Capturing judgment policy on customers’ creditworthiness: A lens model and SDT approach

Designing an intervention program to train human operator's decision-making process and subsequently to improve their performance requires a thorough understanding of the relationships between the environment, the human operator, and cues on which decisions are based. Understanding these relationships provides opportunities to make better decisions when human operators encounter novel situations. This study investigated the operator's decision-making process from the lens model perspective and signal detection approach, in which operators make judgments on customers’ creditworthiness using three types of information such as (1) days that accounts were unpaid, (2) probability of the customers damaging the institution based on the customers’ credit histories, and (3) case-based reasoning (CBR) scores. Results showed that the operators were conservative, which resulted in extremely low performance while maintaining a good level of domain knowledge. Analyses of the relative cue weights also showed that the aggregated judgment strategy represented by the cue weights was similar to the validity of the cues to the ecology, reflected in the high level of policy matching (rm). Specifically, the probability and the days cues were more and equally emphasized over the CBR score.Relevance to industryCapturing the human operator's judgment policy can be used to identify and develop customized training needs for individual operator to make better judgments on customers’ creditworthiness or the environment of interest in general.