|Publication date:||24 January 2020|
|Who this is for:||Users and producers of statistics|
|Type:||Harmonisation guidance and principles|
Harmonisation is the process of making statistics and data more comparable, consistent and coherent. This harmonised principle sets out how to collect and report statistics to ensure comparability across different data collections in the Government Statistical Service (GSS). This produces more useful statistics that give users a greater level of understanding about this topic.
Unpaid care is a topic of interest across government. The 2021 Census in all four countries of the United Kingdom will collect data on this.
Questions on this topic aim to capture information on unpaid caring activities. These are neither funded by individuals nor provided by government or associated agencies. Care is provided because the person being cared for has needs due to physical or mental health condition(s) or illness(es), or problems related to old age. Different organisations may have varying definitions to be considered in question design.
Adult social care and unpaid care are currently being investigated in more detail by government agencies. Examples of these are given in the further links section. The work of the GSS harmonisation team complements this and it is an area that the team will be working on over the next year.
The harmonised question(s) on unpaid care are designed to collect basic information, for use in the majority of surveys. They are not designed to replace questions used in specialist surveys where more detailed analysis is required
If it meets your needs, we recommend that you use the question from the 2019 Census Dress Rehearsal in England and Wales. This question has been fully tested and Census data for England and Wales will be available for comparison. It is not anticipated that there will be changes to this question after the Census rehearsal.
|Question used in the 2019 Census Dress Rehearsal in England and Wales||Response options|
|Do you look after, or give any help or support to, anyone because they have long-term physical or mental health conditions or illnesses, or problems related to old age?
Exclude anything you do as part of your paid employment
Yes, 9 hours a week or less
Yes, 10 to 19 hours a week
Yes, 20 to 34 hours a week
Yes, 35 to 49 hours a week
Yes, 50 or more hours a week
When using the England and Wales Census question, you should bear in mind that:
- It is one combined question asking both if a person cares, and hours of care together
- Advice on what should be thought of as care is given in the accompanying notes, not the questions
- It does not specify who is being cared for e.g. a specific group of people, or only for people living with you
- It is for care not undertaken as part of paid employment
- It does not include an option for the hours of care varying
If you are based in Scotland or Northern Ireland, then you may wish to harmonise with the Census question in that location. In Scotland, in addition to the Census, there is also a question on unpaid care in the Scottish Surveys Core Questions set. This is asked on the Scottish Health Survey, The Scottish Household Survey and the Scottish Crime and Justice Survey.
Northern Ireland have proposed a question with the same question stem as England and Wales. Their first hours category is Yes, 19 hours a week or less, which combines the first two categories of the England and Wales questions. This is not available on the website.
Scotland’s Census have recommended that their proposed question will not change from the 2011 Census.
In all four nations census questions need parliamentary approval. This will be sought during 2020. It is not anticipated that any of the questions will change on unpaid care.
The team will continue to update this page as more results from the Census tests are published.
This is for self-completion surveys. The GSS harmonisation team are proposing to develop an equivalent question for interviewer-led data collection.
No consistent question on unpaid care is currently used across the UK. If it meets your needs, we recommend that you use the question from the 2019 Census Dress Rehearsal in England and Wales. This question has been fully tested and Census data for England and Wales will be available for comparison. It is not anticipated that there will be changes to this question after the Census rehearsal.
A review in 2019 identified 23 surveys that collected information on unpaid care. These were:
- Adult Psychiatric Morbidity Survey: Survey of Mental Health and Wellbeing – England (2014)
- Census – England and Wales (2011)
- Census – Northern Ireland (2011)
- Census – Scotland (2011)
- Community Life Survey (2016-17)
- European Quality of Life Survey (2016)
- Family Resources Survey – Great Britain (2017-18)
- Family Resources Survey – Northern Ireland (2017-18)
- Health Survey for England (2017)
- Health Survey Northern Ireland (2017/18)
- National Survey for Wales (April 2019 to March 2020)
- Northern Ireland Life and Times Survey (2015)
- Northern Ireland Young Life and Times Survey (2015)
- Personal Social Services Survey of Adult Carers in England (2016-17)
- Poverty and Social Exclusion Living Standards Survey (2012)
- Scottish Crime and Justice Survey (2017-18)
- Scottish Health and Care Experience Survey (2017/18)
- Scottish Health Survey (2016)
- Survey of Carers in Households (2009-10)
- The lives of young carers in England (Omnibus, one-off – January 2017)
- Understanding Society (Wave 11 (2019-22)
- Wealth and Assets Survey (Wave 2: July 08-June 10)
- Workplace Employee Relations Survey (2011)
There are multiple factors to consider when collecting data on this topic. This means that there are a variety of ways of wording and structuring a question. As such, no consistent question is currently used across these surveys. This makes comparability across data sources difficult; comparing data from multiple sources should be done with caution.
The GSS Harmonisation Team are keen to understand the reasons each survey uses the question(s) they do and will be contacting survey managers. The team are working in this area to help data users to be able to more easily compare data sources.
When using data on this topic, analysts should consider the following factors when assessing comparability between data sources:
|Question structure||There are generally two approaches to this:
1. Two part – asking if a person is a carer and then asking hours or care for those who say yes
This can be more of a consideration in collecting, rather than analysing data.
|Direct or indirect question||1. Direct question - does the survey ask directly if a person considers themselves a carer?
2. Indirect question - does it specify a list of activities and ask if a person does those without using the word ‘carer’– e.g. so the survey designer sets what is considered a caring activity. How were those activities shown to respondents – e.g. showcard?
|Demographics of respondents||Who is asked about caring and hours of care? All people; adults; younger people (Northern Ireland Young Life and Times Survey); an existing sample of carers (Survey of Carers in Households)?|
|Number of people cared for on whom data collected||Surveys can ask for information about different numbers of people. From asking about the primary person that is cared for only (Personal Social Services Survey of Adult Carers in England), to a specific number of people (Health Survey for England); to any care provided.|
|Does the person cared for need care, based on an assessment?||There may be a distinction between whether someone cares for a person who meets the needs assessment to receive social care; or not. This assessment is done by the Local Authority in Great Britain and the Health and Social Care Trust in Northern Ireland.|
|Residence of caring||Does the data consider where the people live who are caring and being cared for? For example, for people inside the household, or outside the household, or both?|
|Relationship between carer and those receiving care||Data can be collected where care is given to family, friends, neighbours or others. In some cases, these groups are specified; in others the question is open.|
|Regularity of care||Some survey questions specify “regular” care (Understanding Society)|
|Inclusions and Exclusions||For example, the 2011 Census question specified not to include any care given as part of paid employment - this is not explicit in all surveys; others specify to include travel time when thinking about time spent caring (Health Survey Northern Ireland).|
|Number of hours spent caring||This is collected by either free text or a list of options. This options list varies between surveys. With some collected from 1 to X hours and others 0 to X hours as a base category. The highest category also varies from 35 hours or more to 100 hours or more.|
|Varying hours of care||This is collected by either free text or a list of options. This options list varies between surveys. With some collected from 1 to X hours and others 0 to X hours as a base category. The highest category also varies from 35 hours or more to 100 hours or more.|
|Time to consider when thinking about caring||Surveys can ask ‘in the last month’ or ‘in the last week’ ahead of the question|
We are always interested in hearing from users so we can develop our work. If you use or produce statistics based on this topic, get in touch: firstname.lastname@example.org.
This page will be reviewed annually.
|24 January 2020||Harmonised principle updated in January 2020 to match the proposed census 2021 question. The previous principle was not widely adopted. We also removed the need for a person to be identified as disabled.|