Last updated on 2020-01-30
by Paul Northrop
This CRAN task view gives information about packages with features
that are designed to assist with the teaching of Statistics.
It is not concerned with the teaching of R itself.
A few of these packages are listed in other task views, but only
the Bayesian task view has a section devoted explicitly
to teaching (Bayesian) Statistics.
The packages are grouped into three broad topics:
teaching, examination and packages associated with Statistics books.
The latter is for books that are general enough to be of potential
interest to a wide audience of teachers of Statistics. They should
concern models and methods with wide applicability and not be tied
closely to a particular application.
If you think that a package is missing from the list, or have any other
comments or suggestions, then please contact the maintainer.
- Rcmdr provides a GUI for R, based on the tcltk package.
A point-and-click interface loads data and calls R functions to perform
the kinds of analyses involved in introductory Statistics courses.
More advanced and specialized analysis are also available, some of them
via plug-ins. The R commands are shown in the console.
The R Commander homepage for more information.
- swirl uses the R console to provide an interactive learning
environment for students to learn Statistics. Students select courses
to download from the swirldev/swirl_courses GitHub page
and are provided with immediate feedback as they work. A variety of
topics are available, under the general headings of
Exploratory Data Analysis, Statistical Inference and Regression Models.
Teachers can author and share their own swirl courses using the swirldev/swirlify
package (currently on Github only).
See also the swirl home page.
- mosaic contains a wide range of tools to assist in
teaching of basic, and more advanced ideas and techniques in mathematics,
statistics, computation and modelling. Key aspects are the provision
of functions that enable beginners easily to perform tasks that would
otherwise be difficult and the use of simulation to illustrate
See the Project MOSAIC homepage
for more information.
- xplain can be used to provide bespoke interactive
interpretations of the output from statistics functions. This information
needs to be provided by the instructor in XML format and may contain R
code, to tailor the explanation to the specific results.
See the xplain
website for a tutorial and cheatsheet.
- animation provides functions to produce animations relating to
a wide range of topics in Statistics, Data Mining and Machine Learning.
These animations, or a sequence of images generated by the user, may be
exported to a variety of formats.
- gganimate animates plots produced by ggplot2.
It can be used to render the plots into an animation, such as a GIF
or MP4 video .
- TeachBayes provides visualizations to illustrate the basic ideas of Bayesian inference, the roles of prior and posterior distributions in particular. Key teaching examples are used, namely inference for: a mean, a proportion, two proportions and which of several multi-faced dice have been thrown in an experiment.
- smovie provides movies to illustrate concepts in Statistics.
Topics covered are: probability distributions; sampling
distributions of the mean (cf. central limit theorem), the maximum
(cf. extremal types theorem) and the (Fisher transformation of the)
correlation coefficient; simple linear regression; hypothesis testing.
- visualize provides graphs of the pdf or pmf of various continuous and discrete probability distributions, annotated with the mean and variance of the distribution. Shading is used to indicate an interval (lower tail, upper tail, two-tailed or a user-supplied interval) within which the random variable lies with a user-supplied probability.
- LearnBayes provides functions and to illustrate the
essential ideas of Bayesian inference, such as the roles of the prior,
likelihood and posterior; posterior predictive checking and predictive
inference, and several example datasets.
- TeachingDemos Provides a wide range of static and
interactive plots to demonstrate statistical concepts, including:
coin tossing and dice rolling; confidence intervals; various aspects of
hypothesis testing; the central limit theorem;
maximum likelihood estimation; scatterplot smoothing; histograms;
correlation and simple linear regression; Box-Cox transformation.
- distrTeach provides plots to illustate the Central Limit
Theorem (CLT) and the Law of Large Numbers (LLN). The effects on the CLT plots of changing inputs can be shown using a Tcl/Tk-based widget.
- learnstats uses a console-based menus and
shiny apps to provide interactive plots that illustrate
key statistical concepts. Topics covered include probability areas on
density functions, binomial, normal, t and F distributions, p-values,
QQ-plots and simulation of time series with different behaviours.
- AtelieR uses a GTK GUI to help teach some key
statistical concepts. Includes the sampling distributions of the mean
(cf. central limit theorem) and variance, probability calculator for
common distributions, Bayesian inference for proportions, multinomial
counts, means and variances.
