Description This library provides three sets of functions for the following tasks Model estimation Model diagnostics Model summary References
This library enables the examination of political communication by politicians and ordinary users on social media. It uses the news stories shared by users to estimate the news-sharing ideology of users and the ideology of the content that they share.
(1) model estimation, (2) model diagnostics, and (3) model summary.
The core estimation function, mediascores
, takes as an input a
matrix or data.frame where the rows represent social media users and the
columns represent the news media domains shared by each user. Each cell of
the matrix represents the count of each domain shared by each user. If, for
example, a user shares 10 news stories from the New York Times, the value of
the cell indicated by the row of that user and the column of the New York
Times would be a 10.
The object outputted from the estimation function mediascores
is a stanfit-class
object from the
Stan Bayesian inference engine. All methods for
the analysis of Stan objects are therefore useable for those outputted by the
mediascores
function. A helper function rhat
is
provided to give a quick summary of R-hat values to assess the most common
convergence statistics for one's parameters of interest.
Point estimates and credible intervals for the quantities of interest
(i.e. the ideology of social media users and news organizations) are
accessible through the point_est
function.
Stan Development Team (2018). RStan: the R interface to Stan. R package version 2.18.2. https://mc-stan.org
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.