10 Wrapping Up

10.1 Effective Data Storytelling

As we’ve progressed throughout this book, you’ve seen how to work with data in a variety of ways. You’ve learned effective strategies for plotting data by understanding which types of plots work best for which combinations of variable types. You’ve summarized data in table form and calculated summary statistics for a variety of different variables. Further, you’ve seen the value of inference as a process to come to conclusions about a population by using a random sample. Lastly, you’ve explored how to use linear regression and the importance of checking the conditions required to make it a valid procedure. All throughout, you’ve learned many computational techniques and focused on reproducible research in writing R code and keeping track of your work in R Markdown. All of these steps go into making a great story using data.

As the textbook comes to a close, we thought it best that you explore what stellar work is being produced by data journalists throughout the world that specialize in effective data storytelling. We recommend you read and analyze this article by Walt Hickey entitled The Dollar-And-Cents Case Against Hollywood’s Exclusion of Women. As you read over it, think carefully about how Walt is using data, graphics, and analyses to paint the picture for the reader of what the story is he wants to tell.

In the spirit of reproducibility, the members of FiveThirtyEight have also shared the data that they used to create this story and some R code here. A vignette showing how to reproduce one of the plots at the end of the article using dplyr, ggplot2, and other packages in Hadley’s tidyverse is available here as part of the fivethirtyeight R package (Ismay and Chunn 2017). Great data stories don’t mislead the reader, but rather engulf them in understanding the importance that data plays in our lives through the captivation of storytelling.

Concluding remarks

If you’ve come to this point in the book, I’d suspect that you know a thing or two about how to work with data in R. You’ve also gained a lot of knowledge about how to use simulation techniques to determine statistical significance and how these techniques build an intuition about traditional inferential methods like the \(t\)-test. The hope is that you’ve come to appreciate data wrangling, tidy datasets, and the power of data visualization. Actually, the data visualization part may be the most important thing here. If you can create truly beautiful graphics that display information in ways that the reader can clearly decipher, you’ve picked up a great skill. Let’s hope that that skill keeps you creating great stories with data into the near and far distant future. Thanks for coming along for the ride as we dove into modern data analysis using R!