2019 Citizen Voting Age Population (CVAP) Data for Georgia from the American Community Survey (ACS) 5-Year Estimates (2015-2019) Disaggregated to 2020 Census Blocks ##Redistricting Data Hub (RDH) Retrieval Date 05/13/21 for 2019 CVAP Data 10/16/21 for 2010 Public Law 94-171 Data 04/14/21 for Voting Election and Science Team (VEST) Crosswalk Tabulations ##Sources CVAP Data retrieved from the Census Citizen Voting Age Population by Race and Ethnicity website: https://www.census.gov/programs-surveys/decennial-census/about/voting-rights/cvap.2019.html Public Law 94-171 Data was retrieved using the Census API, variables are available here: https://api.census.gov/data/2010/dec/pl/variables.html VEST Crosswalk tabulations were retrieved here: https://dataverse.harvard.edu/file.xhtml?fileId=4513650&version=2.0 ##Fields Field Name Description GEOID20 Unique Identifier CVAP_TOT19 CVAP Estimate for Total CVAP_NHS19 CVAP Estimate for Not Hispanic or Latino CVAP_AIA19 CVAP Estimate for American Indian or Alaska Native Alone or In Combination CVAP_ASN19 CVAP Estimate for Asian Alone or In Combination CVAP_BLK19 CVAP Estimate for Black or African American Alone or In Combination CVAP_NHP19 CVAP Estimate for Native Hawaiian or Other Pacific Islander Alone CVAP_WHT19 CVAP Estimate for White Alone CVAP_AIW19 CVAP Estimate for American Indian or Alaska Native and White CVAP_ASW19 CVAP Estimate for Asian and White CVAP_BLW19 CVAP Estimate for Black or African American and White CVAP_AIB19 CVAP Estimate for American Indian or Alaska Native and Black or African American CVAP_2OM19 CVAP Estimate for Remainder of Two or More Race Responses CVAP_HSP19 CVAP Estimate for Hispanic or Latino C_TOT19 Citizen Estimate for Total C_NHS19 Citizen Estimate for Not Hispanic or Latino C_AIA19 Citizen Estimate for American Indian or Alaska Native Alone or In Combination C_ASN19 Citizen Estimate for Asian Alone or In Combination C_BLK19 Citizen Estimate for Black or African American Alone or In Combination C_NHP19 Citizen Estimate for Native Hawaiian or Other Pacific Islander Alone C_WHT19 Citizen Estimate for White Alone C_AIW19 Citizen Estimate for American Indian or Alaska Native and White C_ASW19 Citizen Estimate for Asian and White C_BLW19 Citizen Estimate for Black or African American and White C_AIB19 Citizen Estimate for American Indian or Alaska Native and Black or African American C_2OM19 Citizen Estimate for Remainder of Two or More Race Responses C_HSP19 Citizen Estimate for Hispanic or Latino ##Processing The processing for this dataset was completed in multiple parts. Part 1: Processing the 2019 Block Group CVAP data CVAP data was retrieved at the block group level from link in "Sources" above. The Georgia data was extracted from the national data to create a statewide dataset. The data was pivoted from narrow to wide data based on GEOIDs so that one row is one block group, and each field represents a particular race/ethnicity. The fields were renamed to fit character length requirements. Processing was primarily completed using the pandas library. To improve the usefulness of the data, we have modified three categories to correspond with the Office of Management and Budget (OMB) racial categories. The "Alone" categories for American Indian or Alaska Native (fields with "AIA"), Asian (fields with "ASN"), and Black or African American (fields with "BLK") represent an encompassing racial category that is inclusive of all categories that include that race. For example, CVAP_AIA19 would be the sum of the original "Alone" CVAP_AIA19, CVAP_AIB19, and CVAP_AIW19. For CVAP_BLK19, the field would be the sum of the original CVAP_BLK19, CVAP_AIB19, and CVAP_BLW19. For CVAP_ASN19, the field would be the sum of the original CVAP_ASN19 and CVAP_ASW19 These fields are also noted in the description as being "Alone or In Combination". No other estimate categories were modified. All of the racial estimates provided are Non-Hispanic. Breakdowns for Hispanic/Non-Hispanic by race are not provided in the CVAP special tabulation. Part 2: Disaggregating the 2019 Block Group CVAP data to 2010 Blocks The 2019 CVAP data was disaggregated using 2010 PL 94-171 data. Using the Census API, the following variables were retrieved (some were summed, as indicated, to better resemble the OMB categories the CVAP modifications): P004001 CVAPTOT19 P004003 CVAP_NHS19 P004007 CVAP_AIA19 P004008 CVAP_ASN19 P004006 CVAP_BLK19 P004009 CVAP_NHP19 P004005 CVAP_WHT19 P004014 CVAP_AIW19 P004015 CVAP_ASW19 P004013 CVAP_BLW19 P004018 CVAP_AIB19 P004011 CVAP_2OM19 P004002 CVAP_HSP19 P002001 C_TOT19 P002003 C_NHS19 P002007 C_AIA19 P002008 C_ASN19 P002006 C_BLK19 P002009 C_NHP19 P002005 C_WHT19 P002014 C_AIW19 P002015 C_ASW19 P002013 C_BLW19 P002018 C_AIB19 P002011 C_2OM19 P002002 C_HSP19 After retrieving the data from the API, the OMB categories were created using the same method as the 2019 block group CVAP data. For each block in each column (e.g. CVAP_TOT19) the value for the 2010 block retrieved from the API was multiplied by the 2010 total for the block group. The totals for the block group were retrieved by grouping the block data by block group ID. This provides a ratio of block to block group population for each individual racial category. The ratios for each block in each column were multiplied by the 2019 CVAP total for the block's corresponding block group. In the scenario where no blocks in a block group contain a value for a particular race/ethnicity combination in 2010, but have a value in the 2019 estimates, total CVAP population or total citizen estimate population were used as a proxy, depending on the variable. In the scenario where no blocks in a block group contain a value for either the C_TOT19 category or CVAP_TOT19 categories, the 2019 population is divided evenly amongst all blocks in the block group. Part 3: Transforming 2010 Blocks to 2020 Blocks To translate the 2010 blocks with 2019 CVAP data to 2020 blocks, the VEST crosswalks were used (see link above in source for more information). For each 2020 block, a weighted dictionary was retrieved from the 2010 data using the crosswalk tabulations. For example, 2020 Block 001 could have a weighted dictionary that may look like {001: 1.0, 002: 0.5} where for all columns, the 2020 Block 001 should be assigned the entirety of the 2010 Block 001 and half of 2010 Block 002. The results were rounded to the nearest hundredth. All results were verified throughout the process to ensure that data was not being omitted and the totals of disaggregated data on 2020 blocks is nearly the same as the original 2019 block group totals (very slight differences may occur due to the lack of rounding until the final product). ##Additional Notes For more information on the ACS CVAP documentation please refer to the ACS link above in Sources as well as the ACS technical documentation included in this folder (also available at this link: https://www2.census.gov/programs-surveys/decennial/rdo/technical-documentation/special-tabulation/CVAP_2015-2019_ACS_documentation.pdf ). For more information on OMB racial categories, please see this link: https://obamawhitehouse.archives.gov/omb/bulletins_b00-02/#n_1_ For more information on VEST crosswalk tabulations, please refer to their documentation here (note that the files used in this process is block10block20_crosswalks.zip): https://dataverse.harvard.edu/file.xhtml?fileId=4575858&version=2.0 Visit the RDH GitHub and the processing script for this code here: https://github.com/nonpartisan-redistricting-datahub/ Please direct questions related to processing this dataset to info@redistrictingdatahub.org.