Do you teach introductory statistics or data science? Need some help planning your fall class?

I apply the 10 Principles of Burning Man in the design and conduct of all my undergraduate and graduate-level courses, including **my introductory statistics class (which has a heavy focus on R and data science)** at JMU. This means that I consider learning to be *emergent*, and as a result, it often doesn’t follow a prescribed path of achieving specified learning objectives. However, in certain courses, I still feel like it’s important to provide a general structure to help guide the way! This also helps the students get a sense of our general trajectory over the course of the semester, and do readings in advance if they’re ready.

Since several people have asked for a copy, here is the SYLLABUS that I use for my 15-week class (that also uses the “informal” TEXTBOOK I wrote this past spring). We meet twice a week for an hour and 15 minutes each session. The class is designed for undergraduate sophomores, but there are always students from all levels enrolled. The course is intended to provide an introduction to (frequentist) statistical thinking, but with an applied focus that has *practical data analysis* at its core.

**My goal is simple. At the end of the semester, I want students to be able to:**

- Quickly become intimate with a new collection of data, using charts, graphs, and exploratory data analysis
- Construct effective research questions and select appropriate (frequentist) statistical techniques to answer those research questions
- Recognize that frequentist methods are just ONE way of answering those research questions
- Appreciate the message of Alex Reinhart’s amazing book, Statistics Done Wrong

Please let me know if this syllabus is helpful to you! I’ll be posting my intensive (5-session) version of this tomorrow or the next day.

Feel free to join our class Facebook group at https://www.facebook.com/groups/262216220608559/ if you want to play along at home.

Categories: Applied Statistics, Data Science, Education, innovation, R

I’m interested to read about your class. I’ve been thinking and writing much more about data analytics lately, though I am still a relative novice with all of it. I’ve been hearing and reading about R for two years now, but I always think I need to learn how to do more advanced projects with other statistical packages first before I’m ready to start learning R. Needless to say, the idea that you are using R to teach introductory statistics students makes me think that perhaps I should dive into it sooner rather than later.

I’m in the middle of writing an article for Journal of Statistics Education (the approach I take was not only informed by Burning Man, but also the last 20 years of research into reform in this area!) but that’s not going to be out for a couple/few months… so let me tell you right now what happened when I tried to shift students to R! 🙂

Years ago, I taught using Minitab – which to me is a dead easy package. You know, pop your data in, browse around the menus, select your analysis approach, click some buttons, and voila! But students were having a hard time with it anyway. As R became more and more powerful (and stable, and widely used) I realized that I could feel better about introducing my students to a FREE and open source package that they’d be able to use in the early part of their career. For about a year, I let students choose Minitab or R, and taught using Minitab. The really advanced/go-getter students ALL gravitated towards R, so it was a nice way to get a sense of who they were. Still, complaints about Minitab… “it’s so HARD!” The year after that, I taught in R but allowed the students to pick Minitab if they thought R was too difficult. AND THAT’S WHEN A MAGICAL THING HAPPENED. Most of the students were fine with R, because that’s what I was showing them. The students who were struggling with R (about 15% of the classes) would go to Minitab. AND THEY WOULD FIGURE OUT MINITAB EASILY AND WITH NO COMPLAINT. Then the story became “I’m so glad I can use Minitab… it’s so easy.” There was definitely a contrast effect.

Now I teach 100% in R, not just for my intro class, but also for an intro programming class (for students who plan on becoming intelligence analysts) and the more advanced ones too (which include simulation and applied AI/intelligent systems/machine learning). I prepare them for the R learning curve by saying “don’t worry, I’ll give you the code for all the basic stuff you need to do in this class… your job is to figure out how to use it on YOUR data, and then figure out how to make informative charts and graphs.”

Do you provide a lot of that sample code in your book? I think that’s one thing I dread is that I would start using it and try to do something basic that I’m able to do easily in SPSS, and it would take me way too long to figure out the right code to use for R. I know there is quite a user community out there, as well as many blogs and other resources, but sometimes you just want to go in and produce a table quickly to answer a question.

I definitely TRIED to provide all of it… it really frustrates me when I get a book, and I’m like “yes!! that’s what I want to do!” and then they’ve missed some sort of “this is such an obvious step even beginners would of course know how to do it” which of course I can’t figure out at that moment. And then I start feeling bad when I can’t reproduce their code, because of course it’s got to be my skill level and not their communications ability… bla bla bla self-talk into downward spiral… but no, my skill level is fine.

Thus, any omissions are completely unintended. Also, I did things “the long way” and not “the elegant way” in several places, which will irritate the purists but comfort the students.

Along this same line, I also provided all the code I used to produce each plot in the book right next to the plot. Not perfectly relevant in every given place, but if you see a plot and you want to do the same thing, at least you’ll know how it was produced.

Hi Nicole,

I will be receiving your book later on today (yipee!) and I’m looking forward to digging in. With respect to the discussion that you and Dean had, I’m taking the Coursera Data Science Specialization that uses R. I had learned & used SAS 10-15 years ago, but started looking at R a few years ago on my own. I found that although some of the packages use slightly different syntax from R and most of the mainstream packages, R was pretty easy to pick up. I found that Code School has a free set of lessons (http://tryr.codeschool.com/levels/1/challenges/1), Digithead provides great material (http://digitheadslabnotebook.blogspot.com/p/r.html) and Chi Yau provides a great set of introductory tutorials at http://www.r-tutor.com/r-introduction.

So is there a way to give you feedback about the book in case I have any?

Yay! Sure, just email me with feedback (my email address is inside the front cover + also in the Preface). Probably will not be making any major changes or additions to this book until next year (have two more that have to be finished) but I’ll put your comments in my file, and they will be integrated into future work! Thanks so much.

Hi Nicole I need some tips on projects based on dmaic can u share some ppt

Can you be more specific? I MIGHT have something but not sure what you’re looking for.

Some dfss ppt that is approved by asq