# Can I think of strata as a (small) geographical unit?

**URL:** <https://forum.ipums.org/t/can-i-think-of-strata-as-a-small-geographical-unit/6581>\
**Category:** INTERNATIONAL\
**Created:** [June 9, 2025, 6:36am UTC](https://forum.ipums.org/t/can-i-think-of-strata-as-a-small-geographical-unit/6581 "2025-06-09T06:36:20Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Ming\_Fang](https://avatars.discourse-cdn.com/v4/letter/m/c2a13f/32.png) [@Ming\_Fang](https://forum.ipums.org/u/Ming_Fang)\
**Post date:** [June 9, 2025, 6:36am UTC](https://forum.ipums.org/t/can-i-think-of-strata-as-a-small-geographical-unit/6581/1 "2025-06-09T06:36:20Z")

</div>

This is related to this [previous post](https://forum.ipums.org/t/what-households-are-grouped-in-a-single-strata-in-ipums-international/1832), but my question is for another country (China) and specific about how we may use the strata variable.

My question is: when analyzing the 1990 Chinese census from IPUMS International, can I consider the variable strata as effectively the unique id of some small geographic units (such as a village or a small county, etc.).

Here is what I already learnt from the IPUMS documentation:  
(1) Description of the variable _strata_ (in my case _cn1990a\_strata_): This variable is the strata identifier for the sample. Strata is a constructed variable that captures the implicit geographic stratification resulting from the sample design. It is created by assigning a unique identifier to groups of between 10 and 19 adjacent households. Additional documentation is available on the Variance Estimation page.  
(2) In an excel file named “Strata” that describes the strata variables of each country/datasets, the description for China 1990 goes like this: CN1990A / China\_1990 / CN1990A\_0030 / No recoding. I presume that this means the variable _cn1990a\_strata_ comes from a source variable called _CN1990A\_0030_, without any recoding. However, when I try to search for more information for _CN1990A\_0030_, I cannot find out what its specific meaning is (i.e. is it a source variable which initially means village or county in the 1990 Chinese census?).

Thanks very much!

---

<div class="post-metadata">

**Author:** ![Ivan\_Strahof](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.ipums.org/ivan_strahof/32/626_2.png) [@Ivan\_Strahof](https://forum.ipums.org/u/Ivan_Strahof)\
**Post date:** [June 19, 2025, 2:59pm UTC](https://forum.ipums.org/t/can-i-think-of-strata-as-a-small-geographical-unit/6581/2 "2025-06-19T14:59:45Z")

</div>

[STRATA](https://international.ipums.org/international-action/variables/STRATA) is created by IPUMS-International by assigning a unique identifier to groups of geographically adjacent households in order to improve accuracy of standard error estimation. Households grouped into the same strata do not necessarily reside within the same small geographic area such as village or town; in the China 1990 sample I can only confirm that they reside in the same county ([GEO3\_CN1990](https://international.ipums.org/international-action/variables/GEO3_CN1990)). You can read more about the construction of strata in[this IPUMS working paper](https://international.ipums.org/international/resources/misc_docs/cleveland_davern_ruggles_variance.pdf).

Note that you can find the source variables (e.g., [CN1990A\_0030](https://international.ipums.org/international-action/variables/CN1990A_0030#description_section)) for each integrated variable (e.g., [STRATA](https://international.ipums.org/international-action/variables/STRATA#source_variables_section)) by referring to the integrated variable’s _Source Variables_ tab. In the case of STRATA, you can see in this tab that CN1990A\_STRATA (the name assigned to CN1990A\_0030) is the source variable for STRATA. In this case, the source variable was created by IPUMS-International and will be identical to the data in the integrated variable.