The Family Resources Survey (FRS) is a continuous household survey which collects information on a representative sample of private households in the UK. Detailed information is recorded on respondents’ income from all sources:
- housing tenure
- caring needs and responsibilities
- spending on housing
- pension scheme participation
- family circumstances
- child maintenance
The survey is conducted in respondents’ homes. In the latest 3-year period, 2014/15 to 2016/17, the FRS sample consisted of just over 58,000 households in the UK. It has a financial-year survey period with surveys conducted throughout the year and is cross-sectional (a ‘snapshot’ of households over the year). Individuals are not followed up for later surveys.
Estimates are subject to sampling error and non-sampling bias. The FRS only covers private households. Therefore, individuals who live in communal accommodation like care homes or have no fixed address (for example, who are homeless) are not included in these results.
Households where every resident aged 16 years and over responds to the questions are classed as fully co-operating. Only households with full co-operation are included in the analysis. This may introduce some error, as the characteristics of individuals living in households which do not respond fully may be different to those who fully co-operate. Therefore, the sample will be more representative of those who do respond fully. Overall, 54% of households fully co-operated with the survey, meaning there was a sample size of 19,387 households in 2016/17.
Results derived from a low number of responses are more likely to be affected by statistical variation, so observed changes might not reflect real differences. As such, caution is needed when interpreting short-term trends in the data, especially for subgroups (for example, a specific ethnic group, age group and gender). Using a 3-year average for income minimises the risks due to uncertainty.
As the data is presented as a 3-year average, no statistical tests have been carried out to determine whether the estimates taken from the survey are statistically significantly different from one another.
In other words, as the results are based on a sample of the population, without further testing it isn’t possible to determine whether any differences observed would likely be seen across the entire population. This means that conclusions about differences between groups cannot be drawn. However, the data is still useful because it indicates where differences between ethnic groups might exist.
Weighting is used to adjust the results of a survey to make them representative of the population and improve their accuracy.
For example, a survey which contains 25% women and 75% men will not accurately reflect the views of the general population, which we know has an even 50/50 split.
Statisticians rebalance or ‘weight’ the survey results to more accurately represent the general population. This helps to make them more reliable.
Survey weights are usually applied to make sure the survey sample has broadly the same gender, age, ethnic and geographic make up as the general population.’
Suppression rules and disclosure control
Any values based on fewer than 100 responses have been suppressed.
Percentages are given to the nearest whole percentage point. Due to this rounding, some figures may not add up to 100.
Family Resources Survey with previous years' data and background informationQuality and methodology information