- This event has passed.
BSU Seminar: ‘Phases of research for statistical methods’
January 31 @ 2:00 pm - 3:00 pmFree
Speaker: Dr Tim Morris, MRC Clinical Trials Unit at UCL
Abstract: Research into statistical methods should provide an evidence-base that allows applied researchers to choose between methods. Unfortunately, the appetite of funders and journal editors for novelty and innovation means that there are many new methods but the evidence base is often too pithy to serve applied researchers’ needs.
How does a method go from ‘neat idea’ to mature and trustworthy, so we can say when it should be used or avoided? By analogy to the familiar phases of drug-development, I sketch four possible phases of research for statistical methods. This framework helps us to consider the contribution of a piece of statistical research to our overall understanding of a method. I hope taking this perspective achieves the following:
- Helps frustrated researchers doing early-phase work a framework to understand why their method has not been universally adopted;
- Recognises ‘late-phase’ statistical methods research as valuable;
- Gives applied researchers a way to articulate their hesitation about using new methods.
This will be a free hybrid seminar. To register to attend virtually, please click on this link: https://us02web.zoom.us/meeting/register/tZEtd–hrzItHtTIezm0p763Lsc95VI0Q254