Mexico 2015 person and household weights are the same, why?

Hello, I was wondering if anyone knows why the Mexico 2015 person and household weights are all the same? We had been doing some analysis and using the harmonised variables our population estimates looked correct, now we were testing something with source variables and the estimates came out quite differently.
Can anyone give me any info on this?
Thanks!

Yes, I’ve been able to replicate your observation that HHWT==PERWT in the Mexico 2015 sample. I’ll have the IPUMS International Team look into this as soon as possible.

We like to reward users who bring errors such as this to our attention with an IPUMS coffee mug. To claim yours, please send your mailing address to ipums@umn.edu.

I am following up with some additional information after the IPUMS International Team has looked into this question.

IPUMS International received the 2015 Mexican sample that was roughly 19% of the total population. In processing this data, the IPUMS International Team systematically selected 1 in 2 households from this original sample data. Thus, the weights we offer in HHWT and PERWT are double the original expansion factor.

I suspect this is the reason for the discrepancy reported in this question. Please, let me know if you have any other questions.

PERWT should be equal to HHWT in the 2015 Mexican sample, as IPUMS International are sampling households and taking all persons from each household.

Hi Jeff,

To follow-up on the above question, I see the same for Mexico 2010 census as well. Is this because of same reason mentioned above?

Thanks,
Avinandan

Yes, for the 2010 Mexico Census, it is true that PERWT = HHWT for the same reason that Jeff mentioned earlier.

1 Like