Information

Scottish Household Survey 2020: methodology and impact of change in mode

The methodology report for the Scottish Household Survey 2020 telephone survey which discusses the impact of the change in mode.

This document is part of 2 collections


Appendix 3: Seasonal effects

The 2020 face-to-face fieldwork was undertaken between January and 16 March 2020. The 2020 push-to-telephone/video fieldwork took place during the months of October 2020 and January to March 2021, whereas SHS face-to-face surveys normally run throughout the year.

The SHS is designed to provide results that are representative on an annual basis and not on a quarterly basis.

This appendix looks at the level of fluctuation in key SHS estimates on a quarterly basis. These fluctuations are due to three different drivers:

  • Sampling error and the natural propensity for estimates from survey samples to vary.
  • Fieldwork practicalities. Once the sample is drawn, all addresses are batched into workable assignments and scheduled for a particular month. This means that for each quarter the addresses worked are not necessarily representative of Scotland as a whole. While we would expect fieldwork in each local authority to be carried out throughout the year, it is possible that for any one quarter, the addresses will be clustered in parts of the council area. Additionally, weighting is carried out only on an annual basis and not on a quarterly basis.
  • Real seasonal effects, for example in employment rates.

A range of geographical, household and random adult measures from the 2019 SHS (or the 2018 SHS for biennial even questions) have been analysed by quarter, and the results are presented in Tables A3.1 to A3.4.

Table A3.1 Geographical measures by quarter and the 95 percent confidence intervals
Quarter 1 Quarter 2 Quarter 3 Quarter 4
Urban/rural indicator
Large Urban 34% ± 2% 35% ± 2% 36% ± 2% 34% ± 2%
Other Urban 38% ± 2% 33% ± 2% 34% ± 2% 39% ± 2%
Accessible Small Towns 10% ± 1% 9% ± 1% 8% ± 1% 8% ± 1%
Remote Small Towns 4% ± 1% 4% ± 1% 4% ± 1% 3% ± 1%
Accessible Rural 11% ± 1% 11% ± 1% 11% ± 1% 11% ± 1%
Remote Rural 4% ± 1% 8% ± 1% 7% ± 1% 5% ± 1%
SIMD Quintile
Most deprived 19% ± 2% 21% ± 2% 22% ± 2% 22% ± 2%
2nd 21% ± 2% 19% ± 2% 22% ± 2% 20% ± 2%
Middle quintile 19% ± 2% 23% ± 2% 19% ± 2% 20% ± 2%
4th 21% ± 2% 19% ± 2% 19% ± 2% 20% ± 2%
Least deprived 20% ± 2% 19% ± 2% 19% ± 2% 18% ± 2%
Table A3.2 Household measures by quarter and the 95 percent confidence intervals
Quarter 1 Quarter 2 Quarter 3 Quarter 4
Tenure
Owner-occupied 62% ± 2% 63% ± 2% 62% ± 2% 59% ± 2%
Social Rented 23% ± 2% 25% ± 2% 23% ± 2% 23% ± 2%
Private Rented 14% ± 2% 11% ± 1% 13% ± 1% 17% ± 2%
Other 1% ± 0% 1% ± 0% 1% ± 0% 1% ± 0%
Length of time at address[61]
Less than a year 12% ± 2% 8% ± 1% 11% ± 1% 14% ± 2%
1-3 years 18% ± 2% 20% ± 2% 20% ± 2% 21% ± 2%
4-15 years 35% ± 2% 35% ± 2% 35% ± 2% 36% ± 2%
Over 15 years 36% ± 2% 36% ± 2% 34% ± 2% 30% ± 2%
Property type
House 66% ± 2% 66% ± 2% 68% ± 2% 63% ± 2%
Flat 34% ± 2% 33% ± 2% 32% ± 2% 37% ± 2%
Other 0% ± 0% 0% ± 0% 1% ± 0% 1% ± 0%
Household type
Single adult 21% ± 2% 20% ± 2% 18% ± 2% 23% ± 2%
Small adult 21% ± 2% 19% ± 2% 20% ± 2% 21% ± 2%
Single parent 4% ± 1% 4% ± 1% 4% ± 1% 5% ± 1%
Small family 13% ± 2% 13% ± 1% 13% ± 1% 13% ± 2%
Large family 5% ± 1% 5% ± 1% 6% ± 1% 5% ± 1%
Large adult 9% ± 1% 8% ± 1% 9% ± 1% 9% ± 1%
Older smaller 13% ± 2% 15% ± 2% 14% ± 1% 13% ± 2%
Single pensioner 14% ± 2% 17% ± 2% 15% ± 1% 12% ± 1%
Household working status
Single working adult 19% ± 2% 19% ± 2% 19% ± 2% 21% ± 2%
Non-working single 26% ± 2% 29% ± 2% 25% ± 2% 25% ± 2%
Working couple 29% ± 2% 28% ± 2% 32% ± 2% 30% ± 2%
Couple, one works 11% ± 1% 10% ± 1% 11% ± 1% 11% ± 1%
Couple, neither work 15% ± 2% 14% ± 2% 13% ± 1% 12% ± 2%
Net annual household income
GBP 0 to GBP 10,000 8% ± 1% 9% ± 1% 8% ± 1% 8% ± 1%
GBP 10,001 to GBP 20,000 27% ± 2% 27% ± 2% 27% ± 2% 26% ± 2%
GBP 20,001 to GBP 30,000 20% ± 2% 22% ± 2% 21% ± 2% 22% ± 2%
GBP 30,001 to GBP 40,000 16% ± 2% 14% ± 2% 16% ± 1% 15% ± 2%
GBP 40,001 and above 28% ± 2% 27% ± 2% 30% ± 2% 28% ± 2%
Whether struggling financially
Struggling financially 9% ± 1% 9% ± 1% 9% ± 1% 9% ± 1%
Satisfaction with housing
Very/fairly satisfied 91% ± 2% 92% ± 2% 89% ± 2% 89% ± 3%

