I am proud to announce the release of our free textbook Learning Statistics with JASP: A Tutorial for Psychology Students and Other Beginners by Danielle J. Navarro, David R. Foxcroft, and Thomas J. Faulkenberry. This textbook, which is freely downloadable from https://learnstatswithjasp.com, continues the series of open-source adaptations of the popular text Learning Statistics with R by Danielle Navarro. Learning Statistics with JASP maintains the lively, conversational style of Navarro’s original text, but uses JASP as the tool for calculation and development of core statistical concepts. In my own experience as a statistics teacher, I have often found that textbooks tend to be expensive and also have difficulty achieving a balance between readability and rigor. I think our Learning Statistics with JASP book also achieves this balance, with an added benefit of being free to download. What more could a student (or teacher) want?
The list of topics covered in Learning Statistics with JASP will not be a surprise to anyone who has taught a traditional statistics course in the psychological sciences curriculum. It is based on a fairly standard sequence of topics, including describing and plotting data, basic probability, quantifying uncertainty, and null hypothesis testing. The latter part of the book focuses on the theory and practice of the standard hypothesis testing procedures that are covered as part of the undergraduate psychological statistics sequence: chi-square tests, t-tests, correlation and regression, and ANOVAs (including factorial ANOVA). Finally, the book concludes with a quick chapter introducing Bayesian analyses. Personally, I feel that this is currently the weakest part of our book, but it is one that will be greatly improved upon in an upcoming release. Certainly, JASP is built for easy implementation and communication of Bayesian analyses, so this is one place where Learning Statistics with JASP will begin to diverge from its earlier Learning Statistics with R roots. Stay tuned!
Now, I have a confession to make. I have used JASP in my research methods courses at Tarleton State University since 2015. But I will admit that I am a bit of a traditionalist with my introductory statistics courses — I have always avoided using software to teach the course (there are various reasons for this, but this is not the place to get into them). Thus, I’ve never used JASP as the primary tool for teaching statistical concepts to my beginner students. That changed this summer, when I discovered the power of using JASP’s ability to create computed columns. To explain, let me briefly describe one of the activities that you’ll find in Chapter 7 of Learning Statistics with JASP.
The activity in question falls under the topic of sampling distributions, which is a core concept for the beginning statistics student to learn. Traditionally, I teach this in two stages: first, I get students thinking about the action of drawing samples and computing means by using concrete objects (e.g., poker chips with numbers, dice, or whatever else is handy). Then, I ask them to internalize these actions and think about what would happen if I drew samples from a normal distribution. This has always been difficult for me, though, because it is difficult to do physical draws of things like IQ scores, etc. It usually ends with me gesticulating wildly and drawing lots of pictures, and I’m never sure if I’m getting the point across in an effective manner.
I discovered that JASP will demonstrate these random draws quite nicely with a combination of a pre-made .jasp file and a little bit of R code. The full description is in Chapter 7, section 7.3.1 of Learning Statistics with JASP. Briefly, here’s how it works. First, I ask the students in my class to download a file called IQsim.jasp, which is provided for download on the textbook’s web page. This file contains 10,000 rows of a single computed variable called IQsim. By clicking on the function symbol beside the variable name, students can see a simple bit of R code that generated these 10,000 draws from a normal distribution with mean 100 and standard deviation 15. I simply tell the students that this R code will let us take draws of any size. I usually suggest that we start with samples of size 5. Then, every student in class changes the 10000 to 5 and clicks “Compute column” (see figure below).
What results is exactly 5 draws (representing 5 random IQ scores drawn from the population with mean 100 and standard deviation 15). Importantly, each student gets a different sample. And, since it is all done in JASP, each student can quickly compute the mean of their sample using the “Descriptives” button. For a typical class (I usually teach 50-75 students in my sections), we get a healthy sample of draws to begin plotting distributions of these means. In approximately 30 minutes, we can draw a sketch of a sampling distribution for N=5, N=25, and N=100, and importantly, see the critical pattern that emerges; namely, that the distributions get less and less variable. My experience has been that students really enjoy this activity, and it illustrates how we can leverage the power of JASP to teach core concepts to our beginning statistics students.
I hope the readers of this blog who teach statistics to beginners will seriously consider using Learning Statistics with JASP. As described above, it is a fun read with a good mix of conversation and rigor, and JASP is used as a tool for teaching, not just mere computation. And, since the book is open source (all source files for the textbook are hosted in a Github repository), you can remix the book and make it your own. If you find errors, obvious omissions, or things that you would like to see in the next release, please let me know. Feel free to log an issue on the Github repo, or you can simply send me an email at faulkenberry@tarleton.edu.
Like this post?
Subscribe to our newsletter to receive regular updates about JASP including our latest blog posts, JASP articles, example analyses, new features, interviews with team members, and more! You can unsubscribe at any time.About the author
