Peer Review

R Psychologist

Material Detail page

Reviewed:: Dec 10, 2018 by Psychology

Bookmark

Ratings

Overall Rating:

Content Quality:

Effectiveness:

Ease of Use:

Overview:

Kristoffer Magnusson's Blog about R, Statistics, Psychology, Open Science, Data Visualization is focused on open science, evidence-based treatments, statistics, and methodology in relation to the evaluation of psychological treatments. R is a free software environment for statistical computing and graphics, and the blog offers a forum for sharing methodology. Magnusson provides packages for computing power, visualization of r and r squared, and Cohen’s d. These visualizations allow one to manipulate and see relationships among relevant factors. Additional information on the statistics debate includes P-curve, interpreting confidence intervals, and archives of many other types of simulations.

Type of Material:

Simulation

Recommended Uses:

The visualizations work well in class. Assignments for advanced statistics courses could easily reference the appropriate simulation.
The blog also lends itself to individual research.

Technical Requirements:

The site is licensed under a Creative Commons 4.0 International License.
Two peer reviewers successfully surfed the site using multiple browsers (Mozilla Firefox and Google Chrome).

Identify Major Learning Goals:

Learn the tools to visualize statistical techniques that are used in psychology.
Use visualizations to see relationships among relevant factors with Power, r, r squared, and Cohen’s d.

Target Student Population:

The site is most useful for researchers or instructors of statistics and research methods to demonstrate issues with power, effect size, and related issues.
Students in advanced classes--College Upper Division, Graduate School--would be best served by being directed to particular simulations with additional instruction.

Prerequisite Knowledge or Skills:

A solid knowledge of advanced statistics is necessary to use the simulations.

Content Quality

Rating:

Strengths:

This site contains a large collection of blogs that go over tools and visualizations of advanced statistical techniques as they relate to psychology. The site's description of the statistics contains accurate information and summarizes the concepts well.
The author has included a description of the technique accompanied by R code that assists in the visualization of the statistics. The code is available for use by those who are familiar with R. Some of the pages are interactive and others contain screenshots of the results.
While the site covers a large variety of information, it all revolves around a central theme of open science.
Power is a difficult concept for many researchers, so it definitely presents a challenge to students. The discussions are useful, but the visualizations allow one to play with relevant factors and directly see the effects on research design.

Concerns:

It would be helpful to have links to reference materials that explain the underlying ideas behind what is in each article. Although there are some interactive activities, most of the newer articles just have the code and some explanation and not the interactive activity directly on the blog page.

Potential Effectiveness as a Teaching Tool

Rating:

Strengths:

By providing the code for R as well as the graphics, one can run the programs in R and check their work or simply use the information provided to understand the material presented. The interactive visualizations allow students to quickly and easily see overall relationships among factors. For example, the effect of Power, alpha, sample size, and tails on effect size shows the overlap between samples as well as the probability of a Type I error or Type II error. (http://rpsychologist.com/d3/NHST/)
This site contains several articles that can be useful to graduate students and researchers who want to use or adjust the code that the article contains. Each article begins with a brief discussion of the learning goals for the article. There is also a section on the bottom of each for comments and questions. The author is quick to reply to the questions that are posted. The code is well documented so that others can use it and figure out how to modify it for their specific use.

Concerns:

Since this is a collection of articles, there will be significant work required to incorporate them into a course.

Ease of Use for Both Students and Faculty

Rating:

Strengths:

When browsing through the site, there were no broken links and the pages all loaded properly. The headers at the top: Home, About, Archives were also helpful.
The current blog posts are organized with tags of Power, R Packages, Statistics, Longitudinal, Multilevel, Linear Mixed-Effects Models, and LME4. The Blog Archives are organized by year. Overall, the information is relatively easy to use with basic knowledge of statistics.

Concerns:

Students will likely find the site overwhelming without additional direction. It would be beneficial if the visualizations were separated into their own links.
The articles are organized by date, which is not as useful as had they been organized by topic. Someone who is looking for an activity for a specific statistical method will have to sort through all of the articles to find it. There is also no search field, so finding items that go over a specific topic can be cumbersome.

Other Issues and Comments:: R Psychologist is a site that offers discussion on many relevant issues in statistics.
Creative Commons: