Inconsistency on Wave 2 related to Raceldv compared to other waves
First of all, congratulation for the fantastic database and space for discussion.
I have been working on an SEM using Ethnicity, Socioeconomic Status, and Health.
The earlier observed variable (Race) seems strange compared to other waves. When I restrict my analysis to employees above 60 years old ( b_dvage < 60 ) receiving a salary(b_payg), the racial distribution(b_racel) of white respondents drops significantly when compared to waves 1 and 3 to 5.
If you want to mimic, I further restrict it to non-proxy interviewees and thoroughly answered questionnaires.
I checked across all waves a model that worked flawlessly from wave 1 to 5. Was there a change in methodology for waves 6 and 7?
Once I finish my work I am happy to share it to you, if it interests.
#1 Updated by Stephanie Auty over 1 year ago
- Status changed from New to In Progress
- Assignee changed from Felippe Oliveira to Stephanie Auty
- % Done changed from 0 to 10
- Private changed from Yes to No
Many thanks for your enquiry. The Understanding Society team is looking into it and we will get back to you as soon as we can.
Stephanie Auty - Understanding Society User Support Officer
#2 Updated by Alita Nandi over 1 year ago
- Assignee changed from Stephanie Auty to Felippe Oliveira
- % Done changed from 10 to 80
The ethnic group information is collected only once for each adult (16+), the first time that person is interviewed. So, while this information should be available for every adult who responded in Wave 1 (except for those who said refused/don't know), in Wave 2 this will only be available for those who were interviewed for the first time in that wave. And so on for every subsequent wave. We combine all this information collected from every wave and produce a variable RACEL_DV available in XWAVEDAT.
Have you used the XWAVEDAT file before? This file includes all time invariant information collected across all waves for every individual in the sample. For example, SEX_DV, BIRTHY, RACEL_DV, BORNUK_DV etc. This is an extremely useful file! You can merge the data from XWAVEDAT with W_INDRESP files using PIDP.
About Wave 6 & 7: In Wave 6 we added a new sample, the Immigrant and Ethnic Minority Boost sample. Could this be the reason? Try running your model without this sample which can be identified by W_HHORIG = 8. You can find out about the different samples in
the FAQ: https://www.understandingsociety.ac.uk/help/faqs Look for "What are the different sub-samples?"
the Get started page: https://www.understandingsociety.ac.uk/help/getting-started
the Main User Guide: https://www.understandingsociety.ac.uk/sites/default/files/downloads/documentation/mainstage/user-guides/mainstage-waves-1-7-user-guide.pdf