Category: R

Ozancan Ozdemir replicates a style:

I think everyone agrees on the fact that the Economist magazine produces very-well designed graphics, sometimes the best in the world. The success behind their graph lies on the ability of explaining complex matters in a simpler way by employing traditional data visualization techniques such as line graph or bar plot. They put emphasis on the message they want to convey rather than the aesthetics of the graph itself. They also have a clear hiearchy in their plots and use colors, fonts and lines which represents the brand identity of the magazine.

In this tutorial, we are going to create an Economist-style graph in R by using ggplot2, ggthemes, showtext, ggtextand grid packages. I am going to use a dataset that I have been collecting since 2014 about the poverty line and minimum wage in Turkey, but you can adopt these codes to any dataset you want to visualize.

Click through to learn how.

Isabella Velasquez is feeling creative:

If you use ggplot2, you are probably used to creating plots with geom_line() and geom_point(). You may also have ventured into to the broader ggplot2 ecosystem to use geoms like geom_density_ridges() from ggridges or geom_signif() from ggsignif. But have you ever wondered how these extensions were created? Where did the authors figure out how to create a new geom? And, if the plot of your dreams doesn’t exist, how would you make your own?

Enter the exciting world of creating your own ggplot2 extensions.

The post looks a lot like a series of slides, and it takes you through the process of creating a new geom. H/T R-Bloggers.

Tomaz Kastrun has fun generating a triangle:

What motivates human behaviour can be captured in the Maslow’s hierarchy of needs (source: Wiki). Maslow and psychologists have articulated these needs in a form of a Pyramid, and ever since the concept had been widely adopted (also criticised), and yet, another adaptation is the Pyramid of R needs

Read on for Tomaz’s take, as well as how to generate such a pyramid.

Thomas Williams builds a chart:

I also wanted to get a little interactive with my analysis, and came across Dygraphs for R https://rstudio.github.io/dygraphs/ which wraps the “venerable” (according to creator Dan Vanderkam https://github.com/danvk) javascript charting library of the same name, first released in 2006.

I used Dygraphs in an R script file (it can work equally well in R Markdown) to quickly chart my time series data, loaded from the CSV file. Dygraphs were simple to use, are a solid pick among other charting libraries and very functional for being free and open source.

Read on for a few examples of charts, as well as the entirety of Thomas’s code.