On the equivalence of ‘experimental’ and ‘correlational’ statistics
Document Type
Article
Date of Original Version
7-1-1989
Abstract
Wright, Ayton, and Fowler have noted that the use of either experimental or correlational statistics on a data set would lead to the same conclusions. In fact, the similarity is even more fundamental than the authors suggest, since both experimental and correlational statistics are derived from the general linear model. It can be shown, for example, that the t-test and the point-biserial correlation coefficient are mathematically equivalent. Several simple equations are given which demonstrate the relationship between experimental and correlational statistics. Finally, it is emphasized that properly choosing a statistical technique is dependent on what question is being asked. © 1989 Taylor and Francis Group, LLC.
Publication Title, e.g., Journal
International Journal of Mathematical Education in Science and Technology
Volume
20
Issue
4
Citation/Publisher Attribution
Rossi, Joseph S.. "On the equivalence of ‘experimental’ and ‘correlational’ statistics." International Journal of Mathematical Education in Science and Technology 20, 4 (1989): 571-573. doi: 10.1080/0020739890200411.