General Lifestyle Survey
(formerly called the General Household Survey)
The General Lifestyle Survey (in its previous incarnation as the General Household Survey) used to be one of the most used government surveys. However, with new, larger surveys introduced over the last decade, particularly the Family Resources Survey, it is now used much less often. In effect, its main uses are when Great Britain-wide results are required and none of the other surveys have the required data. This means that its uses are largely limited to analyses by:
- Limiting long-standing illness or disability.
- Available from: UK data archive.
- Registration required: yes.
- First survey available: 1971.
- Frequency: annual.
- Updated: May.
- Scope: Great Britain-wide (i.e. does not include Northern Ireland).
- Format: SPSS, STATA or TAB.
- Files: two files per year, one at the individual level and the other at the household level.
- Documentation: comprehensive.
- Weighted or unweighted: weighted from 1998/99 onwards but unweighted for earlier years.
- Household income data: yes, equivalised.
Note that, prior to 2008, the equivalent survey was called the General Household Survey (GHS).
Also note that:
- Prior to 2005, each dataset covered a financial year rather than a calendar year.
- There was no survey in either 1997/98 or 1999/00.
In addition to the dataset itself, ONS publishes an annual report giving many statistical analyses from the dataset.
Which software to use
As the annual dataset is around 30,000 records for individuals and 10,000 records for households, it can be exported into Excel.
When to use the individual and household datasets
Use the household dataset when the desired analysis relates to households. When the desired analysis relates to individuals use either the individual dataset and/or the household dataset depending on which has the variables that are required. If the required variables are in the household dataset then the analysis can be undertaken in either of two ways:
- by linking the two datasets together using the household serial number as the key; or
- by using the household dataset but multiplying the household weights by the number of people in the household to get individual weights.
Note that children have a null weight in the individual dataset so any analyses at the individual-level are actually analyses for adults only.
Because the dataset is for Great Britain only, the graphs below are also for Great Britain only.
|Working-age limiting long-standing illness or disability||individual||first three||Use ONS mid-year population estimates as necessary to translate proportions into absolute numbers.|
|fourth||Data actually obtained from table 7.5 of the ONS annual report rather than from the dataset itself|
|Older people limiting long-standing illness or disability||individual||first three||Use ONS mid-year population estimates as necessary to translate proportions into absolute numbers.|
Data actually obtained from table 7.5 of the ONS annual report rather than from the dataset itself
|Overcrowding||household||first||For the individual-level analysis, multiply the household weight by the number of people in the household.|