newer way to get full USA ACS data by table and fips get the ACS 5year data for selected tables and fips or fipstype

Usage

get_acs_new(
  tables = ejscreen_acs_tables,
  fips = "blockgroup",
  yr = acsdefaultendyearhere,
  fiveorone = "5",
  return_list_not_merged = TRUE
)

Arguments

tables: vector of ACS data table numbers like "B01001" etc. Note some tables used by EJSCREEN are only available at tract resolution, namely "C16001" for detailed specific languages as % of residents, and "B18101" for % with disability. All resolutions get returned if return_list_not_merged=TRUE, but not if FALSE, since those tables would prevent clearcut merging to a single table of places based on fips.
fips: "blockgroups" for all US bg, or a vector of fips codes. can also be "county", "state", "tract", or vector of one of those fips code types. May support these but untested: "REGION", "American Indian Area/Alaska Native Area/Hawaiian Home Land", "MSA", "CSA", "Urban Area", "Congressional District", "ZCTA". If a fips type (e.g., "tract"), defines the SUMLEVEL variable in the ACS data (e.g., "140").
yr: end year of 5 year ACS summary file data, such as 2023 for the 2019-2023 survey released by Census Bureau Dec. 2024.
fiveorone: optional 1 or 5, where 5 is the 5-year sample - only 5-yr tested here
return_list_not_merged: set to FALSE means return a single merged table from all the requested ACS tables, and otherwise a list of data.tables. See "tables" parameter for more.

Value

list of tables or merged single table, with estimates and margins of error and fips and SUMLEVEL

Examples

 x = get_acs_new(yr=2022, tables = ejscreen_acs_tables[1],
   fips="county")
 x[[1]]

 # acs22 = get_acs_new(yr=2022, tables = ejscreen_acs_tables )
 # acs23 = get_acs_new(yr = 2023, return_list_not_merged = FALSE)

 if (FALSE) { # \dontrun{
   ##### EXAMPLE OF GETTING ACS DATA
   ##### FOR ALL US BLOCKGROUPS AND CALCULATING INDICATORS

   ### See more complete code for this in the EJAM package!
   ## -- below is just a very simplified look:

   library(EJAM)
   library(data.table)

   # x <- get_acs_new() # has problem where not all tables have same number of rows
   # even for the blockgroup ones, and last 2 tables are tract resolution
   ## so this is easier for getting the bg part:

   bg     <- get_acs_new(tables = ejscreen_acs_tables[1:13], return_list_not_merged = FALSE)

   acsdata <- list()
   acsdata <- EJAM::calc_ejam(
     bg,
     formulas = EJAM::formulas_ejscreen_acs$formula,
         keep.old = c("fips", "pop")
   )
   data.table::setnames(acsdata, "fips", "bgfips")

   #  dput(setdiff(names(acsdata) , names(blockgroupstats)) )

   keep <- intersect(names(acsdata), names(EJAM::blockgroupstats))
   acsdata <- acsdata[ , .SD, .SDcols = keep]

   t(acsdata[1:2,])

   # save(acsdata, file = "~/Downloads/acs2023 bg via just ACSdownload pkg example.rda")
 } # }