Scottish Climate Survey: technical report
Technical report supporting the publication 'Scottish Climate Survey: main findings'
Part of
Weighting
Design and purpose
Weighting is required to reduce the bias in survey estimates and weights are produced to make the weighted achieved sample match the population as closely as possible.
The weighting procedures for the Scottish Climate Survey used calibration weighting to correct for non-response bias. Calibration weighting derives weights such that the weighted survey totals match known population totals. For the Scottish Climate Survey the population totals used were the National Records of Scotland (NRS) Mid-2023 Population Estimates of Scotland and the Scotland Census 2022 data[1].
The data required two sets of weights to take account of the different base units used for analysis. Since more than one respondent per household could participate in the survey, for any questions requiring a factual response the results were based on households (rather than all respondents), to avoid double counting. The base for the other questions was ‘All respondents’[2].
For questions where the household was the main base of analysis, the data were rim weighted in order by tenure, rurality and Scottish Index of Multiple Deprivation (SIMD) quintile. Questions where the base was ‘All respondents’ were rim weighted in order by age and gender (interlocking), work status and gender (interlocking), education and gender (interlocking), tenure, rurality and SIMD.
To reduce the variance of the weights, and hence increase the efficiency of the sample, the weights were trimmed at below 0.13 and above 4. The weighting procedures applied to the survey data resulted in a weighting efficiency of 55% for the all respondent measures, and 71% for the household measures, for national-level estimates. This efficiency reflects the adjustment because of the disproportionate sampling design and because of differential response by different groups. The lower efficiency for the respondent measures reflects within household selection.
Profile of survey respondents
Tables 7.1 and 7.2 show the weighted profiles of the demographics against to Scottish population profile.
Table 7.1 Weighted sample profile versus population (all respondents)
| Demographic | Unweighted profile (all respondents) | Weighted profile (all respondents) | Scottish population profile |
|---|---|---|---|
|
Age by gender |
|
|
|
|
16 – 34 Male |
4.5% |
11.7% |
13.9% |
|
35 – 54 Male |
10.0% |
14.3% |
14.8% |
|
55 – 69 Male |
17.5% |
11.7% |
11.7% |
|
70+ Male |
13.8% |
7.8% |
7.7% |
|
16 – 34 Female |
6.9% |
12.9% |
14.2% |
|
35 – 54 Female |
13.0% |
15.5% |
15.7% |
|
55 – 69 Female |
17.6% |
12.3% |
12.5% |
|
70+ Female |
12.3% |
9.4% |
9.6% |
|
Prefer not to say / not stated |
4.4% |
4.4% |
- |
|
Working status by gender |
|
|
|
|
Full time – Male |
17.7% |
23.7% |
24.7% |
|
Part time – Male |
3.4% |
3.4% |
3.6% |
|
Not working – Male |
3.8% |
7.7% |
9.5% |
|
Retired – Male |
20.9% |
10.6% |
10.3% |
|
Full time – Female |
15.3% |
16.5% |
16.5% |
|
Part time – Female |
8.0% |
9.9% |
10.3% |
|
Not working – Female |
6.4% |
10.6% |
12.2% |
|
Retired – Female |
20.2% |
13.1% |
13.0% |
|
Don’t know / prefer not to say / not stated |
4.5% |
4.6% |
- |
|
Qualifications by gender |
|
|
|
|
Degree or higher – Male |
21.1% |
14.0% |
14.2% |
|
Other quals – Male |
20.0% |
23.4% |
26.3% |
|
No quals – Male |
3.0% |
6.5% |
7.5% |
|
Degree or higher – Female |
24.0% |
17.9% |
18.3% |
|
Other quals – Female |
21.2% |
23.1% |
24.6% |
|
No quals – Female |
3.0% |
7.5% |
9.1% |
|
Don’t know / prefer not to say / not stated |
7.9% |
7.7% |
- |
Table 7.2 Weighted sample profile versus population (households)
| Demographic | Unweighted profile (households) | Weighted profile (households) | Scottish population profile |
|---|---|---|---|
|
Tenure |
|
|
|
|
Owned outright / Buying with Mortgage |
77.2% |
61.7% |
63.2% |
|
Social renter |
10.6% |
22.0% |
22.5% |
|
Private renter |
8.4% |
12.6% |
12.9% |
|
Other |
1.4% |
1.4% |
1.4% |
|
Don’t know / prefer not to say / not stated |
2.5% |
2.4% |
- |
|
Rurality |
|
|
|
|
1. Large urban |
26.9% |
38.6% |
38.6% |
|
2. Other urban |
24.2% |
33.4% |
33.4% |
|
3. Accessible small town |
12.8% |
8.3% |
8.3% |
|
4. Remote small towns |
9.8% |
2.8% |
2.8% |
|
5. Accessible rural |
13.8% |
11.4% |
11.5% |
|
6. Remote rural |
12.6% |
5.5% |
5.5% |
|
SIMD |
|
|
|
|
1 (most deprived) |
11.8% |
20.9% |
20.9% |
|
2 |
16.3% |
20.3% |
20.3% |
|
3 |
23.3% |
20.1% |
20.1% |
|
4 |
23.3% |
20.4% |
20.4% |
|
5 (least deprived) |
24.5% |
18.4% |
18.4% |
[1] National Records of Scotland (NRS) Mid-2023 Population Estimates of Scotland : https://www.nrscotland.gov.uk/publications/mid-2023-population-estimates/; Scotland Census 2022 : https://www.scotlandscensus.gov.uk/2022-results/
[2] For more details of the two different bases, see the Data validation and management chapter.
Contact
Email: emily.creamer@gov.scot