Measuring national well-being is about looking at ‘GDP and beyond’ to measure what really matters to people. The UK’s Measuring National Well-being programme was launched in 2010 and is based at the Office for National Statistics (ONS). The aim is to produce accepted and trusted measures of the well-being of the nation - how the UK as a whole is doing.
A wide range of 41 different headline measures are monitored regularly in areas such as the natural environment, personal well-being, our relationships, health, what we do, where we live, personal finances, the economy, education and skills, and governance. All of this helps us to measure the progress and well-being of the nation - how the UK as a whole is doing.
ONS published the second ‘Life in the UK’ report in March 2014, giving the latest snapshot of the nation’s well-being. The report was accompanied by updated national well-being measures data, and interactive wheel of measures and interactive graphs which provide regional breakdowns for selected measures. A full list of outputs from the measuring national well-being programme is available here.
Find out more about our approach to measuring national well-being on the measuring national well-being pages of the ONS website.Back to table of contents
Personal well-being is also known by the more scientific term of ‘subjective well-being’. It is about people evaluating their own lives. There are several ways in which this is looked at by researchers who study well-being. These include asking people to evaluate how satisfied they are with their life overall, asking whether they feel they have meaning and purpose in their life, and asking about their emotions during a particular period. The ONS measures of personal well-being ask people to assess each of these aspects of their lives.
Focus groups with members of the public conducted by ONS in 2013 found that the term ‘personal well-being’ is clearer and simpler to understand than ‘subjective well-being’. In the light of this, both the questions and findings from them have been referred to by ONS as ‘personal well-being’ since then.Back to table of contents
Personal well-being data can contribute to better decision making, for policy makers, individuals, communities, business and civil society. It complements other measures of progress and quality of life such as unemployment and household income and is used by the Office for National Statistics to better monitor and understand national well-being.
The uses of personal well-being data are varied, but four main uses have been identified including:
overall monitoring of national well-being,
use in the policy making process which includes:
- identifying needs and targeting policies
- policy appraisal
- evaluation and monitoring of policy
public decision making.
Although the Measuring National Well-being programme at ONS was only launched in late 2010, the data and measures from the programme are increasingly used across government departments as well as the private sector and third sector organisations.
For further details of a wide range of well-being work across UK government departments, please go to ‘Wellbeing: Policy and Analysis’.
The Office for National Statistics is also working with other national statistical offices and international organisations such as the Organisation for Economic Cooperation and Development (OECD) and Eurostat to develop comparable international measures of personal well-being. An example of this is a recent publication by the OECD providing guidance to national statistical offices on the measurement of subjective well-being. This included many examples from the ONS Measuring National Well-being programme.Back to table of contents
There is no difference between personal well-being and subjective well-being. Focus groups with members of the public conducted by ONS in 2013 found that ‘personal well-being’ is clearer and simpler to understand than ‘subjective well-being’. In the light of this, both the questions and findings from them have been referred to by ONS as ‘personal well-being’ since then.Back to table of contents
Personal well-being is one of many ways in which the Measuring National Well-being programme at ONS aims to assess the progress of the nation, looking both at standard objective measures such as income and health and at people’s own views about their lives. Monitoring personal well-being across the nation year on year will help to show how people feel their quality of life changes in relation to changes in circumstances, policies and wider events in society. It will also show how people in the UK evaluate their lives compared to people in other countries across the world.Back to table of contents
The personal well-being domain includes the four ONS personal well-being questions on life satisfaction, worthwhile, happiness and anxiety. In addition, the domain also includes a ‘population mental well-being’ measure. This measure uses the Warwick-Edinburgh Mental Well-being Scale (WEMWBS) to assess mental well-being out of a total possible score of 35. The data source for WEMWBS data is Understanding Society. The WEMWBS data is provided alongside the other four personal well-being domain measures as part of the Measuring National Well-being, ‘Life in the UK’ releases.Back to table of contents
ONS uses four survey questions to measure personal well-being. People are asked to respond to the questions on a scale from 0 to 10 where 0 is ‘not at all’ and 10 is ‘completely’. The four questions are:
“Overall, how satisfied are you with your life nowadays?”
“Overall, to what extent do you feel the things you do in your life are worthwhile?”
“Overall, how happy did you feel yesterday?”
