Coding of x_jbxpcha in Waves 6 and 8
I am using the variable x_jbxpcha as part of a longitudinal study and noted that response options include a 'does not apply' option in Waves 6 and 8 (this option is not present in Waves 2 and 4) and have seen previous feedback on USoc User Support (Support #991) on this issue which suggested that this might apply to self-employed respondents. However, my sample is limited to employees only (i.e. x_jbsemp == 1) but I still have respondents who answered 'does not apply'. I am at a loss as to who these respondents are and whether they can be classified under 'yes' or 'no' in a binary variable.
#1 Updated by Stephanie Auty about 1 year ago
- Private changed from Yes to No
- % Done changed from 0 to 10
- Assignee set to Stephanie Auty
- Status changed from New to In Progress
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 Stephanie Auty about 1 year ago
- % Done changed from 10 to 20
- Assignee changed from Stephanie Auty to Nhlanhla Ndebele
Dear Nhlanhla Ndebele,
Please can you tell me in which wave you found this problem? I have looked at Waves 6 and 8, and if I restrict to individuals where w_jbsemp==1 then I do not find any inapplicable (-8) responses to w_jbxpcha.
#3 Updated by Nhlanhla Ndebele about 1 year ago
Thank you for the response, I think I was not quite clear with my question. The issue I had was why respondents might choose the 'doesn't apply' option for w_jbxpcha (e.g. might it be that they are at the top of their grade or band with no prospects for progression?), so their responses are valid and not missing. Previous views were that this category might be for self-employed respondents, however my sample excludes the self-employed (see attached document). If these were self-employed I do not expect a frequency for the 'doesn't apply' category. Rather than classifying them as missing and losing cases, the question is if I were to recode the variable into a binary variable would they be classified under 'yes' or 'no'? Some questions in the study have additional instructions that make it explicit how to classify respondents, e.g. for w_jbrise respondents at the top of their pay scale are classified as 'yes'.
How do I classify this group to include in meaningful analyses as 'doesn't apply' does not tell me much?
#4 Updated by Stephanie Auty about 1 year ago
- % Done changed from 20 to 80
- Status changed from In Progress to Feedback
Thanks for explaining.
We don't have any more information about why people may choose that option, but some possibilities are that they are already in the best job their employer offers for the type of work that they do, or they are an employee-director of their own company.