Project

General

Profile

Support #1715

Longitudinal Weighting of Non-Movers Only from UKHLS (in Wave 1)

Added by Sue Easton 4 months ago. Updated about 2 months ago.

Status:
Feedback
Priority:
High
Category:
Weights
Start date:
06/13/2022
% Done:

70%


Description

Hi, I have searched and can't find these key words in any posts.

Due to limitations of time I need to limit my analysis to individuals who have not changed location since they entered the survey in Wave 1 (UKHLS sample and any others in from Wave 1 with more than 1 wave).

This means some people's data will be right censored due to household moves.

How will this affect weighting?

Will I need to calculate new weights? As variables such as age are highly likely to be correlated with the "risk" of moving home.

Thanks.

Sue EAston

#1

Updated by Olena Kaminska 4 months ago

Sue,

Thank you for your question. Do you mean you want to study the group who has not moved their house in the last 12 years (for example)? If so, this is a substantive group within a population and is well represented with the existing weights (same as any other subgroup of the population).

Hope this helps,
Olena

#2

Updated by Sue Easton 4 months ago

Hi Olena,

Well sub-population 1) is the residents who have not moved at all.

However, is it possible to also include subset 2) residents/respondents who have moved up to the point they moved e.g. those respondents who were in from wave 1, but moved house in to 2007, using the same longitudinal weights at wave k? (i.e. the ones who will have right-censored data because they moved location before 2020/21).

For this sub-population who moved, I will only be including their data up to the point of moving, so are their weights at wave k still appropriate?

Thanks.

Sue E.

#3

Updated by Understanding Society User Support Team 2 months ago

  • Category set to Weights
  • Status changed from New to Feedback
  • % Done changed from 0 to 70
  • Private changed from Yes to No
#4

Updated by Olena Kaminska about 2 months ago

Sue,

Yes, such analysis with a longitudinal weight from wave k will be fine. Having said that some models that allow right censoring (e.g. survival analysis), also correct for attrition within the model. In such case you could use wave 1 weight relying on the model for attrition correction.

Best wishes,
Olena

Also available in: Atom PDF