Randomised Controlled Trials: (Almost) Everything You Need to Know

Standard statistical techniques in the social sciences are good at uncovering relationships between variables, but less good at establishing whether these relationships are causal. If A and B are correlated, does that mean A "causes" B? That B "causes" A? Or could both A and B be driven by a third factor C?

Randomised controlled trials are a type of study often considered to be the gold standard in uncovering this kind of causality. Many students and early-career researchers avoid RCTs, assuming they are complex and expensive to run. However, that need not be the case. This module will explain the theory of RCTs, how they are implemented, and will encourage participants to think about how they might design an RCT in their own field of work.

This workshop is targeted at researchers from different background and different levels of experience. No prior statistical knowledge, or knowledge of RCTs is required; RCTs are a research design, not a statistical method. A maximum of 25 participants can take part in this course.

Prior to the workshop, participants will be asked to think of a policy or intervention that they are interested in evaluating. This could be one they have evaluated in the past, one they will evaluate, or a fictitious policy. You will need to come to the workshop with this idea and describe the policy.

Students will need to download the (free) software called PowerUp! http://web.missouri.edu/~dongn/PowerUp.htm, specifically: PowerUp_tool download (Updated on May 29, 2017)

By the end of the course, participants will: 1. understand when it is appropriate to use RCTs as an evaluation method; 2. be familiar with the elements involved in designing a robust RCT; and 3. understand some of the practical considerations in running RCTs in real life.

A mixture of presentations, pair/group discussions, practical examples and practical exercises

Click the "Booking" panel on the left-hand sidebar (on a phone, this will be via a link called Booking/Availability near the top of the page).

There may be an online open-book test at the end of the module; for most students, the test is not compulsory.