جو خدمات و تعهد سازمانی : اهمیت ارتباطات با مشتری

Service employees are simultaneously concerned with their own and their customers' well-being. Managing these dual concerns is of strategic importance in a customer-oriented service firm. This research tests a model comparing overall and customer-linked antecedents and consequences of employee affective organizational commitment. The study determines (a) how service climate variables influence both overall and customer-linked job satisfaction, (b) the contribution of both job satisfaction measures to commitment, (c) the relationship between commitment and both overall citizenship behaviors and customer-linked behaviors, and (d) the influence of commitment with regard to employee intentions to recommend the organization to prospective employees and customers. Co-worker support and the perception of fair treatment are precursors of customer-linked job satisfaction (not overall job satisfaction). Customer-linked job satisfaction is more related to organizational commitment than is overall job satisfaction. Organizational commitment influences both overall citizenship behaviors and customer-linked behaviors as well as intentions to recommend the organization.

Customer-linkage research (Schneider and Bowen, 1985) has recently received increasing attention in both the management and marketing literatures (Bansal et al., 2004 and Pugh et al., 2002). Implicit with this customer-focused approach is the assumption that the organization can design and manage workplace conditions (a service climate) that creates employee satisfaction, commitment to the organization and subsequent behaviors that facilitate the creation of customer satisfaction, perceived service quality and loyalty (Borucki and Burke, 1999, Dietz et al., 2004 and Heskett et al., 1994). The concept of employee organizational commitment is both theoretically and managerially central to customer-linkage research. In this paper, we first focus on employee perceptions of the service climate as antecedents of job satisfaction and organizational commitment and second, on employee behaviors as consequences of organizational commitment. In the present study we compare these overall associations to those that are specific to customer-linked job satisfaction and customer-linked employee behaviors. We address four key questions. How are different service climate variables related to overall and customer-linked job satisfaction? How do these two measures of job satisfaction compare as antecedents of affective organizational commitment? How is affective organizational commitment related to both organizational citizenship and to customer-linked behaviors? Finally, how is employee commitment related to word-of-mouth behaviors leading to the attraction of new customers and new employees? We first describe a research model (Fig. 1) of the antecedents and consequences of affective organizational commitment from an overall and a customer-linked perspective. Second, we develop the conceptual framework leading to the research hypotheses derived from the model. Third, using structural equation modeling (SEM), we test the proposed model. Fourth, with additional data from outside the model, we use analyses of variance (ANOVA) to evaluate the influence of employee commitment on word-of-mouth (WOM) behaviors related to the willingness to recommend the hospital to potential patients and influencers of potential patients as well as to prospective employees. Fifth, the results are presented in the light of the proposed hypotheses. Finally, the discussion treats the theoretical and managerial importance of the findings as well as the limitations and future research opportunities.

Even though most of the multi-item scales used in this study were adapted from well-established measures in the literature, their psychometric properties were still assessed. Statistical procedures used to validate measures included evaluation of dimensionality, reliability, as well as convergent and discriminant validity. We also assessed psychometric properties of the measures by means of confirmatory factor analysis procedures using the EQS structural equation modeling software (Bentler, 1992). Unidimensionality occurs when a set of items forming an instrument all measure just one thing in common. Cox and Cox (2002) tested the unidimensionality of their scales by performing principal components analyses. The same procedure was followed and the results indicate that unidimensionality was achieved for each construct. Confirmatory factor analyses (CFA) using EQS were also conducted to provide a more thorough validation. The results indicate significant loadings (i.e., > 0.50) of all scale items on their respective latent constructs (Table 1). Furthermore, the norm fit index (NFI) of all the measurement models varied between 0.93 and 0.99, which is satisfactory. The comparative fit index (CFI) ranged from 0.95 to 0.99, which constitutes another good indication that each measurement model represented an adequate fit to its respective data. Reliability is defined as the similarity of results provided by independent but comparable measures of the same object, trait, or construct. Based on the recommendations of Churchill (1979), corrected item-total correlations were first computed. Results indicate that all the items had corrected item-total correlations greater than 0.35, which represents the generally accepted cut-off. Subsequently, Cronbach alphas were computed. The alphas varied between 0.80 (for the “organizational commitment” construct) to 0.93 (for the “co-worker support” construct), and were deemed acceptable (Peterson, 1994). Fornell and Larcker's (1981) procedures were followed to evaluate convergent validity of the multiple-item constructs. Convergent validity is established if the average variance extracted for each item accounts for 0.50 or more of the total variance. As shown in Table 1, the average variance extracted for the factors were: 0.61 for co-worker support, 0.57 for supervisor support and for organizational commitment, 0.56 for customer-linkage behaviors, 0.50 for overall citizenship behaviors, and 0.48 for job characteristics (which remains very close to the 0.50 level). Overall, convergent validity was confirmed for each dimension. Moreover, Andersen and Gerbing (1988) note that convergent validity is also demonstrated by statistically significant path coefficients. In this study, all coefficients are significant at the p < 0.05 level. According to Fornell and Larcker (1981), discriminant validity is established if the average variance extracted is larger than the squared correlation coefficients between factors. Results in Table 2 show that this criterion was met across all pairs of multiple-item independent constructs (albeit marginally in the case of co-worker and management support). Furthermore, results from LaGrange Multiplier tests indicated no significant cross-loadings for measurement items with non-hypothesized constructs thus supporting discriminant validity.