In-person Delivery: University of Birmingham, Gisbert Kapp N424 (What3Words for Building Entrance: https://w3w.co/drift.wire.dance) 

Aimed at: All Years

Duration:  1 day

CENTA Training Credits: 2

Training Area(s): Theme Science and Technical Skill

Trainer: Dr Josh Larsen

What causes what in science?

It is extremely easy to ask this basic question in research, but far more difficult to convincingly answer it. There can be data restrictions, model assumptions, or even poor conceptual frameworks that all get in the way of trying to get to a satisfactory understanding of causality in our research questions.

We will take a step back and look at what we even mean by causality, and how we can approach our questions and research program with both informal and formal methods to make causal understanding our main priority. This will take the form of some introductory ideas and discussion, followed by some more formal data analysis options that increase in complexity, such as analyses that take advantage of special circumstances in data, and directed acyclic graphs (DAGs). We will mention and discuss, but not formally interrogate, similar challenges in process-based models. No formal training in modelling, statistics, or maths is required, but of course exposure to basic ideas such as Bayesian statistics, multiple linear regression methods, and basic calculus are an advantage. Be prepared to discuss and even show your research in the context of causal relationships, and even bring your data or ideas to either demonstrate your current approach, or even try some new approaches.

***This is CENTA organised training; if you are a CENTA student you may reclaim your expenses to attend this training in line with the University of Birmingham Expenses Policy. All others should consult with their own funder's regulations.***