Analysis of census participation rates by race/ethnicity

The Center for Urban Research at the City University of New York Graduate Center has analyzed 2010 Census “participation rates” through the second week of the Mail-out/Mail-back operation by key variables, including race and Hispanic origin (ethnicity), using its Census 2010 Hard To Count Interactive Mapping Website (www.CensusHardToCountMaps.org).

Key findings through week 2 include:

• Counties with higher percentages of White population have higher participation rates, on average, than counties with higher percentages of Black and Hispanic populations.

• Within counties, the correlation between greater percentages of Black population and lower participation rates is even stronger. At the census tract level, higher percentages of Blacks, Hispanics, and Asian Americans all were associated with lower participation rates.

• Several major cities stood out from these trends. For example, the association between higher percent Black population at the tract level and lower participation was much stronger in cities such as St. Louis, Boston, and Oakland (CA), while much weaker in cities such as Houston, Dallas, and Georgia.

• Detroit stood out altogether. Participation rates in Detroit tended to be lower in tracts with a greater percentage of Whites, and higher in tracts with a greater percentage of Blacks.

• In three cities with substantial Hispanic populations (Miami; Newark, NJ; and New York), higher percentages of Hispanics at the tract level were associated with higher participation rates.

The analysis continues to track the correlation between tract level hard-to-count (HTC) scores and mail response. The Center’s analysis through Week 1 of the mail-out/mail-back operation is posted at www.urbanresearch.org/news/resources/census2010participation.

The Mail-out/Mail-back phase ends on April 19; after that, the Census Bureau will begin to compile the addresses that census takers must visit in the massive door-to-door operation (Nonresponse Follow-Up), scheduled for May 1 – July 10. The Census Bureau might be able to remove some “late mail returns” received in the last week of April from the lists to be used in Nonresponse Follow-Up but cannot guarantee that census takers will not visit those households.

For further information:
Steven Romalewski (CUNY), 212-817-2033, SRomalewski@gc.cuny.edu
Terri Ann Lowenthal (FCI), 203-353-4364, TerriAnn2K@aol.com