Annex A - Weighting
As two members per household are allowed to participate in the Knowledge Panel, a design weight is added to correct for unequal probabilities of selection of household members. Calibration weights are also applied using the latest population statistics relevant to the surveyed population to correct for imbalances in the achieved sample. The calibration weights were applied in two stages:
the first set of variables used the ONS 2019 mid-year population estimates as the weighting targets to create an interlocked variable of gender by age and region;
the second set of variables used the ONS 2019 mid-year population estimates and the ONS Annual Population Survey as the weighting targets for education, ethnicity, Index of Multiple Deprivation, and number of adults in the household.
Table 1: Weighting targets for age and gender
| Age & gender |
Male |
Female |
In another way |
Prefer not to say |
| 18-34 |
13.7% |
13.6% |
0.3% |
0.1% |
| 35-44 |
7.4% |
7.7% |
0.1% |
0.0% |
| 45-54 |
8.3% |
8.8% |
0.0% |
0.0% |
| 55-64 |
8.0% |
8.5% |
0.1% |
0.0% |
| 65-74 |
6.2% |
6.8% |
0.1% |
0.0% |
| 75+ |
4.3% |
6.1% |
0.0% |
0.0% |
Table 2: Weighting targets for Scotland's regions
| Scotland – region |
| Central Scotland |
12.1% |
| Glasgow |
13.1% |
| Highlands and Islands |
8.3% |
| Lothian |
14.6% |
| Mid Scotland and Fife |
12.3% |
| North East Scotland |
14.1% |
| South Scotland |
12.6% |
| West Scotland |
12.9% |
Table 3: Weighting targets for ethnicity
| Ethnicity |
| White |
94.4% |
| Non-White |
4.7% |
| Prefer not to say/Not stated |
0.8% |
Table 4: Weighting targets for education
| Education |
| Degree level or above |
28.5% |
| Below degree level |
70.7% |
| Prefer not to say/Not stated |
0.8% |
Table 5: Weighting targets for
SIMD quintiles
| SIMD quintiles |
| 1 |
20.0% |
| 2 |
20.0% |
| 3 |
20.0% |
| 4 |
20.0% |
| 5 |
20.0% |
Table 6: Weighting targets for number of adults in the household
| Number of adults in the household |
| One adult |
21.7% |
| Two or more adults |
78.3% |
