Influence of Commitment on Customer Discretionary Behaviour : A Survey in Retail Sector

Retail sector is one of the fastest growing sector in Indian economy. It is the largest source of employment after agriculture and has deep penetration into rural India and generating more than 10 percent of India’s GDP. Retail industry is heavily people-based, requiring interaction and coordination of customers and service providers. Due to those unique characteristics of services, customer’s role in organization has become indispensable. The present study examines the factors of customer citizenship behaviour (CCB) and impact of commitment on customer citizenship behaviour. To analyze the factors, exploratory factor analysis (EFA) has been applied and later on confirmatory factor analysis conform the factors that has been explored through EFA. To measure the impact of commitment on CCB structural equation modeling technique has been used. The findings of the article provide insight into multiple helping behavior of customers which are though discretionary in nature, but help in the overall functioning of the organization.


Introduction
Organizational citizenship behaviour (OCB) has predominantly focused on the service employee behaviour and its focus has been targeted on their behavior towards customer needs, promoting the organizations and its people and to improve organizational effectiveness.Only recently, organizational citizenship behavior has been extended to customer domain.Customers act as productive resources, as contributors to quality, satisfaction and value and as competitors to the organization (Song and Adams 1993;Bitner 1995;Romero and Molina 2011).They make suggestions for service improvement, cooperate during the service encounter, engage in positive word-of-mouth, buy additional services, make recommendations to others and increase price tolerance (Lengnick-Hall 1996; Bettencourt 1997; Fang, Palmatier and Evans 2008; Schmitt2011).It is positive, voluntary, helpful and constructive behaviors enacted towards the firm and other customers (Bove et al. 2009;Gruen, 1995;Godwin and Kalpana 2013).A series of studies recognize the role of positive customer functions and proposes concepts such as customer citizenship behavior and voluntary performance.
Considering the importance of positive behavior of customers in company's growth and performance it is necessary to investigate what promotes the citizenship behavior in customers.However, researchers have paid little attention to what are the factors that affect customer citizenship behavior.There is lack of censuses among researchers and academics regarding dimensions of citizenship behavior.Against this background, this study attempts to study extends by embedding the dimensions of citizenship behavior.

Customer Citizenship Behaviour
Groth (2005) defines customer citizenship behavior as "voluntary and discretionary behavior by individual customer that is not directly or explicitly expected or rewarded, but that aggregates into higher service quality and promotes the effective functioning of service quality and promotes the effective functioning of service organizations (e.g., helping another customer or providing suggestions to the service organization)".Customer citizenship behaviour (CCB) is based on theory of social exchange where customer reciprocates positive behavior from a sense of personal obligation or gratitude.Yi, Gong and Lee (2013) argued that when customers engage in citizenship behaviour, they foster citizenship behavior among other customers.CCB is a form of customer extra-role behavior, and is helpful, kind, considerate or thoughtful acts voluntarily performed.
Hsieh, Yen and Chin (2004) deem such helpful behaviors which are performed by customers as customer voluntary Influence of Commitment on Customer Discretionary Behaviour: A Survey in Retail Sector performance (CVP).According to Rosenbaum and Massiah (2007) customers who display CVP by exhibiting instances of loyalty, participation or cooperation are actually indulging in organizational citizenship behavior.Therefore customers' citizenship is in which customer spreads positive word of mouth about a firm, acts as a firm's partial employee and co-operates with the firm's employees.These include act of cooperation, helpfulness and similar gestures of kindliness which are often only indirectly related to the organization's core work activities (Lengnick-Hall, Claycomb and Inks 2000).Bettencourt (1997)  Benevolent Act of Service: Waiting for some extra time for their service, helping employees in performing service.