Table A3.3 HIH measures by quarter and the 95 percent confidence intervals

Quarter 1 Quarter 2 Quarter 3 Quarter 4
HIH Banded age
16-24 5% ± 1% 4% ± 1% 4% ± 1% 6% ± 1%
25-44 30% ± 2% 28% ± 2% 31% ± 2% 33% ± 2%
45-59 30% ± 2% 28% ± 2% 28% ± 2% 29% ± 2%
60+ 36% ± 2% 40% ± 2% 37% ± 2% 32% ± 2%
HIH Gender
Man/Boy 56% ± 2% 57% ± 2% 58% ± 2% 59% ± 2%
Woman/Girl 44% ± 2% 43% ± 2% 42% ± 2% 41% ± 2%
HIH Economic status
Self employed 7% ± 1% 7% ± 1% 8% ± 1% 8% ± 1%
Employed full time 44% ± 2% 42% ± 2% 45% ± 2% 45% ± 2%
Employed part time 6% ± 1% 7% ± 1% 7% ± 1% 8% ± 1%
Looking after the home/family 2% ± 1% 2% ± 1% 2% ± 1% 2% ± 1%
Retired from work 29% ± 2% 32% ± 2% 28% ± 2% 24% ± 2%
Unemployed 3% ± 1% 2% ± 1% 2% ± 1% 3% ± 1%
In further/higher education 3% ± 1% 2% ± 1% 2% ± 1% 4% ± 1%
Permanently sick or disabled 5% ± 1% 5% ± 1% 5% ± 1% 5% ± 1%
Short-term illness or injury 1% ± 1% 1% ± 0% 0% ± 0% 1% ± 1%
Table A3.4 Random adult measures by quarter and the 95 percent confidence intervals
Quarter 1 Quarter 2 Quarter 3 Quarter 4
Banded age
16-24 10% ± 1% 10% ± 1% 11% ± 1% 13% ± 2%
25-44 34% ± 2% 31% ± 2% 34% ± 2% 33% ± 2%
45-59 26% ± 2% 25% ± 2% 24% ± 2% 27% ± 2%
60+ 30% ± 2% 34% ± 2% 31% ± 2% 27% ± 2%
Gender
Man/Boy 49% ± 2% 46% ± 2% 49% ± 2% 49% ± 2%
Woman/Girl 51% ± 2% 54% ± 2% 51% ± 2% 51% ± 2%
Ethnicity
White Scottish/British 89% ± 1% 90% ± 1% 88% ± 1% 87% ± 2%
White other[62] 7% ± 1% 7% ± 1% 7% ± 1% 7% ± 1%
Minority ethnic groups[63] 4% ± 1% 3% ± 1% 5% ± 1% 5% ± 1%
Highest educational attainment
None 15% ± 2% 16% ± 2% 16% ± 2% 15% ± 2%
Level 1 - O grade etc 19% ± 2% 17% ± 2% 17% ± 2% 16% ± 2%
Level 2 - Higher, A 16% ± 2% 16% ± 2% 17% ± 2% 18% ± 2%
Level 3 - HNC/HND 14% ± 2% 12% ± 2% 12% ± 1% 14% ± 2%
Degree or prof qual 31% ± 2% 32% ± 2% 33% ± 2% 32% ± 2%
Other qualification 5% ± 1% 6% ± 1% 5% ± 1% 5% ± 1%
General health
General health bad or very bad 8% ± 1% 9% ± 1% 8% ± 1% 9% ± 1%
Disability
Disabled 25% ± 2% 25% ± 2% 22% ± 2% 26% ± 2%
Non-disabled 75% ± 2% 75% ± 2% 77% ± 2% 73% ± 2%
Greenspace
Within 5 mins of greenspace 66% ± 2% 66% ± 2% 66% ± 2% 65% ± 2%
Personal use of the internet
Used internet for personal use 88% ± 2% 85% ± 2% 87% ± 2% 90% ± 2%
Culture and Heritage
Cultural attendance 81% ± 2% 79% ± 2% 82% ± 2% 82% ± 2%
Cultural participation 74% ± 2% 75% ± 2% 76% ± 2% 75% ± 2%
Cultural engagement 90% ± 1% 90% ± 1% 90% ± 1% 91% ± 1%
Physical Activity and Sport
Participated in sport in last 4 weeks 79% ± 2% 79% ± 2% 81% ± 2% 79% ± 2%
