Answers are grounded in this paper and cite their sources.
With great interest I read the May 2002 editorial in the Journal for Research in Mathematics Education (JRME) (King, 2002) regarding changes to the 5th edition of the Publication Manual of the American Psychological Association (APA, 2001). Of special note to me, and of great import to the field of mathematics education research, are changes regarding the reporting of results from statistical analyses. In particular, researchers are urged to use effect and confidence interval (CI) reporting in conjunction with null hypothesis statistical significance testing (NHST). The APA Publication Manual now states that for a reader to understand the importance of a reported p-value . .. it is almost always necessary to include some index of effect size (APA, 2001, p. 25). The manual further advises: confidence intervals combine information on location and precision and can often be directly used to infer significance levels, they are, in general, the best reporting strategy (APA, 2001, p. 22). Because this kind of statistical reporting has not been commonplace in mathematics education research and because a spirited exchange of views on methodological issues can facilitate progress in the field, I offer here some discussion of NHST, effect size, and CI reporting to characterize some of the potential contributions of these latter methods to educational research.
No takes yet. Share an insight, caveat, or question.
Robert M. Capraro (2004) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: