Project

General

Profile

Support #2074

Longitudinal weights

Added by Joe Mattock 2 months ago. Updated about 1 month ago.

Status:
In Progress
Priority:
High
Category:
Weights
Start date:
03/09/2024
% Done:

20%


Description

Hi,

I'm conducting an analysis specifically over waves 2, 3, 6 and 9 for Understanding Society, as relating to the voteintent variable which is only included in these waves. I would just like to ask about the weighting procedure for this case. I am examining how an independent variable (gentrification, as measured by an index) affects voting intention at the LSOA-level.

My understanding is that I need to take the longitudinal weight from the final wave used in my analysis and apply it to all respondents (i_indscub_lw - I believe). However, given that my dependent variable of interest is not observed in consecutive waves, I wanted to ask whether this principle applies in the same way.

I also wanted to ask how this weighting would be applied in practice. I am slightly confused about the order of things. For example, would you remove all wave-specific prefixes, merge LSOA indicators with the Understanding Society data, and then apply the relevant weight for each respondent?

Much appreciated,

Joe

#1

Updated by Olena Kaminska 2 months ago

Joe,

Before I answer your question, could you clarify whether the voting intention you study is asked to all in a wave, or only to some following the timing of elections? If the latter, could you give me a few details (if you know) on who specifically is asked, as weighting will depend on that.

Also, can you clarify whether you study each wave separately? And you are not looking in dependency of voting in earlier and later waves, for example?

Thank you,
Olena

#2

Updated by Joe Mattock 2 months ago

Hi Olena,

Voting intention is asked about in the waves 2,3,6,9 and 12. To my understanding, it is asked to all respondents of the self-completion questionnaires and face-to-face interviews under the political engagement question module. The specific wording of the question asked is as follows:

"How likely is it that you will vote in the next general election?" (on a scale of 1-10)

Variable: https://www.understandingsociety.ac.uk/documentation/mainstage/variables/voteintent/

There are also values applied to respondents that cannot vote, or are missing, inapplicable, proxy, refused or don't know. I am considering excluding those who cannot vote (aged <18) from the analysis.

I am studying all waves together through a fixed effects methodology, with dummy variables included for the wave number and cross-wave identifier. To do this, I have created a wave variable that specifies the wave number that corresponds each respondent.

Thanks,

Joe

#3

Updated by Understanding Society User Support Team about 2 months ago

  • Status changed from New to In Progress
  • % Done changed from 0 to 20
  • Private changed from Yes to No
#4

Updated by Olena Kaminska about 2 months ago

Joe,

Thank you. Questions 14-16 here https://www.understandingsociety.ac.uk/wp-content/uploads/working-papers/2024-01.pdf can answer many of your questions.

If you are pooling cross-sectionally, you should use cross-sectional weights. In terms of levels LSOA should be your highest level, followed by psu, individual and time point. Use self-completion cross-sectional weights (preferably scaled) for your analysis. You probably would want to create a new weight variable and assign the relevant weight values to it.

Hope this helps,
Olena

#5

Updated by Joe Mattock about 2 months ago

Hi Olena,

Thank you very much for the response, all is clear on the data merging front.

I should clarify that I would be looking at the dependency of voting in earlier and subsequent waves. I have begun going through the online course regarding tailored weights. Specifically, I will be focusing on voting intention in waves 3, 6, and 9, and if you could provide some guidance on how I might approach this, that would be much appreciated.

Thanks as always,

Joe

#6

Updated by Joe Mattock about 2 months ago

I would also just like to ask if you would recommend any resources for coding tailored weights in R Studio?

Joe

#7

Updated by Olena Kaminska about 1 month ago

Joe,

In any software, in including R, you would just need a function for running a logistic regression in order to create a tailored weight. In your situation you may want to start at wave 3, and use the longitudinal wave 3 weight as your base weight.

Hope this helps,
Olena

Also available in: Atom PDF