Abstract
In the present study, we explored the ability of statistics-naïve students to perform an intuitive analysis of variance and compare their performance with that of more experienced statistics students. Participants were shown several sets of data that varied with respect to within group variability and/or between group variability. They were asked to rate the
strength of evidence provided by each dataset in support of a hypothetical theory. Results indicate that statistics-naïve students are able to perform a rudimentary form of analysis of
variance with some accuracy, demonstrating at least a partial understanding of the importance of both within and between group variability. In one instance, statistics-naïve students actually performed in a more expert-like manner than did statistics-experienced students. However, statistics-naïve students also displayed a tendency to overweigh the relative importance of evidence provided by between group variability. This tendency persisted in statistics-experienced students.
Trumpower, D.L., & Fellus, O. (2008). Naïve statistics: Intuitive analysis of variance. Proceedings of the 30th Annual Meeting of the Cognitive Science Society, 499–503. Washington, DC. https://cloudfront.escholarship.org/dist/prd/content/qt4sv5n2v4/qt4sv5n2v4.pdf