Scottish Social Attitudes Survey 2025: Technical Report

4 Sample design

From 1999 to 2015, the survey was conducted with adults aged 18 or over. In 2016, the age range for the survey was extended to include 16- and 17-year-olds to reflect the lowering of the age limit for voting in Scottish elections.

The survey is designed to yield a representative sample of adults aged 16 or over living in private households in Scotland. The sample frame is the Postcode Address File (PAF), a list of postal delivery points compiled by the Post Office.

In 2025 a stratified sample of 15,078 (12,564 in the main sample and 2,514 in the reserve) unclustered addresses was drawn from the PAF Multiple Residency file. Stratification enhances the efficiency of sample design by increasing precision, reducing variability, and ensuring representativeness by controlling for differences between subgroups. The method involves dividing the population into distinct subgroups, known as strata, based on specific characteristics relevant to the study. Each stratum is then sampled separately. For SSA 2025, strata were created for sampling based on SIMD score quintiles whereby the two most deprived SIMD quintiles were oversampled.

Prior to selection of the sample, all PAF addresses were sorted within each stratum by: (a) population density at Local Authority (LA) level; (b) tenure profile (percentage owner occupation) at Output Area level; (c) then within the percentage of owner occupied addresses by postcodes and (d) within postcodes by addresses.

A systematic (1 in N) random sample of addresses was then drawn from each stratum. The list of sampled addresses was then split into a main sample (n= 12,564) and a reserve sample (n=2,514), the latter of which were to be issued, if necessary, to meet the targeted number of completed interviews.

The initial invitation to participate in the online survey was made by post. Consequently, in instances where the selected address contained more than one dwelling unit (DU) or household it was not possible to make a random selection of a single DU/household. Instead, the selected household was effectively the one which first opened the invitation letter and decided to take part. The overall proportion of such addresses is very small (around 1% at the national level) and the impact on survey estimates is expected to be minimal.

Invitations to take part were issued to the 12,564 addresses selected and up to two adults in the household were invited to take part in the survey. The invitation letter did not need to be sent out to the reserve sample as the target response rate was achieved during the main fieldwork period.

A random selection of individuals within a household is difficult to operationalise accurately in an online survey (i.e. where an interviewer is not physically present to verify who is taking part). Therefore, to mitigate the possible effect of selection bias within households (that is, that those who choose to take the survey are distinctive in their attitudes) up to two people aged 16 or over at each address were invited to take part in the survey.

Allowing up to two people to take part still means that not everyone in households with more than two people aged 16 years or over could take part, potentially resulting in selection bias. However, such households account for only around 11.9%^[2] of all households, so the effect will be minimal. This potential discrepancy was also corrected for during the weighting process to ensure that people in larger households were not underrepresented within the final data. This is discussed in more detail in the weighting section.

Following fieldwork completion, a validation check was undertaken of respondent postcodes (collected at the beginning of the survey) against the sample postcodes to ensure that it was only the intended sample that took part in the survey.