Perceived Fairness , Emotions , and Intention of Fast Food Chain Restaurants Customers in Indonesia

This study aims to investigate interrelationships among perceived service fairness, emotions, and behavioral intentions in a fast food chain restaurant context. we use terms that are commonly use on the study of fairness or justice perception. This study uses purposive sampling and the survey method to generate our sample which consists of 800 respondents from big cities in Indonesia, namely: Jakarta, Semarang, Surabaya, Medan, and Makassar. The data are analyzed using Structural Equation Modeling (SEM). The results show different roles for each fairness perception in relation to peoples’ emotions and behavioral intentions, based on the Mehrabian-Russel model. Three fairness variables (price fairness, outcome fairness and interactional fairness) have positive and significant effects on customers’ positive emotions, while, procedural fairness does not influence the formation of a positive emotion. Furthermore, a positive emotion has a positive influence on a customer’s behavioral intention. Data collected in this study are limited to the context of the restaurant industry, therefore, precaution must be taken when generalizing these results to other industries. The results of this study can serve as guidelines for managers in the restaurant industry to develop effective and efficient strategies for ensuring their services’ perceived fairness and its impact on both customers’ retention rates and the companies’ financial gains.


Introduction
Over the last decade, fairness principles (distributive fairness, procedural fairness, and interactional fairness) have been applied to numerous studies such as management, psychology, and organizational attitudes (Azar and Darvishi 2011;Blader et al. 2013), as well as legal and political studies (i.e.Zmerli and Castillo 2015).For example, Azarand Darvishi (2011) created a measure for organisational fairness based on the fairness theory from Folger and Cropanzano (1998), where they measure organizational fairness based on fairnesses' perception.They show that the use of a different measurement for organisational fairness will produce a different result.
Academics and practitioners' interest in conceptualizing service fairness has increased along with the development of service marketing literature (Aggarwal and Larrick 2012).The studies by Heo and Lee (2011), Liang et al. (2017), Wang et al. (2018) all show that when customers are aware of any deviation from the fairness principles, it may trigger a perception of unfairness.Moreover, consumers' individual characteristics will influence their perception of fairness depending on what they experience.In line with these studies, Aggarwal and Larrick (2012) show that consumers' responses toward perceived interactional fairness are contingent on the form of their social relationship (communal and exchange), including the conditon of perceived distributive fairness.Linking its effects on emotions and attitudes, Smith and Bolton (2002) argue that customers are involved in a psychological contract with the service provider who gratifies their needs and treats them fairly (Chalmers 2016).When a service provider neglects fairness when providing a ser-vice to their customers, it may create a negative emotion among the customers, which could lead them to boycott the service provider, or make public their unpleaseant experience (Kaura et al. 2015;Liang et al. 2017;Liu et al. 2017) For customers, fairness and quality service are inseparable points (Sekhon et al. 2016).Fairness is not merely one dimension of service, rather, it embraces all the dimensions of good quality service (Aggarwal and Larrick 2012;Naquin et al. 2015).Giovanis et al. (2015) and Nikbin et al. (2016) argue that poor service is not a service that equates with low quality service, it is a distinct phenomenon.In relation to the quality of service, the fairness principle offers a specific framework to understand the process of a service's consumption related to it's fairness, and this may be use as a substantial component in evaluating the service (Cropanzano et al. 2008;Heo and Lee 2011).Therefore, it is worthwhile studying the appropriateness of fairness as one of the instruments when evaluating any service (Aggarwal and Larrick 2012).
It has been widely recognized that considerations regarding fairness elicit emotional responses, particularly, when unfairness occurs (Cohen-Charash and Spector 2001; Barclay et al. 2005) and finally lead to behavioral reactions such as complaints, negative word-of-mouth recommendations, or support (Adams 1966).Furthermore, despite previous fairness theory and other studies having shown that emotion forms a part of the relationship between experiencing unfairness and wanting to retaliate, there are only a few studies that have tested the relationship between fairness, emotion, and behavioral intention (Cohen-Charash and Spector 2001; Barclay et al. 2005).
For example, Cropanzano and Stein (2009) show that organizational unfairness is capable of influencing the ocurrence of employees' negative emotions, which can lead to their unethical behavior as a result of the unfairness they experienced.Tzafrir and Hareli (2009) conducted a study into the influence of organizational justice on the formation of employees' positive and negative emotions when they receive a promotion.Their study only focused on two variables of organizational fairness, which were procedural justice and distributive justice.Reuben and van Winden (2010) argue that a perceived unfairness by an individual will form a perceived negative emotion (guilt and humiliation) but has no consequence on the formation of certain behavior.Radke et al. (2013) studied the irrational behavior carried out by individuals who experience depression as a result of perceived unfairness towards their negative emotions.They employed a unidimensional fairness concept in their research, aside from the fairness concept, which comprised of procedural, distributive, and interactional fairness.A study by Uludag (2014) on the infuence of fairness toward verbal and physical aggresion does not fully explain the measure of the dimension of fairness towards aggresion as a form of negative emotion.Moreover, the study by Silva and Caetano (2016) into organizational fairness across different cultures shows that differences in the perception of an organization's members toward organizational fairness is related to the culture where the organization belongs.They state that the consequences that arise from organizational unfairness are positive and negative emotions which are expressed by the employees both verbally and nonverbally.
According to the previous studies, as mentioned above, this study concluded that those studies tended to be conducted in some type of organizational setting and not in the area of service hospitality organizations, such as hotels or restaurants.The fairness concept used mostly in those studies is based on the concept of the equity theory proposed by Adams (1966); while this study uses the fairness concept proposed by Seiders and Berry (1998).Consumers perceive and assess fairness or unfairness based on the situation related to the potential of maximizing profits by minimizing personal investment or sacrifice (Chou et al. 2009).Therefore, according to this study, the benefits received by customers and their sacrifices are related to the actual context of the fairness theory (distributive fairness, procedural fairness, and interactional fairness).Thus, this study hopes to obtain a better understanding concerning service's fairness that captures all the various service contexts based on consumers' sacrifices and the benefits they receive.Monetary and time costs, which are associated with the experience of customers' consumption, are consider as constructs representing sacrifices made by customers, while products (services non-interactions) and personnel (the way customers are treated) obtained by customers while consuming products or services are consider as constructs representing customers' benefits.
In order to fill this research gap, this study proposes a theoretical model which relates to service fairness, emotion, and behavioral intentions, and examines this empirically in the context of the service industry, especially for restaurants.The definition of service fairness used in this study is the service fairness perception of customers related to the service behavior of organizations or companies (Namkung and Jang 2010).The background of this study is the customers of fastfood chain restaurants in Indonesia.Based on the data from Euromonitor International: Fast Food in Indonesia year 2016, there are 7,236 fast-food chain restaurants throughout Indonesia, which are mainly located in the big cities.According to data from AAFC Kanada, part of the Ministry of Farming and Agrifood, in 2015, the fast-food chain industry in Indonesia is predicted to grow at roughly around 9.2 percent between 2015 and 2019.Therefore, it is appropriate to select fast-food restaurants for our study's setting.
This study was conducted in the big cities in Indonesia, such as Jakarta, Semarang, Surabaya, Medan, and Makassar, where some of these fast-food chain restaurants are located.The selection of research locations is based on the areas that have high economic growth and big populations.These fast-food chain restaurants have always emphasized they satisfy their customers with the quality of their service.
Based on the previous discussions, the main objectives of this study are to determine the relationship between service fairness, emotions, and behavioral intentions, and to discern the level of importance of each dimension of service fairness in the context of restaurant services, and to provide information for fast-food chain restaurant entrepreneurs to evaluate their service's qualities in terms of their service's fairness.