“Overall, how anxious did you feel yesterday?”
ONS first added these four questions to the Annual Population Survey (APS), in April 2011. The APS is the source of the national estimates of personal well-being in the UK that are published annually by ONS.Back to table of contents
The four personal well-being questions gained national statistics status in September 2014. Prior to this, and still ongoing ONS have conducted a range of quantitative testing and qualitative testing.
The papers that explain in depth the validity of testing the four questions are as follows. The paper Overview of ONS phase three cognitive testing of Subjective Well-being Questions (328.2 Kb Pdf) gives an overview of the qualitative testing of the questions, and paper Measuring Subjective Well-being (240.8 Kb Pdf) covers the background and rationale behind the ONS 4 questions.
There are also several other papers documenting the quantitative testing of the ONS 4 questions. These are:
The personal well-being estimates from the Annual Population Survey are updated annually and are published in a statistical bulletin called, ‘Personal Well-being in the UK’. Recent efforts to reduce the time lag of the publication of the headline Personal Well-being estimates have resulted in an earlier publication of July. Further and more detailed analysis will be released on a more ad hoc basis throughout the year.Back to table of contents
ONS uses 4 questions to measure personal well-being, and does not produce one composite measure. The four questions were designed to measure distinct aspects of personal well-being (Evaluative, Eudemonic and Affective). It is therefore not appropriate to combine these questions as they are all individually important, and together they give a balanced approach to well-being. For more information on these approaches, and the on the background to the 4 personal well-being questions please see "How were the four personal well-being questions developed?" and "What is meant by 'evaluative', 'eudemonic' and 'affective well-being?'.Back to table of contents
The ONS personal well-being questions were developed as part of the Measuring National Well-being Programme. ONS sought advice from experts working in the field of subjective well-being and consulted with specialists on the National Statistician’s Measuring National Well-being Advisory Forum and Technical Advisory Group. Based on this, as well as an extensive programme of question testing, four questions were designed which provide a concise and balanced approach to the measurement of subjective well-being drawing on three main theoretical approaches (Dolan et al 2011, ONS, 2011b).
For more information about the development and testing of the four ONS personal well-being questions please see the report Measuring Subjective Well-being (240.8 Kb Pdf).Back to table of contents
These are academic terms used to describe different aspects of well-being; the personal well-being questions used by ONS ask about each of these.
The ‘evaluative’ approach asks individuals to step back and reflect on their life and make a cognitive assessment of how their life is going overall, or on specific aspects of their life.
The ‘eudemonic’ approach, sometimes referred to as the psychological or functioning/flourishing approach, draws on self-determination theory and measure such things as people’s sense of meaning and purpose in life, connections with family and friends, a sense of control and whether they feel part of something bigger than themselves.
The ‘experience’ approach focuses on people’s positive and negative emotional experiences (or affect) over a short timeframe to measure personal well-being on a day-to-day basis.
More information on the theoretical approaches underpinning the four ONS personal well-being questions is available in the paper First Annual ONS experimental subjective well-being results.Back to table of contents
ONS have no plans to change the personal well-being questions. To be sure that the questions are fit for purpose, ONS have done extensive testing using data from the Opinions and Lifestyle Survey, as well as cognitive interviews with members of the public. Academic experts and policy makers have also been consulted as members of our Technical Advisory Group (TAG). The results from the Opinions and Lifestyle Survey, the cognitive interviews and minutes from the Technical Advisory Group meetings are included within this link.
Together, the findings from this work suggest that the questions do provide robust measures of personal well-being. In future, the value to be obtained from any proposed changes will have to be weighed carefully against the value of maintaining consistent measures that enable comparisons over time.Back to table of contents
The four ONS personal well-being questions were added to the Harmonisation Programme standards as an ‘interim standard’ for the measurement of personal well-being in February 2014. The expectation is that they will be included as ‘finalised standards’ after they are accredited as National Statistics (planned for later in 2014). The interim harmonised standards are available to download.
The Harmonisation Programme is a UK cross-governmental programme of work looking into the standardising the way information is collected and the way results are presented and has been led by ONS for several years. The aim of the programme is to make it easier for users to draw clearer and more robust comparisons between data sources.Back to table of contents
No, the Annual Population Survey is one of many surveys in the UK that now include the ONS personal well-being questions. Other ONS surveys which include the personal well-being questions are: the Wealth and Assets Survey, the Living Costs and Food Survey, the Crime Survey for England and Wales and the Opinions and Lifestyle Survey.
