Support #2308
open
Choose a correct weight for analysis
50%
Description
Hi team,
Happy New Year!
I am analyzing wave 1-14 UKHLS data as a longitudinal data to examine effects of a policy. I may use RDD or DID study design. Accroding to "6614_main_survey_user_guide_weighting_variables.pdf"[please see screenshot 1], if I am correct, I probably should use 'w_indpxus_lw'. But by the "6614_main_survey_user_guide.pdf" [please see screenshot 2], this weight is for GPS&EMB sample. Will I treat other samples wrongly if I use this weight?
Should I use a different weight for different wave to account for all samples in each wave correctly? If so, how to do so when I pool all wave's data to do analysis?
Please let me know if my question is unclear or my understanding is wrong.
Appreciate all your help!
Best,
Bing
Files
Updated by Understanding Society User Support Team 4 days ago
- Category set to Weights
- Status changed from New to Feedback
- % Done changed from 0 to 50
- Private changed from Yes to No
Hello Bing,
As the user guide mentioned (https://www.understandingsociety.ac.uk/documentation/mainstage/user-guides/main-survey-user-guide/selecting-the-correct-weight-for-your-analysis/), specifically in the example box, if your analysis pools multiple waves, you should select the longitudinal (“*_lw”) weight from the last wave included in your analysis. So, for a longitudinal analysis using Waves 1–14, you would use a Wave 14 *_lw weight.
Which specific Wave 14 *_lw weight you should use depends on the instrument and the level of analysis. For example, if you are analysing individuals using the individual questionnaire, you should use o_inding2_lw. This is the most comprehensive longitudinal individual weight, as it has all UKHLS samples (i.e., GPS, EMBS, BHPS, IEMBS, and GPS2).
I hope this information is helpful
Best wishes,
Roberto Cavazos
Understanding Society User Support Team