METHOD
The 2018 release of IPUMS data was the bedrock of this project (Steven Ruggles, Sarah Flood, Ronald Goeken, Josiah Grover, Erin Meyer, Jose Pacas and Matthew Sobek. IPUMS USA: Version 9.0 [dataset]. Minneapolis, MN: IPUMS, 2019. https://doi.org/10.18128/D010.V9.0). Specifically, we selected the following variables from the IPUMS data: YEAR (Census year), SAMPLE (IPUMS sample identifier), SERIAL (Household serial number), CBSERIAL (Original Census Bureau household serial number), HHWT (Household weight), CLUSTER (Household cluster for variance estimation), STATEFIP (State (FIPS code)), PUMA (Public Use Microdata Area), STRATA (Household strata for variance estimation), GQ (Group quarters status), PERNUM (Person number in sample unit), PERWT (Person weight), BPL (Birthplace [general version]), and BPLD (Birthplace [detailed version]). For each PUMA, we calculated the percentage of each foreign-born population (e.g., for Brazil, Canada, Nigeria, etc.). We then examined each foreign-born population, and found the PUMA that contained its highest percentage, using this as a proxy of diaspora presence. This measure is certainly not perfect, and due to the variation in PUMA shape, we may have missed some important immigration hubs. Additionally, because this measure exclusively focuses on foreign-born individuals, second- and third-generation communities are less likely to be represented in our research.