A new correlation analysis method used in behavioural science research

摘要

Human behaviour involves complex cognitive, affective, and experience-based processes. It is empirically measured by a self-reported questionnaire associated with labeled scale items given before or after completion of a cognitive test. Traditional correlation analysis seems to be defective in dealing with such behavioural measures. This study proposes a new correlation analysis method to measure human information behaviour. The proposed method is based on the grey relational analysis accompanied by the average linkage operation. An experiment concerning unemployed adult novices' e-learning behaviour was conducted. Through the empirical study, we can identify certain user characteristics and behaviour intentions of such learning environment with respect to individual differences in gender, age, and educational level. The proposed method can be used to quantify the attributive correlations among different classified groups, from individual analysis of the relational grades to the holistic understanding of the groups' behavioural correlations.