Diego Torrealba

DISCLAIMER. This is the opinion of someone who has coded and taken classes mainly in Python. That said, I have also coded in Java, JavaScript and R, so this blog aims to offer advice to people learning to code for the first time.

Whether you are curious about data analysis or need it for a specific application, many people will direct you to learn either Python or R as a good first language because of their support for advanced statistics and relatively simple syntax. But which one should you learn, and why?

The most important factor in your decision should be what you plan to do outside of data analysis, if you plan on doing anything outside of it at all. Choose R if you are interested strictly in data analysis or a quick way to do advanced statistics with little setup. However, choose Python if you want to do more than just data analysis, such as building websites or developing software (though this is not to say this cannot be done in R).

The blog could end here for most readers, but maybe a few more reasons could influence your decision. If you prefer to have a larger community of support to help troubleshoot your coding issues, choose Python. While the R community continues to grow in membership because of the popularity of big data and artificial intelligence, Python consistently ranks at the top with other programming languages like C, Java and JavaScript for its versatility. Nonetheless, both Python and R have robust libraries capable of making your time with data much easier than otherwise.

However, Python lacks the ease with which someone can pick up R, install the Tidyverse packages, and immediately begin manipulating, analyzing, and visualizing data. The tidy tools manifesto leaves a sense of simplicity and cohesion often lacking from Python's overwhelming number of packages, which can vary immensely in form and usage. While package managers like Conda streamline this process in Python, it requires more installation steps than R’s Tidyverse.

Despite their similarities and differences, one thing is sure: As interpreted languages, Python and R are slow! Be prepared to wait several minutes for specific scripts to run with large datasets, especially if they involve machine learning. Coding to reduce time complexity, or the rate at which time increases as the data inputted into the algorithm rises, is an art in itself. But that’s for another blog post!

Have any questions or want to reach out to me?

Please reach out to me through my socials to talk about any of my blog posts! I'd love to hear your thoughts or feedback.