Discrimination and Harassment
Experienced either discrimination or harassment 10% ± 1% 8% ± 1% 9% ± 1% 11% ± 1%
Satisfaction with local services
Satisfied with local health services (excluding no opinion) 82% ± 2% 82% ± 2% 78% ± 2% 78% ± 2%
Satisfied with local schools (excluding no opinion) 72% ± 3% 74% ± 3% 73% ± 3% 74% ± 3%
Satisfied with public transport (excluding no opinion) 68% ± 2% 69% ± 2% 66% ± 2% 70% ± 2%
Satisfied with all three services (no opinion for up to two) 53% ± 2% 55% ± 2% 50% ± 2% 53% ± 2%
Outdoors
One+ visits to the outdoors 58% ± 2% 55% ± 2% 57% ± 2% 55% ± 2%
Social capital
Feels lonely some, most, almost all or all of the time[64] 19% ± 2% 22% ± 2% 22% ± 2% 22% ± 2%
Meets socially at least once a week[65] 72% ± 2% 73% ± 2% 72% ± 2% 74% ± 2%
Volunteering
Volunteered 26% ± 2% 26% ± 2% 27% ± 2% 25% ± 2%
Provided unpaid help to improve their local environment[66] 4% ± 1% 5% ± 1% 5% ± 1% 4% ± 1%
Rating of neighbourhood
Rating of neighbourhood as very good 56% ± 2% 57% ± 2% 58% ± 2% 56% ± 2%
Rating of neighbourhood as fairly good 37% ± 2% 38% ± 2% 36% ± 2% 38% ± 2%
Community belonging
Very/fairly strong feeling on belonging to immediate neighbourhood 76% ± 2% 80% ± 2% 79% ± 2% 77% ± 2%
Agreement with statements about local neighbourhood
If I was alone and needed help, I could rely on someone in this neighbourhood to help me 85% ± 2% 87% ± 2% 86% ± 1% 83% ± 2%
If my home was empty, I could count on someone in this neighbourhood to keep an eye on my home 85% ± 2% 86% ± 2% 86% ± 1% 82% ± 2%
I feel I could turn to someone in this neighbourhood for advice or support 79% ± 2% 80% ± 2% 79% ± 2% 76% ± 2%
In an emergency, I would offer to help people in my neighbourhood who might not be able to cope well 90% ± 1% 91% ± 1% 90% ± 1% 88% ± 2%
This is a neighbourhood where people are kind to each other 83% ± 2% 84% ± 2% 84% ± 2% 81% ± 2%
This is a neighbourhood where most people can be trusted 78% ± 2% 80% ± 2% 79% ± 2% 76% ± 2%
There are welcoming places and opportunities to meet new people 52% ± 2% 53% ± 2% 51% ± 2% 50% ± 2%
There are places where people can meet up and socialize 57% ± 2% 57% ± 2% 58% ± 2% 56% ± 2%
This is a neighbourhood where people from different backgrounds get on well together 69% ± 2% 71% ± 2% 70% ± 2% 66% ± 2%
This is a neighbourhood where local people take action to help improve the neighbourhood 57% ± 2% 58% ± 2% 59% ± 2% 54% ± 2%
I can influence decisions affecting my local area 19% ± 2% 17% ± 2% 18% ± 2% 17% ± 2%

