Over spring break I was writing R for one of my IS2000 assignments on the lanai of my grandmother’s home in Florida. My grandmother happened to see what I was writing in and I was surprised to hear that she recognized it as R. She used to work in the health insurance department of a hospital and retired in early 2000s. First of all, I was amazed that R was that old (obviously, I didn’t tell her that) but I was also surprised that she used it in her profession. She explained that they use to use it to analyze and create different pricing models based on various health factors. I would have thought they would have used a GUI or something like Excel but she said that she thought that Excel or Lotus would not be able to handle the amount of calculations and factors. She also said she had used S programming language which apparently is where R was adapted from. This is most likely why she would have used R instead of Python.
Before learning R this year, I had limited experience in python which in many ways has the same functionality and purpose as R. The things I’ve noticed from personal experiences has led me to believe that while Python is more popular, R has some advantages. For example R has functionality that makes it way easier to import different types of data as well as graphically model them. It also has CRAN which is a massive package library of different functions and tools.The closest thing that Python has for public repository is PyPi and that isn’t anything comparable to the amount of functions that CRAN offers. From what my grandmother said, she was provided with a proprietary R package that handled import from various company data and also added formulas that made calculations simple. This type of extensibility is probably what makes it so useful and widely adopted in data science.
I learned to use python for quick text file manipulation and data retrieval. I thought of it as a simple functional programming language for quick data manipulation. After learning R, I think of python as something I would use for quick tasks and R as something I might use for long-term data explorations. For example, I might use python as a backend to retrieve weather data from multiple sources and compile an average, more accurate weather forecast whereas I would use R to calculate different probabilities that someone might die based on mass data sets and health factors.