# Calculating unemployment rate by state-year

**URL:** <https://forum.ipums.org/t/calculating-unemployment-rate-by-state-year/4770>\
**Category:** USA\
**Created:** [July 8, 2022, 9:14pm UTC](https://forum.ipums.org/t/calculating-unemployment-rate-by-state-year/4770 "2022-07-08T21:14:00Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![cj\_0001](https://avatars.discourse-cdn.com/v4/letter/c/dc4da7/32.png) [@cj\_0001](https://forum.ipums.org/u/cj_0001)\
**Post date:** [July 8, 2022, 9:14pm UTC](https://forum.ipums.org/t/calculating-unemployment-rate-by-state-year/4770/1 "2022-07-08T21:14:00Z")

</div>

I am trying to calculate unemployment rates for California and Nevada between 2001-2015 using ACS 1-year estimates; however, my estimates appear to be wildly different from conventional estimates found elsewhere. I know this is partly expected but the numbers seem to be way off. Before I share what I have, can someone confirm the standard procedure? Here’s what I did:

First, I filtered the data down to California and Nevada. Then, looking at each state-year grouping, using the EMPSTAT variable, I used the person-level weights to calculate the population at each level of the variable: (1) employed; (2) unemployed; (3) not in the labor force. I then created a rate by dividing (2) unemployed by the population in the labor force (by adding together the employed and unemployed population, that is, excluding the population not in the labor force).

For each state-year, the equation is: unemployment\_rate = unemployed/(employed + unemployed)

Are there other filters/factors that are typically considered when calculating unemployment rates, such as age, when using 1 year ACS estimates from IPUMS?

I appreciate any help you can provide.

---

<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:** [July 13, 2022, 9:38pm UTC](https://forum.ipums.org/t/calculating-unemployment-rate-by-state-year/4770/2 "2022-07-13T21:38:27Z")

</div>

That is generally the correct approach. Since the universe of EMPSTAT only includes individuals over age 16, same as the standard definition of unemployment, this shouldn’t be causing your problem. As you noted, it is expected that the estimates from ACS and CPS will differ due to different methodologies. You might find this paper [Comparison of ACS and CPS Data on Employment Status](https://www.census.gov/library/working-papers/2011/demo/SEHSD-WP2011-31.html) and [this fact sheet on Survey Differences in Labor Force Estimates](https://www.census.gov/topics/employment/labor-force/guidance/survey-differences.html) helpful.