Introduction to data analysis and linear regression in R

This repository contains content for a 5-day course for new PHD students (and other interesting people), run within the School of Biological, Earth & Environmental Sciences (BEES) at the University of New South Wales.

Details for this session are as follows:

• Dates:
  • Monday 03 February to Tuesday 04 February (9.00am to 5.00pm) - Data Manipulation and Visualisation
  • Monday 10 and Tuesday 11 February – Introduction to design and analysis + Introduction to linear modelling
• Audience: New HDR or Hons students in BEES
• Venue: BEES Teaching Lab 3, Ground Floor D26
• What to bring: your laptop
• Presenters
  • Daniel Falster (BEES)
  • Will Cornwell (BEES)
  • Dony Indiarto (BEES)
  • Eve Slavich (Stats Central)
  • Gordana Popovic (Stats Central)

Aims & Content

Day 1 – Introduction to R (for new beginners) [ Will Cornwell ]

Getting started with R

• Introduction to Rstudio
• Introduction to coding in R
• Getting data in and out of R - R objects and classes
• Packages

Days 2-3 Project management, data manipulation & data visualisation [ Daniel Falster, Will Cornwell, Dony Indiarto ]

Topics

• Projects: Organising and managing data - Reproducible research with Rmarkdown Data manipulation & visualisation with the tidyverse
• Data manipulation with the tidyverse
• Data visualisation with ggplot

Lesson plan (Day 1)

• 9:30 Intro (Dan)

• 9:45 Getting organised: Projects, path names, folders (Dan)

• 10:30 Rmd files (Dan)

• 11:00 MORNING TEA

• 11:15 Reading data with readr (Dan)

• 11:45 Data manipulation with dplyr (Dan)

  • filter, select, mutate, rename, arrange, summarise,
  • pipes

• 12:30 LUNCH

• 13:30 Imagine your plot (Will)

• 14:30 Intro to data visualisation with ggplot (Dony)

• 15:15 AFTERNOON TEA

• 15:30 Exercises

Lesson plan (Day 2)

• 9:30 Tidy Data concept (Dony)

  • pivots

• 10:00 Advanced data manipulation with dplyr (Dan)

  • group_by (summarise, mutate),
  • join

• 11:00 MORNING TEA

• 11:15 Advanced data visualisation with ggplot (Will)

  • (extend plots from Day 1 in various ways)
  • facets
  • styles: themes, scales, labels, palettes
  • multiple plot layouts with patchwork

• 12:30 LUNCH

• 13:30 Data wrangling & visualisation challenge (Dan)

• 15:15 AFTERNOON TEA

• 15:30 Extensions

  • ggplot in talks (Rose O'Dea)
  • ggplot extensions (Will)
  • Reproducible research (Dan)

Day 3-4 Introduction to design and analysis and linear modelling [ Eve Slavich and Gordana Popovic]

Introduction to statistics

• Which method do you use when? - Statistical inference
• Two-sample t-test

Introduction to Experimental design

• Sample sizes
• Treatments

Linear regression

• Linear regression
• Equivalence of two-sample t and linear regression

Linear models

• Multiple regression
• Analysis of variance (and equivalence to multiple regression)

Weirder linear models

• Blocked and paired designs - ANCOVA
• Factorial experiments
• Interactions in regression

Installation instructions

The course assumes you have the R software and the development environment RStudio installed on your computer.

R can be downloaded here.

The Desktop version of RStudio can be downloaded here.