Support #2106

Selecting appropriate weights and using different weights

Added by Julia Diniz about 1 month ago. Updated 23 days ago.

In Progress
Start date:
% Done:



Hello! I am using data from the USoc to evaluate the impact of childcare on children's outcomes. In looking at the data and the weights, I have the following questions I was hoping to get some clarity on:

1. Is it okay to use different weights across different datafiles which will be appended (long format) and which information across datafiles will be used in the same analyses? For instance, the weights in datafiles child and indresp are different (e.g. indresp weights include indpxus_lw while child weights include psnenus_lw).

2. Why do some weights disappear after certain waves? For example, psnenub_xw disappears in wave 9 in the child datafile, but psnenui_xw appears - is it okay to use these different weights across waves? Equally if I am using the data longitudinally and cross-sectionally?

3. The USoc guidance says to use longitudinal weights if using more than one wave, and cross-sectional weights if using a single wave. If I want to use more than one wave but analyse them cross-sectionally, should I still use the longitudinal weights?

Thank you very much.



Updated by Understanding Society User Support Team 28 days ago

  • Category set to Weights
  • Status changed from New to In Progress
  • Assignee changed from Understanding Society User Support Team to Olena Kaminska
  • 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. We aim to respond to simple queries within 48 hours and more complex issues within 7 working days.

Best wishes,
Understanding Society User Support Team


Updated by Olena Kaminska 23 days ago


Thank you for your question.
1. You should use one weight in one model (or one type of weight in one model for pooled analysis). On pooling read questions 14 and 15 here:
2. Question 6.6 will answer how to select the xw weight:
The reason is that with each boost our cross-sectional representation improves - hence new weight.
3. This sounds like a pooled analysis. Use cross-sectional weights in this situation. Again, read here: for our advice on it.

Hope this helps,

Also available in: Atom PDF