knitr::opts_chunk$set(echo = TRUE)
This package contains only datasets---the tidy datasets for POS 5737 found on the data page of the course website.
# install package devtools::install_github("pos5737/pos5737data") # see datasets help(package = "pos5737data") # load data data(parties, package = "pos5737data")
New datasets require the following:
make-data/get-raw/get-raw-*.R
that obtains the raw data (ideally from Dataverse and hopefully versioned) and saves the original format and filename to raw-data/
. If raw data are not automatically gettable, then save in raw-data-loc/
.make-data/clean-raw/clean-raw-*.R
that tidies the raw data into a easy-to-use data frame and saves it to data/*.rda
. Variable names and factor levels should be interpretable. Factor levels should be in a reasonable order. Dates should be store appropriate as date or date-time objects. R/*-documentation.R
that documents the data set. Include brief description of the data set, the dimmensions and descriptions of each variable (it's okay to point users to the source paper for the details), a link to the raw data files, a link to the published paper(s) using the data, and an example only uses base R functions (and ideally replicates a key result from the original paper).tests/testthat/test-*.R
(usethis::use_test()
initiates the file) to test that the data works as expected. Ideally, the test that key results from the paper reproduces.README.Rmd
.In the usual workflow, one might download the data file parties.rds
into the data/
directory. They might load the data with
parties <- readr::read_rds("data/parties.rds)
This package allows you to skip the download step.
To install the package, run
devtools::install_github("pos5737/pos5737data")
You only need to run this once.
Once you have the package installed on your computer, you can load the datasets with
# load data data(parties, package = "pos5737data")
Alternatively, you can use
# load packages library(pos5737data) # load data data("parties")
To see the available datasets, use
help(package = "pos5737data")
parties
: Clark and Golder (2006)# load packages library(tidyverse) library(broom) # load data data(parties, package = "pos5737data") # regression model from their table 2, pooled analysis, whole sample fit <- lm(enep ~ log(average_magnitude)*eneg + upper_tier*eneg + en_pres*proximity, data = parties) # create table texreg::screenreg(fit)
therms
: ANES Feeling Thermometers# load data data(therms) ggplot(therms, aes(x = ft_democratic_party, y = ft_republican_party)) + geom_point()
anscombe
data(anscombe, package = "pos5737data") ggplot(anscombe, aes(x = x, y = y)) + geom_point() + facet_wrap(vars(dataset))
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.