View source: R/bayes_curves_functions.R
| plot_bayes_information_gain | R Documentation |
Plots the information gain chosen in make_bayes_curves() for each
lineup response: by default the absolute change in the probability of guilt,
or the reduction in entropy when gain = "entropy".
plot_bayes_information_gain(bayes_obj, selected_responses = NULL)
bayes_obj |
List output from make_bayes_curves() |
selected_responses |
Character vector of response categories to plot. If NULL (default), plots all responses. |
The vertical axis follows the gain used in make_bayes_curves().
With gain = "probability" (Wells & Lindsay, 1980; Wells et al., 2015)
it is the absolute change |posterior - prior|. With gain = "signed" it
is posterior minus prior: positive for responses that point to guilt,
negative for responses that point to innocence. With gain = "entropy" it is the reduction in
uncertainty in bits: a response that points to guilt gains most at priors
above 0.5 and one that points to innocence at priors below 0.5, and either can
be negative where it moves belief towards 0.5. The legend lists responses from
the most to the least indicative of guilt.
A ggplot2 object
data(lineup_example)
bayes <- make_bayes_curves(lineup_example)
plot_bayes_information_gain(bayes)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.