Commitment
Commitment entails a desire to develop a stable relationship, a willingness to make short-term sacrifices to maintain the relationship and confidence in the stability of the relationship (Fullerton 2003).Customers who perceive a social exchange relationship as evidenced by higher levels of commitment are likely to be more cooperative and willing to put forth effort on behalf of the firm (Bettencourt 1997).It is considered as an essential ingredient for successful long-term relationships (Morgan and Hunt 1994).In a model of commitment developed by Meyer et al. (2001) the three approaches were labelled 'affective', 'continuance' and 'normative' commitment respectively.Among all commitments the one most related to identification is affective commitment, which is defined as 'positive feelings of identification with, attachment to and involvement in the work organization (Crameli 2005).It reflects customers' psychological attachment to a firm and their belief in the importance of the relationship.It implies a higher level of obligation to make a relationship succeed and to make it mutually satisfying and beneficial in a reciprocating manner (Gunlach et al. 1995).
Commitment is one of the main concepts in relations' marketing in social-exchanges theory (Dwyer, Schurr and Oh 1987; Hennig-Thurau2002).Commitment is a long-term exchange of parties for maintain valuable permanent relations (Moorman Zaltman and Deshpande 1992; Morgan and Hunt 1994).This tendency may be a result of lateral advantages of long-term relations with service organizations such as increase in confidence and mutual social relations (Gwinner, Gremler, Bitner 1998).
Committed customers are likely to perform voluntary behaviors such as CCB because they identify with the firm's goals and values and are interested in the welfare of the organization (Bartikowski and Walsh 2011).Bove et al. (2009) investigated the link between commitment and customer citizenship behavior.Their findings approved positive significant relationship in commitment and citizenship behaviour.Commitment leads to functional behavior patterns characterized by personnel sacrifice made for the purpose of organization, persistence in behavior which is not dependent on reward and punishment and a devotion of personnel time to organizational activities (Chonko 1986).On the basis of above lines, we can hypothesize that: H1: Commitment has a positive impact on customer citizenship behaviour.

Objectives of the Study
The present study is planned with following research objectives: 1. To determine the factors affecting customer citizenship behaviour.2. To study the impact of commitment on different dimensions of customer citizenship behaviour.

Research Methodology Table 1. Sample characteristics
The present study has been carried out to know the dimensions of customer citizenship behaviour and role of commitment on customer citizenship behaviour.A sample of 558 customers who visits supermarkets for their daily needs were taken.Table 1 provide description of sample.A pre-tested structured questionnaire was used to collect primary data which contained question regarding customer citizenship behaviour and commitment.Each statement was measured on a 7-point Likert scale ranging from 'Very Strongly Agree to 'Very Strongly Disagree'.Exploratory Factor analysis has been used to determine the factor affecting customer citizenship behaviour and then confirmatory factor analysis has been applied to conform the factors.Structural equation modeling has been applied to know the impact of commitment on customer citizenship behaviour dimensions.

Analysis and Discussion of Results
To explore the dimensions of citizenship behaviour, consumers of supermarkets were asked 13 statements on 7 point Likert scale ranging from 7 (very strongly agree) to 1 (very strongly disagree).Factor analysis has been employed to using PASW statistics 18.0.Reliability analysis has been conducted and the Cronbach alpha is determined as 0.845 which indicate the reliability of the scale.Further factor analysis is applied to 11 variables.Table 2 shows the Kaiser -Meyer -Olkin (KMO) measure of sampling adequacy and Bartlett's test of sphericity.
The value of KMO is determined as 0.847 and the Bartlett's test of sphericity as 4084.770(significant at 1 per cent, p< 0.000) that justify the usage of factor analysis.Factor analysis has been performed using the principal component extraction method with varimax rotation, on the basis of Eigen value not less than 1 for any factor.Table 3 shows the results of the factor analysis by depicting rotated values of factor loadings, Eigen values, percentage of variance explained by factors extracted and cumulative percentage of variance explained.Nomenclature of the factors extracted has been given on the basis of highest factor loadings of the variables onto a particular factor.Influence of Commitment on Customer Discretionary Behaviour: A Survey in Retail Sector Table 3 demonstrates the results of factor analysis by documenting the three factors.Three items have loaded on 'Providing the Feedback' factor with the variance of 27.202 per cent.'Helping the Company' with four items contributes a variance of 26.184 per cent.'Helping Customer' contributes a variance of 24.493 with three items, altogether explaining the cumulative variance of 77.878 per cent.This finding is in line with Bartikwoski and Walsh (2011) who found that customers engage in extra-role behaviors, such as helping other customer, improving the service delivery and providing the feedback.This finding confirms that of Yi, Natarajan and Gong (2011) customer citizenship behaviour influence employees indirectly by enhancing role clarity and fostering a sense of clarity.These findings are in consonance with Johnson and Rapp (2010) and Yi and Gong (2012) who argued that customer citizenship behaviour is multi-dimensional construct.

