========================================================================================

Introduction to Statistical Modelling
========================================================================================

Thursday 28^th June 2018 (09.30 - 12.30)

Seminar Room, Wellcome Trust Clinical Research Facility, Western General Hospital, Edinburgh, Scotland, UK

Trainer: Dr Helen Brown

Course Fee:
• Student Rate: £24 (with a valid matriculation card)
• NHS/Uni Early Bird Rate paid before 17 May 2018: £36

• NHS/Uni Rate paid after 17 May 2018: £48

This workshop will introduce the popular types of classical statistical model including ANOVA, regression, the general linear model, logistic regression and other models suitable for non-normal data. While basic statistical tests form essential building blocks for statistical analysis, statistical modelling allows data structure beyond simple groupings to be taken into account and more sophisticated questions to be addressed. For example, patients grouped within hospital clinics or health boards, clinical trials carried out across several centres, multiple measurements made on the same patient. Sometimes data may be unbalanced, e.g. in terms of deprivation, gender or age - here a statistical model can often reduce the bias this may cause when comparing groups, by adjusting for these effects. In other situations a model may be constructed to assess whether there is an interaction between two effects – e.g. does the effect of a treatment effect vary depending on location, gender, age, etc?

The workshop is suitable for those who already have some experience of statistical analysis and who wish to be able to interrogate their data further using statistical modelling.

To book a place, please email the Edinburgh CRF Education Team at [log in to unmask] or register online using the following link

========================================================================================

Introduction to Mixed Models
========================================================================================

Thursday 28^th June 2018 (13.30 - 16.30)

Seminar Room, Wellcome Trust Clinical Research Facility, Western General Hospital, Edinburgh, Scotland, UK

Trainer: Dr Helen Brown

Course Fee:
• Student Rate: £24 (with a valid matriculation card)
• NHS/Uni Early Bird Rate paid before 17 May 2018: £36

• NHS/Uni Rate paid after 17 May 2018: £48

This workshop will provide an introduction to mixed (or multilevel) models and emphasise their practical application to medical data with particular reference to clinical trials. Mixed models are a useful tool for situations where there is more than one source of random variation (e.g. between and within patients) or a pattern of correlation between observations. Conventionally medical data is analysed using fixed effects models (e.g. ANOVA, regression, general linear models, logistic regression), however benefits may often be gained by using a mixed model. For example: allowance can be made for the correlations occurring between repeated observations on the same patient even if there are missing values; in hierarchical datasets (e.g. patients grouped within hospitals) standard errors are more appropriate and incorporate all sources of variation; in some situations more accurate estimates will be achieved compared to a fixed effects model.

The workshop will be most useful to those who already have some knowledge of statistical modelling techniques and wish to expand their knowledge to encompass mixed models. Attendance at the preceding Introduction to Statistical Modelling workshop is recommended for those requiring a recap on classical statistical modelling methods.

To book a place, please email the Edinburgh CRF Education Team at [log in to unmask] or register online using the following link

The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.