The Effect of Fairness on Service Marketing
Fairness is defined as an action which is deemed appropriate from an individual's perception (Aggarwal and Larrick 2012).The term fairness is often used with regard to the ability to make unbiased, concrete and spe-cific judgements about a particular case without referring to one's own feelings (Goldenberg et al. 2016;Faullant et al. 2017).
In connection with the marketing of services, Bagozzi (1975) introduces the concept of fairness/equity into the marketing exchange theory.He examines the relationship of fairness in the context of dyadic reciprocity and the equality needed to maintain the process of ongoing exchanges between the customer and the marketer.Aggarwal and Larrick (2012) and Blader et al. (2013) also discuss the consumers' perceptions of marketing practices which are unfair, and so have consequences Seiders and Berry (1998) summarize the relevant principles based on three categories of fairness (procedural, distributive, and interactional) and show these principles are appropriate for customers to assess the fairness they have experienced.Aggarwal and Larrick (2012) and Lee et al. (2013) also found that three types of fairness are likely to be relevant for background services, such as those found in banks, medical facilities, fast-food restaurants, and luxury restaurants, and hence these may be used as a basis for assessment by consumers.
The role of fairness in the context of the recovery of service failures has also been the focus of attention in many studies.According to Chou et al. (2009) the role of fairness, in the context of the recovery of service failures, is particulary significant for some service industries, such as hairdressers, restaurants or cafes, supermarkets, and banks.By adding a moderating effect of loyalty, Hur et al. (2014) found that both types of fairness (distributive fairness and procedural fairness), when performing recovery services, have a significant influence on the reactions of loyal customers.Although the principles of fairness have provided a better understanding concerning the influence of fairness, especially in the context of service failures and service recovery, yet other areas of research outside these two contexts are limited and not well developed.

