Visualising variability and uncertainty in R by Jack Taylor

About the speaker

Jack Taylor is a PhD student at the University of Glasgow. He is in interested in how we represent words, particularly how we represent and access concepts associated with words, like emotion and imageability. Jack is also interested in the extent to which we predict the visual features of words, given a semantic context. Jack also has a keen interest in supporting reproducible research practices, including data visualisation among others, and you can find some useful links below. Jack can also be found on Twitter @JackEdTaylor.

Useful links:

• LexOPS is an R package I’ve written for generating word stimuli, to use in Psychology experiments.
• Hack Your Data Beautiful (HYDB) was an introductory workshop for R, sponsored by the Scottish Graduate School of Social Science.
• Shiny Tutorials introduce Shiny to R users with a focus on Psychology research.
• GitHub is where I share most of my code.
• Twitter is where I am occasionally spotted.

About the talk

A picture is worth a thousand words, and good data visualisation is worth a thousand summary statistics. I’ll argue that good data visualisation is a key component of open and transparent science. I’ll highlight some example ways of visualising data transparently, focusing on presenting individual observations, visualising uncertainty, and embracing variability. Because it’s well-known, I’ll show some implementations in the ggplot2 package of R, highlighting some really useful functions and extensions for presenting informative features of data and statistical models.