UQRUG 48

meeting
Overview: Identify Data Problems using visdat.
Questions: GLMM Help, two y-axis plot, ggplot, splitting data into hours
Published

March 27, 2024

2024-03-27: UQRUG 48

R Overview of the Month

This month at UQRUG, before we jump into our help session for everyone to ask questions, this month Nick will look at how we can use some preliminary exploratory data visualisations to identify data issues using the visdat package.

You can download the R file here and then convert it to Rmd after downloading.

Attendees

Add your name, where you’re from, and why you’re here:

Name Where are you from? What brings you here?
Nicholas Wiggins UQ Library Here to help!
Bel QAEHS Here to learn
Zoe MacLean HASS Research Assistant Hang out while coding
Luke Gaiter Library Here to learn/help/hangout
Christopher Sexton Poche Learn and improve R coding
Pia Wohland SENV/HABS Learn and help if Ican
Kar Ng Senior Analyst Bank of Queensland Hangout
Grant Taylor HAB Learn and Improve in R
David Green UQ RCC To help and to learn :-)
David Chau QAAFI see how to use visdat
Julia Wood HABS PhD Candidate Learn and help if I can :)

Questions

Q1 - Julia Wood - Help with GLMM

Need to run some different statistical methods, and need resources on”

  • Generalised Linear Mixed Effects Models
  • Underdispersion in a binomial generalised mixed effect model

Answers


Q2 - Nick - More resources for data checking

Nanair provides a deeper dive into the kinds of missing data that visdat can provide

Answers

  • https://github.com/njtierney/naniar

Q3 - Jayne - Plotting two lots of data on one graph and splitting by time

  • regarding plotting of two lots of data on the one graph
  • plus splitting the graphs into smaller time lots

Answers

Plotting two y axes * reminds me of some data I had analyse … correlating events with denser sets of points

  • This site has details for plotting a ggplot2 graph with two y axes: https://r-graph-gallery.com/line-chart-dual-Y-axis-ggplot2.html

Splitting data by time * Create a new variable that gives a value to each “block” of time and create the facet wrap on that * Perhaps using the Lubridate function hour() to extract the hour into a new column