The purpose of this course is to introduce participants to the R environment for statistical computing. The course focuses on entering, working with and visualising data in R as well as regression modelling in R, including linear and general linear models.
 

Learning Outcomes

Participants will be able to use R to:

• Direct themselves around the R interface in an efficient way
• Import and export their own data from spreadsheets and a number of other data storages to R
• Summarise the data with R's built-in summary statistic functions
• Plot data in interesting ways
• manipulate data in ways such that they can efficiently analyse data

Participants will be able to:

• Have a thorough understanding of popular statistical techniques
• Have the skills to make appropriate assumptions about the structure of the data and check the validity of these assumptions in R
• Be able to fit regression models in R between a response variable
• Understand how to apply said techniques to their own data using R's common interface to statistical functions
• Be able to cluster data using standard clustering techniques

Topics Covered

Topics covered include: 

• Introduction to R: A brief overview of the background and features of the R statistical programming system
• Data entry: A description of how to import and export data from R
• Data types: A summary of R's data types
• R environment: A description of the R environment including the R working directory, creating/using scripts, saving data and results
• R graphics: Creating, editing and storing graphics in R
• Summary statistics: Measures of location and spread
• Manipulating data in R: Describing how data can be manipulated in R using logical operators
• Vector operations: Details of R's vectors operations
• Basic hypothesis testing: Examples include the one-sample t-test, one-sample Wilcoxon signed-rank test, independent two-sample t-test, Mann-Whitney test,teo-sample t-test for paired samples. Wilcoxon signed-rank test
• ANOVA tables: One-way and two-way tables
• Simple and multiple linear regression: Including model diagnostics
• Clustering: Hierarchical clustering, k-means
• Principle components analysis: Plotting and scaling data

Target Audience

This course is ideally suited to anyone who:

• Is familiar with basic statistical methods (e.g. t-tests, boxplots) and who want to implement these methods using R
• Has used menu-driven statistical software (e.g. SPSS, Minitab) and who want to investigate the flexibility offered by a command line package such as R
• Is already familiar with basic statistical methods in R and would like to extend their knowledge to regression involving multiple predictor variables, binary, categorical and survival response variables
• Is familiar with regression methods in menu-driven software (e.g. SPSS, Minitab) and who wish to migrate to using R for their analyses

Assumed Knowledge

The course requires familiarity with basic statistical methods (e.g. t-tests, box plots) but assumes no previous knowledge of statistical computing. 

Each participant will need to bring their own laptop installed with the R software (which can be downloaded free for Linux, MacOS X or windows from http://www.stats.bris.ac.uk/R/)

The course tutor will be one of the following:

Dr Colin Gillespie

Dr Colin Gillespieis a statistics lecturer at Newcastle University and is often employed as an R consultant by Jumping Rivers. He has been using R since 1999 and teaching R programming for the last eight years. Colin has authored a number of R packages and regularly answers R questions and is a top contributor on stackoverflow.

Dr Jamie Owen

Dr Jamie Owen has a PhD in Statistics, with a focus on scalable Bayesian inference. Since leaving academia, he has applied these skills in a number of commercial settings. His focus is using complex data sets, to give meaningful insights. 

Additionally, he has delivered R training courses ranging from introductory to advanced programming and big data analytics at a wide range of companies and academic institutions in both the UK and Europe, including the Francis Crick Institute and the MoD.