The Implications of Choosing a Type of Quantitative Analysis in Interlanguage Research

Much interlanguage research has contributed to L1 and L2 theoretical literature through empirical testing and validating theoretical constructs. Quantitative methods have been widely employed as a useful apparatus for L2 empirical research. Yet, only a limited range of quantitative methods such as mean comparison methods have actually been applied, at least in the domain of formal approach to interlanguage. By testing the Unaccusative Hypothesis in relation to derivational morphology, this study investigates how exploratory statistical techniques would complement the mean comparison methods. Particularly, the knowledge of split intransitivity of native and non-native speakers was explored independently, without the predetermination of verb classes. This study illustrated that comparative quantitative analyses are inherently unable to (1) show how L1 and L2 data can be used to test linguistic hypotheses and to (2) explain precisely what interlanguage itself looks like. This paper emphasizes the incorporation of exploratory statistical analyses such as cluster analysis into interlanguage research, complementing the widely-used comparison analyses and therefore, contributing to the understanding of theoretical constructs and bringing a broader picture of the status of interlanguage.