- BetaBit provides games for students to play in the R console, including one that involves data-cleaning and regression modelling. See the BetaBit home page.
- DALEX provides functions to explore and understand predictive models. The DALEX GitHub page includes two teaching-related showcases.
- exams provides a framework for the automatic random generation of exams and self-study materials from a pool of exercises composed using either Sweave (.Rnw) or R markdown (.Rmd) formats. R code can be used to generate exercise elements dynamically. Questions can be formatted for use in a variety of e-learning platforms or output as documents, for example a PDF file, for which. Scans of PDF answer sheets can be marked automatically.
See also the R/exams homepage
- ProfessR creates multiple choice exams from a pool of exercises organised in ASCII test files. Multiple versions of an exam can be created by randomizing the questions and the choices of answers.
- TexExamRandomizer enables the randomization of questions created using LaTeX's document class for preparing exams. Spreadsheets containing students' answers can be marked automatically.
Packages associated with Statistics books
The following packages are associated with textbooks that are of
potential interest to a general statistical audience, rather than being
specific to a particular application area. The general principle for
inclusion is that package is likely to be of direct use in the teaching
of statistical methods. Official publisher links are provided where
possible and, in some cases, a link to further resources.
- AER: Kleiber, C., Zeileis, A. (2008),
Applied Econometrics with R, Springer Verlag, New York. Further resources.
- ACSWR: Tattar, P.N., Suresh, R., and Manjunath, B.G. (2016),
A Course in Statistics With R, John Wiley and Sons, Inc.
- BaM: Gill, J. (2014),
Bayesian Methods: A Social and Behavioral Sciences Approach, New York: Chapman and Hall/CRC.
- BayesDA: Gelman, A., Carlin, J., Stern, H., Dunson, D.,
Vehtari, A., Rubin, D. (2013), Bayesian Data Analysis, Third Edition. New York: Chapman and Hall/CRC. Further resources.
- Bolstad: Bolstad, W. M. and Curran, J. M. (2016),
Introduction to Bayesian Statistics, Third Edition. John Wiley and Sons, Inc.
- car, carData, effects: Fox, J, and Weisberg, S. (2019), An R Companion to Applied Regression, Springer Verlag, New York. Further resources.
- faraway: Three books by Julian Faraway:
Practical Regression and ANOVA in R (CRAN document),
Linear Models with R (2014), CRC Press,
Extending the Linear Model with R (2016), CRC Press.
- HH: Heiberger, R. M. and Holland B. (2015),
Statistical Analysis and Data Display: An Intermediate Course with Examples in R, Second edition. Springer-Verlag, New York.
- HSAUR3: Hothorn, T. and Everitt, B. S. (2014),
A Handbook of Statistical Analyses using R, Third edition. New York: Chapman and Hall/CRC.
- ISwR: Dalgaard, P. (2008),
Introductory Statistics with R, Second Edition, Springer Verlag, New York.
- MASS: Venables, W. N. and Ripley, B. D. (2002)
Modern Applied Statistics
with S, Fourth Edition, Springer, New York. Further resources.
- moderndive: Ismay, C. and Kim, A. Y. (2019)
ModernDive: Statistical Inference via Data Science. See also infer.
- MPV: Montgomery, D.C., Peck, E. A. and Vining, G. (2012),
Introduction to Linear Regression Analysis, John Wiley and Sons, Inc.
- msos: Marden, J. (2015)
Multivariate Statistics: Old School, CreateSpace Independent Publishing Platform. Free PDF version.
- openintro: Three open-source books published by
Introductory Statistics with Randomization and Simulation,
Advanced High School Statistics.
- regtools: Matloff, N. (2017), Statistical Regression and Classification: from Linear Models to Machine Learning, New York: Chapman and Hall/CRC.
- resampledata: Chihara, L. M. and Hesterberg, T. C. (2018), Mathematical Statistics with Resampling in R, Second Edition, John Wiley and Sons, Inc. Further resources.
- Sleuth2 and Sleuth3: Ramsey, F. and Schafer, D. (2013), The Statistical Sleuth: a Course in Methods of Data Analysis,
Brooks / Cole Cengage Learning.
- SMPracticals: Davison, A. C. (2003),
Cambridge University Press. Further resources.
- vcd: Friendly, M. and Meyer, D. (2015), Discrete Data Analysis with R, New York: Chapman and Hall/CRC. Further resources.