Support #2012
open
Added by Margherita Agnoletto about 1 year ago.
Updated 10 months ago.
Description
Dear Understanding Society Team,
I am currently examining the relationship between flexible work arrangements (FWA) and some employees' outcomes.
Given that questions about FWA are asked every two waves, I have chosen to conduct a longitudinal analysis (FE) using waves 2, 4, 6, 8, and 10. Some of my outcomes come from the self-completion questionnaire.
As I understand, it is recommended to use the appropriate longitudinal weight from the last wave in my analysis (i.e. i_indinus_lw). However, I observe a significant loss of observations.
Given that my panel is unbalanced, could I use the corresponding longitudinal weight from the last available wave for each individual? For instance, if an individual 'i' has information until wave 8, I propose imputing the appropriate longitudinal weight from wave 8. Similarly, if individual 'k' has information until wave 6, I suggest imputing the weight from wave 6.
Thank you for your attention.
Kind regards
Margherita,
Thank you for your question. Unfortunately, this isn't an option. The results will be wrong if our weights applied from different waves in the described way. This is because each set of weights represents a population, but taking different values from different waves and putting them together looses this representation.
If you are concerned about numbers you could create a tailored weight for your analysis. Our online course describes it (unfortunately the link seems to be broken today): https://www.understandingsociety.ac.uk/help/online-training/creating-tailored-weights .
Hope this helps,
Olena
- Status changed from New to Feedback
- % Done changed from 0 to 50
- Private changed from Yes to No
Thank you very much, Olena, for your help, I understand the point.
If I wanted to use Understanding Society weights, is it correct to use the longitudinal weight from the last wave (i_indinus_lw or, if I also use variables from the self-completion questionnaire, i_indscus_lw) even if my variables of interest are only present in the even waves?
In both cases, I find significantly fewer observations for males than for females; I wonder what the reason might be.
Thanks for your availability.
Margherita,
Thank you for your question. Yes, you should use the longitudinal weight from the last wave in your analysis even if you don't use all waves.
There are a few subgroups that tend to have higher nonresponse and attrition in all surveys, including UKHLS, males being one of such groups. Weights correct for this nonresponse, so you should always use them.
Hope this helps,
Olena
- Status changed from Feedback to Resolved
- % Done changed from 50 to 100
Also available in: Atom
PDF
Updated by 
Updated by 
Updated by