For more information about other surveys across government and more widely that include some or all of the ONS personal well-being questions, please see: Surveys using the four ONS personal well-being questions (241.9Kb Pdf) This also provides details of topic coverage and how the data from each survey can be obtained.Back to table of contents
The following is a brief summary of some of the key features of the APS dataset to help researchers decide whether this is the best source of data for their analysis. Further details can be found in the Personal Well-being user guide.
Sample size: over 158,000 adults aged 16 and over responding to the PWB questions in the year ending March 2016, making this the UK’s largest survey containing the ONS personal well-being questions
UK coverage: the APS provides a representative sample of adults (aged 16 and over) living in residential households in the UK.
Levels of geography included: Unitary Authorities/Counties in England and Wales, Local Authorities in Scotland, Local Authority Districts in England and District Council Authorities in Northern Ireland. (Please note that level of geography available on the dataset to the approved researcher depends on the level of user license agreement. Please refer to the Personal Well-being user guide for further details).
Topic coverage: The APS is primarily a labour market survey so it includes an extensive range of questions which are important for understanding labour market participation, many of which are also useful for the analysis of personal well-being. For example, it includes a wide range of social and demographic questions as well items about housing, employment and education.
Mixed interview modes: The APS is a mixed mode survey in which people are interviewed either face-to-face or by telephone. The method by which people are interviewed appears to affect responses to the personal well-being questions. This may be particularly relevant to researchers wishing to make comparisons between areas north and south of the Caledonian Canal because all interviews north of the Caledonian Canal are by telephone. More information on interview modes is available under section 31. Please see the Personal Well-being QMI for further details.
Cross-sectional: Whilst the APS reflects a panel design and attempts are made to interview respondents on more than one occasion, we recommend analysis of Personal Well-being only on a cross sectional basis.
Time series: Personal Well-being as been collected on an annual basis since the financial year ending 2012.
The sample size for the Personal Well-being responses on the APS is over 158,000 adults aged 16 and over per year. The sample covers the whole of the UK for the household population but does not include young people under the age of 16 or adults living in communal establishments (for example nursing homes, homeless hostels, prisons etc).Back to table of contents
Yes, there are a range of ways in which the data are made available to the public. A regular set of key estimates from the data are available in Excel spreadsheets published alongside the Personal Well-being statistical bulletins. For the most recent personal well-being data by various personal characteristics, please see the bulletin ‘Personal Well-being in the UK’, published July 2013. For personal well-being data by place, please see the bulletin published in October 2013 ‘Personal Well-being Across the UK’. From September 2014 onwards, the data tables will all be released as part of a single annual statistical bulletin. The data are currently available in Excel spreadsheets, published alongside the statistical bulletin(s).
For UK-based academics designated as Approved Researchers, the APS Personal Well-being dataset is also deposited with the UK Data Service about 6 weeks after publication of the Personal Well-being statistical bulletins. It is available from the UK Data Service in both End User License and Special License versions. For further information about these options and details of how to access the data, please contact the UK Data Service directly.
Academics not based in the UK can also access the data either through the ONS Virtual Microdata Laboratory (VML) or through the Secure Data Service of the UK Data Service. In both cases, overseas researchers will need to be approved to use the data and must travel to the UK to use these facilities. Please contact either the UK Data Service or the APS team at ONS for further details.
ONS can also provide the data directly to UK civil service statisticians and government researchers. Government analysts interested in this option should please contact the APS team directly.Back to table of contents
ONS included estimates for the first time at local authority district level in the supporting reference tables accompanying the October 2013 statistical bulletin. Sample sizes vary considerably between local authority districts and the data therefore are more robust in some areas than others.
Two different approaches are used to help users assess the quality of data at this level. Where sample sizes are small and could be potentially disclosive, standard disclosure control processes are followed and relevant figures are suppressed. Additionally, estimates with a coefficient of variation (CV) of 20% or greater are also suppressed. The CV indicates the amount of variation in the average (arithmetic mean), that is, the higher the CV, the more the responses are spread out around the average value. Estimates that are provided are colour coded in the data tables in order to indicate the quality of the data based on the CV values.
