Australasian Marketing Journal 11 (3), 2003 33
Background of the problem
An empirical study among 439 customers of a financial services conducted to test the hypothesis of our model and compare it with rival models. Structural (SEM) results revealed that affective commitment can best be explained by the involvement of the position. In addition, affective commitment is the key determinant of mouth, purchase intention and price sensitivity. Managerial implications, limitations and directions for future research are provided. Affective commitment is generally referred to as a perpetual commitment to the desire to maintain a relationship (Morgan and Hunt 1994). The concept of commitment is similar to the concept of long-term orientation that consists of desire and utility of the buyer to have a long term relationship with the seller (Gruen 1995). Along with Day (1969), Dick and Basu (1994) states that the stronger the commitment, the buyer is more likely to overcome potential obstacles in the buyer-seller relationship, resulting in customer loyalty. Similarly, Beatty et al. (1988) stated that commitment and loyalty are related, but according to different definitions of construction, with a commitment leads to loyalty. In line with Allen and Meyer (1990), Geyskens (1998, p 50) states that "the use of measures that measure global commitment intention to continue the relationship without consideration of the underlying motivation may disrupt or mask different, and perhaps even the opposite effect." In this study, we focus on one particular type of commitment: affective commitment. Since it has been shown that affective commitment has the most significant effect on loyalty (Harison-Walker 2001). Affective commitment can be defined as "the desire to continue the relationship because the relationship for its own sake enjoyment, aside from the instrumental value and because they have a sense of loyalty and ownership" (Allen and Meyer, 1990; Geyskens et al. 1996).
The purpose of research
The main objective of this study was to assess the effects of three psychological predecessor (engagement position, will choice and complexity of information) on affective commitment in the regulation of financial services. The purpose of this study was two-fold. First, we assessthree predecessors psychological commitments in financial services regulation. Second, we investigated the effect of affective commitment on customer loyalty. In the organizational psychology literature of different types of commitments that have been identified.
Formulation of the problem
Based on the above background, the question of whether the research is to analyze how much influence aektif commitment to customer loyalty?
Methodology
Data collected from a sample of customers of three bank branches of a Belgian bank, in three medium cities in the Flemish part of Belgium.In April 2001, customers were randomly asked to complete a questionnaire. Four hundred thirty-nine completely filled questionnaires were collected. The sample was found to be representative for a bank customer focus in terms of gender, education, age, and occupation. Multipleitem questionnaire design is based on the measurement scale that has been validated and proven reliable in previous studies. All construction is measured on a seven-point Likert scale ranging from strongly disagree to completely agree.The items measuring different constructs (will of choice, the position of engagement, the complexity of information, affective commitment, rates,insensitivity, purchasing behavior, and complaint behavior).
Result
To test our conceptual model, structural equation modeling (SEM) with observed variables performed using maximum likelihood estimation in LISREL 8.3. A variance covariance matrix is computed using PRELIS program LISREL's companion and used as input for the path analysis.We then analyze the overall model, the measurement model and structural model.
Overall Evaluation Model
In the overall model, chi-square values were significant (428.98 with 179 degrees of freedom), the findings are not unusual with large sample sizes (Doney and Cannon 1997). The ratio of chi-square to degrees of freedom is 2.39, which can be considered as adequate.Value of GFI (0.91) and AGFI (0.88) are slightly lower than those of the CFI (.94), IFI (0.94) and NNFI (0.93). This result is mainly attributable to the former being more easily affected by sample size and model complexity. In general, indicated a good fit, including RMSEA, a .060, and SRMR, being 0.056. Given the purpose of our study, the adequacy of the index, given the fact that the model developed on the basis of the theory, and given the relative complexity of the model, no re-specification of the model were made.
Evaluation Measurement Model
We assess the quality of the measurement model (see Table 1) in unidimensionality, convergent validity, reliability, and discriminant validity. Evidence for the unidimensionality each construct was based on the principal component analysis revealed that the corresponding items loaded at least 0.60 (except for one item of information complexity) for each hypothesized component, with a load no greater than 0.30 in other components. Convergent validity is supported by the model fit was good overall, all phonetically yangSimakBaca load significantly (p <0.01), and the SMC's exceeded 0.50 (Hildebrandt 1987). Reliability demonstrated by the reliability of composite measures all exceeded 0.70 except for the complexity of the information and complaint behavior. Discriminant validity was tested in a series of nested confirmatory factor model comparison the correlation between latent constructs constrained to 1, and the chi-square difference was significant for all comparison models (p <0.01).In addition, the average percentage of variance extracted for constructs almost all greater than 0.50, except for information complexity and behavioral complaints. In short, the measurement model (Table 1) were clean, with evidence of unidimensionality, convergent validity, reliability, and discriminant validity.
