With just over a week to go until the 2018 KSU election, SDM lead Pulse by a huge margin of 22 percentage points, a survey carried out by the Yuppie shows. In answer to the question ‘If the election was tomorrow, would you vote for…’ 38% of campus said they support SDM, 16% supported Pulse, and 26% said they intend to cast a mixed ballot. 20% of campus said they would not be casting a vote.
The 180 respondent survey was carried out over Wednesday 11th to Monday 16th April. Unfortunately, Kris Bajada’s announcement that he is also contesting only came after 141 responses were gathered. Participants on subsequent days were told that they could include him in their responses, but over the remaining 39 responses, no one opted to.
Extrapolating the results to when the “won’t vote’s” are excluded ends up giving SDM 47.9% of the votes and Pulse 19.4% of the votes. This result is similar to the historical average of KSU elections over the past decade.
As a percentage of block votes, which also extrapolates away mixed votes, this means SDM is on track to win 71%, with Pulse getting the remaining 24%.
Interestingly, the results indicate a situation that is more or less come to be expected, with Pulse’s decision not to contest for the past two years having had seemingly little effect. In fact with these statistics, one can make the argument that the decision has backfired somewhat.
The student parties last went head to head in 2015, and SDM managed to secure 50.6% of all votes cast then, while Pulse obtained 25.5% of all votes cast. The polled estimates indicate that Pulse has lost 4.1% over the three years compared with SDM’s 2.7%. The increase was made up for by students who said they’d cast a mixed vote, which jumped up by 7.1% since 2015.
This doesn’t mean it’s game over for Pulse
With over a week to go, anything can happen. This exercise simply obtained the mood of campus two weeks prior to the election after the respective teams of candidates were announced, and it’s entirely possible that Pulse might swing the tide in their favour.
Indeed this survey was carried out before any proposals, by either side, were released and before the official roles for each candidate were announced. One could argue in fact that with the percentage of people stating that they intended to cast a mixed ballot currently standing at a substantial number, that same number may decrease and turn into block votes for either side upon the release of the respective manifestos.
Things can also end up being different because of the independent candidate contesting, and since the announcement only came after the bulk of this survey was completed, it probably didn’t accurately capture the entire picture.
Majority of Students Think KSU is doing a good job
In answer to the question ‘Do you think KSU has done a good job this year’, 61% said yes, 34% were unsure, and only 4% said no.
Limitations of This Exercise
Because of a number of constraints, this survey used a variation of a convenience sample that closely resembles the mall intercept method, were people are asked if they would like to participate. Since this is not a genuine probability sample, where each student has an equal and random chance of being chosen, you will have to take its findings with a pinch of salt.
Despite this, attempts are made to match the demographics of the sample with the actual population of university. For instance, if we know that 11% of University attends the faculty of Arts, we’ll aim to match this in our sample. Here are the demographic splits for faculty and sex compared to the latest statistics from University:
The first column indicates the percentage of students that each faculty has as a proportion of the entire campus. The second column is that observed proportion in our sample, while the third is the deviation of our sample from the population. For instance while we managed to accurately represent the faculties of Dental Surgery and the Built Environment, we over represented Laws and underrepresented Social Wellbeing.
The second main limitation is the sample size, which could stand to be larger. As a side note, if our sample was a probability sample, a size of 180 respondents would give a margin of error of 7%.
Thank you to all the students who took part in the survey! The Yuppie will continue to provide daily coverage on the upcoming KSU Elections.
Author – Charles Mercieca