# 2018 CPS ASEC Health Insurance Variables Don't Match Census Report

**URL:** <https://forum.ipums.org/t/2018-cps-asec-health-insurance-variables-dont-match-census-report/4920>\
**Category:** CPS\
**Created:** [October 20, 2022, 1:30pm UTC](https://forum.ipums.org/t/2018-cps-asec-health-insurance-variables-dont-match-census-report/4920 "2022-10-20T13:30:34Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Michael\_Evangelist1](https://avatars.discourse-cdn.com/v4/letter/m/b38774/32.png) [@Michael\_Evangelist1](https://forum.ipums.org/u/Michael_Evangelist1)\
**Post date:** [October 20, 2022, 1:30pm UTC](https://forum.ipums.org/t/2018-cps-asec-health-insurance-variables-dont-match-census-report/4920/1 "2022-10-20T13:30:34Z")

</div>

Hello, I have a question about matching the 2018 CPS ASEC Health Insurance population counts with the published totals in Table 1 of [Health Insurance Coverage in the United States: 2017](https://www.census.gov/content/dam/Census/library/publications/2018/demo/p60-264.pdf).

I can replicate every row for the 2017 column of Table 1 except for Employer-based (181,036) and Direct-purchase (51,821) coverage. In both cases, my totals are substantially less (172,647 and  
50,607, respectively).

However, if I use my employer-based and direct-purchase variables with the variable _out_, I am able to match the “any private plan” total. I suspect that the Census may allocate some respondents in the _out_ category back to the employer-based and direct-purchase totals, but I didn’t see any mention of this in the documentation.

I was hoping someone may know how to match the employer-based and direct-purchase totals exactly. Here is my code.

gen hi\_employer = 0  
replace hi\_employer = 1 if grpdeply == 2 | grpownly == 2 | grpoutly == 2 | grpcoutly == 2

gen hi\_direct = 0  
replace hi\_direct = 1 if dpdeply == 2 | dpownly == 2 | dpoutly == 2 | dpcoutly == 2

gen hi\_medicare = 0  
replace hi\_medicare = 1 if carely == 2

gen hi\_medicaid = 0  
replace hi\_medicaid = 1 if himcaidly == 2

gen hi\_va = 0  
replace hi\_va = 1 if hichamp == 2

gen hi\_private = 0  
replace hi\_private = 1 if (grpdeply == 2 | grpownly == 2 | grpoutly == 2 | grpcoutly == 2) | (dpdeply == 2 | dpownly == 2 | dpoutly == 2 | dpcoutly == 2) | out == 2

gen hi\_public = 0  
replace hi\_public = 1 if carely == 2 | himcaidly == 2 | hichamp == 2

gen hi\_unins = 0  
replace hi\_unins = 1 if hi\_private == 0 & hi\_public == 0

\* Matches Census Table 1  
table hi\_private if year == 2018 [pw=asecwt]  
table hi\_public if year == 2018 [pw=asecwt]  
table hi\_unins if year == 2018 [pw=asecwt]  
table hi\_medicare if year == 2018 [pw=asecwt]  
table hi\_medicaid if year == 2018 [pw=asecwt]  
table hi\_va if year == 2018 [pw=asecwt]

\* Does not Match Census Table 1  
table hi\_employer if year == 2018 [pw=asecwt]  
table hi\_direct if year == 2018 [pw=asecwt]