/nsm3data

Additional Datasets to Accompany Hollander, Wolfe, and Chicken - Nonparametric Statistical Methods, Third Edition

Primary LanguageR

nsm3data

Additional Datasets to Accompany Hollander, Wolfe, and Chicken - Nonparametric Statistical Methods, Third Edition

Designed to add datasets which are used in the textbook.

THIS IS NOT a substitute for the textbook. You will almost certainly not be able to use anything in this package without having the textbook. Rather, this package is supplemental and is only designed to save you time that you would otherwise spend typing data into R or into a spreadsheet.

Installation

Installing from CRAN (Preferred)

This is the preferred installation method. This package is available in CRAN as a regular package. Assuming you have not modified your package repository list, you can install it using the package manager in software like RStudio (search for package "nsm3data") or by using

install.packages("nsm3data")

After the package is installed, load the library into your current environment using the library() function.

library(nsm3data)

devtools method

Install the package from GitHub using the devtools library.

install.packages("devtools") # if it isn't already installed on your system
devtools::install_github("kim3-sudo/nsm3data")

After the package is installed, load the library into your current environment using the library() function.

library(nsm3data)

Naive method

To install the package from a release instead of with the devtools library, download the tar.gz file from the latest GitHub release. Then use

R CMD INSTALL nsm3data_X.Y.ZZZZ.tar.gz

in a terminal window, replacing X.Y.ZZZZ with the actual version number of the release. After the package is installed, load the library into yur current environment using the library() function.

Start R and type library(nsm3data) to see that it was installed, then try out some of its functions.

Usage

nsm3data contributes no additional functions, but it does have data and documentation. You need not specify the location of a dataset; simply use the data() function to read the data into your current environment after you have loaded the library.

All datasets have accompanying descriptions that have varying degrees of helpfulness. The package itself also has a help description. To view this description, use the help() function in R to view the object's documentation. For example, use

help(nsm3data)
help(pokeweed)

to view the pokeweed dataset documentation.

Table of Contents

nsm3data contains its own table of contents for the datasets that it includes. This table of contents may not be up-to-date, and we would appreciate your contributions if you find a dataset that doesn't have an entry. Instructions on how to do so are in CONTRIBUTING.md.

To view the table of contents, use the help() function in R to view the toc object documentation

help(toc)

Please note that the toc object doesn't actually exist in any meaningful way, so trying to reference it will result in just a message.

Loading Data

All datasets in nsm3data are provided as R datasets, so to use a dataset, simply use the data() function after you have loaded in the library. Consult the table of contents to find the name of a dataset. For example, use

data(pokeweed)

to load the pokeweed dataset into your environment. The data will be loaded in as a dataframe, so you can use any method to view or manipulate that data as you would any other dataframe, including using tidyr:: functions. If you are new to R, you need not fear overwriting the library datasets, since you are editing data that is in your own environment.

Get started with:

View(pokeweed)
head(pokeweed)
str(pokeweed)

Need help?

Please use the GitHub issues feature to get help. The maintainer encourages you to use the pink "question" label to get help.