The relationship between lexical complexity measures and language learning beliefs

Abstract: Although the relationship between language proficiency and learner beliefs is generally viewed as weak, indirect, and distant, there are empirical findings which show that the relationship between syntactic complexity measures and language learning beliefs is statistically tangible. Since syntactic complexity is only one constituent of the linguistic complexity system, it seems plausible to question whether other constituents of the system are also in statistically measurable relationships with language learning beliefs. This research project explores the relationship between 25 lexical complexity measures (Lu 2012; 2014) and four subscales of language learning beliefs that are suggested for Horwitz’s (2013) Beliefs about Language Learning Inventory—BALLI 2.0 (Kovačević 2017). For three semesters (Fall 2014, Spring and Fall 2015), 152 freshman students at the International University of Sarajevo responded to BALLI 2.0 and wrote in-class exam essays which were converted into an electronic format. The results show 15 statistically significant correlation coefficients between 14 lexical complexity measures and three BALLI 2.0 subscales. Overall, it may be concluded that the relationship between lexical complexity measures and language learning beliefs is statistically detectable. The findings imply that the lexical complexity framework offers valuable opportunities for exploring how and to what extent particular individual differences manifest in foreign language production.