To be used ideally with emmeans::contrast. Each contrasts is tested (sum to 0?) and scaled so that all positive weights sum to 1 (and all negative weights to -1).
Usage
contrast_weights(..., .name = "custom", .adjust = NULL)
cw(..., .name = "custom", .adjust = NULL)Arguments
- ...
Can be:
Unnamed scalars.
(Possibly named) vectors if equal length
- .name
The label as it will appear in the results in
emmeans.- .adjust
Gives the default adjustment method for multiplicity (used in
emmeans).
Examples
data(mtcars)
mod <- lm(mpg ~ factor(cyl) * am, mtcars)
my_contrasts <- data.frame(
"squares" = c(-1, 2, -1),
"4 vs 6" = c(-30, 30, 0),
check.names = FALSE
)
(my_contrasts2 <- cw(my_contrasts))
#> squares 4 vs 6
#> 1 -0.5 -1
#> 2 1.0 1
#> 3 -0.5 0
my_contrasts3 <- cw(my_contrasts, .adjust = "fdr")
library(emmeans)
(emms <- emmeans(mod, ~ cyl + am))
#> cyl am emmean SE df lower.CL upper.CL
#> 4 0 22.9 1.751 26 19.3 26.5
#> 6 0 19.1 1.516 26 16.0 22.2
#> 8 0 15.1 0.875 26 13.3 16.8
#> 4 1 28.1 1.072 26 25.9 30.3
#> 6 1 20.6 1.751 26 17.0 24.2
#> 8 1 15.4 2.144 26 11.0 19.8
#>
#> Confidence level used: 0.95
contrast(emms, method = my_contrasts, by = "am")
#> am = 0:
#> contrast estimate SE df t.ratio p.value
#> squares 0.30 3.61 26 0.083 0.9344
#> 4 vs 6 -113.25 69.48 26 -1.630 0.1152
#>
#> am = 1:
#> contrast estimate SE df t.ratio p.value
#> squares -2.34 4.24 26 -0.552 0.5858
#> 4 vs 6 -225.25 61.59 26 -3.658 0.0011
#>
contrast(emms, method = my_contrasts2, by = "am") # estimate is affected!
#> am = 0:
#> contrast estimate SE df t.ratio p.value
#> squares 0.15 1.80 26 0.083 0.9344
#> 4 vs 6 -3.77 2.32 26 -1.630 0.1152
#>
#> am = 1:
#> contrast estimate SE df t.ratio p.value
#> squares -1.17 2.12 26 -0.552 0.5858
#> 4 vs 6 -7.51 2.05 26 -3.658 0.0011
#>
contrast(emms, method = my_contrasts3, by = "am") # p value is affected
#> am = 0:
#> contrast estimate SE df t.ratio p.value
#> squares 0.15 1.80 26 0.083 0.9344
#> 4 vs 6 -3.77 2.32 26 -1.630 0.2303
#>
#> am = 1:
#> contrast estimate SE df t.ratio p.value
#> squares -1.17 2.12 26 -0.552 0.5858
#> 4 vs 6 -7.51 2.05 26 -3.658 0.0023
#>
#> P value adjustment: fdr method for 2 tests
# Also in interaction contrasts
contrast(emms, interaction = list(cyl = my_contrasts2, am = "pairwise"))
#> cyl_custom am_pairwise estimate SE df t.ratio p.value
#> squares 0 - 1 1.32 2.79 26 0.474 0.6393
#> 4 vs 6 0 - 1 3.73 3.09 26 1.206 0.2386
#>