Research, statistics and mathematics educators in Nigeria: effect size perspective

Abstract

Over reliance on the perspective of a dichotomous reject or fail-to-reject outcome from a null hypothesis testing framework to answer research questions has become a worrisome issue to research methodologists and statistics experts. Thus, the Journals of Mathematical Association of Nigeria, Abacus (2013 & 2014) were surveyed to investigate the effect size reports practice among Mathematics Educators in Nigeria. The study showed that majority (60%) of questions of research interests of Mathematics Educators were answered by Null Hypothesis Significant Testing (NHST) and less than one-fifth (16%) of this empirical studies reported Effect sizes (standardized) for their findings. However, the survey further revealed that though Effect sizes were mostly not reported but can be conveniently estimated from the associated descriptive statistics reported by the researchers except in the analyses which seek for relationship in categorical data. Recommendations made included that Editorial Policies and guidelines of Journals in Education, especially in Mathematics Education should include Effect sizes and Confidence Intervals reports for authors.