This paper presents a set of guidelines for reporting on five types of quantitative data issues: (1) Descriptive statistics, (2) Effect sizes and confidence intervals, (3) Instrument reliability, (4) Visual displays of data, and (5) Raw data. Our recommendations are derived mainly from various professional sources related to L2 research but motivated by results from investigations into how well the field as a whole is following these guidelines for best methodological practices, and illustrated by L2 examples. Although recent surveys of L2 reporting practices have found that more researchers are including important data such as effect sizes, confidence intervals, reliability coefficients, research questions, a priori alpha levels, graphics, and so forth in their research reports, we call for further improvement so that research findings may build upon each other and lend themselves to meta-analyses and a mindset that sees each research project in the context of a coherent whole.Keywords confidence intervals; effect sizes; language; L2; mean; meta-analysis; methodology; quantitative; raw data; reporting practices; standard deviation
IntroductionThorough and consistent reporting practices are critical to advancing and informing second language (L2) theory and practice. In this chapter we call for quantitative L2 researchers to adopt a synthetic mindset-understanding that no research stands alone, that progress as a field happens best if we can pool research findings, and that reporting practices can either facilitate or impede
Larson-Hall and PlonskyWhat Gets Reported and Recommendations for the Field this type of research. However, currently researchers have little in the way of L2-specific guidance or direction on what to present and how. As a baseline from which to assess L2 quantitative research, we will first introduce reporting practices based on recommendations from various sources related to L2 research. Next, we will review how well the field as a whole follows these guidelines by presenting results from secondary/meta-analytic research. We then propose a set of recommendations for improving the means by which quantitative L2 data are reported and interpreted. We do this for five main areas of quantitative reporting: (1) Descriptive statistics and other measures of study quality such as power, (2) Effect sizes and confidence intervals (CIs), (3) Instrument reliability, (4) Visual displays of data, and (5) Raw data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.