Scottish Household Survey (SHS)
SHS is substantial general household survey for Scotland. As such, it can be used for a wide variety of analyses of household circumstances and views in Scotland.
It also surveys a random adult in each household and, as such, can be used for analyses of individual circumstances and views in Scotland.
- Available from: UK data archive.
- Registration required: yes.
- First survey available: 1999.
- Frequency: annual.
- Updated: December.
- Scope: Scotland.
- Format: SPSS, STATA or TAB.
- Files: 5 files per year but the only one that is substantive is the household-level file.
- Documentation: comprehensive.
- Weighted or unweighted: weighted.
- Household income data: yes but unequivalised only.
- Although the SHS dataset is published on an annual basis, the files for a particular year sometimes also contain the records for the previous year.
- Although there is no individual-level file, each record in the household-level file contains some information about a 'random adult' in that household.
Which software to use
As the annual dataset is around 15,000 records, it can be exported into Excel.
What weights to use
There are two weights: a household weight and a weight for the random adults who were interviewed and care should be taken to choose the right one for the analysis required. If the variables to be analysed relate to the household, use the household weight but if the variables to be analysed relate to individual adults then use the random adult weight. Furthermore, if the analysis is using the random adult weight, then those records with a null weight should be excluded.
What topics to analyse
Note that not all questions are asked every year. So, for example, the questions about involvement in the community have not been asked since 2002 and the questions about worries about crime were not asked prior to 2002.
|Longstanding illness or disability||first||random adults|
|Without home contents insurance||all||households|
|Access to transport||all||random adults|
|Access to essential services||all||random adults|
|Dissatisfaction with public services||all||random adults||Question not asked in 2003 or 2004.|
|Dissatisfaction with local area||all||random adults|
|Burglaries||all bar first two||random adults|
|Feeling unsafe out at night||all||random adults|
|Working in a voluntary capacity||all||random adults|
- For those analyses which use the random adult weight, exclude those with a null weight.
- For those analyses which are by household income, allocate each record to a household income quintile, calculating the quintile thresholds required to achieve this.