Social Capital Project
Recent PostsThe Numbers Behind the Opioid Crisis: Revised Utah Edition Inactive, Disconnected, and Ailing: A Portrait of Prime-Age Men Out of the Labor Force All the Lonely Americans? Visualizing the Distribution of Social Capital across America Supplemental Data for Social Capital Index The Geography of Social Capital in America Rising Unwed Pregnancy and Childbearing across Educational and Racial Groups Further Thoughts on Volunteerism Trends and Data Issues A Future Without Kin? Love, Marriage, and the Baby Carriage: The Rise in Unwed Childbearing
About the Project
Apr 23 2018
Vice Chairman’s Staff, Joint Economic Committee
The Social Capital Project is pleased to provide new social capital index and subindex estimates for metropolitan and micropolitan areas (Core Based Statistical Areas, or CBSAs) and for commuting zones (CZs).
CBSAs are agglomerations of counties. They include a dense core of at least 10,000 people plus neighboring counties that are linked to the core as evidenced by commuting patterns. Metropolitan areas have a core with at least 50,000 people; the rest are micropolitan areas. CZs were defined by Tolbert and Sizer (1996)  using 1990 census data to group counties together that are bound by commuting patterns. Unlike CBSAs, they include rural areas, covering the entire United States. CZs were recently popularized by Raj Chetty and his colleagues in a series of papers using “big data” to study intergenerational mobility.
We obtained a county-to-CBSA crosswalk file from the Missouri Census Data Center, using their online MABLE/Geocorr14 tool. The crosswalk maps counties as defined in 2014 to CBSAs as defined in 2010. The county-to-CZ crosswalk file is from Chetty’s Equality of Opportunity Project. It maps counties as defined in 2010 to CZs as defined in 1990. We made small adjustments in five instances to address subsequent changes to county definitions.
Since counties aggregate directly into CBSAs and CZs, we estimate social capital scores and subindex scores as the population-weighted average of the county-level scores across a CBSA’s or CZ’s counties. We use the 2012-16 population estimate for each county from the American Community Survey 2016 5-year estimates. If any county within a CBSA or CZ has a missing score, the entire index or subindex is set to missing.
Methodological details regarding the county social capital index and subindex scores are available on the Social Capital Project Social Capital Index website.
1 Charles M. Tolbert and Molly Sizer, “U.S. Commuting Zones and Labor Market Areas: A 1990 update.” Economic Research Service Staff Paper 9614 (1996).
2 See, e.g., Raj Chetty, Nathaniel Hendren, Patrick Kline, and Emmanuel Saez, “Where is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States,” Quarterly Journal of Economics 129(4): 1553-1623 (2014).
4 See http://www.equality-of-opportunity.org/data/descriptive/replicate_2014.zip. The crosswalk file is “county_2010” and is found within the zipped archive in the directory, \clean_public_data\crosswalks.
5 Wade Hampton Census Area, Alaska was changed to Kusilvak Census Area, Alaska in 2015, and the FIPS code changed. Shannon County, South Dakota was changed to Oglala Lakota County, South Dakota in 2015, and the FIPS code changed. Bedford (independent) city, Virginia was incorporated into Bedford County, Virginia in 2013. We combined the data for these two areas.
In 2013, parts of Petersburg Census Area, Alaska and Hoonah-Angoon Census Area, Alaska were combined to form Petersburg Borough, Alaska. However the portion of Hoonah-Angoon Census Area that was added to Petersburg Census Area was negligible. About 15 percent of the population of Petersburg Census Area was moved to Prince of Wales-Hyder Census Area, Alaska. This division of the Petersburg Census Area affects the social capital index estimates for the Juneau and Ketchikan CZs, so we give these CZs missing values.
6 American FactFinder Table DP05; universe is the civilian noninstitutionalized population.