Integrated Household Survey (Experimental statistics): January to December 2014

Integrated Household Survey

The Integrated Household Survey (IHS) was previously formed from "core" questions asked by a number of our household surveys. This year the IHS is based solely upon the Annual Population Survey (APS), following the removal of IHS questions on the Living Costs and Food (LCF) survey in January 2014.

Data for this bulletin were obtained from the APS for the survey months January to December 2014.

These data have been prepared using the established IHS approach of imputing and calculating weights using IHS specific methodologies. This will be the last independent IHS dataset to be released.

IHS variables will be added to the APS dataset from January – December 2015. For this dataset IHS variables will be subjected to the APS processing methodology.

The 2015 dataset, to be released in 2016, will be the first APS dataset to include IHS variables. Further information on this change can be obtained by emailing socialsurveys@ons.gov.uk

Further information on the IHS can be found on the ONS website.

1.1 Consultations

We have conducted a series of consultations with users about main decisions relating to the IHS. The outcome of each consultation can be found in the links:

• Proposal to reduce the number of datasets produced using the IHS

• Proposal to rotate the Sexual Identity questions on the IHS

• Proposal to stop IHS core questions in the Living Costs and Food Survey

Notes on the themes

2.1 Sexual identity

The IHS asks a question on self-perceived sexual identity of adults in the UK. This question was developed as part of the ONS Sexual Identity Project, which developed, tested and implemented the question to satisfy the increasing demand for data on sexual orientation to meet the Equality Act 2010.

While the Equality Act, and relevant legislation in the past, referred to sexual orientation, we have focused on collecting data on sexual identity. The sexual identity question measures how people see themselves and not how others see them.

The question on sexual identity was developed and tested on a number of surveys in 2008 and was added to the IHS in 2009. The data have been collected to provide accurate statistics to underpin the equality monitoring responsibilities of public sector organisations and to assess the disadvantage or relative discrimination experienced by the lesbian, gay and bisexual population.

The sexual identity question was asked of respondents aged 16 years and over; it was not asked by proxy. Proxy interviews are defined as those where answers are supplied by a third party, who is usually a member of the respondent’s household.

The sexual identity question was asked in both face-to-face and telephone interviews, at first personal contact. During the face to face interviews, adults were asked; "Which of the options on this show card best describes how you think of yourself?" For telephone interviews, a slightly different way of collecting the information was used: "I will now read out a list of terms that people use to describe how they think of themselves". The list was read out to respondents twice. On the second reading, the respondent had to say "stop" when an appropriate term they identified with was read out. In both modes, the order in which the terms appeared, or were read out, was unique for each household’s respondent to ensure confidentiality.

The "Other" option on the question was included to address the fact that not all people will consider they fall in the first 3 categories, that is, heterosexual/straight, gay/lesbian or bisexual.

Processing of sexual identity data

Due to the changing composition of surveys forming the IHS between 2009 and 2014, a series of minor changes were applied to the processing of sexual identity responses from the APS (in January 2012 and January 2014). These changes were applied to the rotation of previous responses to the sexual identity question. These have led to minor differences in the numbers of non-responses recorded in the APS compared with previous survey estimates.

Non-response and sexual identity

The voluntary nature of the survey means that people who do not wish to take part in the survey can refuse to do so. Although every effort is made to minimise the impact of non-response to the IHS component surveys, non-response can appear in several stages of the survey:

• household non-response: where our interviewers are unable to obtain an interview with anyone in the household

• unit non-response: where one or member of a household is not available to take part in our interview

• item non-response: where an individual who has agreed to take part in our survey does not provide an answer to one or more question (but takes part in the majority of the survey)

For the sexual identity question, there are two main sources of non-response which are reported in our estimates:

• non-responders: these are adults who are part of an eligible household who were not present when the sexual identity question was asked, and therefore are classed as non-responders

• don’t know or refusals: these are adults who have taken part in our survey who have declined to provide a valid response to the sexual identity question

Our initial analysis has suggested the characteristics of eligible responders and non-responders to the sexual identity question are broadly similar. Therefore the non-response to this question should not lead to a significant bias in our estimates (with the distribution of the main characteristics of non-responders broadly following the distribution of our known population). To provide a transparent set of results, non-responders are included as a category in our results.

