It is human nature to pursue fairness or equity in any situation [1]. In the marketplace, consumers have fairness expectations regarding results or benefits they should receive in a service delivery situation. Those fairness expectations are formed based on consumers’ own experience with a service firm, knowledge of the company and its competitors, word-of-mouth, and advertisements [2]. Consumers judge fairness by comparing their fairness expectations with the service they received.

Leonard Berry, the leading service quality expert indicated that the fairness issue is embedded in the five service quality dimensions that he and his colleagues proposed. In a restaurant situation, customers expect to eat in a clean, comfortable environment (tangibility dimension), receive the correct order of food and beverages (reliability), be greeted and served in a timely manner (responsiveness), be helped by knowledgeable staff (assurance), and have their needs are taken care of by friendly servers (empathy). If a restaurant fails to deliver these service quality dimensions, customers would feel they are not treated fairly [3,4].

Berry [3] also suggested when customers feel fairness expectations are violated, it can have an immense effect on their trust toward a service provider, which can subsequently demolish the relationship between a firm and its customers perpetually [5]. Therefore, treating customers fairly is essential for a company to build and sustain a longterm relationship with them.

Previous research [2,4,6] has suggested that service fairness consists of three dimensions: procedural fairness, interactional fairness, and distributive fairness. Much research has tested these three service fairness constructs and confirmed their significance in measuring service fairness. Yet, some studies [7,8] suggested that these fairness dimensions may not be universal, and different sets of fairness dimensions might be needed for different types of business.

In order to refine the measurement of service fairness for the restaurant industry, this study attempts to test procedural fairness, interactional fairness, and distributive fairness in a restaurant situation and to explore new fairness constructs that may also be important to be included in the measurement.

Equity theory and justice theory have been used in service fairness research. This section provides a review of the theories and discusses previous service fairness studies.

Equity/Justice theory

The concept of fairness is almost identical with equity, and these two terms have been used interchangeably [8]. Fairness/equity concerns the equitable distribution of physical objects such as money and metaphorical objects such as job opportunities, punishments, and equal chances of winning. It is fair when those objects are divided equally and/or when people receive what they deserve [9]. The theory of equity was developed by Adams [10,11] and received attention in sociology, psychology, and organizational behavior research [12]. Bagozzi [13] introduced the concept of equity to the marketing and consumer behavior literature and raised questions regarding the consequences of inequity in marketing exchange. Consequently, several studies [2,4,8,12,14-16] attempted to investigate the fairness/ equity issue in a marketing context.

The equity theory suggests that parties in an exchange relationship compare their input-to-outcome ratio with that of others in the relationship. Fairness occurs when an individual’s input-to-outcome ratio matches an exchange partner’s input-to-outcome ratio. A mathematical equation of the input-to-outcome ratio was used in fairness/equity research; however, this method was found to have many restrictions. For instance, some inputs or outcomes cannot be quantified, such as the efforts and frustration that consumers invested during their purchasing processes. Additionally, people do not always know the exact inputs and outputs of their exchange partners [17]. Therefore, the calculation of the input-to-outcome ratio has received little interest in the fairness literature. Instead of using the inputto- outcome ratio, many studies [12,14-16] used scenario-based experimental design to measure fairness, and several researchers [1,8] developed measurement scales.

Dimensions of service fairness

Previous research [2,4,6] suggested that there are three bases for judging fairness. The first one is distributive fairness. This dimension refers to results of a service delivery. In a restaurant situation, it can be correctness, quantity, and quality of food and beverage. Additionally, consumers compare the results they receive with their investments, such as price paid, time, and efforts. Thus, this dimension is similar to consumers’ perceived value.

The second dimension is procedural fairness, which relates to the process of delivering a service. It includes the components such as handling of service problems, speed of responding to service requests, and customers’ waiting time [2,4].

The third dimension is interactional fairness which concerns consumers’ evaluations of service employees’ interpersonal manners. Expression of respect, friendliness, honesty, concern, sensitivity, courtesy, politeness, and empathy are important characteristics of interactional justice [2,4].