Mehrabian-Russell Model
This study is based on a model developed by Mehrabian and Russell (1974) for environmental psychology.They proposed that environmental stimuli (S) lead to an emotional reaction (O) which, in turn, drives consumers' behavioral response (R), based on the Stimulus-Organism-Response (S-O-R) paradigm.They posited that consumers have three emotional states in response to environmental stimuli: Pleasure, arousal, and dominance.According to their study, these emotional responses result in two contrasting behaviors: Approach or avoidance.Approach behavior involves a desire to stay, explore, and affiliate with others in the environment, whereas avoidance behavior includes escaping from the environment and ignoring any communication attempts from other people (Jang and Namkung 2009;Namkung and Jang 2010).Using Mehrabian and Russell's model, many studies have been conducted on the role of environmental stimuli as a predictor of emotional responses, such as pleasure or arousal, and as a predictor of consumers' behavior, such as the extra-time they spend in a store and their actual incremental spending (Jang and Namkung 2009).
Based on the Mehrabian-Russell model, environmental stimuli will affect the individuals' emotions and as a result will affect their response in the form of approach or avoidance (Namkung and Jang 2010).In their research model, stimuli are the stimuli is lo-cated externally from people and consist of various elements of physical atmosphere.Furthermore, the organism refers to structure and internal processes which consequently will intervene in the relationship between the external stimuli of people and the attitude being demonstrated or responded to.This suggests that the impact of the stimuli on human behavioral intentions is mediated by emotion.
According to Mehrabian and Russell (1974), there are three forms of emotion, namely: Pleasure, passion, and power.Furthermore, the response to the environment can be classified as either an approach or avoidance.Approach behavior includes such things as: Desire to stay, observe the area around you, explore the environment, and communicate with others in the environment.Avoidance behavior is the opposite of approach behavior.Our study applies the model of Mehrabian and Russell (1974) and modifies it in accordance with the background of this research (see Figure 1).
In line with Mehrabian and Russell (1974), some justice researchers rely on the social exchange theory to predict relations between perceived fairness and its outcome (Cropanzano et al. 2007;Cropanzano et al. 2008;Cropanzano and Stein 2009).The social exchange theory views organizations as arenas for long-term, mutual social transactions between employees and their organization (Cropanzano et al. 2007;Cropanzano et al. 2008;Cropanzano and Stein 2009).Justice is considered to be an input of the organization to the exchange relationship and derives from either (a) the organization or (b) the direct supervisor (Cohen-Charash and Spector 2001).Futhermore, according to Cohen-Charash and Spector (2001), attitudinal and affective reactions toward the orga-nization were predicted by fairness models, where procedural fairness predicted cognitive reactions toward the organization, distributive fairness predicted affective and cognitive reactions toward the outcomes (Cohen-Charash and Spector 2001).
There is some research based on the Mehrabian-Russell model, namely: Jang and Namkung (2009), who undertook research to address the lack of a comprehensive evaluation of restaurants' quality.Their study extends Mehrabian and Russell's stimulus-organism-response framework by incorporating restaurant-specific stimuli and including restaurant-specific measures of emotion.Lee (2010) investigated how technology products' attributes influence consumer responses.He proposed a research model of consumers' use of technology products by applying the Stimulus-Organism-Response (S-O-R) framework (Mehrabian and Russell 1974) to suggest that technology products' attributes elicit consumers' cognitive (attitude) and affective (pleasure and arousal) responses, leading to their approach-avoidance behavior.Ltifi and Gharbi (2012) studied the impact of emotional states and the perceived risk of remote purchasing on e-satisfaction during Internet shopping, as well as the influence of e-satisfaction on e-loyalty.Kim and Lennon (2013) extended Mehrabian and Russell's stimulus-organism-response model to include both external (i.e.reputation) and internal sources of information (i.e.website quality) as stimuli which affect consumers' response systems.Therefore, the purpose of their paper was to test a more comprehensive model consisting of reputation and website quality (stimuli), cognition and emotion (organism) and purchase intention (response).Further, Chen et al. (2015) examined diners' luxury restaurant consumption behavior by incorporating diner expectations into a modified Mehrabian-Russell model.They argue that consumers dine at luxury restaurants for reasons beyond fulfilling their basic needs.
The results from all this research shows that the framework, based on the environmental psychology model of Mehrabian and Russell (1974), could be used to describe the sources for the various environments as inputs/stimuli, and their effects on people's emotions and behavior.Accordingly, this research considers fairness as an input of the organization to predict affective and behavior reactions toward outcomes (Cohen-Charash and Spector 2001).

