This function lets you easily compute differences in conditional expectations between all pairs of specified racial groups.
Arguments
- x
A
birdiemodel object.- subgroup
If
TRUE, return subgroup-level (rather than marginal) disparity estimates.- races
A character vector of racial groups to compute disparities for. The special value
TRUE, the default, computes disparities for all racial groups.
Value
A data frame containing a row with every possible disparity for the
specified races, which are identified by columns race_1 and race_2.
The reported disparity is estimate_1 - estimate_2.
Examples
data(pseudo_vf)
r_probs = bisg(~ nm(last_name) + zip(zip), data=pseudo_vf)
fit = birdie(r_probs, turnout ~ 1, data=pseudo_vf)
disparities(fit)
#> # A tibble: 60 × 6
#> race_1 race_2 turnout estimate_1 estimate_2 disparity
#> <chr> <chr> <chr> <dbl> <dbl> <dbl>
#> 1 black white no 0.358 0.301 0.0573
#> 2 black white yes 0.642 0.699 -0.0573
#> 3 hisp white no 0.392 0.301 0.0908
#> 4 hisp white yes 0.608 0.699 -0.0908
#> 5 asian white no 0.613 0.301 0.312
#> 6 asian white yes 0.387 0.699 -0.312
#> 7 aian white no 0.778 0.301 0.477
#> 8 aian white yes 0.222 0.699 -0.477
#> 9 other white no 0.254 0.301 -0.0468
#> 10 other white yes 0.746 0.699 0.0468
#> # ℹ 50 more rows
disparities(fit, races=c("white", "black"))
#> # A tibble: 4 × 6
#> race_1 race_2 turnout estimate_1 estimate_2 disparity
#> <chr> <chr> <chr> <dbl> <dbl> <dbl>
#> 1 black white no 0.358 0.301 0.0573
#> 2 black white yes 0.642 0.699 -0.0573
#> 3 white black no 0.301 0.358 -0.0573
#> 4 white black yes 0.699 0.642 0.0573