As the APS is a mixed mode survey, researchers should also consider whether interview mode effects may be relevant to their analysis. People who answer the personal well-being questions by telephone appear to give higher ratings on average than those who answer in face-to-face interviews. As those living north of the Caledonian Canal are interviewed exclusively by telephone, care should be taken when comparing areas north of the Caledonian Canal to those South of the Caledonian Canal. Researchers may wish to disregard any differences between such areas which are only marginally statistically significant.Back to table of contents
Yes, the APS Personal Well-being dataset includes information about the respondent’s occupation and industry sector using the National Statistics Socio-Economic Classification (SOC2010 based) (NS-SEC) and Standard Industrial Classification 2007 (SIC 2007) respectively.
ONS published personal well-being data by broad occupational group in July 2012 from the APS Personal Well-being dataset covering the period from April 2011 to March 2012. In February 2012, ONS also published data on personal well-being by NS-SEC covering the period from April 2011 to September 2011.
For details of other occupation-related variables that are available on the APS Personal Well-being Dataset, please see the Personal Well-being Survey User Guide.Back to table of contents
The personal well-being questions can be analysed by various labour market variables using the Annual Population Survey dataset. This includes: economic activity, information about main job and second job where applicable, self-assessed health, travel to work, hours worked, employment patterns, union representation, details of unemployment, benefit entitlement, education and training, health and injury and personal earnings. Demographic variables available on the dataset include respondent characteristics such as sex, age, relationship status, ethnicity, religion and housing tenure. Geographic variables are also available to allow users to conduct analysis below the UK level. This includes: UK countries, English regions and Local Authorities. A full list of variables available on the Annual Population personal well-being dataset is available in the Personal Well-being user guide.Back to table of contents
Datasets from all of the ONS surveys which include the ONS personal well-being questions are either currently available or will be made available through the UK Data Service to approved researchers by the end of 2014. For further information about these datasets and details of how to access the data, please contact the UK Data Service directly or the relevant survey team at ONS for further details.
Opinions and Lifestyle Survey (OPN): email@example.com
Wealth and Assets Survey (WAS): firstname.lastname@example.org
Effects of Taxes and Benefits on Household Income Personal Well-being dataset (ETB): email@example.com
Crime Survey for England and Wales: firstname.lastname@example.org
The Opinions and Lifestyle (OPN) Survey is cross-sectional and includes a sample of approximately 1000 respondents per month across Great Britain. From April 2014, the survey will run every month of the year apart from March, June, September, and December. It is a multi-purpose social survey covering different topics of immediate interest each month it is conducted. Government organisations, academic institutions and charities can commission modules on a monthly basis and these have included topics such as public attitudes to road congestion, smoking, drinking, disability and contraception. Demographic information such as age, sex, and marital status is asked of every respondent. As OPN is the main survey used by ONS for testing variations on the personal well-being questions, the standard ONS personal well-being questions are not asked of the full sample each month. Please see the Opinions and Lifestyle Survey guidance for further details.
The Wealth and Assets Survey is a longitudinal survey covering Great Britain (excluding north of the Caledonian Canal in Scotland and the Isles of Scilly). Approximately 30,000 household interviews in the first wave of interviews were achieved and approximately 20,000 household interviews in the second wave. This is around 41,000 interviews at the individual level that can be linked across the two years. As answers by proxy are not accepted for the personal well-being questions, the sample size available will be fewer that 41,000.
WAS collects information about the economic well-being of households by gathering information on level of assets, savings and debt, saving for retirement, distribution of wealth amongst households or individuals and factors affecting financial planning among others. Standard demographic information is also included such as age, sex, marital status, and educational attainment. Further details are available in the Wealth and Assets Survey User Guide.
The Effects of Taxes and Benefits on Household Income Personal Well-being Dataset covers Great Britain and is based on a household survey including approximately 8,000 individuals. As with all other ONS surveys, proxy responses are not accepted for the personal well-being questions so the final sample size for personal well-being data is smaller than the full sample size.
The ETB also includes information on the socio-demographic characteristics of household members (age, sex, marital status) and relationships to one another (e.g. spouse, son/daughter, partner). Information is also collected on regular items of expenditure, such as mortgage/rent, insurance, utility bills, education fees and infrequent purchases such as vehicles. Details about employment and income, including receipt of benefits, allowances and other financial assets are also included. Further details about the Effects of Taxes and Benefits on Household Income Dataset are available on the ONS website.
