# Creating kenya population estimates

**URL:** <https://forum.ipums.org/t/creating-kenya-population-estimates/1053>\
**Category:** INTERNATIONAL\
**Created:** [October 4, 2015, 3:55pm UTC](https://forum.ipums.org/t/creating-kenya-population-estimates/1053 "2015-10-04T15:55:00Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![margitti](https://avatars.discourse-cdn.com/v4/letter/m/b782af/32.png) [@margitti](https://forum.ipums.org/u/margitti)\
**Post date:** [October 4, 2015, 3:55pm UTC](https://forum.ipums.org/t/creating-kenya-population-estimates/1053/1 "2015-10-04T15:55:00Z")

</div>

Hi, I’m new to using and analyzing IPUMS data.

I’m trying to use IPUMS data for Kenya to estimate population densities by district in Kenya over time. (And am also interested in other variables in association with population).

Is there a way to use the IPUMS data to make population density estimates? I am guessing that I will have to arrange the data in Stata first by district, though I’m not sure how to do that.

Also, when exporting data from Stata into Excel or .cvs to be used in arcGIS, I know that there are case number limitations. Will this be solved if I arrange the data differenty (such as by district?)

Thanks so much

---

<div class="post-metadata">

**Author:** ![Tim\_Moreland](https://avatars.discourse-cdn.com/v4/letter/t/f1d935/32.png) [@Tim\_Moreland](https://forum.ipums.org/u/Tim_Moreland)\
**Post date:** [October 5, 2015, 9:44pm UTC](https://forum.ipums.org/t/creating-kenya-population-estimates/1053/2 "2015-10-05T21:44:00Z")

</div>

It is not possible to directly estimate district-level population density for Kenyan districts over time purely using IPUMS-International microdata. Rather, you can estimate the population of a district with IPUMS data and then use a non-IPUMS source to determine the area of the district. The population of a district is simply the sum of the [PERWT](https://international.ipums.org/international-action/variables/PERWT#codes_section) values for each respondent in that district.

[Note](https://international.ipums.org/international-action/variables/GEO2_KEX#comparability_section) that the boundaries of Kenyan districts have changed over time. The IPUMS district variable, [GEO2\_KEX](https://international.ipums.org/international-action/variables/GEO2_KEX#codes_section), harmonizes districts by name, but it does not account for the inconsistent boundaries. As a result, GEO2\_KEX is not perfectly comparable over time.

As for exporting from STATA to Excel for use in ArcGIS, yes, you will want to group your output by districts, instead of exporting one observation per respondent.

Hope this helps.