A poll released Friday has put Naheed Nenshi 17 points ahead of Bill Smith in the mayoral race leading up to Monday’s election in Calgary.
The poll by Forum Research and commissioned by The University of Calgary and several other universities comes the same day as a Mainstreet Research poll that gave a sizeable advantage to Smith.
The Forum study found 49.8 per cent of Calgarians polled said they would support Nenshi.
WATCH BELOW: Mainstreet Research president discusses new Calgary election poll
Smith would receive 33 per cent support, according to the polling data, and Andre Chabot came in a distant third with only 4.1 per cent.
The poll also found that 12.5 per cent of Calgary voters polled were undecided.
It found Nenshi’s support was highest among female voters at 52.2 per cent and in voters 18 to 34 years old where he received 61.8 per cent.
Smith’s support was highest among voters aged 65 and older.
On the flip side, when asked who they “absolutely would not vote for,” 30.5 per cent named Nenshi.
Another 28.8 per cent said they would not vote for Smith.
How much stock should Calgary voters put in election polls?
There have been a number of polls leading up to the election with radically different results.
Independent pollster Janet Brown said Friday that determining which methodology is likely to produce the most accurate results is a difficult task for the average voter.
“I think this is a cautionary tale for voters to make up their mind based on the policies of a candidate and don’t rely on these polling numbers,” she said.
“Unless you have a statistical background and you really feel confident that you can differentiate a good poll from a bad poll.”
LISTEN: Janet Brown on why it can be difficult to trust polls done during municipal elections
Brown said strategic voters may rely on polls to determine which candidate has the best chance of ousting a mayor or councillor they’re unhappy with.
She said it can be difficult to accurately predict the outcome of an election, especially in municipal politics, because of the likelihood voters will wait until the last minute to make their decision.
“Without parties … to figure out who I should and shouldn’t be voting for, it may just be sort of natural for people to leave it to the last weekend,” she said.
Brown said it can also be harder to obtain a representative sample in polls done at the local level because there is a smaller population size to draw from.
“There’s more error in local elections than there is in provincial or federal elections,” she said.
“We’re just not going to know for sure who did the better job constructing their sample until this is all over.”
Data was collected between Sept. 28 and Oct. 12. Respondents were recruited through random digit dialing via telephone and by online survey, using a total sample size of 843 randomly selected declared Calgary voters. Results based on the total sample are considered accurate +/- 3.37 per cent 19 times out of 20.