Revisiting Problem Gamblers’ Harsh Gaze on Casino Services: Applying Complexity Theory to Identify Exceptional Customers Arch G. Woodside Boston College Catherine Prentice Swinburne University of Technology Anders Larsen Copenhagen Business School ABSTRACT This study revisits the theory, data, and analysis in Prentice and Woodside (2013). The study here applies fuzzy-set qualitative comparative analysis (fsQCA) to customer service–evaluation data from seven mega casinos in the world’s gambling capital—Macau. The study includes contrarian case analysis and offers complex algorithms of highly favorable customer outcomes—an alternative stance to theory and data analysis in comparison to the dominant logic of statistical analyses that Prentice and Woodside (2013) report. Contrary to their principal hypotheses, Prentice and Woodside (2013) report a negative main effect between problem gambling and casino service evaluations. The findings in the reanalysis here include more complex, nuanced views on the antecedent conditions relating to high problem-gambling, immediate service evaluations, and desired customer behavior measures in casinos. Counter to the findings using symmetric testing via multiple regression analysis in Prentice and Woodside (2013), the present study, using asymmetric testing via fsQCA, recognizes the occurrence of causal asymmetry, and draws conclusions from different algorithms leading to high scores for favorable and unfavorable outcome conditions. The findings indicate that not all problem gamblers gaze on casino services harshly; the minority of problem gamblers who view casinos positively versus harshly may be the most valuable customers for the casinos—the casinos’ exceptional customers. © 2014 Wiley Periodicals, Inc. Using the “Problem Gambling Severity Index” (PGSI) and grouping gamblers into categories by severity, Prentice and Woodside (2013) propose that problem gamblers are unique in antecedent conditions and have a positive relationship with service evaluations of the service providers (i.e., the casinos). Using data col- lected from inside seven casinos in the world’s largest gambling market (Macau), their study tests the two hypotheses. The study finds statistically significant relationships between demographic and gambling be- havioral antecedent and problem gambling. The study also finds statistically significant negative relation- ships with problem gambling and casino service evalu- ations, suggesting that problem gamblers view casino services harshly. This negative relationship finding was contrary to the hypothesis of their study, and Prentice and Woodside offer revisions of the theory by adapt- ing Nataraajan and Goff’s (1991) motive-control con- tinuum: the harsh view is due to a blame of the service provider for likely negative outcomes (e.g., gambling losses) rather than oneself. Compulsive buyers may both want to buy and to avoid such behavior (Frost, Stekettee, & Williams, 2002). Nataraajan and Goff (1991) suggest two types of compulsive buying based on this distinction: obsessive–compulsive disordered buy- ing and impulse-control disordered buying. How each problem gambler comes to understand and manage such a personal desire–avoidance conundrum likely re- sults in a wider variance of responses to problem gam- blers’ evaluations of casino services than captured by reporting only a negative (or positive) statistically sig- nificant main effect. Prentice and Woodside (2013) both oversimplify Nataraajan and Goff’s (1991) theory and miss reporting on valuable information available by Psychology and Marketing, Vol. 32(1): 65–77 (January 2015) View this article online at wileyonlinelibrary.com/journal/mar © 2014 Wiley Periodicals, Inc. DOI: 10.1002/mar.20763 65