Service Fairness
For decades, fairness has been viewed as a three-dimensional construct, comprised of distributive fairness, procedural fairness, and interactional fairness (Seiders and 1998;Cropanzano et al. 2008;Aggarwal and Larrick 2012;Hur et al. 2014).Although the principles of fairness developed in social psychology may be useful in the contexts for which they are developed, they may have limitations when applied to specific consumption situations.For example, Deutsch (1985) criticizes the measurement capabilities of the original theory of fairness, because it is difficult for consumers to assess the input and output values at the same time.Chou et al. (2009) in their study also argued that when equity is accounted for in the process of changing inputs into outputs, with a certain outcome for a customer, it will create problems with the operationalization of fairness itself.Since there are conflicting views about the dimensions of fairness, a consensus is growing that the perception of fairness is a dependent context, which is not only appropriate for the service failure situation, but it may also be applied in other situations (Azar and Darvishi 2011;Aggarwal and Larrick 2012;Blader et al. 2013).
According to Seiders and Berry (1998), restaurant patrons who are dissatisfied with the food served to them, due to its poor quality, which they believe is not equivalent with the price they are paying, will question the distributive fairness of the transaction.If these patrons are kept waiting for a table despite having a reservation, and also for food and beverage service (once seated), they will question the procedural fairness.Moreover, if they experience insulting treatment or a disrespectful attitude from the service staff, they will also question the interactional fairness.From the perspective of the customers' sacrifice in consuming the product or service, the price to be paid (price fairness) and time spent (procedural fairness) are substantial factors (Seiders and Berry 1998).In general, price will appear in any situation, and in par-ticular it is the magnitude of the economic costs sacrificed in any transactions.Thus, price fairness is defined as the customers' assessment of the difference between what they expect to receive and what they actually receive (Kahneman et al. 1986).That is, consumers generally have internal standards set for the actual price, and compliance with these standards will be verified when a judgment is made as to whether it is equitable or not.
The perception of fairness is the judgment of whether or not customers accept an outcome and/or a transaction process as being reasonable, acceptable, and just (Heo and Lee 2011;Su et al. 2015;Su et al. 2016).Since assessments of price fairness are subjective, some studies have focused on consumers' perceptions regarding relative prices (Haws and Bearden 2006;Gielissen et al. 2008;Xia et al. 2004).A study by Homburg et al. (2014) shows that the price stated in the menu may be assessed differently by a customer, depending on the stimulus (stimuli) provided by the restaurant.Daskalopoulou (2008) found evidence that a price's fairness perception is based on the acceptance level of the individual and the prevailing social standards.The concept of price in this study is different from the concept of price from previous studies, which stated that prices are fixed and absolute (Kukar-Kinney et al. 2007).Thus, to expand the understanding of consumers' perceptions of prices' fairness, this study proposes that the definition of price fairness is the assessment of the overall consumers' price based on a comparison between the actual price and the acceptable price, according to social standards and conformance levels (Homburg et al. 2014).
As one of the aspects of customers' sacrifices, time is viewed as a limited resource (Croonen 2010;Liang et al. 2017;Chun et al. 2018).Customers' evaluations toward the waiting process are not only altered objectively but also subjectively (Azarand Darvishi 2011).When there are two customers who assess time differently, but who are kept waiting for the same duration, they will differ in paying for the services they receieved (Blader et al. 2013).In restaurants, the waiting time derives from conditions which require extra time, such as service preparation (i.e.how to serve the food) or a level of demand beyond the system's capacity (i.e.customers arrive when the requirement for service already exceeds the providers ability to accomodate it).Thus, with regard to this discussion, the concept of procedural fairness can be defined as the accuracy of time and the efficiency of the service system as a part of the customers' sacrifice, since waiting time and delayed service are categorized as drawbacks.
On its relationship with the benefits perceived by customers, studies into fairness show that fairness behavior, as demonstrated by the service provider, is not only important for the instrumental aspect, but also important for the relational aspect (Namasivayam 2004;Croonen 2010;Azarand Darvishi 2011).Thus, service evaluations by customers do not only come from fairness, regarding the tangible outcome of fairness, but also from the intagible outcome of fairness, which is interactional fairness (Hur et al. 2014).According to Gronroos (2007) providing exceptional products and service is one of the main interests of the service business.However, he argues that fairness regarding the perceived outcome of fairness will affect the emotions of the customers, and their behavioral intentions.Furthermore, in restaurants there are some human intearctions.Lee et al. (2013) and Hur et al. (2014) define a service contact as a period of time where customers di-rectly interact with the service.The service contact is the critical moment of truth when customers often attain unforgettable impressions concerning a business (Lee et al. 2013;Hur et al. 2014).This condition only lasts for a few minutes, and may satisfy the customers, or frustrate them (Liang and Zhang 2011).Therefore, interactional fairness in this study refers to the interest and respect shown by service providers to customers, and as a result, customers will feel that they are treated fairly during the interaction (Aggarwal and Larrick 2012).Thus, in general, our study develops a model of fairness services consisting of four variables of fairness, namely: Price fairness, procedural fairness, outcome fairness, and interactional fairness, all of which will be tested empirically in the context of restaurant service, specifically fastfood chain restaurants.

The Effect of Price Fairness on Positive Emotions
Positive and negative emotions felt during the process of fairness may be characterized in various ways; they may stand alone (positive or negative) or as a whole (positive and negative); or they are experienced or expressed (Heo and Lee 2011;Lee et al. 2013).Some studies of fairness also use the emotional approach in isolation, which shows that the perception of price unfairness is associated with feelings of disappointment or anger (Xia et al. 2004;Xiaand Monroe 2010).Similarly, in the study of social fairness, the effect of procedural fairness on separated emotional responses describes emotions of happiness in the form of happiness, joy, and feelings of pride, along with emotions of sadness in the form of disappointment, anger, and frustration (Barclay et al. 2005;Blader et al. 2013).Based on the descriptions above, the first hypothesis proposed in this study is: The Effect of Procedural Fairness on Positive Emotions Lind and Tyler (1988) argue that the perceived fairness of outcomes was not only considered as a determinant of perceived organizational justice, but the perceived fairness of the process in which the outcomes were achieved was also considered to be important.In some cases, the fairness of the process was considered to be the most important determinant of perceived organizational justice (Blader et al. 2013).Futhermore, Hur et al. (2014) show that procedural justice, as the fairness of the process in which outcomes are determined, is considered to exist when there are procedures which embody certain types of normatively accepted principles.Accordingly, when a process leading to a certain outcome is perceived to be unfair, a person's reactions are predicted to be directed at the service provider rather than at the specific outcome in question (Namkung and Jang 2010).Therefore, the second hypothesis proposed in this study is: H 2 : Perceived procedural fairness has a positive effect on positive emotions.

