Introduction to Regression Analysis
10:00 am, 20 January 2020 to 4:30 pm, 21 January 2020
This course provides an overview of different regression types and details the application of multiple linear regression. Day one of the course focuses on the theory behind regression analysis, in particular linear regression, and covers the formulation, interpretation and validation of linear regression models. A second, optional day allows delegates hands-on use of a statistical package (SPSS) to see how the theory can be applied to answer a specific research question.
Centre for Applied Statistics Courses
Often in research it is important to look at what factors (usually more than one) are associated with, or predict a particular outcome, while also being able to control for the influence of other variables. Regression analysis is a very powerful technique that allows investigation of the combined associations between one or more predictors and an outcome. Some examples where this is helpful are:
i) Within a trial we may wish to adjust for factors that differ between treatment groups to gauge the true effect of treatment
ii) In observational studies we might want to take into account differences between the demographics or health behaviours of two or more subgroups
iii) Considering the combined effects of different factors may facilitate understanding of variation in outcome
The first day of this course focuses on the theory behind regression analysis; starting by introducing different types of regression analysis and how we choose between them, then focusing on linear regression models. Delegates will see how linear regression models are formulated, interpreted and finally how they are checked. Interaction terms, diagnostic tests and model assumptions are all introduced and explained.
On the second optional day, an example dataset will be introduced, regression models will be run through SPSS and used to answer a research question of interest. Delegates will see how the theory taught on the first day can be applied to data to answer important question. This analysis will cover simple and multiple linear regressions; we will interpret the output to determine the best model for the problem and assess the goodness-of-fit via examination of residuals and outlying measurements. The dataset will be available to all delegates whether they choose to attend the second day or not so they can work through the analysis on different statistical packages.
Please note that this course assumes a basic understanding of statistical concepts such as p-values and confidence intervals. Although day two of the course takes place on SPSS, the theory taught throughout this course is applicable to other statistical packages. If attending day two of the course, basic understanding of SPSS is desirable (offered in our one-day 'Introduction to SPSS' course).
The second day of the course will take place in a cluster room. Delegates are welcome to bring their own laptops and access to the UCL Guest network will also be provided. Everyone wishing to bring their own computer should ensure the software is licensed before attending. Where possible, we recommend using a recent version of SPSS for maximum compatibility with the notes provided during the course.
External Delegates (Non-UCL)
£150.00 (£275 for both days)
UCL Staff, Students, Alumni
£75.00* (£137.50 for both days)
Staff and Doctoral Students from ICH/GOSH
† Limited free spaces available. If there are no free places remaining, Staff and Doctoral Students from ICH/GOSH can still register at the UCL rate.* Valid UCL email address and/or UCL alumni number required upon registration. Please note, this category does not include hospital staff unless you hold an official contract with the university.
Prices include printed course materials, refreshments (and lunch for non-ICH/GOSH participants 12.45pm - 1.45pm on the first day)
Finally, please note that no refunds will be given for non-attendance or cancellations made within 5 working days of the start of the course. For delegates attending courses on funded places, a £50 fee will be charged for late cancellation, non-attendance or partial-attendance. This fee is needed to cover printing, catering, etc. costs that are ordered no later than 5 days before the course and are based on the number of people registered at that point in time.
- Future Dates
Dates Time Apply 20-21 Jan 2020 10.00am - 4.30pm (10.00am - 12.45pm, day 2)‡ TBC 10.00am - 4.30pm (10.00am - 12.45pm, day 2)‡
"Thought it was an excellent course, really helpful to have somethings I have had to teach myself from the internet explained by someone with an actual solid grounding in statistics. Also very useful to have some common misuses cleared up (e.g. use of multivariate when they mean multiple/multivariable and the simple regression to select predictors for a multiple regression model technique). Hard to speak for everyone but pitched at the perfect level for me."
"The presenter was very good and very engaging. confident in the topic she was teaching. I really liked the interaction asking questions throughout the sessions and practical work which helps consolidate the learning."
"I thought this course was perfect; I was worried I would get lost but it was pitched at the right level for me. The instructor was very clear and approachable, which made the course very interactive. Even with all the discussions she still managed to cover the whole set of notes and finished just in time"
"Really enjoyed the course, which is something I didn't think I would ever say about statistics. Originally signed up as I was trying to do multiple linear regression and there were certain aspects that I couldn't quite get my head around. The course was perfectly pitched for my level of previous understanding and covered exactly the sort of thing I was looking for (the practical example was essentially the same as my research but just with slightly different predictors - so ideal). The lecturer was excellent, explained things really well, engaged everyone in the room, and kept it interesting throughout. Days were just the right length and practical exercises and breaks were well spaced to stop people zoning out. Overall a massive help."
"I found the course very helpful. As a nurse who has done one stats module for an Mres I didn't find it intimidating and very accessible . The lecturer was delightful, I will definitely recommend the course to my fellow PhD students"
"Outstanding quality, excellent preparation"
"Teacher was very good, it was clear that she had an in depth understanding of the course content and was able to explain it in a very clear and consistent way. Best stats teaching I've had to date."
"This was a very good course, good value and excellent teaching."
"The course was very well presented and delivered. The course materials were excellent. [The teacher] was very competent and approachable. It was an excellent idea to focus on the theory and not get embroiled in software. Unlike other courses I really appreciated the focus on interpretation of results.