Designing Experiments and Evaluations

Add to your list(s) Download to your calendar using vCal

• Christopher Baber
• Wednesday 06 March 2019, 12:00-14:00
• Computer Science Sloman Lounge.

If you have a question about this talk, please contact Fatma Faruq.

This is an overview of some key points to consider in the design of data collection exercises – experiments, evaluation trials etc. – which you are using to compare two or more conditions. I will talk about framing testable hypotheses – by which I mean, making sure that the two conditions that you wish to compare can be fairly tested (so you don’t introduce bias into the testing), that the two conditions can be described with data that are amenable to statistical testing, that the data themselves can be amenable to parametric tests etc. In Computer Science, we use R for running statistical analysis (and I give an introduction to R in the ‘Evaluation Methods and Statistics’ module in the Autumn term).

This talk is part of the SoCS PhD Research Training Sessions series.

This talk is included in these lists:

• Computer Science Sloman Lounge
• SoCS PhD Research Training Sessions

Note that ex-directory lists are not shown.