The Effect of Outcome Fairness on Positive Emotions
Emotion is an individual's reaction to an event or stimulus (Cassar and Buttigieg 2015) and it is fundamentally a social phenomenon (Tangney and Fischer 1995).Social scientists acknowledge that issues of fairness can lead to emotional reactions, particulary when unfairness occurs.Adams (1966) stated that when a violation of fairness occurs, it will cause an individual to feel unhappy, and he/she will try to eliminate that negative feeling (Barclay and Skarlicki 2009).Homans (1961) also argues that when the outcome gained is lower than what was expected, it will lead to resentment in the recipient (Blader et al. 2013).In relation with the scheme of positive emotions and negative emotions, Lazarus (1991) states that positive emotions, such as pleasure or pride, arise in situations where individuals' needs are fulfiled.In contrast, negative emotions, such as shame and anger, arise when the fulfillment of an individual's need is blocked (Cassar and Buttigieg 2015).Therefore, the third hypothesis proposed in this study is:

The Effect of Interactional Fairness on Positive Emotions
Liiand Sy (2009) found that failure in the application of interactional fairness standards in service contacts will result in negative emotional responses (Cassar and Buttigieg 2014).In the study of emotion in the customer service area, Dube and Menon, as cited in Swartz et al. (1998), illustrate the importance of the role of socialization in the formation of exchange relationships between customers and service providers (Schlett and Ziegler 2014).Customers' satisfaction with service contacts is generally influenced by the level of loyalty, and consistency from the service provider in service fairness (Skarlicki et al. 2008).Thus, it should be noticed that emotions may be generated from stimuli that are relevant to the service.Therefore, service fairness plays a role as a producer of individuals' emotions, as presented in this research model.Our fourth hypothesis proposed in this study is: H 4 : Perceived interactional fairness has a positive effect on positive emotions.

Behavioral Intention
Modification of the relationship between a stimulus and its emotional response will lead to consumer behavior.Zaltman et al., as cited in Hunt (1978), examined consumers' behavior when faced with unfair marketing practices at different stages, namely at the stage prior to purchase; the time of purchase; and the time after the purchase (Skarlicki et al. 2008).Their study shows that an increase in the awareness of unfair practices and fraud will increase the tendency of consumers to complain (Skarlicki et al. 2008).White et al. (2012) proposed a model of consumers' complaint behavior and concluded that perceived fairness influenced the consumers grumbling, word-of-mouth opinion, and patronage behavior.The patronage be-havior is measure with the intention to stay and intention to deliver positive opinions of the products to family, friends, and colleagues.This behavior have the ability to predict consumer purchase behavior and Customer's fairness perceptions are key driver of this behavior (White et al. 2012).Behavioral intention is defined as the rate at which individuals consciously want to display or not display certain behaviors in the future (Tzafrirand Hareli 2009;Fishbeinand Ajzen 2010;White et al. 2012).Therefore, the intention to behave is a proxy for certain behaviors (Fishbein and Ajzen 2010).Thus, this study also applied the intention to behave as a construct of the perceived result, where the intention to behave is influence by positive emotions.

Emotions' Influence on Behavioral Intention
Numerous studies have shown that affection is a significant factor in consumer behavior (Yang and Chang 2008;Tzafrir and Hareli 2009).The underlying concept of this theory is people often make judgments concerning external events based on their affective reactions that arise at that time.Therefore, they use affection as the basis of their assessment when they assess something they like and are feeling happy (Williamson and Williams 2011).Positive emotions (such as joy, excitment, and happiness) have been recognized to have a consistent effect on an individual's behavior (Jangand Namkung 2009).When positive emotions are consistent during service contacts, the expectation is to be able to deliver a good experience for the consumer (Tu 2004).Associated with perceived fairness, people will use positive emotions to evaluate their assessment of their preferred service company (Nguyenand Klaus 2013).Furthermore, they will be willing to spread a positive opinion of the company by word-of-mouth to other people or to re-visit their preferred service company (Nguyen and Klaus 2013).As stated by Adams (1966), the negative feelings resulting from unfairness are an emotional statement to restore the sense of fairness (Jangand Namkung 2009; Namkung and Jang 2010).Thus, the fifth hypothesis proposed by this study is:

H5: Positive emotions have a positive influence on the customer's behavioral intentions.
The hypotheses in this study can be visualized as in Figure 2.

