Integrated Household Survey Frequently Asked Questions
What is the Integrated Household Survey (IHS)?
The Integrated Household Survey (IHS) is a composite survey which brings together in a single dataset some of the data from a number of Office for National Statistics (ONS) social surveys (called “modules”). The aim of the IHS is to produce high-level estimates for particular themes to a higher precision and lower geographic level than current ONS social surveys. The first dataset available is for the year April 2009 - March 2010.
Which surveys (or “modules”) are included in the IHS?
Current surveys (called “modules”) included in the IHS are:
- Labour Force Survey (LFS) and associated boosts (APS)
- General Lifestyle Survey (GLF) (formerly the General Household Survey)
- Living Costs and Food Survey (LCF) (formerly the Expenditure and Food Survey)
- English Housing Survey (EHS)
- Life Opportunities Survey (LOS)
The Opinions survey (OPN) (formerly called the Omnibus survey) was initially included in the IHS but was dropped in January 2010; the OPN is entirely financed by sponsors and including it in the IHS reduced the space for questions from sponsors.
Why do we need an Integrated Household Survey if these data are included in other data sets already available?
The IHS contains only some of the variables that are included in its constituent modules; the IHS in 2009-10 (End User licence version) contains 125 variables. The IHS aims to produce high-level estimates for particular themes to a higher precision and lower geographic level than current ONS social surveys. It has a sample size of approximately 450,000 individuals from interviews undertaken in Great Britain and Northern Ireland. This large sample size and UK coverage means various geographical breakdowns are possible in the IHS and it is possible to use a geographical hierarchy to drill down to lower level detail within an area.
How does the IHS link data from different surveys?
Answers to Core IHS questions from different modules are presented in a single variable in the IHS. The Core questions within the IHS are designed to facilitate the combining of the same question on different ‘modules’ of the IHS. Samples for each IHS module are chosen independently.
Can users identify from which module (the component surveys) the data came?
No, because weights in the IHS are designed to be used with the whole of the IHS and do not allow analysis within component surveys.
Is it possible to link IHS data with that of some of the component surveys?
Not using the published IHS data. Researchers with a strong research case for linking IHS data to another data set should contact the ONS.
Does the IHS contain panel data?
Which years of the IHS are available?
The first year of data available is for 2009-2010. The IHS is an ongoing continuous survey. Annual datasets will be released every quarter on a rolling basis. For an up-to-date list of available datasets please refer to the Dataset Titles page.
A pilot study for the IHS was conducted in 2008-2009. It included 3 modules initially in January 2008 and a fourth was added in April 2008. The IHS data for this period was not published as it does not provide better quality than existing surveys except for smoking prevalence.
What population is covered in the IHS?
The IHS covers the UK including Northern Ireland. Note that different modules have different geographical coverage:
- The LFS and associated APS cover the whole of the UK
- The GLF, LCF, LOS and OPN each cover Great Britain (they exclude Northern Ireland)
- The EHS covers England only
What format are the data available in?
The IHS is available in SPSS, STATA and ASCII tab-delimited formats.
Does the IHS include information at the person level?
Yes. Information is held at the person level.
Are children included in the IHS?
Children are not interviewed in the IHS but adults are asked to give proxy responses to questions on behalf children in their household for some but not all questions.
What is the most detailed geography I can analyse the data at?
The data are available at the level of Government Office Region in the End User Licence version of this dataset.
More detailed geographic variables are available in the Special Licence version. Variables include county, unitary/local authority, Nomenclature of Territorial Units for Statistics 2 (NUTS2) and NUTS3 regions and Training and Enterprise Councils (TECs). Users should note that the user guide also mentions variables that are not included in either the EUL or SL datasets held at the Archive.
Before I order, how do I find out what questions/variables are included?
Variable lists and PDF user guides (including questionnaires) are freely available via the Doc column in the Dataset Titles page. A variable search tool is also available which allows users to find out whether a specific topic or question is covered in each of the IHS surveys.
Will household income be present in the IHS?
Not at present as it could not be harmonised with the LFS.
Where do the questions about happiness appear in the IHS, and what has been done to ensure their validity?
There are four questions about happiness. They appear in a block of subjective questions near the start of the questionnaire, after a question about sexual identity but before questions about employment. The ONS has consulted with academics about which questions about happiness to use. As data are experimental at the moment, the ONS are open to discussion about all methodological issues.
Do I need to use weights to analyse the IHS?
Yes. The household weight is called HHWT092. NPWT092 is a non-proxy weight provided solely for analysis of the sexual identity question on the IHS. For further information about weights in the IHS see the User Guide.
How reliable are the weights in the IHS at this stage (as at Spring 2011)?
The process of weighting the IHS is ongoing. More comparison of estimates needs to be done using other social surveys.
Can I use the IHS to analyse change over time?
There is currently is only a single year of data available (for 2009-2010). When future data sets become available, these data sets will provide a series of repeated cross-sectional surveys which may be used to analyse change over time.
Is there a more general FAQ?
There is a generic FAQ for all surveys available.