Open AccessMediation Analysis in Linguistics: Essentials for Good Practice
Author(s) -
Kamel Boustani
Publication year - 2021
Publication title -
journal of english literature and language
Language(s) - English
Resource type - Journals
ISSN - 2767-2964
DOI - 10.31829/2767-2964/jell2021-2(1)-106
Subject(s) - mediation , sobel test , bootstrapping (finance) , computer science , outcome (game theory) , linguistics , inference , causal inference , psychology , econometrics , artificial intelligence , mathematics , sociology , social science , machine learning , philosophy , mathematical economics , path analysis (statistics)
This review article attempts to provide a comprehensive view of mediation analysis with an emphasis on its use in linguistic research. It presents an overview of the basic statistical techniques and tools necessary for the study of the mechanisms underlying the relationships between a predictor, an outcome, and an intermediate variable(s). Traditional methods of inference (e.g., the four-step approach of Baron & Kenny, the Sobel test, and the Structural Equation Modelling) and bootstrapping are described. Direct, indirect, and total effects are defined and the difference between them is clearly shown through examples. This paper, also, tries to focus on some of the most important criterion that should be considered when conducting mediation analysis in order to avoid some critical mistakes that may bias the results of the analysis (e.g., the timing criterion, confounding, sample size) and provides a short review showing the lack of mediation-based research in the field of Linguistics.