Calculate Diversity and Segregation Indices

Implements common measures of diversity and spatial segregation. This package has tools to compute the majority of measures are reviewed in Massey and Denton (1988) . Multiple common measures of within-geography diversity are implemented as well. All functions operate on data frames with a 'tidyselect' based workflow.

divseg

divseg implements common measures of diversity (within-geography) and segregation (across-geographies).

Installation

You can install the released version of divseg from CRAN with:

install.packages("divseg")

You can install the released version of divseg from GitHub with:

pak::pak('christopherkenny/divseg')

Example

The basic workflow relies on a tibble where each row represents a geography and has columns that represent some form of population data.

library(divseg)
#> 
#> Attaching package: 'divseg'
#> The following object is masked from 'package:base':
#> 
#>     interaction

divseg comes with two example datasets. de_county contains 2010 Census data on the counties in Delaware. de_tract likewise has 2010 Census data on the tracts in Delaware.

data('de_county')
data('de_tract')

A pretty standard function call returns a vector, where the first entry is a tibble and the second is tidyselect language.

ds_blau(.data = de_county, .cols = starts_with('pop_'))
#> [1] 0.5155228 0.5570435 0.4052769

More importantly, if you specify an argument to .name, all functions are pipe-able.

de_county |>
  ds_blau(starts_with('pop_'), .name = 'blau') |>
  ds_delta(starts_with('pop_'), .name = 'delta') |>
  dplyr::relocate(blau, delta)
#> Simple feature collection with 3 features and 22 fields
#> Geometry type: MULTIPOLYGON
#> Dimension:     XY
#> Bounding box:  xmin: -75.78866 ymin: 38.45101 xmax: -75.04894 ymax: 39.83901
#> Geodetic CRS:  NAD83
#> # A tibble: 3 × 23
#>    blau delta GEOID NAME     pop pop_white pop_black pop_hisp pop_aian pop_asian
#>   <dbl> <dbl> <chr> <chr>  <dbl>     <dbl>     <dbl>    <dbl>    <dbl>     <dbl>
#> 1 0.516 0.345 10001 Kent… 162310    105891     37812     9346      916      3266
#> 2 0.557 0.345 10003 New … 538479    331836    124426    46921      984     23132
#> 3 0.405 0.345 10005 Suss… 197145    149025     24544    16954      924      1910
#> # ℹ 13 more variables: pop_nhpi <dbl>, pop_other <dbl>, pop_two <dbl>,
#> #   vap <dbl>, vap_white <dbl>, vap_black <dbl>, vap_hisp <dbl>,
#> #   vap_aian <dbl>, vap_asian <dbl>, vap_nhpi <dbl>, vap_other <dbl>,
#> #   vap_two <dbl>, geometry <MULTIPOLYGON [°]>

Each function has a partner that can go inside calls to dplyr::mutate() by dropping the ds_ prefix:

de_county |>
  dplyr::mutate(herf = hhi(starts_with('pop_'))) |>
  dplyr::relocate(herf)
#> Simple feature collection with 3 features and 21 fields
#> Geometry type: MULTIPOLYGON
#> Dimension:     XY
#> Bounding box:  xmin: -75.78866 ymin: 38.45101 xmax: -75.04894 ymax: 39.83901
#> Geodetic CRS:  NAD83
#> # A tibble: 3 × 22
#>    herf GEOID NAME           pop pop_white pop_black pop_hisp pop_aian pop_asian
#>   <dbl> <chr> <chr>        <dbl>     <dbl>     <dbl>    <dbl>    <dbl>     <dbl>
#> 1 0.484 10001 Kent Count… 162310    105891     37812     9346      916      3266
#> 2 0.443 10003 New Castle… 538479    331836    124426    46921      984     23132
#> 3 0.595 10005 Sussex Cou… 197145    149025     24544    16954      924      1910
#> # ℹ 13 more variables: pop_nhpi <dbl>, pop_other <dbl>, pop_two <dbl>,
#> #   vap <dbl>, vap_white <dbl>, vap_black <dbl>, vap_hisp <dbl>,
#> #   vap_aian <dbl>, vap_asian <dbl>, vap_nhpi <dbl>, vap_other <dbl>,
#> #   vap_two <dbl>, geometry <MULTIPOLYGON [°]>