their supporters and asking them to participate in the survey. However, the relative stability of the graph in the second part of the survey period suggests that this had only a limited effect on the overall result, and the broad picture of Labour in the lead ahead of a chasing pack of three does actually represent student opinion fairly accurately, not just CULC’s superior ability to pester its supporters. The Lib Dem drift upwards in the later part of the survey is perhaps due to the effect of their leaflet campaign. So overall this does reinforce the theory that the race for Cambridge MP is a good deal closer than the Ashcroft poll indicates.
The survey also asked students for their college, year, and subject, so we can look at how those factors correlated with political opinion. Of course we are dealing with some pretty small sample sizes in some cases. I’ve only included groups that had at least 20 responses, but even so, random variation will have a significant effect here, so take this with a pinch of salt. Here are the results for the colleges where at least 20 students responded, sorted by the size of the Labour vote:
Girton comes out as the most pro-Labour college, just ahead of traditional left-wing hot-bed King’s – though UKIP’s candidate Patrick O’Flynn as well as Labour’s Daniel Zeichner studied there. Clare, Julian Huppert’s college, has the largest Green share of the vote, while Trinity is by far the most Conservative, and Jesus has the largest Lib Dem share.
Here are the vote shares by year:
There’s not a great deal of variation here; Green support seems to diminish slightly with age, at least in the first three years, and Lib Dem support increases slightly, but not a lot. There is a good deal more variation by subject, however:
As noted above, these are just the subjects where at least 20 students responded. Labour support is strongest amongst English, HSPS and History students; the Conservatives do best with mathematicians, economists and medics; geographers and NatScis like the Lib Dems most, and the Greens are particularly popular with musicians. Nobody likes UKIP very much.
One thing this survey doesn’t tell us much about, and which will be vital to the Cambridge result, is how many of the student supporters of each party will actually get to the polls on Thursday. Often younger people have a lower turnout than older people – and this could be crucial to the Cambridge result which will be declared at the Guildhall in the early hours of Friday 8th May.