A User Researcher’s Journey to Learning R
As a User Researcher coming from an undergrad psychology background, I had a lot of practice with qualitative data. Talking to people and asking questions was my “bread and butter.” However, when it came to quantitative data, I had only taken one stats class in the beginning of my college career. Because of this, I had a lot of fears and doubts about learning quantitative analysis, let alone learning how to apply it using the R language. R is a programming language and free software program that aids in statistical analysis and visual representation of data. I wanted to learn more about quantitative analysis to help figure out what areas of the company we need to focus and also help support my qualitative research. My co-workers had much more experience with quantitative data and R from their master’s and doctorate programs. In undergrad, my only experience with quantitative data was to simply plug in the numbers my professors gave me into SPSS, a statistical tool that didn’t require any code. Experienced researchers or data analysts will tell you that R is more effective if you just learn the language, because of how it handles larger data sets and creates better data visualizations. R is also more flexible given the functions you are able to use and the way you are able to manipulate data. However, with only a basic understanding of HTML programming, I knew that leaning R was going to be a challenge. After all, my co-workers who are fluent with R have been implementing it for years, whereas I am just getting started.
Honestly, getting started was by far the hardest part. I had no clue where to start, and all the articles and YouTube videos on R just confused me even more. After doing a lot of digging on where to start, I landed on a Udemy beginners course called R Programming A-Z™: R For Data Science With Real Exercises! in December of 2020. The course broke down the simplicities of R and walked me through the basics of how to download R and RStudio to my machine. The course explained definitions and names of what I would be using and gave the “why” behind it. For example, I had heard about “vectors” before but never really knew what they were or how they were used. The course explains a numeric vector and provides great visual to better help understand what they are. The course walked me through creating my first vectors and using vectorized operations. It also taught me how to set up my work environment so that I would be comfortable using the user interface that I was going to be working in.
The picture below is what RStudio looks like when you first download the program:
I have changed my RStudio to have a darker look just so it is easier on my eyes and I like how dramatic the colors are to distinguish variables better:
The program showed how to install packages into R that make my job easier and make the presenting of data more visually appealing. The most valuable part of this course was the actual implementation of the R program. It gave me real life examples of data sets so I could relate to what I had learned and implement it in the real world. I believe it’s very important to learn R by doing, as opposed to learning R by reading the documentation. If you just read how to use R, its use case is not always intuitive. However, once you start applying functions in real life scenarios, it becomes apparent how and when the functions should be used.
Even though I was gaining confidence by using the data sets in the course, I still felt a little lost regarding how to use it in my actual work. There was only one time I had used R in my work and felt successful; that was thanks to a co-worker who helped me sort and pull the correct participants data I needed for a project. R did not start to click for me until I got my own small data project and I was able to collaborate with a co-worker to figure out the question we were looking to solve. With the functions she gave me, I was able to build more confidence by doing a lot of the processes on my own. My co-worker and I would meet up to three times a week, for an hour each, over a two-week span. During these meetings, I really figured out how and when to apply certain functions and started seeing these functions in an abstract manner.
I’ve realized through my journey with R that you don’t need to be an expert in quantitative data analysis or know every feature that R has to offer. I also don’t think there is ever truly an end to learning R. I still meet with my co-worker once a week to learn new aspects and capabilities of the program. By practicing R, my understanding of quantitative data became better because I was learning how to apply and manipulate data effectively through functions built into R. When given a data set and a project with questions to be answered, you can just simply apply the function and receive the result. For example, I wanted to see if users who had training in the Buildertrend program correlated with the amount of usage of our schedule features. I was able to run the following function to see that there was a correlation between how many trainings a user had and how much they used the schedule feature.
cor.test(Trained_Users$ScheduleItems, Trained_Users$TrainingCount, method=c("pearson"))Also, Google is a helpful tool when you don’t know what function to use or what an error message means. Nowadays both R and internet forums have been around long enough that most questions have already been asked and answered. Additionally, I highly suggest finding a mentor or partner in R so you can always “phone a friend” to help you or, at least, share the struggle of finding the answer with you; it’s more fun than it sounds. My fears and doubts of R have lessened since I first started my journey, but it has taken me about a year to get to this point. I am nowhere near close to being an expert in R. But my main advice is to give yourself time and grace because it is not always an easy task to learn a new language and a new way of thinking, but it is a fun one!
