ABSTRACT
The chi-square test (also written χ2 test) is a type of inferential statistic for analyzing frequency counts of nominal data. It is used to determine whether the counts of two nominal variables are associated with each other. It is useful when questions about relationships among variables cannot be answered by means of a correlational analysis, such as Pearson’s correlation coefficient or Spearman’s rho, or by means of a comparative analysis, such as a t-test or ANOVA. The following are some examples of questions that have been answered by using the chi-square test:
Liu (2011) investigated whether the occurrence of phrasal verbs relates to register. Specifically, does English used in spoken interaction, fiction writing, magazines, newspapers, and academic writing differ in terms of the frequency of the use of phrasal verbs?
Bell (2012) asked whether playful engagement with language facilitates the learning of meaning and form as compared to non-playful engagement.
Laufer and Waldman (2011) sought to address the question of whether native speaker status relates to the production of noun-verb collocations, and whether learners’ different levels of proficiency relate to their production of deviant collocations.
