sm_bland_altman {smplot2}

A Bland Altman plot

Description

This function generates a Bland-Altman plot. This function requires two paired data sets as input (same length), and uses sm_statBlandAlt() to compute statistical values necessary for a Bland Altman plot. For more information on these values, please type ?sm_statBlandAlt.

The plot automatically uses sm_classic() theme. The upper dashed line indicates the upper limit (mean_diff + 1.96*sd), the middle dashed line indicates the mean difference between the two samples, and the lower dashed line indicates the lower limit (mean_diff - 1.96*sd).

To add a legend, you will need to add sm_classic(legends = TRUE). To customise the figure, you can add more geom objects.

Usage

sm_bland_altman(first, second, point_size = 3.3, diff_ci = TRUE, ...)

Arguments

first

Data from the first repetition/session

second

Data from the second repetition/session

point_size

The size of the individual points. The default is set to 3.3.

diff_ci

If set TRUE, then it will draw a shaded region that represents the 95 confidence interval of the difference between the two sessions from one-sample t-test. If the region (i.e. confidence interval) overlaps with zero, then there is no significant bias/difference between the two sessions/datasets. If it does not overlap with 0, then the measurement variability is significantly large.

...

Parameters of geom_point(), such as 'color', 'fill', 'shape', etc.

Value

Prints a figure, which is the Bland-Altman plot (ggplot2 object).

Examples

library(smplot2)
library(tibble)

first <- rnorm(20)
second <- rnorm(20)
df <- as_tibble(cbind(first,second))
sm_bland_altman(df$first, df$second)
# when all 3 dashed lines are not shown, extend the range of the y-axis.