The Crime Survey for England and Wales (CSEW), formerly known as the British Crime Survey (BCS), is a continuous face-to-face victimisation survey in which people aged 16 and over resident in households in England and Wales are asked about their experiences of a range of crimes in the 12 months prior to the interview. Respondents to the survey are also asked about their attitudes towards different crime-related issues, such as the police and the criminal justice system and perceptions of crime and anti-social behaviour. Until recently the survey did not cover crimes against those aged under 16, but since January 2009 we have also interviewed children aged 10 to 15.
The CSEW has a sample size of 35,000 interviews across the year with adults aged 16 and over resident in households in England and Wales and an additional 3,000 10 to 15 year olds identified through screening at households interviewed for the main sample. Standard demographic information is also included such as age, sex, marital status, and educational attainment. Further details are available in the Crime Survey for England and Wales Technical Report.Back to table of contents
Where data is potentially disclosive due to small sample sizes, standard ONS statistical disclosure control processes are applied and the relevant figures are suppressed.
Any estimate which has a coefficient of variation (CV) of 20% or greater is also excluded from the published reference tables as part of data quality assurance processes. The CV indicates the amount of variation in the average (arithmetic mean): the higher the CV, the more the responses are spread out around the average value. Estimates that are provided in the Excel reference tables are colour coded to indicate the quality of the data based on the CV values.
For more information, please see the Personal Well-being Survey User Guide.Back to table of contents
On the APS, it takes on average about a minute and a half for the interviewer to ask the four questions and respondents to give their answers. This includes the interviewer reading out the introduction to the personal well-being questions.Back to table of contents
There is no evidence that the personal well-being questions have had any adverse effects on response rates on the surveys in which they have been included at ONS.
Additionally, all of the personal well-being questions have very low rates of item non-response (typically around 1% non-response for each of the four questions).
ONS would welcome further feedback on response rates from others who have used the questions on their surveys. Please contact us at QualityOfLife@ons.gsi.gov.ukBack to table of contents
Ideally, the four personal well-being questions should be placed at the beginning of the survey, immediately after the key demographic questions. This is where the questions appear on the Annual Population Survey. This was selected as the best place for the questions following extensive qualitative testing and consultations with academic experts. Key considerations are allowing sufficient time for the development of rapport between the interviewer and the respondent before asking potentially sensitive questions about well-being, and asking the questions early enough in the interview to avoid any bias from preceding questions on other topics such as health or employment.
For more details please see the Personal Well-being Survey User guide.Back to table of contents
Yes, all responses to the personal well-being questions are kept even if the respondent has not answered all four questions.Back to table of contents
All responses are retained even if a respondent gives the same response to all four questions.Back to table of contents
The Annual Population Survey uses both telephone and face-to-face data collection methods. This use of mixed data collection methods makes it unfeasible to use show cards consistently among all respondents to the APS.Back to table of contents
Different data collection methods do appear to affect how people respond to the personal well-being questions. Findings from the APS show that on average, people rate each aspect of their well-being more positively when interviewed by telephone than when interviewed face to face by an interviewer. For example, in 2012/13, higher ratings were given on average for the life satisfaction, worthwhile, and happy yesterday questions and lower ratings to the anxiety question during telephone interviews compared to face-to-face.
The relationship between the mode of interview and average responses to the personal well-being questions has also been examined using regression analysis to hold other possible influences on personal well-being constant. This shows the same pattern found in descriptive statistics; on average, people give more positive responses when interviewed by telephone than when interviewed face-to-face.
The findings are statistically significant for each question, suggesting they are not due to chance alone. The mode effect is smallest for the question about anxiety yesterday which people rate 0.05 points higher on average on the 0-10 scale when interviewed face-to-face compared to telephone. The effect is greatest on ratings of life satisfaction which people rate 0.17 points lower on average on the 0-10 scale when interviewed face-to-face compared to telephone.
It is challenging to account for mode effects when using statistics. Mode effects can be controlled for in regression models by including mode of interview as an independent variable, and we advise researchers to do this in any regression analysis involving the personal well-being data.
