Find Out Now have conducted a hypothetical Scottish Independence Referendum voting intention poll for the Alba Party finding that 48.6% would vote Yes, 45.2% No, and 6.2% Don’t know. This gives Yes 52%, and No 48% when Don’t know responses are removed. Like our last two Independence Referendum polls, we used more up to date population quota estimates from the ONS, taken from the 2021 census, as well as including quotas for the 2014 Independence Referendum, taking ageing into account.

For ease of comparison with our older Scottish Independence polling, we have also included results without using 2014 Independence Referendum polling quotas in the balancing of the sample:

If there were a referendum on Scottish independence with the question, “Should Scotland be an independent country?”, how would you vote?

With 2014 vote quotas:

• Yes – 48.6%
• No – 45.2%
• Don’t know – 6.2%

Without 2014 vote quotas:

• Yes – 48.0%
• No – 43.7%
• Don’t know – 8.3%

The survey asked 5,313 Scottish adults how they would vote, and how likely they would be to vote, before outputting a representative sample size of 842 based on ONS’s 2021 population estimates for age, gender, Scottish region, social economic group, EU 2016 vote, GE 2019 vote, Indyref 2014 vote (where stated), and turnout likelihood. The past vote quotas were also adjusted for an ageing population, namely increasing the “could not vote” quotas, and decreasing the other quotas accordingly.

Tables

Turnout adjusted voting intention results with 2014 IndyRef quotas
Turnout adjusted voting intention results without 2014 IndyRef quotas
Turnout adjusted turnout likelihood results with 2014 IndyRef quotas
Turnout adjusted turnout likelihood results without 2014 IndyRef quotas

Non-turnout adjusted voting intention results with 2014 IndyRef quotas
Non-turnout adjusted voting intention results without 2014 IndyRef quotas
Non-turnout adjusted turnout likelihood results with 2014 IndyRef quotas
Non-turnout adjusted turnout likelihood results without 2014 IndyRef quotas

Important notes

Not everyone who answers a poll about voting, would actually vote in a real election or referendum. It is important to note that unlikeliness to vote is not the same as selecting “I don’t know”. We adjust the results to the voting intention question (Yes, No, or I don’t know) by only including those who say they are most likely to turn up to vote. Our voting likelihood adjustment sets the threshold for counting a response as a vote or not so that the overall turnout would be 85%, as it was in the actual Independence Referendum in 2014. in this case, the threshold is set at 5, so anyone who says they are “5” to “10 – Extremely likely” to vote are included.

The voting intention results before turnout likelihood is adjusted for is reversed, a marginal lead for No.

Yes respondents answer that they are more likely compared to No respondents, so more of their voting intention answers are included in the turnout adjusted results. We covered this in more detail in our last Scottish Independence poll article.

It is unclear whether this would accurately reflect the respondents’ behaviour in a real world referendum. When faced with a real-world outcome they do not like, they may become more motivated to act.

In addition, our (and others’) hypothetical voting intention polling does not, and cannot, take into account the campaigning that would precede a real-world referendum or election. No could perform better than expected, and tip the balance back, but the reverse could also be true.

You may also be wondering why the number of interviews (5,313) is so high compared to the final number of respondents used in the tables (842). This is because our methodology differs from other polling companies. We over-sample until we have enough respondents for each demographic quota. Typically, 18-24s take the longest to sample, so we end up with more responses from older age groups. However, the “Filtered results” contain the correct number of respondents for each quota within plus or minus one percent. More about our methodology here.

Finally, it is important to note that the sample contains no 16-17 year olds, as our panel is 18 years old and older only. The 2014 Independence Referendum allowed 16-17 to vote.

The fieldwork was conducted between the 11th and 24th of January 2024.