An empirical analysis of behavioral maintenance for organizational change in Ethiopia through machine learning techniques

Behavior maintenance for organizational change is the continuous behavior performance following an initial intentional change. This research examines the importance of factors that influence behavioral maintenance for organizational change. This study proposes a research model incorporating self-determination, regular-fit, self-concept, and habit theories to identify potential influencing factors of behavioral maintenance for organizational change in Ethiopia and quantify the importance level of these factors using ML techniques. A survey study was carried out in Addis Ababa, Ethiopia, with 310 valid responses. The comparison of five different ML techniques shows that Naive Bayes (GaussianNB) outperforms the other classification model. Naive Bayes (GaussianNB) model-based feature importance analysis shows that perceived competency, perceived enjoyment, and perceived autonomy are the most prominent contributor to behavioral maintenance for organizational change. The results confirmed that the quality of individuals' motivation affects the extent to which individuals will engage in, and persist with, behaviors. Key Words:Behavioral maintenance, organizational change, machine learning, quantitative study, Naive Bayes (GaussianNB)