Overview

This online course gives participants an overview of commonly used regression methods to examine the relationship between an outcome of interest and an explanatory variable. Participants will be introduced to classical linear regression and generalised linear models (e.g. logistic, Poisson, ordinal/multinomial models) depending on the distribution of the outcome. The course covers the basic concept, formulation, interpretation, and validation of the models. Real-world data will be used to demonstrate the practical applications of these models.

This course is specifically designed to enhance knowledge, self-confidence and expertise among researchers who wish to develop their skills in analysing complex longitudinal biosocial data. It makes key concepts and approaches accessible to quantitative researchers from a wide range of disciplines and sectors, with particular emphasis on how these can be practically applied in their work. It is part of the RADIANCE programme, which has been developed by an expert team from UCL and the University of Manchester to offer comprehensive, state-of-the-art longitudinal data science training all in one place.

Who this course is for

This course is aimed at quantitative researchers working with biomedical and social data who wish to use regression models in their research.

Course content

The course provides an introduction to the following topics:

• Selection of regression models depending on the outcome of interest
• The principles and assumptions behind each model
• Determining the relationship between an outcome and one or more explanatory variable
• Interpreting and communicating results when using regression models

Teaching and structure

The course consists of four pre-recorded lectures, four computer practicals (for which either R or Stata can be used) and solutions.

Participants will need to be familiar with either R or Stata (the course can be followed using either software).

Certificates

A certificate of completion will be available to download once the required activities have been completed.

Entry requirements

Participants will need to have a prior understanding of basic statistical concepts (i.e. descriptive statistics, mean, standard deviation, confidence intervals etc.), quantitative data structures and types of variables, as well as being familiar with the software they intend to use (either Stata or R).

Learning outcomes

By the end of this course you will be able to:

• understand different regression models and when they are applicable for your research question
• specify and perform regression analyses
• propose, select and evaluate models
• recognise confounding in statistical analysis
• interpret and communicate results

Cost and concessions

The standard fee for this course is £200.

A 10% discount is available for UCL students and staff – please email radiance@ucl.ac.uk for details.

Course team

Prof Paola Zaninotto

Paola is a Professor of Medical and Social Statistics at UCL. Paola’s areas of expertise include longitudinal data analysis, time-to-event methods, structural equation models, multiple imputation, and multistate life tables methods.

Dr Giorgio Di Gessa

Giorgio is a Lecturer in Data Science at UCL. Giorgio’s area of expertise include regression analysis, longitudinal data analysis, multi-level modelling, and multiple imputations.

Dr Andrea Aparicio-Castro

Andrea is an Associate Lecturer in Statistics at UCL. Andrea’s areas of expertise include multilevel and hierarchical modelling, data harmonisation and integration, estimation and forecasting, empirical Bayesian methods, data visualisation, and multivariate analysis.

Meredith Martyn

Meredith is an Associate Lecturer in Statistics at UCL. Meredith’s areas of expertise include study designs, hypothesis testing, survival analysis and regression analysis.

Course information last modified: 25 Jan 2024, 15:31