Method of Phase 1
This study consists of two phases, namely: Phase 1 which is the qualitative phase, and Phase 2 the quantitative phase.In Phase 1, the researchers conduct an explanatory study.This stage aims to determine the form of the percieved fairness felt by customers of fast-food chain retaurants, and the positive emotions that are formed when someone perceives the fairness, and their behavioral intention in the future related to their preference for fast-food chain restaurants.
The first step was to conduct a preliminary interview with customers of fast-food chain restaurants located in the capital cities of 2 Indonesian provinces, namely Jakarta (3 restaurants) and Surabaya (2 restaurants).The number of customers selected for these initial interviews were 15 customers from each restaurant.The determination of the customers to be interviewed at this stage was based on the following considerations: (1) They were eating in the restaurant; and (2) they were willing to engage in the research.In addition to these preliminary interviews, the customers were also asked to write down on a piece of paper how often they ate there, what impressions they have regarding the service provided when they are eating at the restaurant, and why they chose to eat at that restaurant.
From these interviews, the items mentioned by at least 10 percent of the respondents (Fishbein and Middlestadt 1995) were considered to be essential factors in the development of a questionnaire to be use for conducting the confirmatory factor analysis.According to Fishbein and Middlestadt (1995), those attributes or outcomes that are mentioned most frequently by the population must be considered.

Results of Phase 1
After the questionnaire was developed, 200 copies of it were left in the fast-food chain restaurants selected as the sample for our research.Subsequently the questionnaires were collected and the results from them processed using SPSS software; the results are in Table 1.Based on the results of the confirmatory factor analysis (in Table 1), it is clear that the impressions for each variable written by the customers may be used as the basic indicators for each of the variables (factor loading > 0.7).Therefore, each variable was given a name as follows: Variable 1 (price fairness), Variable 2 (outcome fairness), Variable 3 (procedural fairness), Variable 4 (interactional fairness), Variable 5 (positive emotion), and Variable 6 (behavioral intention).The result from the factor analysis is the formation of these 6 constructs which will be tested in Phase 2.

Method for Phase 2
Phase 2 was the quantitative phase, and aimed to discover the effect of service fairness on emotions and their impact on behavioral intentions.In this phase, the questionnaire was arranged based on the results of interviews with key informants from Phase 1.The arrangement of the questionnaire was based on the previous research conducted by Cropanzano et al. (2008), Jang and Namkung (2009), Fishbein and Ajzen (2010), Xia andMonroe (2010), andXia et al. (2010).Furthermore, the questionnaire was reviewed by the respective managers of 4 fast-food chain restaurants included in our sample, and by 3 professional academics in the field of services, in an attempt to test the validity of the content.The questionnaire contains fairness variables, namely: Price fairness, procedural fairness, outcome fairness, and interactional fairness; positive emotions; and behavioral intentions.
After all the items in the questionnaire receieved a back translation, the Socially Desirable Response test (SDR) was conducted due to the potential for the respondents to provide dishonest information to this study, which would jeopardize our result.This problem has been one of the concerns for this study.In short, Socially Desirable is confirming good things which is in accordance with social norms and hiding bad things which is not appropriate by social norms (Sjostrom and Holst 2002).Respondent has the tendency to answer questionnaire with socially desirable point of view which sometimes different from the truth or from what they are really feeling or think or do.This sort of responds will create problems and make the results of the study inappropriate.Thus, we conduct SDR test to minimize this potential problem.The result from the SDR test was appropriate, as expected.The phase of mea-suring the construct's validity (convergence and discriminant) and the path analysis to analyze the effect of service fairness on emotions that may impact the behavioral intention was conducted subsequently.

Population and Sample
The data colletion method used in this study was the survey method.The fast-food chain restaurants in our sample are located in the capital cities of 5 Indonesian provinces, namely: Jakarta (15 restaurants); Semarang (10 restaurants); Surabaya (10 restaurants); Medan (8 restaurants); and Makassar (7 restaurants).A total of 1,000 questionnaires were dissiminated to the sample's restaurants and 920 questionnaires were returned to us.Futhermore, a check of all the questionnaires received was conducted and all those questionnaires with missing values or incomplete answers were dropped.In the end, there were 800 completed questionnaires which met the sampling requirements.Therefore, the response rate was 80 percent and this is considered to be adequate for this study.The sampling method applied in this study was purposive sampling.Sampling was conducted from August 2016 to February 2017.

Sample Characteristics
Table 2 presents the demographic characteristics.Descriptive information showed that 54.38 percent of the respondents (n = 800) were male, and 45.63 percent were female.The majority of respondents (55.13%) were married.Regarding their education levels, 38.75 percent of the respondents have at least a bachelor's degree, 28.75 percent and 32.50 percent hold master's and doctoral degrees, respectively.The respondents aged between 25 and 30 formed the largest group (28.75%).The majority of respondents work as employees in private companies (32.50%).Twenty-three point seven five percent of the respondents have monthly expenses which range between Rp 2,500,001 -Rp 5,000,000.

Result Of Phase 2 (The Research Instruments Testing) Socially Desirable Response Test
A Socially Desirable Response Test (SDR) was conduted on non-paired samples for each indicator of the 6 constructs measured.In doing so, this study entrusted 30 questionnaire to a restaurant in Jakarta with direct questions and 30 questionnaire to a restaurant in Surabaya with indirect questions.The test was performed with non-parametric statistics using SPSS 16 (Mann-Whitney test).The test results showed the p value is greater than 0.05 and concluded that those two samples (non-paired) derived from the population which has the same average (mean) or expectation.Thus, the average answer of the respondents from these two samples are similar.See more detailed results in Table 3.

