Is R just for data scientists?

Take a look at most R courses and they're pitched to aspiring data scientists. They say things like, "learn R and you can get a great job working at one of the many companies that use it to do data science."

Take a look at the bios of many of the most prominent R users and you'll see that they are data scientists (or talk about using data science in some other field).

Take a look at the most famous book for learning R and what is it called? R for Data Science.

So, is R only for data scientists? No!

When I started learning R, I was initially put off by how synonymous it is with data science. I’m not a data scientist — outside of my R teaching, I’m an anthropologist turned evaluator turned data visualization consultant. I don’t do machine learning. I don’t build neural networks. The only random forests I’m familiar with are the ones where I hike on weekends. Given all of this, I worried I wasn’t a “real” R user.

Over time, I've come to terms with how I use R. I'm ok with being what seems like the rare non-data scientist using it. I use R for descriptive statistics and data visualization, and for these purposes it is the best tool out there.

My inspiration for creating R courses was, and is, to help non-data scientists learn to use it. I want to help the rest of us learn R. The two sentences on the top of the R for the Rest of Us website have become a short-form manifesto:

You don't need a PhD in statistics or years of coding experience to learn R. Anyone can learn the most powerful tool for data analysis and visualization.

What does this mean for the courses that I'm creating?

First, it means that I meet you where you are. I do assume that you’re familiar with the basics of working with data, but I don’t assume you have any programming experience. And I don’t dump you into the statistical deep end. My courses use simple exercises — things like calculating means, doing counts, and the like — in order to teach R. I want you to focus on what’s new (R) and eliminate all other potential distractions.

Second, the progression of material in the course assumes you want to do work with R, not necessarily understand all of the inner workings of the software. When I started learning R, I was overwhelmed with so material that started by teaching me about data structures in R. I figured, well, I guess I've got to learn the ins and outs of vectors, matrices, and lists before I do anything in R. As I've become a more experienced R user, I've realized how wrong this approach is.

Yes, you do need to understand data structures to work with R. But you do not need to learn about them first. To the degree that I teach about these behind-the-scenes concepts, I do so in the context of a problem you might encounter while using R for some other purpose. Why, for example, is my filter not working? Well, perhaps it's because your data is not set up in a way that works for the type of filter you're attempting to do. I think this approach is more pedagogically appropriate, and far more interesting than the typical "you've got to learn boring stuff first" approach.

Third, I only cover the fundamentals of using R. If you want to learn to use R for machine learning, I fully support you. My courses can be a great first step on your journey. But, if you want to use R for descriptive stats, data visualization, and improved reporting, that's also great, and my courses are designed with you in mind. You don't need to do super complex work to make it worth learning R. R is the best tool for the types of data work that many of us do on a daily basis.

In designing the Getting Started and Fundamentals of R courses, I rely on my master's degree in education and experience teaching. I'll be back tomorrow to share how I apply lessons learned from the time I spent earlier in my career working with young students to teaching R to folks like you. See you then.