Structural Model Evaluation
We could not find a significant relationship between information complexity and affective commitment. Affective commitment clearly positively influenced by the involvement and position in the slightest by the will of choice. In addition, affective commitment plays an important role in explaining customer loyalty. These results support the findings Geyskens (1998) states that the use of a global commitment to build masked influence the types of commitments, emphasizing the importance of our focus on affective commitment. In addition, we explain more than 70% of the variance in affective commitment, which seems to be the result of the impact of the strong involvement of the position even to ignore such important antecedent for example satisfaction, beliefs and values (Sirdeshmukh et al. 2002). This means that the involvement of the position is the kind of engagement that should be considered in future research.Furthermore, the focus on the kind of particualr commitment and refinement of its predecessor seems promising as our empirical results show a clear difference in terms of the direction and magnitude of the relationship between the constructs.Rival Model
It is generally agreed that researchers should compare rival models and not only test the performance of the proposed model (Bagozzi and Yi 1988; Bollen and Long 1992). In discussing the affective construct initial commitment, we provide a theoretical basis for a position as a mediating variable affective commitment. Because we hypothesized parsimonious models do not allow a direct path from one of the three that the process for customer loyalty, it indicates central nomological statusto affective commitment. A rival non-parsimonious model would be one hypothesis only direct road from each preliminary process for the four dimensions of affective commitment and customer loyalty.This model makes affective commitment nomological similar to the four dimensions of customer loyalty. Rival models tested (see Figure 3), therefore, there is no direct permission effect, implying that affective commitment is not allowed to mediate a relationship.According to Morgan and Hunt (1994), we compared our hypothesized model of rival models in the following criteria: overall fit, saving, a good percentage of the model parameters are statistically significant, and the R2 for endogenous construction. Regarding the overall fit, CFI rival models is slightly lower than the hypothetical model (0.93 vs. 0.94) and model rival's ratio chi-square to degrees of freedom is higher than the hypothetical model (2.58 vs. 2.39). In addition, eight additional pathways that need to be estimated in the model competition, reducing parsimony rival models. In addition, only 53% of the trail in a significant rival models compared with 86% in the hypothesized model, showing additional road does not mean theoretically or empirically. Finally, the average explained variance of the four dimensions of customer loyalty in the rival models are similar to the average explained variance in the model hypothesis. This suggests that the direct effect plus the effect of affective commitment indirect preliminary process in the same hypothetical model for three straight strong effect on the process in which rival models in explaining the dimensions of customer loyalty. Based on these findings, we believe that the practice of fitting rival models have strengthened the support we found meaningfulness and robustness of our hypothetical model. Besides conceptual position found support affective commitment as a mediating variable in the model hypotheses, rival empirical model shows the value added.Ignoring the role of mediation is building reduces the parsimony and results in a lower percentage of significant path coefficients.
Conclusion\
The results of this study clearly show that affective commitment is an important determinant of word of mouth, purchase intention, price insensitivity and complaining behavior to a less extent. Affective commitment relates to a sense of belongingness, a customer happy, feeling emotionally attached and feel part of a family of financial service providers. Furthermore, with regard to have comfortable and reliable relationships with financial service providers. Improving aspects of affective commitment than a positive influence on the loyalty of financial service providers, therefore, must ensure that customers are treated as individuals who are really where the organization really interested and concerned. This can be achieved by, among others, the actual personalized communication efforts and to demonstrate attachment and with a personal meaning to each and every individual customer. In line with our studies, particularly the influence of the position of engaging in stimulating affective commitment may not be ignored. The position of engagement can be created by ensuring that the image of the financial services provider comes close to the customer's lifestyle and that financial service providers reflects the personality of the customer in terms of values and norms. Vulnerable customer relationships tend to reveal involvement of high-impact positions affective commitment and ultimately loyalty. Limitations and Directions for Future Research Our study should be seen as an initial effort to address an issue that has important implications for marketing theory and practice. Any initial effort will involve a number of limitations. However, recognition of this limitation also suggests new directions for future studies. The first potential weakness in this study is common method bias.Because we are using a single questionnaire to measure all constructs included, the relationship between the construction may be somewhat inflated.
Suggestion
Future studies should assess the generalizability of our findings to other contexts. Furthermore, future research should concentrate on the issues that can help management to identify, to attract and to retain customers with a strong position involvement commitment, high affective as this customer shows a high tendency to be loyal to the financial services provider. In conclusion, it is expected that the results will serve as a stimulus for additional empirical research involving ongoing relationships in service settings
Tidak ada komentar:
Posting Komentar