# Strata for CPS Survey

**URL:** <https://forum.ipums.org/t/strata-for-cps-survey/3945>\
**Category:** CPS\
**Created:** [January 25, 2021, 8:42pm UTC](https://forum.ipums.org/t/strata-for-cps-survey/3945 "2021-01-25T20:42:48Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![David\_Hudson](https://avatars.discourse-cdn.com/v4/letter/d/898d66/32.png) [@David\_Hudson](https://forum.ipums.org/u/David_Hudson)\
**Post date:** [January 25, 2021, 8:42pm UTC](https://forum.ipums.org/t/strata-for-cps-survey/3945/1 "2021-01-25T20:42:48Z")

</div>

I know strata variables are available for some of the IPUMS datasets, but I cannot find one for the CPS data. Is there one?

I am trying to use BSWEIGHTS in Stata to replicate earnwt for my CPS data for 2000-2012.

---

<div class="post-metadata">

**Author:** ![Matthew\_Bombyk](https://avatars.discourse-cdn.com/v4/letter/m/34f0e0/32.png) [@Matthew\_Bombyk](https://forum.ipums.org/u/Matthew_Bombyk)\
**Post date:** [January 26, 2021, 9:57pm UTC](https://forum.ipums.org/t/strata-for-cps-survey/3945/2 "2021-01-26T21:57:48Z")

</div>

There are no strata or cluster variables in the CPS public-use microdata. For ASEC samples starting in 2005, you can incorporate the CPS sampling design into your standard errors using [REPWTP](https://cps.ipums.org/cps-action/variables/repwtp#codes_section). You can find more detail on this process [here](https://cps.ipums.org/cps/repwt.shtml). The Census Bureau also provides replicate weights for some other CPS supplements. These are not available through IPUMS CPS, but can be downloaded from the Census Bureau website ([link](https://www.census.gov/data/datasets/time-series/demo/cps/cps-supp_cps-repwgt.html)).

For basic monthly samples, there is currently no way to incorporate the sample design into the standard errors. IPUMS testing has found that using replicate weights with ASEC data usually increases standard errors slightly, compared to ignoring the sampling design and just using ASECWT to calculate weighted standard errors. So a rule of thumb is that the standard errors calculated with basic monthly CPS data are likely to be slight underestimates.

---

<div class="post-metadata">

**Author:** ![David\_Hudson](https://avatars.discourse-cdn.com/v4/letter/d/898d66/32.png) [@David\_Hudson](https://forum.ipums.org/u/David_Hudson)\
**Post date:** [January 26, 2021, 10:44pm UTC](https://forum.ipums.org/t/strata-for-cps-survey/3945/3 "2021-01-26T22:44:58Z")

</div>

Thank you for taking the time to respond to me. I will take that into consideration.