Politeness is a universal dimension of human communication (Lakoff, 1973; Brown & Levinson, 1987). In practically all communication, a speaker can choose to be more or less polite to their audience. In this package, we provide tools for researchers to measure the markers and effects of politeness in natural language.
You can install politeness from github using the devtools package, like so:
Many of the politeness features containted in this package use some part-of-speech tagging. We recommend Spacy, which is simple to install through python and which has a convenient wrapper function for use through R, SpaCyR. If you do not have SpaCy installed you must do it outside of R first - otherwise, a reduced set of features (i.e. those that do not require part-of-speech tags) is provided by the package.
In the package we have included a dataset, phone_offers, that was collected from an experiment in which participants were told to write offers asking to by a smartphone from a seller on Craigslist. We randomly manipulated their instructions so that their task was to write in a warm or tough style. The dataset includes the text of their messages, as well as the condition assignment.
data("phone_offers")
politeness::politeness(phone_offers$message)
# install.packages("spacyr")
spacyr::spacy_initialize(python_executable = PYTHON_PATH)
politeness::politeness(phone_offers$message, parser="spacy")
politeness::politenessPlot(politeness::politeness(phone_offers$message),
split=phone_offers$condition,
split_levels = c("Warm","Tough"),
split_name = "Condition")