Calculate association measure between a pair of variables

Usage

pair_cor(df, type)

Arguments

df: a dataframe with 2 columns
type: a vector with 2 elements corresponding to types of variables in df. Types can be "numeric", "ordinal" or "factor", but cannot both be numeric.

Value

values and type of association measure, as well as p value from corresponding association test

Details

The following associated measures and tests are implemented dependent on variable type:

factor vs numeric, factor or ordinal: Pseudo R^2 and p value from multinomial regression

ordinal vs ordinal or numeric: GK gamma and GK gamma correlation test

Examples

data1 <- data.frame(x = rnorm(10), y = rbinom(10, 1, 0.5))
# second variable as factor
type1 <- c("numeric", "factor")
pair_cor(data1, type1)
#> Error in pair_cor(data1, type1): could not find function "pair_cor"
# second variable as ordinal
type2 <- c("numeric", "ordinal")
pair_cor(data1, type2)
#> Error in pair_cor(data1, type2): could not find function "pair_cor"