On June 19th, I attended the R-Ladies Lightning Talks event, a series of 5 minute talks presented by several members of the R-Ladies community. This event provided a great way to learn about some of the organizations members, and how they use R at work or on a day-to-day basis. It was an eye-opening experience showcasing several applications of R.

While I can’t cover every single talk presented in this post, I did want to touch on a few.

Using R to Analyze Fundraising Data

The evening started with Alejandra who works in fundraising. In her free time, Alejandra found and explored fundraising data from donorschoose.org. While Alejandra didn’t know a lot about how to work with the data in R, she did know the right questions to ask about the data set. This was enough for her to plot the data and and find answers. Her main question was whether or not donors were coming back. Using dplyr tools, she was able to find that donor retention was low and was the biggest opportunity for improvement.

Alejandra’s talk resonated with me as someone new to data science. Her biggest take away was not to worry about what you don’t know, but rather to focus on what you do know. Running with what you know will get you somewhere.

Maps and More Maps

One big theme of the evening was maps. Ayanthi presented on how R can be used as a GIS (geographic information system) to create maps. The packages needed to do this in R are GIStools and ggplot2. While using R as a GIS makes it harder to do small adjustments and customizations it is free to use, allows you to know each and every step along the way and allows you to create interactive maps.

The evening also consisted of two talks presented by high school groups; a pleasant surprise considering that I didn’t even know what R was until I was in college!

The first group used machine learning and modelling to create spatio-temporal maps in order to predict future suitable habitats for three different species of sloths. The second group looked at land coverage changes over time for the Solomon Islands, and with more data points, hope to have more accuracy for better predictions.

Expand Your Data Science Toolbox

Madison talked about the benefits of using R and Tableau together as a Data Science tool. Tableau, and end-to-end analytics platform, is easy to use with just a few points and clicks to create interactive visualizations and dashboards. However, while Tableau does not have to handle all the R code, it does lack the statistical modelling of R. Integrating the two, meaning running R code directly into Tableau, can create even more sophisticated reports.

Tracking the Global Pangolin Trade

Emma, a research scientist at EcoHealth Alliance, was studying endangered species and specifically the Pangolin which is the most trafficked animal in the world. To understand their movements, she used legal shipment data from the CITES database. With the help of the echarts package for R, she was able to create an interactive visualization not only showing where they were being trafficked to and from, but could also look at trade traffic over time.

Colors in R

The evening ended with a talk from Anna about colors. Anna emphasized that colors are very important when it comes to communicating data through charts. Not only that, but there are several important factors to choosing a color: whether or not you need to use corporate colors, if you have a colorblind audience, what medium the visualizations will be presented on and more. So, what colors are there? To see the list of all built-in colors in R, you can use the colors() function. For those who need a little help with creating aesthetically pleasing color palettes, the RColorBrewer() package provides pre-made colors palettes.

Hope this is a good start when choosing colors for your next R project!