Instructor: Dr. Reza Norouzian, Texas A&M University
Friday, October 16, 2020, 2pm-5pm (EST)
Researchers have long been interested in being able to extend their study findings to a wider population of participants whom they cannot access. The root of this interest lies in the fact that research studies are often conducted with a limited number of participants, and thus their findings may not reflect a fact applicable to a wider population of participants beyond those studied. While this inferential problem has traditionally been addressed via the Frequentist inference, several empirical fields have begun to embrace an alternative called the Bayesian inference. Bayesian inference has and will very likely continue to gain momentum in second language research (Norouzian et al., 2018, 2019). Thus, understanding the underpinnings of Bayesian inference is critical to the understanding of modern quantitative L2 research.
This workshop covers fundamental concepts of modern statistical inference with an emphasis on the Bayesian inference. To achieve our goal, we take five steps. First, we compare the conceptual underpinning of Bayesian and frequentist methods. Second, using real as well as carefully simulated examples, we will learn and apply a variety of Bayesian methods to various designs in L2 research. Third, to promote the use of Bayesian methods in our future L2 research, we use a free Web‐accessed point‐and‐click software package as well as a suite of flexible and visually rich R programs. Fourth, we will use Bayesian methods for conducting meta-analysis in L2 research. Finally, we will discuss the practical as well as the theoretical dimensions of a Bayesian revolution in L2 research.