Discriminant Validity Tests and Research Constructs' Convergence
Discriminant validity and convergence tests were conducted by distributing 200 copies of the questionnaires to restaurants located in Jakarta, Surabaya, and Semarang.The results from these questionnaires were analyzed using Factor Analysis (FA).The indicator measurements appear to present each construct (factor loading > 0.6), indicating that the measurements' constructs have good discriminant validity.Hair et al. (2010) stated that for the confirmatory phase of the research into the measurement scale, values for the factor loading of greater than 0.6 are con-  4.
Table 4 shows the result of the calculation of the convergence validity (Fornell and Larcker 1981;Nunnally and Bernstein 1994).The convergent validity for each construct is properly adequate when the AVE (Average Variance Extracted) values exceed 0.7 (Hair et al. 2010).In line with this result, Table 5 also shows that the value of Cronbach's alpha and the composite reliability of each construct exceeds 0.7.Thus, it can be concluded that all the measures used in this study are reliable.This study used composite reliability as its reliability test method, which has advantages in estimating the internal consistency of a construct (Fornell and Larcker 1981;Nunnally and Bernstein 1994).

Measurement Model and Structural Model Testing
In conducting the test, Structural Equation Modeling (SEM) was applied to test the validity of the proposed research model and its hypotheses with the help of software Amos 21st.The test's results are provided in Table 6.  5 show that the degree of correlation between the constructs is adequate, with the exception of procedural fairness.

Result and Discussion
The current study was undertaken to explore the relationship among the perceived service fairness (price fairness, outcome fairness, interactional fairness, and procedual fairness) of customers of fast-food chain restaurants in Indonesia.Most of the hypotheses proposed were supported, as illustrated in Table 6.Hypothesis 1 (H 1 ) is supported (coefficient = 0.037; t = 1.901; p < 0.1).This implies that when the customer of a fast-food chain restaurant assesses if the money they spent on their food and beverages is pertinent (fair) with the flavor (good), they will feel a sense of fairness which will affect the formation of positive emotions.This finding is generally consistent with Namkung and Jang (2010) and Wen and Geng-qing Chi (2013) who concluded that from a customer sacrifice's perspective, price fairness was found to be a significant predictor of positive emotions and ultimately future behavioral intentions.Futhermore, since it was the only dimension of fairness that had a significant effect on the dependent constructs, the significance of price fairness should be recognized.
Hypothesis 2 (H 2 ) is not supported.This result indicates that when customers assess that the timeliness and efficiency of the service provided to them are presumed to be incompatible (unfair), they will perceive an unfair service which results in the failure to form any positive emotions.This is in line with Aggarwal and Larrick (2012) who argued that, as it is another aspect of customer sacrifice, procedural fairness seems to act as a basic requirement.Furthermore, they stated that qualified procedural fairness should be a basic requirement, in the sense that consumers expect their service's delivery to be efficient and quick without any delays or deception.
Hypothesis 3 (H 3 ) is supported (coefficient = 0.143; t = 3.329; p < 0.001).The result indicates that when the customers of fast-food chain restaurants assess their visit, they are satisfied and content with the outcomes they received, which they consider to be a positive experience, so it will result in the formation of positive emotions.As outlined by Jang and Namkung (2009), in a restaurant setting, many stimuli could influence a customer's emotional state.They stated that these stimuli encompass both tangible and intangible features of the restaurant, such as its products' attributes, the physical environment, and its service aspects.
Hypothesis 4 (H 4 ) is supported (coefficient = 0.149; t = 2.333; p < 0.05).This result implies that the quality of service perceived by the customers will result in a fairness perception which will affect the formation of positive emotions.Thus, this finding verified that customers are often influenced by the knowledge and skills of the employees when forming opinions about the service offered, and inadequate interactions may directly lead to unfavorable behavior (Jang and Namkung 2009;Namkung and Jang 2010).Given this result, it might be useful to convey a perception of fair service that is characterized by friendly and attentive staff who exhibit equal service to all the customers, without any bias.
Hypothesis 5 (H 5 ) is supported (coefficient = 0.497; t = 2.786; p < 0.05).The result indicates that positive emotions are excellent predictors for predicting consumers' behavior in the context of the service industry.This finding supports the view of Jang and Namkung (2009) and Namkung and Jang (2010) who argue that the role of positive emotions should be obvious given the hedonic nature of restaurants.Furthermore, they stated that restaurant managers could improve the probability of favorable behavioral intentions by establishing fair and reasonable prices, consistently good-quality food, and improving the quality of the interaction between customers and service providers, which would elicit positive emotions.
Therefore, the 3 fairness variables (price fairness, outcome fairness, and interactional fairness) have positive and significant effects on positive emotions.In terms of the benefits received by the customer, two variables of fairness have a positive and significant effect on their positive emotions.In terms of the sacrifices made by the customer, one fairness variable has a positive and significant effect on their positive emotions.Meanwhile, procedural fairness, which belongs to the context of customers' sacrifice, does not have a positive and significant effect on positive emotions.It should be noted that emphasizing procedural fairness alone would not be enough to create customers' positive emotions.Our results show that fast-food chain restaurants' customers expect efficiency and timeliness in the service provided, without any delays or errors.Thus, when fast-food chain restaurants fail to meet these expectations, customers will consider this condition to be unacceptable.This result is in line with Aggarwal and Larrick (2012) and Hur et al. (2014) who concluded that when service delays occur, customers assess this condition as a waste of their time and they rate the service negatively.
From the perspective of the benefit received by customers, the satisfactory result (physical) they perceive has a positive and significant effect on the customers' positive emotions (Dzansi 2016;Wang et al. 2016;Xu et al. 2016).Furthermore, when this result (physical) is capable of influencing and increasing the positive emotions, it will, in turn, impact on future purchase intentions.Thus, successfully forming core products and a quality service are substantial requirement that must be met in the restaurant business.Similarly with other fairness variables which are also included in the context of the benefits received by the customer, interactional fairness has a positive and significant effect on the customers' positive emotions.These results indicate that, in general, the customers' opinions are influenced by the knowledge and expertise possessed by the employees associated with the ser vices offered (Aggarwal and Larrick 2012), hence, any adverse interaction will raise the likelihood of customers' negative behavior.Moreover, Kukar-Kinney et al. (2007) show that excellent interactions between an individual customer and the employees will affect the assessment of the services provided.Therefore, based on these results, it could be concluded that the customers' perception of the fairness of the service is influenced by factors such as a friendly athmosphere, full attention being provided by the service employees, and equal service deliveries, without errors, for every customer.
The role of positive emotions may be generated and highlighted through excellent first impressions of the restaurant, through the quality of the service provided during the service contact.Thus, restaurant managers can increase the positive behavioral intention of their customers through the development and provision of fair and reasonable prices for food and beverages, as well as improvements in the quality of the interaction between the customers and the restaurant's employees, which will result in positive emotions.

