Edmonton (Rajeev Sharma): Amid renewed discussion around provincial sovereignty, a clear majority of Albertans would choose to remain part of Canada if a separation referendum were held today, according to a new survey released by the Angus Reid Institute.
The survey found that only 29 per cent of Albertans would vote in favour of separation, with most of them describing their stance as a leaning rather than a firm commitment. In contrast, 65 per cent said they would vote to stay in Canada, and most respondents in this group said their decision was definite.
The findings are part of the first instalment of a three-part Angus Reid series examining separatist sentiment in Alberta and Quebec, as well as the factors shaping public opinion on the future of the country.
Although opposition to separation remains strong, the survey indicates that opinions are not entirely settled. About four in five Albertans said they have mostly made up their minds, suggesting some room for shifts in public sentiment.
Reasons for and against separation appear sharply divided. Albertans who firmly oppose leaving Canada said national identity and practical concerns were the most persuasive arguments for remaining in Confederation. More than 90 per cent cited these factors, including fears that a landlocked and independent Alberta could face serious challenges in exporting its natural resources.
Those leaning toward staying in Canada agreed with these concerns but expressed them less strongly. Many in this group also acknowledged that some separatist arguments resonated with them, particularly claims that Alberta contributes more to Canada than it receives and would benefit from greater control over its resources.
Albertans who support separation said they were largely unmoved by arguments to remain. Instead, they strongly endorsed the view that Alberta should distance itself from federal policies they believe harm the province and gain full authority over its natural resources.
The survey also explored perceptions of political leadership on the issue. Alberta Premier Danielle Smith recorded a net approval rating of minus 17 for her handling of the debate, while Prime Minister Mark Carney posted a net approval of plus seven.
Premier Smith has faced criticism following her government’s decision to lower the threshold for citizen-initiated referendums and extend the time allowed to collect signatures, moves critics argue could make it easier to trigger a separation vote.
Support for separation was notably higher among United Conservative Party voters. Sixteen per cent said they would definitely vote to leave Canada, while an additional 41 per cent said they were inclined in that direction.
The poll also highlighted differences in how Albertans access information on the issue. Those who would vote to stay in Canada said they primarily rely on mainstream media and social media platforms, while respondents leaning toward separation were more likely to depend on alternative media sources and discussions with friends and family. Overall, three quarters of Albertans said they have been closely following news related to separatism in recent months.
The survey suggests that a separation vote could have major demographic consequences. Among Albertans who said they would vote to remain in Canada, nearly three quarters indicated they would consider leaving the province if Alberta became independent, while fewer than one in four said they would stay.
Respondents were also asked about the prospect of Alberta aligning with or joining the United States. Most said they believe the U.S. would likely attempt to exert political or economic pressure on an independent Alberta, though fewer believed military pressure was probable. A majority, particularly those opposed to separation, said becoming part of the United States would be a negative outcome for the province.
The Angus Reid Institute conducted the online survey between February 2 and 6, 2026, among 4,025 Canadian adults who are members of the Angus Reid Forum. For comparison purposes, a probability sample of this size would have a margin of error of plus or minus three percentage points, 19 times out of 20.
