| Abstrakt: |
Research Question Should machines replace interpersonal interactions in service encounters? Current business practice suggests so. While research has long suggested that "service with a smile" improves customer satisfaction (Barger and Grandey 2006), business practice has not shied away from standardizing processes and automating the human part in service delivery. Today, customers conduct their financial business online, order their retail on the Internet, or interact with a self-service kiosk to get their airline boarding-pass. In strive for increased service productivity, businesses often substitute expensive service personnel with self-service technologies (SSTs) and actively "push" customers to these channels (White, Breazeale, and Collier 2012). Considering the broad introduction of SSTs in a vast variety of service industries, a more nuanced understanding of customers' psychological responses to a technology-based selfservice becomes all the more important. This study aims to address this issue. Drawing from the theories of the personsensitivity bias and attribution-biases in human cognition, this research contrasts customers' attribution-processes and subsequent satisfaction judgments between technologybased self-service and personal service channels. More precisely, we aim to address the questions (1) if customer satisfaction with a service outcome differs between technologybased self-services and personal services and (2) how and why customers' evaluations may differ. Method and Data We have examined a large-scale customer satisfaction survey and conducted two experiments to examine the personsensitivity-bias and its boundaries in service settings in detail. In our field study, we draw on customer satisfaction data collected from 157,240 airline customers from January 2009 to July 2010 across different members of an airline alliance. This study provides first support for a person-sensitivity bias in service encounters by demonstrating that customers are in fact less dissatisfied with a wrong seat assignment during check-in when they used a self-service channel instead of a personal-assisted channel. We validate these findings in two experiments. Both experiments were conducted online and relied on different service scenarios, which represented the experimental conditions. Experiment 2A establishes the existence of a person-sensitivity bias in service settings by demonstrating that customers evaluate personal, "high-touch" services in more extreme manners than technology-based, "high-tech" services. In Experiment 2B we demonstrate an important boundary of this effect. By drawing from literature on interdependence vs. independence and extending the main experiment to an intercultural level, this study shows that the channel effects mostly arise in highly independent (Western) cultures, and cease to exist in more interdependent (Eastern) cultures. Summary of Findings This research shows that customers tend to evaluate personal, "high-touch" services in more extreme manners than technology-based, "high-tech" services. Accordingly, when service outcomes are good, customers are more satisfied with a provider when using a personal service instead of a self-service. When service outcomes are poor, however, customers are more satisfied (or less dissatisfied) when using a technology-based self-service instead of a traditional personal service. By drawing from attribution theory, this study further demonstrates that differences in satisfaction levels between service channels arise as customers make different causal inferences for an outcome when using a technologybased self-service instead of a traditional personal service. In particular, the results reveal that customers of a personal service tend to overestimate the power of the service employee to cause a particular outcome and hence assume that the service employee causes the outcome intentionally. Personal service customers consequently attribute more responsibility to the provider than self-service customers regardless of the service outcome. Technology-based selfservice customers, on the other hand, are more egocentric and hence consistently attribute more responsibility to themselves or-when outcomes are poor-to external, situational factors. Key Contributions Findings of this research reconcile seemingly contradictory findings of human-machine and social psychology research. Accordingly, we demonstrate that humans react similarly to humans and machines and hence engage in self-serving attributions in both self-service and personal service settings, as suggested by social response theory. Results, however, also reveal that customers differ significantly in their attribution of responsibility when interacting with a machine instead of a human. Self-service customers are thus more likely to consider external, situational factors as well as their own responsibility that might have contributed to an outcome. As a result, this study also finds support for a person-sensitivity bias as customers react to personal service encounters in more extreme manners than to self-service encounters. This study also uncovers an important limitation of this channel effect. By drawing from literature on interdependence vs. independence and extending the previous experiment to an intercultural level, we show that the channel effects mostly arise in highly independent, i.e. Western, cultures, but not in more interdependent, i.e. Eastern, cultures. Results reveal that one reason for this limitation is the fact that these customers assign responsibility for an outcome differently. [ABSTRACT FROM AUTHOR] |
