I am attempting to merge the youth data from waves 1, 3 and 5. Is it satisfactory to merge these waves of data via spss using the pidp identifier?. Or do I need to include anything further?. I am hoping to examine sdq scores over a series of times points. Im also curious of your suggestions in regard to weighting variables for this type of analysis?. Any information would be greatly appreciated.
Updated by Alita Nandi over 4 years ago
- Status changed from New to Feedback
- Assignee set to Vicky Thompson
- Target version set to X M
- % Done changed from 0 to 90
- Private changed from Yes to No
Yes, it is ok to merge the data using PIDP. There is a known issue with SDQ scores due to a new missing value code of 9 being introduced in Wave 7. This affects other variables in the youth questionnaire in Wave 7 as well. So, please check. Please see post https://iserswww.essex.ac.uk/support/issues/890. We are aware of this issue and the correction will be included in the next release.
Updated by Vicky Thompson over 4 years ago
Thanks for your response. I will merge the waves via the pidp. I had noticed there were some issues with wave 7 SDQ scores. Thanks for confirming this. Do you have any suggestions in regards to which weighting variables would be best for the above analysis?.
Updated by Stephanie Auty over 4 years ago
Apologies for the delay in responding.
There is infomation regarding choosing a weight in the Mainstage User Guide section 3.9.1. You can download the User Guide here: https://www.understandingsociety.ac.uk/documentation/mainstage/user-guide
Understanding Society User Support Officer
Updated by Olena Kaminska over 4 years ago
This is a response with regard to weighting. I am not sure about how you are doing your analysis.
If you are using the data in a longitudinal way and you are looking only at youth questionnaire between waves 1 and 5 then you will have only one year group who will be in your analysis: children you were 10 years old at wave 1 and became 15 at wave 5. For this analysis we don't provide specific weights but you should use enumeration weight with possible additional nonresponse correction (conditional on positive longitudinal enumeration weight from wave 5).
Alternatively if you are pooling information from the 3 waves then you should use cross-sectional weights for youth provided.
Let me know if you have further questions,