مقیاس های اندازه گیری جدید برای ارزیابی برداشت از تکنولوژی به واسطه تجربه خدمات مشتری

Service organizations are increasingly utilizing advanced information and communication technologies, such as the Internet, in hopes of improving the efficiency, cost-effectiveness, and/or quality of their customer-facing operations. More of the contact a customer has with the firm is likely to be with the back-office and, therefore, mediated by technology. While previous operations management research has been important for its contributions to our understanding of customer contact in face-to-face settings, considerably less work has been done to improve our understanding of customer contact in what we refer to as technology-mediated settings (e.g., via telephone, instant messaging (IM), or email). This paper builds upon the service operations management (SOM) literature on customer contact by theoretically defining and empirically developing new multi-item measurement scales specifically designed for assessing technology-mediated customer contact. Seminal works on customer contact theory and its empirical measurement are employed to provide a foundation for extending these concepts to technology-mediated contexts. We also draw upon other important frameworks, including the Service Profit Chain, the Theory of Planned Behavior, and the concept of media/information richness, in order to identify and define our constructs. We follow a rigorous empirical scale development process to create parsimonious sets of survey items that exhibit satisfactory levels of reliability and validity to be useful in advancing SOM empirical research in the emerging Internet-enabled back-office.

Service organizations are increasingly utilizing advanced information and communication technologies, such as the Internet, in hopes of improving the efficiency, cost-effectiveness, and/or quality of their customer-facing operations (Fitzsimmons and Fitzsimmons, 2004, Huete and Roth, 1988, Haynes and Thies, 1991, Hill et al., 2002, Oliveira et al., 2002, Roth, 2000 and Boyer et al., 2002). More of the contact a customer has with the firm is likely to be with the back-office and mediated by technology (e.g., via telephone, instant messaging (IM), or email). Service organizations are finding that interacting with their customers via these new technologies can be a significant challenge (Zeithaml et al., 2002). While Internet-based customer support can potentially reduce costs on a per-transaction basis, customer satisfaction, as well as long-term customer loyalty, can be severely damaged by a bad on-line experience (Barnes et al., 2000 and Heim and Sinha, 2001a). Therefore, the service operations management (SOM) issues related to managing customer contact in these technology-mediated environments warrant attention. The concept of customer contact (Chase, 1978 and Chase, 1981) has long been an integral element of SOM research. Customer contact is generally held to be a function of the interaction between a customer and a service provider (Kellogg and Chase, 1995). Initial work to empirically define the underlying dimensions of customer contact was performed using a hospital setting, an environment in which all contact between customers and employees occurred in-person and face-to-face (Kellogg and Chase, 1995 and Soteriou and Chase, 1998). While extremely valuable, it is unclear whether the results of this research are equally applicable to contexts involving customer contact in technology-mediated (i.e., non-face-to-face) service delivery processes. We extend the abilities of SOM researchers to examine these environments by adapting Chase’s initial ideas of customer contact to these new technology-mediated contexts, using customer perceptions. This paper builds upon prior literature by theoretically defining the conceptual domains of inquiry, constructs, and operational measures specific to advancing SOM research in technology-mediated customer contact situations. This research context is particularly applicable for the emerging area of e-services. The unit of analysis in this research is the customer. We follow a normative two-step process. First, we identify ten theory-based constructs covering three domains that comprise antecedents and consequences of technology-mediated customer contact from a customer’s perspective. Second, because the constructs are latent (i.e. non-observed) variables, we apply a rigorous procedure for ensuring the psychometric adequacy of the resulting new multi-item measurement scales. While Kellogg and Chase’s (1995) seminal work on customer contact theory and the hypothesized Service Profit Chain (Heskett et al., 1994) helps motivate and structure our thinking in the SOM arena, we also draw upon the Theory of Planned Behavior (Ajzen, 1985 and Ajzen, 1991), and incorporate the concept of media/information richness (Daft and Lengel, 1984 and Daft and Lengel, 1986) in identifying and defining our constructs. Once the constructs are defined, we then follow a rigorous empirical scale development process in order to identify parsimonious sets of survey items that exhibit satisfactory levels of reliability and validity. Section 2 presents a brief background of the research context and defines and illustrates the specific constructs for which new measurement scales are developed. The third section provides details on the preliminary scale development methodology and field study database. Section 4 describes and reports on confirmatory modeling results. In Section 5, we conclude with a discussion of the implications of our results and usage of the scales, review the limitations of our study, and offer some concluding thoughts.

The primary contributions of this paper are the definition of new constructs associated with the technology-mediated customer service experience and the development of new multi-item measurement scales for measuring these constructs. Unlike much prior SOM research, our study takes a grounded theory approach using customers’ perceptions. Future SOM empirical research linking these constructs in causal or structural models in a technology-mediated customer contact situation will benefit significantly from the existence of relevant construct definitions and good measurement scales. A secondary contribution of this work is the demonstration of a rigorous empirical scale and item development process. Like any research, our approach and our results have some limitations. First, the use of convenience samples in the pre- and pilot-tests may have limited our insights early in the process. The use of random sampling in the final data analysis, however, alleviated much of the concern regarding this issue. A second limitation is the fact that our CFA resulted in some scales having only two indicators. While this may prove to be a limitation in some applications and some models, identification methods do exist that support their re-use in new models (Bollen, 1989). Moreover, the reader is provided with three or more items for each scale (with the exception of DAB) in Appendix B. While our CFA results indicated that our measurement model fared better with certain items dropped from our final analysis, future researchers may find that these specific items may perform adequately in their research contexts. A third potential limitation centers on the possibility of discriminant validity of the UB and AE constructs. While we feel that a reasonable explanation is that a strong causal relationship creates a large statistical covariance between these two constructs, future researchers who employ the UB and AE scales should pay particular attention to this potential issue when examining their results. Finally, while CFA is generally the preferred method for establishing convergent and discriminant validity for a measurement model, it is not without weaknesses, two in particular are noted by Bagozzi et al. (1991). They indicate that the base structure of CFA has two potential problems. First, it confounds two different sources of variance (measurement error and unique true-score variance other than that explained by the traits and methods) in the observed measures. Second, it assumes that variation in indicator variables is strictly a linear combination of these sources of variance, thereby eliminating any opportunity for discovering trait-method interactions. Given the consistently high reliabilities of our scales, the first weakness should not be of considerable concern. Given that our scales are assessed using a single method, we have no opportunity for detecting whether the linearity constraint is a source of concern. This is a potential limitation of the study, but is consistent with other research that relies on CFA to support claims of construct validity. Despite these issues, scale development, like research in general, is an iterative process. We feel our new multi-item measurement scales based on the B–A–I theoretic framework will provide SOM researchers with robust starting points when investigating technology-mediated customer contact and e-service operations from a customer’s perspective, a relatively new line of research. Both academia and industry are likely to benefit from additional research on how customer contact can be most effectively managed in technology-mediated service environments, and these clearly defined constructs and robust measurement scales will significantly aid future researchers’ investigations on this important topic.