How ideology, partisanship and competence best explain UCP vote in Alberta

Statistical analysis of Vote Compass data reveals the factors that propelled Albertans to vote for the UCP and the NDP.

Analysis of Vote Compass data reveals the factors that propelled Albertans to vote for the UCP and NDP

A woman with her back to camera walks through a glass doorway. At left, a yellow sign that covers a loading zone sign on the wall points to the voting station.
Voters at an advance polling station in Carstairs, Alta. A statistical analysis of Vote Compass data reveals the factors that led Albertans to vote for UCP and NDP. (Jeff McIntosh/The Canadian Press)

Echoing another conservative party leader's come-from-behind-victory three decades earlier, a jubilant Danielle Smith welcomed her cheering United Conservative Party supporters on election night "to another miracle on the prairies."

The UCP's exceptional feat on election day — the portent of its winning campaign — was likely its ability to hold on to almost all of the same ideologically right-of-centre voters that cast a ballot for the party in 2019, according to a statistical analysis of Vote Compass data. 

Despite a series of Smith's gaffes and controversies — and public opinion polls suggesting the UCP was trailing the NDP for two years — the party captured 52.6 per cent of the popular vote.

In 2019's spring election, under the leadership of Jason Kenney, the UCP hoovered up 55.12 per cent of all the votes cast. 

Much of the news, commentary and punditry about the UCP's comeback zeroed in on the razor-thin margins in battleground Calgary, but the small 2.5-per-cent difference between the UCP's vote in 2019 and last Monday's election likely explains a lot more about how the UCP held on to power, albeit with a diminished caucus. 

"The UCP did a better job retaining previous UCP voters than the NDP did … retaining previous NDP voters," explained Alexander Beyer, Vote Compass's lead data scientist, who sifted through the data from respondents to CBC News' voter engagement tool. 

All that Vote Compass data 

Vote Compass gathered a tremendous amount of data from more than 175,000 people who participated in CBC News' civic engagement application during the recent provincial election. Imagine a spreadsheet with hundreds of columns and tens of thousand rows of data. 

"Vote Compass's sample size is so large that we really catch folks from all walks of life," said Beyer in an interview with CBC News.

Anyone with internet access and a digital device (phone, tablet, computer) can participate in the online tool developed by political scientists. 

A post-election statistical analysis of 5,000 Vote Compass responses considered what  attributes — sociodemographic factors such as age, gender, income, education, ideology, plus public policy opinions as well as leadership evaluations — likely influenced Alberta voters when they cast their ballots.

Vote Compass data scientists used a regression analysis to mathematically sort out what characteristics had a statistically significant impact on Alberta voters' decision-making. The statistical method allows social scientists to consider relationships between variables, while also sorting out what matters and what can be ignored. Ultimately, this helps to determine which individual properties systematically coincide with Albertans' vote choice.

UCP keeps its traditional supporters 

Similar to brand loyalty to running shoes, voters tend to be faithful to their preferred political party. 

Not surprisingly, the analysis of Vote Compass data shows that voters who opted for the UCP in 2019 tended to stick with the party this time around. 

Until the NDP's surprise win in 2015, conservatives were a constant in government. The Progressive Conservatives held power for more than 40 years. 

"When you've been voting the same way for generations and generations," Naheed Nenshi, told CBC Radio One's The Current the morning after the election, "it's hard to change." 

"It was hard for people to break habits," added the three-term former Calgary mayor.

Ideology, according to Vote Compass's analysis, also plays a significant role in UCP voters' likely support.

Not all that surprising, people who considered themselves on the conservative side of the ideological spectrum were considerably more likely to cast a ballot for the United Conservative Party on Monday. 

"The UCP," said Beyer, "did a better job attracting the folks that perceived themselves to be on the right end of the ideological spectrum."

Headshots of two white female politicians representing Alberta UCP on the left and Alberta NDP on the right.
Composite illustration featuring Alberta UCP Leader Danielle Smith, left, and NDP Rachel Notley on election night on May 29. (Todd Korol/Reuters, Amber Bracken/Reuters)

Trust in Smith's leadership, notably, did not likely play a big role in motivating UCP voters. 

During the election campaign, Vote Compass data, in fact, found that UCP supporters didn't consider Smith all that trustworthy, but they still planned to vote for her.

Increasingly, partisanship has become intertwined with our identity, becoming a potent force in politics, making party supporters more willing to overlook their leader's transgressions. 

Notably, a belief in Smith's competency turns out to be the biggest predictor of UCP support. 

"The group of voters that perceived her to be competent," said Beyer, "they were like, 'Yeah, I'm going to vote for her, no question.'" 

But it's not clear, admits Beyer, if vote choice or perceptions of competence comes first. 

"Did people vote the way that they voted because they thought the leader was competent," asked Beyer. "Or did they get the impression that the leader was competent because they knew which way they wanted to vote?" 

It's not certain based on the Vote Compass data. 

What motivates NDP support 

Like UCP voters, NDP supporters tended to be repeat customers, having voted for the party in 2019. 

Also not surprising, people who consider themselves right-leaning ideologically, were dramatically less likely to vote for the New Democrats. 

The NDP's campaign focused relentlessly on competency and trustworthiness, painting Smith as insincere and chaotic, while casting its leader, Rachel Notley, as an honest and capable leader. 

Competency weighed heavily in NDP supporters' vote choice, but it wasn't as big a factor as it was for UCP supporters.

Some political watchers believe the New Democrats' relentless attacks on Smith's character  backfired

"The NDP focused this election on Danielle Smith, and her flaws in her character and her competency at their own peril," Michael Solberg, a former staffer in Stephen Harper's government, told The Current after the election. 

"Voters," added the partner with New West Public Affairs, "clearly wanted to talk about the broader issues like the economy, affordability."

But, it turns out, the economy and other public policy issues don't appear to have driven much of the support for either the NDP or UCP. 

It's not all about the economy or other issues 

The UCP pitched itself as the best stewards of Alberta's economy, portraying itself as tax-cutters and the NDP as tax hikers

The NDP, for its part, said its first law would reduce costs for Albetans by capping home electricity rates and offering $500 per year per child to pay for children's activities.

The NDP also touted its promise to fix the province's foundering health-care system. 

All those promises and all the back and forth between the parties over whose plan is best or which party would keep the province's economy motoring appear to matter little in Alberta voters' minds. 

Neither naming the economy nor health-care delivery when asked "what is the most important issue for you this election" are good predictors of why Albertans likely opted for either the NDP or the UCP, according to statistical analysis of Vote Compass data. 

"It just didn't catch on with voters," said Beyer. 

How the Vote Compass data is gathered and interpreted:

Developed by a team of social and statistical scientists from Vox Pop Labs, Vote Compass is a civic engagement application offered in Alberta exclusively by CBC Radio-Canada. The findings in this story are based on 5,000  respondents who participated in Vote Compass from May 1 to May 29, 2023.

Unlike online opinion polls, respondents to Vote Compass are not pre-selected. Similar to opinion polls, however, the data is a non-random sample from the population. Respondents for this analysis have been quota-sampled from all Vote Compass respondents to approximate a representative sample of Alberta's population in gender, age, education, region and partisanship according to census data and other population estimates.


Brooks DeCillia spent 20 years reporting and producing news at CBC. These days, he’s an assistant professor with Mount Royal University’s School of Communication Studies.