Scottish Social Attitudes Survey 2025: Technical Report

6 Questionnaire versions

Each address was allocated at random to one of two versions of the questionnaire, each of which covered a different mixture of topics as described in section 2 of this report. Both versions of the questionnaire collected key demographic information about participants.

6.1 Demographic data on health and disabilities

A routing error resulting in some missing demographic data in the version of the survey containing the Scottish Government funded questions was identified during checks carried out by Data Managers shortly after survey fieldwork began. As a result of this error, approximately 750 respondents were not asked the three questions (below) on general health and long-standing conditions:

• How is your health in general? Would you say it was very good, good, fair, bad, or very bad?
• Do you have any physical or mental health conditions or illnesses lasting or expected to last 12 months or more? (Yes or No)
• [If Yes to above] Does your condition or illness/do any of your conditions or illnesses reduce your ability to carry out day-to-day activities? (Yes, a lot; Yes, a little or Not at all)

The health and disability data is used along with other background data to assess the representativeness of the sample and was intended to be used for analysis of relevant questions in the attitudes to discrimination module. On identifying the error, the web questionnaire was immediately rectified by adding these three questions into the survey for remaining respondents. To mitigate for the missing data a mini follow-up survey containing the three affected questions was fielded to respondents who had 1) given consent to be recontacted about further research and 2) provided an email address. The mini survey was sent out on the 29^th of October and was in field for two weeks, resulting in 181 responses. The final proportion of respondents in this version of the survey with missing data on these three questions was therefore 37% (n = 570).

Analysis was undertaken to compare the demographic profile of the sample who responded to the follow-up survey providing data on these three questions with the full sample on this version of the survey and the version of the survey (funded by ESRC) for whom all respondents were asked these question from the outset. For both unweighted and weighted analyses, the differences across all categories were small (generally within 1-2 percentage points) and showed no systematic pattern. This suggests no evidence of bias in the follow-up response to the mini survey.

6.2 Disability data and survey comparability with previous face-to-face mode

The data on long-standing conditions/disabilities on SSA, alongside other key demographic data, provides evidence on the representativeness of the sample and the comparability of the previous face-to-face mode with the current web mode. For the last two years of the web survey (2024 and 2025) the data has been consistent on this measure and similar to the data from the last years the survey was run face-to-face, giving further reassurance that the web method is robust.

The weighted proportion of respondents with long-standing health condition/disability on SSA 2025 (in the ESRC funded version) was 38% and on SSA 2024 was 41% which remains much more in line with previous SSA face-to-face surveys (where it was usually 44-46%). The figure was lower (30%) in SSA 2023 when the survey first moved to online. This seems likely to have been as a result of question order - SSA 2023 is the only SSA year where the this question did not immediately follow the question on respondent’s general health.