Can early polls influence elections? It's a question raised after Alberta did what they always do and elected a PC majority government. Any other year, this would be no surprise, since that's just what Alberta does, but this year early polls had predicted a close race between Alison Redford's Conservatives and Danielle Smith's Wildrose party, some going so far as to predict Smith would be sitting in the premier's chair.
That didn't happen, which is leading to many people declaring surprise about what is normally a foregone conclusion. Instead, I suspect that the early polls influenced the election more than normally in this instance, and that polls are actually more than just a way to predict who is winning.
Let's compare Alberta this year, and Saskatchewan last year. Here, we had polls predicting Premier Brad Wall taking an easy majority. That is exactly what happened, so the polls were accurate. Still, the election overall had a decreased turnout from four years earlier, which I suggest is a result of those polls. With the prediction of an easy majority for a known quantity, there was not much incentive to rush to the polls. As a result, people didn't, there were fewer people going to the polls. It was a very sedate election.
Alberta, on the other hand, had a party perched on the far right lead which frightened many people within the province. Early polls said this party could take the whole thing, which lead to an entirely different reaction. It was the highest voter turnout since 1993 - along with the first time in years the majority of eligible voters put in a ballot - and that is likely entirely due to Wildrose being positioned as a threat. It was something which caused a large increase in the amount of ballots cast and actually caused an increase in voter interest in the province. Without early polls, the Wildrose party might not have been seen as a major contender and the turnout could have descended even further.
Now, this province still has a greater percentage of voters than our neighbors to the west, even with a foregone conclusion being predicted - 66 per cent compared to 57 - but our voter turnout is decreasing, while Alberta had a sharp jump from an all time low of 41 per cent. It makes sense that this would be due to an increased interest in the election results, something that could be caused by a prediction of a dramatic upheaval.
That upheaval never happened, and this might have even been caused by those early polls. People frightened by the Wildrose could change their votes just to keep them from getting into power, and those who normally wouldn't vote did, and given the numbers they kept the ruling PC party in the same position they have always been. The results might have been consistent with previous years, but there was a change in voter involvement.
It's difficult to predict what the Alberta election might have seen had there been no advance polls, it is beyond the powers of science. However, I believe that the election was decided by the polls, but in the sense that it caused a great number of people to get together and stop the predictions from happening. It makes one wonder if these polls could even manipulate the end results.
