# matching "normal" IPUMS USA to linked census (1880-1900, 1880-1910, 1880-1920)

**URL:** <https://forum.ipums.org/t/matching-normal-ipums-usa-to-linked-census-1880-1900-1880-1910-1880-1920/820>\
**Category:** USA\
**Created:** [March 9, 2015, 1:12pm UTC](https://forum.ipums.org/t/matching-normal-ipums-usa-to-linked-census-1880-1900-1880-1910-1880-1920/820 "2015-03-09T13:12:00Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![ben1981](https://avatars.discourse-cdn.com/v4/letter/b/dec6dc/32.png) [@ben1981](https://forum.ipums.org/u/ben1981)\
**Post date:** [March 9, 2015, 1:12pm UTC](https://forum.ipums.org/t/matching-normal-ipums-usa-to-linked-census-1880-1900-1880-1910-1880-1920/820/1 "2015-03-09T13:12:00Z")

</div>

hello!

I would like to match the “normal” IPUMS Census data to the IPUMS linked samples (1880-1900, 1880-1910, 1880-1920).

Now my question:

i was assuming that an individual with a given serial and pernum in a given census year would be the same individual with that serial\_2 and pernum\_2 in the IPUMS linked sample.

By checking consistency (for example, checking whether age and age\_2 are the same after the match) I find that this seems indeed to be the case for 1910, but not for 1900 and 1920.

Is there a way to match these datasets in a reliable way?

Thanks for your help

---

<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:** [March 10, 2015, 6:20pm UTC](https://forum.ipums.org/t/matching-normal-ipums-usa-to-linked-census-1880-1900-1880-1910-1880-1920/820/2 "2015-03-10T18:20:00Z")

</div>

Unfortunately, it is not expected that these two files should match. IPUMS-USA considers the 1% samples and the linked samples as separate data projects; therefore, we do not maintain a link between the two. You should be able to find most of the variables contained in the 1% samples on the linked datasets.

Hope this helps.