Support #2101

Time variant and invariant variables using general UKHLS and Covid-19 survey

Added by Michael Francis about 1 month ago. Updated about 1 month ago.

Start date:
% Done:



Hi there,

I was wondering if you could provide some advice about how to treat the Covid-19 survey variables in comparison with the general UKHLS waves as I am looking across both surveys - I was thinking simply to create some dummy variables for the Covid survey, e.g., furloughed for > 6 months = 1 else 0, and then having this as a time-invariant IV in the model if that makes sense. Does this sound like an appropriate way to deal with these variables - or would you recommend treating them as distinct waves instead. The object of my research is to find out if certain conditions during Covid-19 affect outcomes of individuals, so in this sense it is sort of control variable.

Any advice you could provide would be greatly appreciated.

Many thanks,

Michael Francis


Updated by Understanding Society User Support Team about 1 month ago

  • Status changed from New to Feedback
  • % Done changed from 0 to 50
  • Private changed from Yes to No

Hello Michael

Our remit is to address questions related to the survey, its details, and the datasets including data management. We are unable to assist with questions concerning analysis, methodology and syntax (unless related to data management or weighting).

For insights into working with both Surveys, you might consider referring to other papers. For instance, the paper by Miall N, Pearce A, Moore JC, et al. Inequalities in children’s mental health before and during the COVID-19 pandemic: findings from the UK Household Longitudinal Study. J Epidemiol Community Health 2023;77:762-769 , provides a relevant example. The authors use both UKHLS main Study and Covid-19 Survey to explore how certain conditions during the pandemic impacted individual outcomes.

I hope this information is helpful.

Best wishes,
Roberto Cavazos
Understanding Society User Support Team

Also available in: Atom PDF