How to compare difference in median for complex survey design?

[Edward Sutanto]

Hi Statalist members,

I am currently using Indonesian Demographic and Health Surveys (thus complex survey design) to compare the number of cigarettes smoke per day between male and female smokers. The number of cigarettes smoke per day for both sexes, however, is not normally distributed. Thus I am more inclined to report the number using median (interquartile range), instead of mean ± standard deviation.

Based on https://www.stata.com/support/faqs/s...r-survey-data/, I know how to estimate median (interquartile range) using the svy command. But how do I compare whether there was a statistical difference in the median of number of cigarettes smoked per day between male and female smokers?

I would appreciate any help on this. Thank you very much.

[Vivian Jenkins]

Hi Statalist members,

I have a similar question, please.

I am a graduate student holding on to Stata (my professors use R or SAS which I have no clue about but will learn later). I am currently using the CDC’s National Health and Nutrition Examination Survey (NHANES) - a complex survey design - to compare several variables across quintiles of an exposure variable e.g. to compare the distribution of BMI across quintiles of sleep duration. However, all of the variables are not normally distributed, thus I am more inclined to report the number using median (interquartile range), instead of mean ± standard deviation. How can I get the median and also check for statistical significance?
I used the epctile code below (user-written code by Stas Kolenikov) and it gives the Median and Confidence Interval, but not the Interquartile Range and an overall p-value. Please can anyone kindly help with this? Thanks for your help!