Northern Arizona University
This study assesses research and reporting practices in quantitative second language (L2) research. A sample of 606 primary studies, published from 1990 to 2010 in Language Learning and Studies in Second Language Acquisition, was collected and coded for designs, statistical analyses, reporting practices, and outcomes (i.e., effect sizes). The results point to several systematic strengths as well as many flaws, such as a lack of control in experimental designs, incomplete and inconsistent reporting practices, and low statistical power. I discuss these trends, strengths, and weaknesses in comparison with methodological reviews of L2 research (e.g., Plonsky & Gass, 2011) as well as reviews from other fields (e.g., education, Skidmore & Thompson, 2010). On the basis of the findings, I also make a number of suggestions for methodological reforms in applied linguistics.
(Received February 18 2012)
(Received June 26 2012)
(Received August 27 2012)
Correspondence concerning this article should be addressed to Luke Plonsky, PO Box 6032, Flagstaff, AZ 86011. E-mail: [email protected]
This study is part of a larger project, my dissertation, carried out at Michigan State University. I am very grateful to the following individuals there and elsewhere for their much-needed guidance, perspective, and comments: Sue Gass, Shawn Loewen, Fred Oswald, Paula Winke, and Spyros Konstantopoulos. Special thanks also go to Kaytlin Moore for her tireless assistance with coding, to three anonymous SSLA reviewers, and to Lourdes Ortega for a particularly insightful and in-depth review of a previous version of this article. Finally, parts of this study were presented at SLRF 2011 and AAAL 2012, and I am very thankful for the comments and suggestions of those in attendance.