Statistical Cognition (SC) is the study of how people understand–or, quite often, misunderstand–statistical concepts or presentations. Is it better to report results using numbers, or graphs? Are confidence intervals (CIs) appreciated better if shown as error bars in a graph or as numerical values?
And so on. These are all SC questions. For statistical practice to be evidence-based, we need answers to SC questions, and these should inform how we teach, discuss, and practise statistics. Of course.
An SC Experiment
This is a note to invite you–and any of your colleagues and students who may be interested–to participate in an interesting SC study. It is being run by Lonni Besançon and Jouni Helske. It’s an online survey that asks questions about CIs and other displays. It’s easy, takes around 15 minutes, and, as usual, is anonymous. To start, click here.
Feel free to pass this invitation on to anyone who might be interested. I suggest that we should all feel some obligation to encourage participation in SC research, because it has the potential to enhance research. Here’s to evidence-based practice! With cognitive evidence front and central.
Statistical Cognition, Some Background
Ruth Marom, Fiona Fidler, and I wrote about SC some years back. The full paper is here. The citation is:
Beyth-Marom, R., Fidler, F., & Cumming, G. (2008). Statistical cognition: Towards evidence-based practice in statistics and statistics education. Statistics Education Research Journal, 7, 20-39.
Some of Our SC Research
Here are a few examples:
Lai, J., Fidler, F., & Cumming, G. (2012). Subjective p intervals: Researchers underestimate the variability of p values over replication. Methodology: European Journal of Research Methods for the Behavioral and Social Sciences, 8, 51-62. Abstract is here.
Coulson, M., Healey, M., Fidler, F., & Cumming, G. (2010). Confidence intervals permit, but do not guarantee, better inference than statistical significance testing. Frontiers in Quantitative Psychology and Measurement, 1:26. Full paper is here.
Belia, S., Fidler, F., Williams, J., & Cumming, G. (2005). Researchers misunderstand confidence intervals and standard error bars. Psychological Methods, 10, 389-396. Abstract is here.
I confess that all those studies date from pre-Open-Science times, so there was no preregistration, and little or no replication. Opportunity!