For most measures, the 95 percent confidence intervals for the estimates for each quarter overlapped. However, for some measures this was not the case.

Table A3.1 shows that the weighted proportion of households from remote rural areas that were interviewed in quarters 1 and 4 was lower than in quarters 2 and 3. This is likely due to the fact that, historically, addresses have been batched so that interviews in remote rural areas, which can be difficult to access in winter, are more likely to be conducted at another time of the year.

Tables A3.2 and A3.4 show that for some household and random measures, there is higher representation in quarter 4 for some groups compared to at other times of the year – young adults; adults from minority ethnic groups[67]; private rented sector households; households where the randomly selected adult has lived at the address for less than a year; flats; single adult households; and households where the highest income householder is in further/higher education. This is likely due to the fact that these households and individuals are less likely to participate in the survey. Interviewers have to work harder to convince these households/individuals to take part, it is less likely that the interviews will take place at first issue, and this results in these groups being over-represented in the later part of the year.

Table A3.4 also shows that there are other random adults measures for which there is some evidence of seasonal effects, with no overlap of the 95 percent confidence intervals for at least one quarter compared to another. In quarter 4, personal use of the internet is higher, and agreement with statements on neighbourhood strengths is generally weaker, than at other times of the year. This is unlikely to reflect genuine seasonal effects, and is more likely due to the over-representation of households/individuals who are hard to reach in quarter 4. Generally, these individuals are more likely to be younger and will, therefore, be more likely to use the internet for personal use, and be less positive about the strengths of their local neighbourhood.

For measures where we may have expected genuine seasonal effects (e.g. participation in physical activity and sport in the last 4 weeks), Table A3.4 shows that there is no evidence of seasonal effects. This could be, in part, due to genuine seasonal variations being offset by the over-representation of hard to reach individuals (who are more likely to be young and to participate in physical activity and sport) in quarter 4.

In conclusion, there is some evidence of seasonal effects in the SHS. However, these are unlikely to represent genuine seasonal effects, and are more likely to be due to (i) the batching of the sample to facilitate the fieldwork and (ii) the over-representation of hard to reach groups in the later part of the year. There is no reason to expect that seasonality would be a major factor in the differences between the 2020 survey and previous years. Any genuine seasonal effects are likely to be small in comparison to the other reasons for the differences, e.g. non-response bias and mode effects.

It would be interesting to conduct this analysis on only first issue responses and control for rurality. This is beyond the scope of this report but might be better placed to identify genuine seasonal effects in the SHS.

Contact

Email: shs@gov.scot

Back to top