When considering the high-level national estimates published by the ONS, the impact of mode is statistically significant but in general small and roughly consistent over the period for which data are available, meaning that mode effects may be unimportant to any substantive conclusions being drawn.
Mode effects may have a greater impact on comparisons between lower-level estimates (such as local area statistics) for two different reasons: different groups may have different balances of telephone and face-to-face response, and the impact of mode may differ by group. As all interviews in Scotland north of the Caledonian Canal are conducted by telephone, we would advise researchers to disregard any marginally statistically significant findings based on comparisons between areas north and south of the Caledonian Canal.Back to table of contents
No. The aim is to ask people to assess their own well-being, so all answers must be given directly by the respondent. Proxy responses (on behalf of someone else in the household) are never accepted.Back to table of contents
ONS decided on an eleven point scale from 0-10 where 0 is "not at all" and 10 "completely" for all of the questions following a review of the literature on the measurement of personal well-being and in the light of expert academic advice. This ensures the scales between the questions are consistent which helps respondents to answer them more easily. Also, eleven point scales are commonly used across other surveys of interest, and using the same type of scale aids comparisons.
For more information about the use of the 0 to 10 scale, please see: Opinions and Lifestyle Survey: Methodological Investigation into Response Scales in Personal Well-being.Back to table of contents
These different ways of looking at the data are useful for different purposes.
Although average ratings are easy to understand and provide a helpful summary measure to compare responses year on year and across areas, the average does not show the proportion of people who gave each response on the 0 to 10 scale. The average cannot shed light on whether responses are evenly distributed along the 0 to 10 scale, or the proportions of people rating their well-being at the highest and lowest levels. For this reason, relying only on average ratings might mask important differences in how people rate their well-being.
This is why the data are also presented in terms of the proportions of people who rated each aspect of their well-being within a particular range on the 0 – 10 scale. For the life satisfaction, worthwhile and happiness questions, ratings are grouped in the following way:
0 to 4, (low)
5 to 6, (medium)
7 to 8, (high)
9 to 10, (very high)
For the anxiety question, ratings are grouped differently to reflect the fact that higher anxiety is associated with lower well-being. The ratings for anxiety are grouped as follows:
0 to 1, (very low)
2 to 3, (low)
4 to 5, (medium)
6 to 10, (high)
The ratings on the 0-10 scale are grouped this way because this results in fairly even groups of people who rate their well-being in different ways. For example, the mean rating for the positive questions is around 7 to 8 and there are similar percentages of people rating their positive personal well-being as 5 or 6, 7 or 8 and 9 or 10. This is helpful for analysis purposes.
Additionally, qualitative testing has shown the mid-point of the scale (5) acts as an important cut-off for respondents for assessing their own well-being, with responses below five seen largely as a negative response.
Ratings are grouped in a similar way for the anxiety question, but as most people tend to rate their anxiety lower than 5 (indicating higher well-being), the ratings are grouped in a way which reflect this difference.
As policy-makers are particularly interested in people reporting the highest or lowest levels of personal well-being, looking at the ratings in this way helps to shed more light on these groups.Back to table of contents
As the personal well-being data have only been collected since April 2011, ONS have tried various methods of analysing and presenting the findings to see which best meets the needs of those who use the data.
For example, in the 'First ONS Annual Experimental Personal Well-being Results', the data were published data for the first time on smaller geographic areas such as unitary authorities and counties. To ensure that the estimates were based on larger sample sizes, the ratings on the 0-10 scale were divided into two groups rather than four. For the questions about life satisfaction, the sense of whether what one does in life is worthwhile, and happiness yesterday, the ratings were grouped together in the following way:
0 to 6 (low / medium) and
7 to 10 (high / very high).
For the ‘anxious yesterday’ question the ratings were grouped as follows:
0 to 3 (very low / low) and
4 to 10 (medium / high).
A different approach was taken in the October 2013 statistical bulletin, ‘Personal Well-being Across the UK’. In this case, the data was published for more local areas for the first time such as unitary authority/local authority level. Instead of combining groups to create a larger sample sizes, the decision was taken to suppress the data where it did not meet pre-defined quality standards (based on the size of the coefficient of variation).Back to table of contents
Contact details for this Methodology
Telephone: +44 (0)1633 455680