Repartee, per se

This blog is a collection of short posts in the data journalism style across a range of topics, from baseball to politics to humor. Most of these pieces include some aspect of data analysis to help illustrate how these methods can relate to real, everyday questions we might come across in our lives.

FEATURED
These are a collection of more long-form journalism articles that provide a broader context of the piece and more in-depth analysis to enhance the story. While posts on the homepage are quick reads that are generally shorter than 500 words, these articles can be upwards of 5,000 words and lace together a much larger narrative - but these are not a short read.

In addition to each post have a certain category, they will also have a more specific tag. Some of these may not make sense unless you are familiar with the R programming language - as I use R for much of my data analysis, I have tagged some posts with the particular packages (such as {ggplot2} , {kableExtra}, or {rvest}) utilized in the visualizations.

This site is built using R, RStudio, Hugo, and the {blogdown} package for R. The code for each post is available on the website GitHub repository at https://github.com/ardomingo/repartee-per-se.

CRAN RStudio {blogdown}

The Hugo theme is Clarity, available at:



PRIMARY CONTRIBUTOR

A.R. Domingo

Andy Domingo pursued an English Degree during his time at the University of California, Riverside, almost purely out of a passion for the subject and an interest in the analytic nature of literary criticism. This led down a path based more broadly in analysis, beginning in revenue management in hospitality more than ten years ago. Now, utilizing a variety of data science tools, provides in-depth consultation and analytics platforms across a wide range of disciplines.