This function constructs a contrast code from a character variable. It is useful when one needs to recode a two-category character variable to a numeric one.
build_contrast(vector, cond_a, cond_b)A character vector.
A character string to be coded -0.5.
A character string to be coded 0.5.
A numeric vector.
The lm method supports factor and character variables by dummy
coding them. Dummy coding can make the interpretation of regression
coefficient difficult or at least more difficult than contrast coding.
Contrast-coded-variable coefficients interpretation is particularly useful
when conducting a joint-significance test.
scale for centering continuous numeric variable.
data(ho_et_al)
ho_et_al$condition_contrast <- build_contrast(ho_et_al$condition,
"Low discrimination",
"High discrimination")
head(ho_et_al)
#> id condition sdo linkedfate hypodescent condition_contrast
#> 1 2 Low discrimination 1.8125 6.000 2.333333 -0.5
#> 2 3 High discrimination 1.5625 5.875 6.000000 0.5
#> 3 4 High discrimination 1.7500 6.625 6.000000 0.5
#> 4 5 Low discrimination 4.2500 5.125 5.666667 -0.5
#> 5 6 Low discrimination 1.9375 4.375 4.000000 -0.5
#> 6 9 High discrimination 2.8750 3.750 4.000000 0.5