2.2 Smoking prevalence

The IHS collects information on adult smokers in the UK, asking questions of adults aged 18 and over. This question has been collected and published on the IHS since 2010. The Opinions and Lifestyle survey also collects information of cigarette smoking and e-cigarette use in Great Britain. These questions are asked of adults aged 16 and over. More information can be found on the Opinions and Lifestyle survey page on our website.

Personal well-being

As part of the Measuring National Well-being programme, in April 2011 we added what are termed "personal well-being" questions to the constituent surveys of the IHS. Personal well-being concerns a person’s self-reported assessment of their own well-being. Survey questions of this nature aim to capture a person’s well-being by measuring how people think and feel, for example, asking about life satisfaction, meaning and purpose, and day-to-day emotions. Used alongside more objective statistics, they provide a fuller assessment of national well-being.

In September 2015, we produced a fourth statistical bulletin measuring National Well-being in the UK, the largest constituent survey of the IHS.

Other sources of data

Information on some of the topics covered in this bulletin can also be found from other sources. These sources include:

• Scotland: Scottish Health Survey and the Scottish Household Survey

• Wales: Welsh Health Survey and the StatsWales website

• Northern Ireland: Health Survey Northern Ireland

• England: Health Survey for England and the Local Tobacco Control Profiles website

Access to the data

IHS data are made available through the End User Licence at the UK Data Service. The End User Licence is an agreement between the user and the University of Essex to provide users with the right to use the data held at the Data Service. The End User Licence dataset can be accessed through the UK Data Service (UKDS) website.

A more detailed file can be accessed if Approved Researcher Status is sought and agreed. More information on the UK Data Service and the Approved Researcher process can be found on the UK Data Service website.

Pre-release access to the IHS bulletin

The phrase "Pre-release Access" refers to the practice whereby Official Statistics in their final form, and any accompanying written commentary, are made available to an eligible person in advance of their publication. The rules and principles which govern pre-release access are featured within the Pre-release Access to Official Statistics Order 2008.

A list of the people who have pre-release access to this bulletin is available on our website.

Experimental Statistics

Experimental Statistics are those which are in the testing phase, are not yet fully developed and have not been submitted for assessment to the UK Statistics Authority. They are published in order to involve customers and stakeholders in their development, and as a means to build in quality at an early stage. The IHS is designated as Experimental Statistics.

Further information on Experimental Statistics can be found on our website.

Quality

Detailed information relating to quality components of the IHS (including sampling variability, confidence intervals, and detailed methodological information) is in the following documents:

• Quality and Methodology Information

• IHS User guidance

These documents are on the IHS web page.

Confidence intervals and statistical significance

Confidence intervals

The IHS publishes information on confidence intervals to give an indication of the range in which the true population value is likely to fall.

For the IHS, confidence intervals are calculated at the 95% level. This means that 19 times out of 20, the true value of the population would fall within the upper and lower boundaries. For example, a confidence interval of ±0.2% means that an estimate of 93.9% would range between 93.7% and 94.1% (19 times out of 20).

Information on confidence intervals can be found in some figures in this report, and in the associated tables available to download.

Statistical significance

Throughout this report, the changes reported are deemed statistically significant. This means that we are confident that 19 times out of 20, differences will be down to real changes in our population, rather than as a result of sampling variability.

In instances where changes between estimates are not statistically significant, these have been highlighted in the bulletin text, or not commented on in this release.

Weighting to national population totals

Population weighting serves 2 purposes:

• it enables sample estimates to sum to known population estimates

• it compensates for differential non-response among different sub-groups in the population

Each sampled case is given a weight which can be thought of as the number of people that case represents in the population

The estimates in this report have been weighted to ensure they are representative of the target population. Two weights have been used in this report, a sexual identity weight for tables in the sexual identity section and a household weight has been used for tables in the smoking section. Further information on the weighting strategy can be found on the IHS web page.

Reference tables

A series of reference tables have been included in this bulletin. These provide background data for each figure in this report, broken down by country. Figures covering main demographic characteristics of age and gender also provide a time series of estimates between 2010 and 2014. Confidence intervals have been provided within these reference tables for the most recent year of data (2014). These can be found in the data section of the publication.

Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: media.relations@ons.gov.uk