Who is the course for?

This tutorial will explain how to undertake inferential analyses that is appropriate for different data types, including hands-on experience of preforming t-tests, chi-square tests, linear and logistic regression analyses. This will also include variable selection procedures based on univariate plots, and how to check important model diagnostics through scatter plots and histograms. Many of the graphical techniques for these procedures will be based upon those from the previous “Descriptive Statistics in R” BRC tutorial.

It will be assumed attendees have had some exposure to basic statistical inference concepts, such as sampling distributions, confidence intervals, hypothesis tests, p-values and regression techniques, although an overview will be provided. It will also be assumed that attendees have attended the previous two R-related BRC courses, or have had some exposure to R before. 

Course description

  • Refresher on concepts of statistical inference – sampling distributions, confidence intervals, and hypothesis tests, p-values and regression techniques;
  • Refresher on R and R Studio – loading the data and appropriate packages;
  • Performing and interpreting basic inferential test – t-test, chi-square tests;
  • Variable selection – plotting univariate scatter diagrams;
  • Linear regression models – performing models, understanding and interpreting output;
  • Logistic regression models – performing models, understanding and interpreting output;
  • Model diagnostics – checking common model assumptions through graphical techniques, possible solutions when assumptions aren’t met.

About the trainer

Bola Coker is the Senior Data Manager for the NIHR Guy’s and St Thomas’ Biomedical Research Centre (BRC). He is based in the King’s College London Department of Population Health Sciences Unit of Medical Statistics and is involved with statistical consultancy and teaching. He runs a team that provides services for the BRC in the areas of databases development, statistical data management and statistical programming.