Abstract
Purpose - Intentions are often included in service research, but researchers have paid little attention to a discussion in psychology in which different intention constructs are distinguished. This study is based on the belief that different intention constructs capture different aspects of the customer's assessments of his or her future repatronizing behavior - and that intentions measures based on different intention constructs are not equally correlated with firstly, the customer's global evaluation of the supplier, such as satisfaction, and secondly, his or her overt repatronizing behavior. The specific purpose is to examine if such variation is at hand in with regards to two specific intention constructs: intentions-as-expectations and intentions-as-wants. Design/methodology/approach - A first questionnaire was used to collect data on satisfaction and intentions in a restaurant setting, and a second questionnaire - distributed to the respondents one month after the first questionnaire - captured behavioral data. These data were used to assess associations between the main variables (satisfaction, intentions, and behavior). Findings - The analysis shows that the two intention constructs produced different strength in the association with customer satisfaction and with repatronizing behavior. In addition, the findings suggest that the two constructs are characterized by different levels of assessment volition, and this also serve as our main explanation of the results. Originality/value - The findings imply that service researchers should pay careful attention to how intentions are conceptualized and operationalized, because an incautious selection of one intention construct over another may affect the role of intentions as mediators of the link between satisfaction and behavior.