Date of Original Version
With recent focus on the state of research in psychology, it is important to assess the nature of the methods and analyses used and reported. To study this, we coded information about the statistical content reported in articles in the four major Canadian psychology journals published in 2013. We first classified whether the articles were quantitative, qualitative, or theoretical in nature. Our main focus was on articles that used quantitative methods; whereby we investigated the prevalence of different statistical procedures, as well as further details of reporting practices. Few articles in any of the journals used qualitative approaches, as 92.9% of empirical articles included a quantitative study. Analysis of variance (ANOVA), t-tests, and multiple regression were the statistical analyses most often reported in the investigated articles. The majority of articles used hypothesis testing, and while most of these tests were accompanied by an effect size, this rarely included a confidence interval. Many of the quantitative studies provided minimal details about their statistical analyses and less than a third of the studies presented on data complications such as missing data and statistical assumptions. Further discussion highlights strengths and areas for improvement for reporting quantitative results. The paper concludes with recommendations for how researchers and reviewers can improve comprehension in statistical reporting.
Counsell, A., & Harlow, L. L. (2016, October 6). Reporting practices and use of quantitative methods in Canadian journal articles in psychology. Canadian Psychology/Psychologie Canadienne. Advance online publication.
Available at: http://dx.doi.org/10.1037/cap0000074