MOTE: The Shiny App to Calculate Effect Sizes and Their Confidence Intervals

Recent developments in the psychological sciences have shown the de-emphasis of p-values with a renewed focus on effect sizes as a measure of the importance of research findings (Cumming, 2014). Even with the shift in focus, report rates for effect sizes are very low (Fidler et al., 2005; Fritz, Scherndl, & Kühberger, 2013). Given what we’ve been told about reporting effect sizes, why are researchers omitting these values in their journal articles? Several effect size calculators currently exist, including Soper’s webpage (2013) as well as macros available for SPSS/SAS (Smithson, 2003; Wilson, 2010). However, the flexibility of these calculators, as well as the extent to which they explain the calculations, varies greatly. One way to encourage a change in report rates of effect sizes is to train the next generation of researchers to include these values as part of or in lieu of the traditional hypothesis test. However, as statistics teachers know, it can be difficult to get students to understand which test to select, much less which effect size then corresponds to that statistical test. In this presentation, we will demonstrate a new application that could be used as a teaching tool in statistics and research method courses. This application is designed to allow the user to select the research design and corresponding effect size through drop down menus. For each effect, users type in relevant numbers to calculate those effects, and the effect size and related statistics are presented in APA style. For teaching purposes, helpful description text and YouTube how-to videos are coupled with each effect size page. A previous version of this application was implemented in statistics classrooms wherein students indicated that the application was easy to use and helpful for their homework. Faculty feedback from presentations of the new application during beta testing have been overwhelmingly positive. We believe this application will aid in teaching and learning in statistics and research methods courses for students at the undergraduate and graduate level.

近期心理学科学领域的发展显示，对p值的重要性逐渐降低，同时研究发现的显著性衡量标准转向了对效应量的重视（Cumming，2014）。尽管研究重点发生了转变，但效应量的报告率仍然非常低（Fidler等人，2005；Fritz，Scherndl和Kühberger，2013）。鉴于我们对报告效应量的了解，为何研究人员在他们的期刊文章中省略这些数值？目前存在多种效应量计算器，包括Soper的个人网页（2013）以及适用于SPSS/SAS的宏（Smithson，2003；Wilson，2010）。然而，这些计算器的灵活性和对计算过程的解释程度存在显著差异。鼓励效应量报告率的转变之一，是培养下一代研究人员将其作为传统假设检验的组成部分或替代品。然而，正如统计学教师所知，引导学生理解选择何种检验，更不用说对应哪种效应量，可能是一项颇具挑战性的任务。在本展示中，我们将演示一种新的应用，该应用可作为统计学和科研方法课程的教学工具。此应用旨在允许用户通过下拉菜单选择研究设计和相应的效应量。对于每个效应，用户输入相关数字以计算这些效应，效应量和相关统计信息将以APA风格呈现。出于教学目的，每个效应量页面都附有有用的描述性文本和YouTube教学视频。该应用的早期版本已在统计学课堂中实施，学生表示该应用易于使用，有助于他们的作业。在beta测试期间新应用展示会上的教师反馈普遍积极。我们相信，该应用将有助于本科生和研究生层次的统计学和科研方法课程的教学与学习。