Tidy estimated marginal means (least-squares means) objects from the emmeans and lsmeans packages

Description

Tidiers for estimated marginal means objects, which report the predicted means for factors or factor combinations in a linear model. This covers three classes: emmGrid, lsmobj, and ref.grid. (The first class is from the emmeans package, and is the successor to the latter two classes, which have slightly different purposes within the lsmeans package but have similar output).

Usage

## S3 method for class 'lsmobj'
tidy(x, conf.level = 0.95, ...)

## S3 method for class 'ref.grid'
tidy(x, ...)

## S3 method for class 'emmGrid'
tidy(x, ...)

Arguments

Details

There are a large number of arguments that can be passed on to summary.emmGrid or summary.ref.grid. By broom convention, we use conf.level to pass the level argument.

Value

A data frame with one observation for each estimated mean, and one column for each combination of factors, along with the following variables:

When the input is a contrast, each row will contain one estimated contrast, along with some of the following columns:

Examples

if (require("emmeans", quietly = TRUE)) {
  # linear model for sales of oranges per day
  oranges_lm1 <- lm(sales1 ~ price1 + price2 + day + store, data = oranges)
  
  # reference grid; see vignette("basics", package = "emmeans")
  oranges_rg1 <- ref_grid(oranges_lm1)
  td <- tidy(oranges_rg1)
  head(td)
  
  # marginal averages
  marginal <- emmeans(oranges_rg1, "day")
  tidy(marginal)
  
  # contrasts
  tidy(contrast(marginal))
  tidy(contrast(marginal, method = "pairwise"))
  
  # plot confidence intervals
  library(ggplot2)
  ggplot(tidy(marginal), aes(day, estimate)) +
    geom_point() +
    geom_errorbar(aes(ymin = conf.low, ymax = conf.high))
  
  # by multiple prices
  by_price <- emmeans(oranges_lm1, "day", by = "price2",
                      at = list(price1 = 50, price2 = c(40, 60, 80),
                      day = c("2", "3", "4")) )
  by_price
  tidy(by_price)
  
  ggplot(tidy(by_price), aes(price2, estimate, color = day)) +
    geom_line() +
    geom_errorbar(aes(ymin = conf.low, ymax = conf.high))
}