Publication
How to use replicate weights in health survey analysis using the National Nutrition and Physical Activity Survey as an example
Publisher:
Cambridge University Press (CUP)
Date:
19-08-2019
DOI:
10.1017/S1368980019001927
Abstract: To conduct nutrition-related analyses on large-scale health surveys, two aspects of the survey must be incorporated into the analysis: the s ling weights and the s le design a practice which is not always observed. The present paper compares three analyses: (1) unweighted (2) weighted but not accounting for the complex s le design and (3) weighted and accounting for the complex design using replicate weights. Descriptive statistics are computed and a logistic regression investigation of being overweight/obese is conducted using Stata. Cross-sectional health survey with complex s le design where replicate weights are supplied rather than the variables containing s le design information. Responding adults from the National Nutrition and Physical Activity Survey (NNPAS) part of the Australian Health Survey (2011–2013). Unweighted analysis produces biased estimates and incorrect estimates of se . Adjusting for the s ling weights gives unbiased estimates but incorrect se estimates. Incorporating both the s ling weights and the s le design results in unbiased estimates and the correct se estimates. This can affect interpretation for ex le, the incorrect estimate of the OR for being a current smoker in the unweighted analysis was 1·20 (95 % CI 1·06, 1·37), t = 2·89, P = 0·004, suggesting a statistically significant relationship with being overweight/obese. When the s ling weights and complex s le design are correctly incorporated, the results are no longer statistically significant: OR = 1·06 (95 % CI 0·89, 1·27), t = 0·71, P = 0·480. Correct incorporation of the s ling weights and s le design is crucial for valid inference from survey data.