Conclusion
The model used in this study considered two goals and theoretical questions simultaneously.The first question investigated the impact of perceived fairness on customers' perceptions and emotional reactions.The 4 types of customers' perceived fairness considered are interactional, procedural, outcome, and price, which are expected to influence customers' positive emotions.The more customers perceive fairness from the service they receive, the more positive emotions they will experience toward the service organiza-tion (Huang 2016;Roy et al. 2016;Kimand Park 2017).This result supports those of previous studies such as by Namkung and Jang (2010) and Wen and Geng-qing Chi (2013).
However, the findings from this study show that procedural fairness does not affect the customers' emotions, while the effect of the other types of fairness is not confirmed.Procedural fairnesses' dimension on positive consumption's emotion is, to some extent, inconsistent.Therefore, the result of this study is in contrast with Namkung and Jang (2010), who state that only the procedural aspect negatively affects negative consumption's emotions, while other aspects do not have any effects.However, Wen and Geng-qing Chi (2013) find evidence that interactional fairness is not effective in predicting positive emotions.Furthermore, a study by Smith and Bolton (2002) claims that all three dimensions of perceived fairness have the ability to affect customers' negative emotions.
The next theoretical question is the effect of positive emotional fairness on customers' intentions.The findings reveal that positive consumption emotions have a positive effect on customers' intentions.Thus, the more positive emotions customers experience from the service, the more likely it is that they will repeat their purchases.Hence, they will be more enthusiastic in sharing positive comments and recommendations through wordof-mouth advertising.This has already been observed in other studies (i.e. Lee 2010;Wen and Geng-qing Chi 2013;Hur et al. 2014;Chen et al. 2015).

Limitations and Suggestions
Regardless of the contribution made and the managerial implications, there are some limitations to this study.The first limitation is concerning the data collected in this study, which is only from fast-food chain res-taurants.Therefore, the possibility to generalize the results of this study into the context of other service industries or other industrial fields is very small.Thus, further research needs to be conducted into different contexts, especially in the restaurant industry (other than fast-food chain restaurants) and other industrial fields, to reveal their issues of service fairness and its related dimensions.From the methodological aspect, further research should conduct a validation test on fairnesses' service dimensions along with the indicators used in this study, and examine their application to assess the quality of service in the restaurant business.
Further research is expected to involve personal and situational characteristics, such as sensitivity to fairness, loyalty, and various contexts regarding service problems, which may moderate the relationship between these constructs.Aggarwal and Larrick (2012), Hur et al. (2014), andChen et al.(2015) state that, based on the personal sensitivity of fairness, each individual has a different level of importance for the perceived fairness of service.There are several questions in this regard, namely: Whether considerations of the fairness of different services are based on the level of sensitivity to the fairness or on contact based services?What level of loyalty to a particular restaurant affects consumers when they evaluate the perceived fairness of the service?Is a loyal customer more tolerant of unfairness than customers who are not loyal?Are customers who often come to visit a particular restaurant antagonistic when they feel the unfairness of service?Therefore, further research is expected to uncover the moderating role of loyalty and the frequency of visits in the relationship between emotions and behavioral intention, in order to gain more knowledge regarding customers' perceptions of service fairness that impact their emotions and behavioral intentions.

H 1 :
Perceived price fairness has a positive effect on positive emotions

H 3 :
Perceived outcome fairness has a positive effect on positive emotions.

Figure
Figure 2. Research Model

Table 2 .
Sample Characteristics

Table 5
, shows the correlation between the