Getting started with ggBA

What this guide covers

This short guide walks through a complete Bland-Altman workflow:

1. reshape long-format data into paired columns
2. compute agreement statistics
3. visualize agreement with one line of plotting code

library(knitr)
library(tidyr)
library(ggBA)

1) Prepare paired measurements

temperature contains one row per animal, visit, and measurement method.
To compare two methods directly, pivot to wide format.

tbl <- temperature |>
  pivot_wider(names_from = method, values_from = temperature)

head(tbl, 6) |>
  kable(digits = 2)

animalID	treatment	visit	rectal	infrared
1	healthy	baseline	33.89	34.82
1	healthy	visit 1	34.57	36.49
1	healthy	visit 2	35.44	36.22
1	healthy	visit 3	35.41	35.72
1	healthy	end of treatment	35.68	35.06
2	vehicle	baseline	37.62	38.59

2) Compute Bland-Altman statistics

stats_tbl <- ba_stat(tbl, infrared, rectal)
stats_tbl |>
  tidyr::pivot_wider(names_from = parameter, values_from = value) |>
  kable(digits = 3)

n	bias	lloa	uloa	bias.lcl	lloa.lcl	uloa.lcl	bias.ucl	lloa.ucl	uloa.ucl
450	0.234	-2.85	3.318	0.088	-3.1	3.068	0.379	-2.601	3.567

The most commonly interpreted quantities are:

1. bias: average difference between methods
2. lloa and uloa: lower and upper limits of agreement

3) Plot agreement

ba_plot(
  data = tbl,
  var1 = infrared,
  var2 = rectal,
  title = "Infrared vs rectal temperature",
  caption = "Lines show bias and limits of agreement with confidence bounds."
)
#> Warning: Using `by = character()` to perform a cross join was deprecated in dplyr 1.1.0.
#> ℹ Please use `cross_join()` instead.
#> ℹ The deprecated feature was likely used in the ggBA package.
#>   Please report the issue to the authors.
#> This warning is displayed once per session.
#> Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
#> generated.

Next steps

After this baseline workflow, explore:

1. transformed analyses (transform = "log" or "logit")
2. grouped analyses via group = in ba_stat() and ba_plot()