|Page tools: Print Page Print All RSS Search this Product|
(a) Respondent's height and/or weight measurement not taken.
SELF-REPORTED HEIGHT AND WEIGHT
In addition to the voluntary measured items, respondents in the 2018–19 NATSIHS were also asked to self-report their height and weight measurements. Of those whose BMI was not measured, 55.8% of adults and 34.3% of children provided both self-reported height and weight measurements. This provided valuable information about the height and weight that was used in the imputation for people with missing values.
HOW IMPUTATION WORKS
In the 2018–19 NATSIHS and both the 2014–15 National Aboriginal and Torres Strait Islander Social Survey (NATSISS) and 2017–18 National Health Surveys (NHS), missing values were imputed using the 'hot decking' imputation method. In this method, a record with a missing response (the 'recipient') receives the response of another similar record (the 'donor').
A number of characteristics (imputation variables) were used to match recipients to donors.
For adults they were:
For example, a female recipient aged 35–39 years who lives in a capital city, has a self-reported BMI category of overweight (calculated using self-reported height and weight), has a self-perceived body mass of healthy, has high cholesterol and lives a sedentary lifestyle will match to a donor record who has the same profile (female, 35–39 years, self-reports as overweight, etc).
For BMI, 80.4% of imputed records used all seven imputation variables to match to a donor record. The remaining 19.6% could not be matched using all seven variables and were therefore matched using fewer variables. For example, 7.4% of imputed records were matched to donors using all imputation variables except part of state.
For children aged 2–14 years, the following variables were used:
For those aged 15–17 years, the same imputation variables were used as for those aged 2–14 years, in addition to level of exercise and self-perceived body mass (only if a person answered for themselves). Cholesterol data was not collected for persons under 18 years of age and so could not be used as an imputation variable.
Single year of age was used in the imputation method for the 2018–19 NATSIHS, which differs from the 2017–18 NHS. The change was made for the 2018–19 NATSIHS as children's height and weight can change rapidly within a short period of time and, to avoid over or under estimating BMI, donors that were of the same age as the recipient were used. A similar change to the imputation method will be investigated for the next iteration of the NHS.
IMPACT OF IMPUTATION
Physical measurement data (BMI, waist circumference and blood pressure) that includes imputed values are of suitable quality for comparisons with previous surveys; however, the Australian Bureau of Statistics (ABS) recommends using proportion comparisons only as imputation was not used on the physical measurement data in previous years.
The table below shows the impact of imputation on BMI estimates.
Measured and imputed Body Mass Index (BMI) results, by age(a)
Whether Body Mass Index (BMI) measured or imputed(a)
These documents will be presented in a new window.