- Fixed typo that prevented geometry from accompanying
get_adi()
results. - Added escape hatches for examples and tests of functions requiring
USpopcenters
andgeosphere
. - Updated email address of maintainer.
- Added
synthetic_population()
- Fixed
geography = "zcta"
behavior so that it will honor non-NULL
values ofstate
. - Fixed incorrect message that only 1 or 2 calls to
tidycensus
functions were going to occur when in fact many might occur. - Removed
shift_geo
as a formal argument from all functions. The user can still pass this via...
. - Removed
censusapi
fromImports
- 2000 decennial ADI has been added back because the US census has restored the data to its API. 1990 ADI is still unavailable.
- Added functions
areas_in_radius()
,closest_n_areas()
,closest_population()
,lon_lat_from_area()
, andappend_dissimilarities()
- Fixed 2015 and 2016 block group pulls
Note that get_adi()
will not work with 1990 nor 2000 Decennial Census data until the US Census restores the availability of those data.
- The 3-factor split of ADI (i.e., the "Berg indices," or "ADI-3") have been incorporated and are now automatically included in
get_adi()
andcalculate_adi()
output. - There is no longer a default for the
year
argument. This will break old code that relied on the default ofyear = 2017
.
- Updated README and documentation
- Lots of comments added
- Minor code fixes for clarity and greater ease in debugging
- Updated versions of Imports and Suggests
- The
seed
argument had been essentially ignored up until now, sinceseed = 500
was hard-coded into the call tomice::mice()
withinsociome::calculate_adi()
. User-specifiedseed
values incalculate_adi()
andsociome::get_adi()
are now properly passed tomice()
. Users desiring to recover old results should specifyseed = 500
.
get_adi()
now pulls the correct variable for "civilian females age 16+ in white-collar occupations" for pre-2010 ACS estimates (C24010_040 instead of C24010_039).
- Accommodations for new CRAN rules on referencing help pages in other packages.
- Since examples wrapped in
\donttest{}
are now run during CRAN checks, all instances were changed to\dontrun{}
.
- Fixed bug wherein
get_adi(dataset = "decennial", year = 2010)
calls did not work unlessgeography = "tract"
- Fixed bug wherein
get_adi(dataset = "decennial", geography = "tract", year = 2010)
returned [incorrect] results for all tracts in all specified states, irrespective of what was passed to thegeoid
orcounty
arguments.
- Added
seed
argument toget_adi()
andcalculate_adi()
. - Code improvements
- Simplified testing
- Made
tidycensus
0.9.2 the minimum version, now that 0.9.6 is on CRAN. Both those versions are compatible, but 0.9.5 is not.
- Currently, the latest CRAN version of
tidycensus
, 0.9.5, is incompatible withsociome
due to atidycensus
bug. This bug is fixed in version 0.9.6, which is available on Github, but in order to accommodate new users ofsociome
, the older CRAN version, 0.9.2, will be required instead. This will be changed once the latest version oftidycensus
becomes available on CRAN.
- Removed reference to now-nonexistent
tigris-package
help page to avoid CRAN warning.
- Makes 2015 and 2016 block group calls use median household income (B19013_001) instead of median family income (B19113_001), which is unavailable at that level of geography during those years.
- Fixed tract-level 2010 decennial calls so that they go county-by-county for the ACS-data-gathering portions
- Made it so
get_adi(dataset = "decennial", state = NULL)
would work.
- Adapted to tidyr's breaking changes
- Added progress messages for the get_adi() calls that require many calls to tidycensus functions.
- Code improvements.
- Households that have zero households are now excluded from ADI calculation
- Updated README and documentation
- Added more tests
- Fixed bug with 1990 decennial calls.
- Added
install.packages("sociome")
to README's installation instructions. - Added CRAN version number and download counter badges to README.
- Made the
geometry
argumentFALSE
by default. - Overhauled internal code to accommodate
dataset = "decennial"
. - Added tests.
- Documentation and README improvements.
- Added "try again" feature and improved calls to tidycensus functions.
- Fixed ACS variable selection for get_adi().
- Made calculate_adi() an exported function again.
- Added warning() that
dataset = "decennial"
is under development.
- Made calls to tidycensus functions always broken up by state to accommodate its inconsistent behavior depending on geography and year.
- ZCTAs now supported, with the addition of
zcta
argument.
- Better code efficiency
checkmate
package no longer an Import.- Fixed bug in which
cache_tables
argument had been ignored
- Created the function get_geoids(), which allows users to quickly create a table of state, county, tract, block group, and block GEOIDs.
- Made calculate_adi() an internal function (no longer exported)
- Made tidycensus 0.9.0 now required (added to Imports)
- Updated README.md.
- Bug fixes.
- Now using psych::principal() instead of psych::fa() under the hood.
- User is warned if attempting to use get_adi() to calculate ADI values for under 30 locations.
- README.md has been updated to reflect the above and has generally been refreshed.
Successfully produces area deprivation indices (ADIs) for different US census designations based on user specifications.