All three days as a package or individual days
The format of these courses: a mix of lectures and computer practicals using SPSS. This will enable participants to gain experience of the theory and practical application of the statistical techniques, as well as interpretation of the accompanying SPSS output.
A basic knowledge of statistics including p-values, hypothesis testing and simple statistical concepts is assumed for all of the following courses. Prior experience of a statistics package or spreadsheet is desirable, but not essential.
The courses would be relevant to clinical and non-clinical researchers and other health care professionals.
Day 1: Modelling continuous outcomes - linear regression
The aim of this course is to introduce participants to more advanced statistical techniques for measured (continuous/discrete) health outcomes. This will enable participants to gain experience of the theory and practical application of the statistical techniques, as well as interpretation of the accompanying SPSS output.
The course will cover the following topics:
- Analysis of variance
- Linear regression
Day 2: Modelling binary outcomes - logistic regression
The aim of this course is to introduce participants to more advanced statistical techniques for binary and categorical health outcomes. This will enable participants to gain experience of the theory and practical application of the statistical techniques, as well as interpretation of the accompanying SPSS output.
The course will cover the following topics:
- Relative risks, odds ratios, and numbers needed to treat
- Logistic regression
Day 3: Modelling time to event outcomes - survival analysis
Survival analysis (or time-to-event analysis) allows the examination of a relationship between patient information recorded at some origin (e.g. referral to secondary care, disease diagnosis) and a binary outcome that may or may not occur at some later point (e.g. death, recovery, hospital discharge). Some underlying theory will be explained, but only enough to facilitate interpretation of the analysis.
Those with no knowledge of multivariable statistical modelling are strongly encouraged to attend the previous two Intermediate Medical Statistics courses (3 and 4) before attending this course.
The course will cover the following topics:
- Censoring and the hazard function
- Estimating the survivor function using the Kaplan-Meier method
- Comparing groups using the log-rank test
- The Cox proportional hazards model
- Model checking procedures