Get correlation value in a tidy way — get

Determine the Pearson correlation coefficient between an outcome variable on the left-hand side of formula and one or more explanatory variables on the right-hand side, using pipeable and formula-friendly syntax.

Usage

get_correlation(data, formula, na.rm = FALSE, wide = FALSE, quiet = FALSE, ...)

Arguments

data: a data frame object
formula: a formula with the outcome variable on the left and one or more explanatory variables on the right (e.g. y ~ x or y ~ x1 + x2 + x3).
na.rm: a logical value indicating whether NA values should be stripped before the computation proceeds.
wide: if TRUE and the formula has more than one right-hand-side variable, pivot the result wider so that each predictor becomes a column. Has no effect on single-predictor formulas. Default FALSE.
quiet: if TRUE, suppress the informational message that points to corrr::correlate() for full pairwise correlation matrices. Default FALSE.
...: further arguments passed to stats::cor()

Value

A tibble. For a single right-hand side variable, a 1×1 tibble (or 1 row per group) with column cor. For multiple right-hand side variables: long format (default) with columns predictor and cor (plus any grouping variables), or wide format (wide = TRUE) with one column per predictor.

Details

For a single right-hand side variable, the result is a 1-column tibble (or one row per group if data is grouped) named cor. For multiple right-hand side variables, the result is a long tibble with one row per (predictor, group) combination by default; pass wide = TRUE to pivot it to one column per predictor.

Examples

library(moderndive)
library(dplyr)

# Single explanatory variable:
un_member_states_2024 %>%
  get_correlation(formula = life_expectancy_2022 ~ gdp_per_capita, na.rm = TRUE)
#> # A tibble: 1 × 1
#>     cor
#>   <dbl>
#> 1 0.585

# Multiple explanatory variables — long format:
un_member_states_2024 %>%
  get_correlation(
    formula = life_expectancy_2022 ~ gdp_per_capita + fertility_rate_2022 + hdi_2022,
    na.rm   = TRUE,
    quiet   = TRUE
  )
#> # A tibble: 3 × 2
#>   predictor              cor
#>   <chr>                <dbl>
#> 1 gdp_per_capita       0.585
#> 2 fertility_rate_2022 -0.815
#> 3 hdi_2022             0.889

# Multiple explanatory variables — wide format:
un_member_states_2024 %>%
  get_correlation(
    formula = life_expectancy_2022 ~ gdp_per_capita + fertility_rate_2022 + hdi_2022,
    wide    = TRUE,
    na.rm   = TRUE,
    quiet   = TRUE
  )
#> # A tibble: 1 × 3
#>   gdp_per_capita fertility_rate_2022 hdi_2022
#>            <dbl>               <dbl>    <dbl>
#> 1          0.585              -0.815    0.889

# Group by one variable:
un_member_states_2024 %>%
  group_by(continent) %>%
  get_correlation(formula = life_expectancy_2022 ~ gdp_per_capita, na.rm = TRUE)
#> # A tibble: 6 × 2
#>   continent       cor
#>   <fct>         <dbl>
#> 1 Africa        0.474
#> 2 Asia          0.658
#> 3 Europe        0.685
#> 4 North America 0.505
#> 5 Oceania       0.703
#> 6 South America 0.262

# Group by two variables:
un_member_states_2024 %>%
  group_by(continent, income_group_2024) %>%
  get_correlation(formula = life_expectancy_2022 ~ gdp_per_capita, na.rm = TRUE)
#> # A tibble: 21 × 3
#> # Groups:   continent [6]
#>    continent income_group_2024      cor
#>    <fct>     <fct>                <dbl>
#>  1 Africa    Low income           0.321
#>  2 Africa    Lower middle income  0.535
#>  3 Africa    Upper middle income  0.395
#>  4 Africa    High income         NA    
#>  5 Asia      Low income           0.396
#>  6 Asia      Lower middle income  0.527
#>  7 Asia      Upper middle income  0.121
#>  8 Asia      High income          0.415
#>  9 Europe    Lower middle income NA    
#> 10 Europe    Upper middle income -0.320
#> # ℹ 11 more rows