Hello! I have been working with USA 1860 data and noticed that sometimes the hhwt is 99 instead of 100. Is that just an accounting reality from sampling? I guess I’m just concerned about how it will affect my data calculations. Can I treat households that are weighted 99 the same as households weighted 100 because they are “close enough?” Thanks!
Just a follow up question to this, is 99 a flag that something is wrong with those data entries?
A value of 99 for HHWT is not a data quality flag. I am looking into this potential issue with the weights and will get back to you when I have more information.
Thanks! Just wanted to check in to see if you found anything? Unfortunately, I have to figure this out before I can proceed with my research.
After looking through both the IPUMS USA and the IPUMS International versions of the samples, our best guess is that this is a rounding/accounting issue from sampling as you suggest. However, I haven’t been able to find documentation confirming this explicitly (though there is some indication that the version of the data available via IPUMS USA has a weight of 99.97 for all households prior to further modifications). I suggest running your analysis using both the provided weights and applying a weight of 100 to all households and ensuring that your results are robust across the change.