Confirmatory Factor Analysis
The first test in the CFA analysis is to test the hypothesized measurement relationships and evaluate the reliability and discriminant validity.The results of the confirmatory factor analysis demonstrated relatively a reasonable fit of model to the data on the basis of a number of fit statistics as shown in figure I.

Model Fit
The output of the AMOS yielded a chi-square value of 172.405, with 59 degrees of freedom and a probability of less than 0.0001 (p = 0.000), thereby suggesting that the fit of the data to the hypothesized model is not entirely adequate.However, according to Bagozzi and Yi (1988)

Construct Reliability
The most widely used measure to assess the internal consistency of constructs is Cronbach's alpha (Nunnally and Bernstein 1994).The generally agreed upon value of Cronbach alpha is 0.70 although it may decrease to 0.60 in case of exploratory research (Hair et al. 2006).In this research the reliability measure for all the constructs range between 0.856 and 0.94, which is acceptable.Hence the construct reliability is satisfactory in this research.

Convergent Validity
Anderson and Gerbing (1988) state that the convergent validity of a model can be assessed by determining whether the path estimates between the measurement items and their respective Latent constructs are significant or not.In case of the AMOS output the standardized estimates of all the measurement items were significant as shown in table 3 Each variable exhibits significant loadings which supports the convergent validity.Table 3 shows that all reliability estimates were greater than 0.70 (Nunnally and Bernstein 1994;Hair et al. 2006).4 shows that the AVE's are greater than inter-construct squared correlations estimates which supports discriminant validity.

Model Fit of Second Order CFA
The output of the AMOS yielded a chi-square value of 183.650, with 61 degrees of freedom and a probability of less than 0.0001 (p = 0.000), thereby suggesting that the fit of the data to the hypothesized model is not entirely adequate.However, according to Bagozzi and Yi (1988)

Limitations and Future Directions
Although the study yields significant insights, some limitations are worth addressing, and some promising areas arise for future research.First, the present research utilized a cross sectional study in a single industry.This study is limited only to supermarkets.This single-sector study raises some concerns about the generalizability of the findings.Future studies could consider to what extent the measures proposed in this study are valid in different service industries and what modifications need to be made in the scale items across different samples and contexts.Future studies replicating this research across multiple industries and sectors would increase the understanding of customer-based corporate reputation and customer citizenship behavior.Second, the sample selection was based on a convenience sampling, which is often used for exploratory work, rather than a random probability sample.Additional research could be conducted using a random probability sample.Third, the sample represented a limited number of supermarkets in retail industry.The study is based on a self-reported questionnaire.It is possible that respondents, may have, at least to some extent, biased the results by responding how they felt when the event occurred.Thus the results of this study should be considered indicative rather than definitive and future research should address this limitation by examining actual behavior instead of self-reported and behavior intentions.
Having identified the limitations of the study, this section then provides some suggestions for future research to extend the current body of knowledge in the literature on corporate reputation, corporate strategy and brand strategy.Four directions for future research can be identified.This research has been applied to examine the retail industry.However, this whole process of theoretical building can be applied to other contexts of study such as another industry or another country, for digging in depth regarding the issue in different contexts.Future research can be structured in a similar way to the theoretical building of this research.Future research as a comparative study, possibly with more than two different industries or countries is recommended in order to understand this theoretical structure more thoroughly.Thus, there is a need to replicate and extend this study to other contexts either in other industries or countries.Future studies could consider to what extent the measures proposed in this study are valid in different service industries and what modifications need to be made in the scale items across different samples and contexts.Further, personal characteristics like age, income, education play an important role in shaping a customer's behavior intentions.Future research may test the moderating effects relationship age, perceived risk and in-role behavior on customer citizenship behavior.

