Skip Header

Guidance for Data Users

Component ID: #ti1739038380

Summary

The CPS ASEC provides the most timely and accurate national data on health insurance with income detail. Hence, it is the preferred source for national analysis. The ACS is preferred for subnational data on health insurance by detailed demographic characteristics, due to its large sample size. The Census Bureau recommends using the ACS for single-year estimates of health insurance at the state level. Users looking for consistent, state-level trends before 2008 should use the CPS ASEC.

For sub-state areas such as counties, users should consider their specific needs when picking the appropriate data source. The SAHIE program produces single-year estimates of number and percent insured/uninsured by some demographic and income groups. SAHIE has standard errors mostly smaller than direct survey estimates. Therefore, SAHIE is the preferred source if it produces estimates for all of your characteristics of interest. Users who need estimates of health insurance coverage by other characteristics such as occupation should use the ACS, where and when available.

The SIPP is the only Census-provided source of longitudinal health insurance coverage estimates. It provides national estimates, as well as estimates for some larger states. As SIPP collects monthly data over 2 ½ to 4 year panels, it is also a source of health insurance coverage estimates for a time period of more or less than one year.

Note that Figure 1 is a summary of general recommendations. There are a number of considerations users should take into account when selecting a data source. For example, it is generally advisable to use the same data source for a particular purpose across all levels of geography.

The National Health Interview Survey (NHIS) is collected by the Census Bureau and released by the National Center for Health Statistics (NCHS). The Census Bureau recommends the NHIS when examining health outcomes. Links to the NHIS and other surveys that currently collect or have collected health insurance data can be found at Related Sites. Please note the list of external surveys is not exhaustive.

More information on each data source is available in the U.S. Census Bureau Health Insurance Data Source Comparison table below.

Component ID: #ti614949475

The Census Bureau produces health insurance data from three surveys and one model-based program. Depending on your needs, one data source may be more suitable than another data source. Below are some basic differentiations between each source.

Component ID: #ti1739038376

ACS

  • is recommended for examining subnational data.
  • can be used to examine local-area estimates including counties, places, metropolitan areas, congressional districts, and other geographies.
    • The one-year ACS data is capable of generating estimates for geographies with populations of 65,000 or more.
    • The three-year ACS data is capable of generating estimates for those with 20,000 or more.
    • The five-year ACS data can be used for estimates at the sub-county level including census tracts and block groups in addition to all counties, places and other geographies.
  • began capturing health insurance data in 2008.

Component ID: #ti1739038377

CPS ASEC

  • is mainly useful for examining timely estimates of the insured and uninsured population at the national level.
  • can be used for state-level estimates, trends, and differences (through multi-year averages), although the large sampling errors of state-level data limit its usefulness.
  • is the most widely used source of data on health insurance coverage in the United States, with a consistent time series of estimates from 1999 to 2012 (i.e. 2000 CPS ASEC to the 2013 CPS ASEC) and another time series that begins with 2013 (i.e. 2014 CPS ASEC).

Component ID: #ti1739038375

SAHIE

  • is recommended for examining single-year recent estimates for all US counties.
  • provides estimates that are more current than the ACS five-year estimates.
  • increases the precision of state- and county-level coverage for detailed demographic groups, and adds additional income categories of relevance to changes in healthcare laws.
  • has an interactive data, mapping, and trending tool that allows data users to customize and export tables, maps, and charts.
  • provides estimates beginning in the year 2000.

Component ID: #ti1739038374

SIPP

  • is useful for examining monthly dynamics across time at the national level (such as how long a person remains uninsured, how many people obtain coverage, and any changes in a person’s coverage within a given year).
  • can be used to examine a limited number of sub-national geographies, including region and certain states.
  • provides rich detail on health insurance and many other topics extending back to 1984.

Component ID: #ti1739038378

The chart below summarizes the recommendations at various geographic levels:

Component ID: #ti1739038379

Figure 1. U.S. Census Bureau Health Insurance Data Source Recommendations

Geographic Level Cross-Sectional Estimates Longitudinal Estimates
United States CPS ASEC for income
ACS for detailed race groups
SIPP
State ACS/CPS ASEC 2-year averages 1 SIPP (selected states)
Substate (areas with populations of 65,000 or more) SAHIE for counties, else 
ACS 1-year period estimates
None
Substate (areas with populations of 20,000 to 65,000) SAHIE for counties, else
ACS  3-year period estimates2
None
Substate (areas with populations less than 20,000) SAHIE for counties, else
ACS  5-year period estimates 3
None
Footnote
1 Use CPS ASEC non-overlapping 2-year averages when examining state trends that include years prior to 2008.
2 ACS recommends using non-overlapping periods for trend analysis with multiyear estimates. For example, comparing 2005-2007 ACS 3-Year estimates with 2008-2010 ACS 3-Year estimates is preferred for identifying change.
3 ACS recommends using non-overlapping periods for trend analysis with multiyear estimates. For example, comparing 2006-2010 ACS 5-Year estimates with 2011-2015 ACS 5-Year estimates is preferred for identifying change.

X
  Is this page helpful?
Thumbs Up Image Yes    Thumbs Down Image No
X
Comments or suggestions?
No, thanks
255 characters remaining
X
Thank you for your feedback.
Comments or suggestions?
Back to Header