Revision of derived variables in new release
I've run my code with the latest version of the special licence data (waves 1-9) and I'm getting slightly different results than when I used the previous release of the data (waves 1-8).
I make use of a few derived variables in my code, which I understand may have been updated in the new release. Is there a list of derived variables that have been revised in the new release so that I can identify what is causing the difference in results?
Updated by Stephanie Auty almost 2 years ago
- Category set to Data documentation
- Status changed from New to Feedback
- Assignee set to Emily Harris
- % Done changed from 0 to 70
- Private changed from Yes to No
There is a document which lists changes since the last release. The most recent version is available on the website here: https://www.understandingsociety.ac.uk/documentation/mainstage/user-guide under the link Mainstage (Waves 1-9) - Revisions since last release. This document is also included in the mrdoc/pdf folder in your data download and named according to which waves it relates to, but the file name will include the word revisions.
Additionally, any variables that rely on age, sex or within-household relationships or use variables that use these are likely to change as we do cross-wave consistency checks and correct errors if and when they are identified. You can check which variables are used to calculate each derived variable in the derived variable notes in the online documentation, here: https://www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/term/derived-variables
Income variables are likely to change due to out imputation methods, and details of this can be found here: https://www.understandingsociety.ac.uk/sites/default/files/downloads/working-papers/2019-08.pdf