Conclusion and Implications
Present research work has been carried out to study the dimensions of customer citizenship behaviour as it lack uniformity (Bove et al. 2008).In addition, this study measured the impact of commitment on dimensions customer citizenship behaviour.This research provide empirical support for that committed customers are motivated to go beyond prescribed roles.Through the performance of extra-role/discretionary behavior, customers indirectly raise service quality and service productivity (Keh and Teo 2001).Hence service managers should view customers as a valuable resource, not in terms of in-role behavior, but also in terms of potential role as partial employees in improving social psychological services cape of the firm (Bove et al. 2009).The findings of the article also provide insight into multiple helping behavior of customers which are though discretionary in nature, but help in the overall functioning of the organization.Based on social exchange theory it is quite possible service workers who have been recipient of positive behaviour from customers end will reciprocate by engaging themselves in greater levels of citizenship behaviour themselves.This scale can be used both by researchers and practitioners to measure the citizenship behaviour that a customer engage in towards the organization at a specific point of time.

Figure 1 .
Figure 1.First order Confirmatory Factor Analysis and Mulaik et al. (1989), the chi-square statistics is not always the best indication of model fit.The literature on model fit indices reports various other indices which reflect model fit.For the measurement model in Figure 1 the other model fit indices are the following: CFI (comparative fit index) = 0.976; GFI (goodness-of-fit index) = 0.955; AGFI (adjusted GFI) = 0.930; PGFI (parsimonious GFI) = 0.619; RMR (root mean square residual) = 0. 06; TLI (Tucker & Lewis index) = 0.968; NFI (normed fit index) = 0.964; RFI (relative fit index) = 0.953; RMSEA (root mean square error of approximation) = 0.05; BIC (Bayesian information criterion) = 374.784.The values of the fit indices mentioned above indicate a reasonable fit of the measurement model with data (Byrne, 2001).

4. 3 . 3 .
Discriminant Validity As proposed by Fornell and Larker (1981) discriminant validity can be assessed by comparing the average variance (AVE) in indicators explained by the constructs and the corresponding inter-construct squared correlation estimates.The table

Figure 2 .
Figure 2. Second order Confirmatory Factor Analysis and Mulaik et al. (1989), the chi-square statistics is not always the best indication of model fit.The literature on model fit indices reports various other indices which reflect model fit.For the measurement model in Figure 1 the other model fit indices are the following: CFI (comparative fit index) = 0.974; GFI (goodness-of-fit index) = 0.951; AGFI (adjusted GFI) = 0.927; PGFI (parsimonious GFI) = 0.638; TLI (Tucker & Lewis index) = 0.967; NFI (normed fit index) = 0.962; RFI (relative fit index) = 0.951; RMSEA (root mean square error of approximation) = 0.06; BIC (Bayesian information criterion) = 373.381.The values of the fit indices mentioned above indicate a reasonable fit of the measurement model with data (Byrne, 2001).Influence of Commitment on Customer Discretionary Behaviour: A Survey in Retail SectorThe results of structural model which is shown in figure III revealed that commitment had a significant impact on customer citizenship behaviour.In examining the hypotheses.Path analysis demonstrates a significant relationship between commitment and citizenship behavior.Hypothesis 1 suggested that commitment is significantly and positively associated to customer citizenship behavior.The result supported this hypothesis (H1: β = 0.53, p<0.001).

Figure 3 .
Structural Equation Model

Table 3 .
Dimensions of Customer Citizenship Behaviour Source: Authors' Calculations'

Table 3 .
Overall Measurement Model Analysis