### Methodology

The data for this measure is taken from the results of Sport England’s Active Lives (AL) Survey. The survey sample is randomly selected from the Royal Mail’s Postal Address File, which has a very high coverage of private residential addresses. A letter is sent to households inviting up to 2 people per household to take part in the survey, either online or by requesting a paper version of the questionnaire.

A random sampling survey design ensures results are representative of the population. There were 162,418 adults aged 18 and older who gave their height, weight and ethnicity in their survey response. Because people often underestimate their weight and overestimate their height, a person’s self-reported body mass index (BMI) is generally known to be lower than it actually is. This measure adjusts for this bias by applying a formula based on observations from several years of the Health Survey for England, which for many respondents included both self-reported and clinically measured BMIs.

Weighting: Weighting is used to adjust the results of a survey to make them representative of the population and improve their accuracy.

For example, a survey which contains 25% women and 75% men will not accurately reflect the views of the general population, which we know has an even 50/50 split.

Statisticians rebalance or ‘weight’ the survey results to more accurately represent the general population. This helps to make them more reliable.

Data has been weighted to ONS population measures for geography and key demographics.

Confidence intervals: The confidence intervals for each ethnic group are available if you download the data.

In 2016/17, 61.3% of adults surveyed reported a height and weight that determined a body mass index (BMI) of 25 or more when adjusted for bias. This is a reliable estimate of the percentage of adults in England who were overweight. However, because the AL survey results are based on a random sample of adults aged 18 or older, it’s impossible to be 100% certain of the true percentage.

It’s 95% certain that somewhere between 61.0% and 61.5% of all adults in England were overweight. 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 in this range (between the upper and lower 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, the number of people from the Chinese ethnic group sampled for this survey is relatively small, so we can be less certain about the estimate for the smaller group. This greater uncertainty for people from the Chinese ethnic group is expressed by the wider confidence interval of between 27.7% and 35.3%.

Statistically significant differences have been determined where the 95% confidence intervals of an ethnic group do not overlap with the national average or with that of the ethnic group being compared. This is a pragmatic but less precise test of the significance of differences between estimates: it is possible that a finding may be significant when confidence intervals overlap slightly.

For further details of the sampling method, weighting and confidence intervals see the Active Lives Survey technical report.

### Suppression rules and disclosure control

None applied.

### Rounding

Data is rounded to the nearest whole number in charts and tables, and unrounded in the data downloads.