For combined Senegal continuous DHS surveys extract, do I need to adjust weights for each survey? If so, how?


I am using an extract with 5 Senegal DHS surveys from 2012-2017. I would like to product an estimate across the surveys. Are the weights already adjusted for the multiple surveys, or do I need to adjust them? If so, do you have any suggestions of the best way to do this?


Each of the surveys from 2012 to 2017 is representative at the national level and for region groups, called West, Center, North, and South. If you are happy with this level of geographic detail, then you are fine comparing across surveys, and the weights are appropriate within each sample. If you want representative estimates at a finer level of geographic detail (each of 14 regions), you need to combine 2012 with 2014, and 2015 with 2016. The combined files have unique weights.

Currently, IPUMS DHS includes only the single-year samples for Senegal. Later this year, we will make it possible for users to select the combined 2012/2014 and 2015/2016 samples as well (with their unique weights).

For more information on weighting the continuous Senegal surveys, this post on the DHS User Forum is valuable.