Methodology
This data presents the results from the question, ‘Overall, how happy did you feel yesterday?’
People were asked to respond on a scale of 0 to 10, where 0 is ‘not at all’, and 10 is ‘completely’. Estimates were produced as mean scores, as well as thresholds.
Personal wellbeing questions are asked to adults age 16 and over, living in private households. Personal wellbeing questions can only be answered in person (they cannot be answered by proxy).
The Annual Population Survey (APS) is a continuous household survey. Most people are interviewed face to face at first contact, and by telephone at subsequent contacts. The sample is formed partly from waves 1 and 5 of the Labour Force Survey (in which selected addresses are contacted every 3 months) and partly from boost cases, which are in the sample for 4 waves, spread one year apart.
The sampling frame is mainly the Royal Mail Postcode Address File (PAF). The NHS communal accommodation list is also used and (in the case of remote parts of Scotland) telephone directories. All eligible individuals found at the selected address may be interviewed. The complex survey design has been taken into account when calculating confidence intervals.
The achieved sample of approximately 158,000 respondents undergoes weighting, which is structured at local authority level and uses age and sex dimensions. The Office for National Statistics population estimates and projections are used as the basis for this weighting process.
Results derived from a low number of survey responses are more likely to be affected by statistical variation, so observed differences may not reflect actual difference. As such, caution is needed when interpreting shortterm trends in the data, especially for sub groups (for example, a particular ethnic group).
Smaller numbers of survey respondents from ethnic minority backgrounds and smaller geographic regions mean that these estimates are less reliable than estimates for White people and larger regions.
Measuring wellbeing requires a number of different approaches in order to capture a range of factors which contribute to an individual's overall sense of wellbeing. The APS uses a number of specific approaches, including the evaluative, eudemonic, experience and individual wellbeing approaches.
This measure involves the ‘experience approach’, which seeks to measure people’s positive and negative experiences (or affect) over a short timeframe to capture their subjective wellbeing on a daytoday basis. The question “Overall, how happy did you feel yesterday?” is a positive affect question, in contrast to the negative affect question “Overall, how anxious did you feel yesterday?”.
Thresholds are used in the report to present dispersion in the data. These show the proportion of responses that fall into 4 groups on a scale of 0 to 10:
 low levels of happiness (scoring 0 to 4)
 medium levels of happiness (scoring 5 to 6)
 high levels of happiness (scoring 7 to 8)
 very high levels of happiness (scoring 9 to 10)
All the differences noted in the text are statistically significant. The statistical significance of differences are approximate because they are determined on the basis of nonoverlapping confidence intervals.
Confidence intervals
Confidence intervals for each ethnic group are available in the ‘download the data’ section.
The Annual Population Survey is based on a sample of people aged 16 and over across the UK. This measure makes a reliable estimate of, for example, the percentage of people aged 16+ with a high level of happiness, but it’s impossible to be 100% certain of the true percentage.
It’s 95% certain, however, that somewhere between 40.1% and 40.8% of all people aged 16+ in the UK reported a high level of happiness in 2016. In statistical terms, this is a 95% confidence interval. This means that if 100 random samples were taken, then 95 times out of 100 the estimate would fall between the upper and lower bounds of the confidence interval. But 5 times out of 100 it would fall outside this range.
The smaller the survey sample, the more uncertain the estimate and the wider the confidence interval. For example, fewer people aged 16+ from the Arab ethnic group responded to the survey than White people aged 16+ , so we can be less certain about the estimate for the smaller group. This greater uncertainty is expressed by a wider confidence interval, for example of between 32.9% and 45.1% for Arab people aged 16+ in 2016.
All the differences noted in the text are statistically significant. The statistical significance of differences are approximate because they are determined where the 95% confidence intervals for 2 groups or time periods don't overlap.
An example of nonoverlapping confidence intervals would be the results for the Chinese ethnic group, which had a confidence interval of between 40.4% and 51.1%, and those for the Bangladeshi ethnic group, which had a confidence interval of between 30.4% and 40.1%.
Suppression rules and disclosure control
Estimates are suppressed if:

the sample size is less than 50

the degree of variability of responses (coefficient of variation) is greater than 20%

the threshold numerator is based on a small number, as defined by the Government Statistical Service (GSS) quality and suppression guidance
Rounding
Estimates of mean scores have been rounded to 2 decimal places; estimates of percentages within thresholds have been rounded to 1 decimal place.
Sample sizes have been rounded to the nearest 10, following Government Statistical Service guidance.
Comparisons have been based on unrounded data.
Quality and methodology information