Weighting, benchmarking and estimation procedures
Weighting is the process of adjusting results from a sample survey to infer results for the in-scope total population. To do this, a weight is allocated to each sample unit; for example, a household or a person. The weight is a value which indicates how many population units are represented by the sample unit.
The steps used to derive person weights and benchmarks are provided in detail in the Survey Design and Operation chapter of this User Guide. Key variations to the weighting procedures for the HLS were that only person-level (not household-level) weights were produced for the population aged 18 years and over, and that the HLS weights were calibrated to the NHS 2017-18 person-level weights for some key variables (i.e. collapsed highest level of education; one-digit country of birth; current daily smoker status; self-assessed health; collapsed disability status; heart, stroke and vascular disease status; and overweight/obese status).
As a voluntary sub-sample, the HLS had an increased risk of non-response bias. A coverage analysis was undertaken to determine the most important variables (including NHS variables) associated with non-response, with calibration undertaken against a set of benchmarks that included these variables. Weights calibrated against population benchmarks ensure that the survey estimates conform to independently estimated distributions of the population by age, sex and area of usual residence, rather than to the distribution within the sample itself. Calibration to benchmarks helps to compensate for over- or under-enumeration of particular categories of persons and households, which may occur due to the random nature of sampling or non-response.
Comparison between HLS and NHS samples(a), Persons aged 18 years and over, 2017-18
|
 | | HLS unweighted
% | NHS unweighted
% |
|
| Age 65+ | 27.2 | 25.6 |
| Male | 46.0 | 46.3 |
Born in Australia
Employed
Person living alone
Has a non-school qualification(b)
Has no disability or restrictive long term health condition
Excellent or very good self-assessed health
Current daily smoker
Overweight/obese(c) | 73.3
61.7
27.4
65.9
71.2
56.1
11.4
70.9 | 69.7
60.7
29.1
59.1
72.2
52.8
15.0
68.5 |
|
|
|
(a) Weighting accounted for differences in characteristics of samples for collapsed highest level of education; one-digit country of birth; current daily smoker status; self-assessed health; collapsed disability status; heart, stroke and vascular disease status; and overweight/obese status.
(b) Excludes persons where level was not determined.
(c) Includes only persons for whom height and weight were measured.
Due to the additional weighting process undertaken it should be noted that the person-level weighted perturbed data for the HLS does not exactly match the weighted perturbed data shown for the NHS 2017-18 results.
Print Page
Print All