Many marketing researchers have tested the validity and predictive power of the three fairness dimensions in various industries. Clemmer’s [1] study is one of the first ones that measures the consumers’ perceptions of fairness issue in a restaurant setting. She found each of the three service fairness dimension had a unique contribution to customer satisfaction and return intentions. Clemmer [18] concluded that the fairness issue needs to be included when studying customer satisfaction.

Fairness/justice theory has received a lot of attention in service recovery research. McCollough [19] studied post service recovery satisfaction and service quality attitude in a hotel setting. The results concluded that distributive justice, procedural justice, and interactional justice had significant predictive power of consumers’ attitude. In a study of effects of consumers’ choice on service fairness, Mattila and Cranage [20] incorporated distributive justice, procedural justice, interactional justice, and informational justice which is a relatively new fairness dimension in the justice literature. Their study found that all four fairness dimensions have significant relationships with post service recovery satisfaction.

The service fairness dimensions were examined for consumers of different cultural backgrounds. Mattila and Patterson [21] compared how American consumers and East Asian consumers perceive distributive justice and interactional justice in a restaurant situation. Their study shows that culture moderated fairness perceptions. East Asian consumers gave better ratings on both dimensions than American consumers did. The researchers suggested that it might be because Asian culture encourages intra group harmony and tends to attribute failures to external factors such as fate or luck. Thus, they are less likely to hold a service provider responsible than American consumers are.

Purpose of the study

There are still many fairness issues that need to be explored. For instance, previous research has suggested that service fairness consists of three dimensions: distributive dimension, procedural dimension, and interactional dimension. The three dimensions have been tested extensively in the literature and are found to be significant fairness constructs. Nonetheless, these dimensions may not fully represent the concept of service fairness.

Bagozzi [7] stated that consumers may compare themselves with other consumers. For instance, restaurant customers may compare the services they receive with those of customers at the next table. Yet, only a few studies have looked at how customers compare themselves with other customers. In Mowen and Grove’s [22] study of participants who paid a higher price than other consumers in a car purchase situation, the inequity issue had a significant negative influence on their satisfaction and behavioral intentions. Collie, Sparks, & Bradley [23] investigated the fairness issue under the conditions of varying knowledge of others’ outcomes in a restaurant context. Regardless of whether consumers were aware or unaware of other people’s outcomes, those who received better interactional fairness treatment had significantly higher satisfaction with service recovery. These studies did not agree on the significance of customer-to-customer fairness. More measurement of this particular fairness issue is needed.

Bagozzi [7] also indicated that consumers may compare the service they receive with that of other service providers. In a restaurant situation, customers compare food, service, and value with those from other similar restaurants. To the knowledge of this author, no study has included how consumers compare similar businesses.

In order to capture the concept of fairness thoroughly, Oliver and Swan [8] suggested that future studies need to include customer-tocustomer and seller-to-seller comparisons when measuring service fairness.

For the purpose of developing and refining the measurement of fairness for the restaurant industry, this study tests the currently well-accepted fairness constructs: distributive dimension, procedural dimension, and interactional dimension. This study also includes two fairness constructs that have been ignored in the literature: customerto- customer fairness and seller-to-seller fairness. Customer-tocustomer fairness refers to consumers’ comparison of their benefits from a service transaction with that of other customers of the same business. Seller-to-seller fairness regards consumers’ comparison of different businesses of the same type.

This study uses overall fairness, customer satisfaction, overall quality, and behavioral intentions as four different outcomes of customers’ reactions to the five fairness issues. Although overall fairness has been mentioned in the literature, very few studies included it in empirical testing. Carr [24] included this variable in a study and confirmed that it is a valid construct. Thus, this study also included overall fairness as one of the dependent variables. Many studies [8,15,19,21,23-26] found strong links between customer satisfaction, overall quality, behavioral intentions and service fairness. Therefore, this study also examines the influence of the five service fairness dimensions on customer satisfaction, overall quality, and behavioral intentions.

Instrumentation

This study included an on-site survey at a table-service restaurant. The survey instrument used for this study was a self-administered questionnaire composed of four sections measuring service fairness, disconfirmation, customer satisfaction, and behavioral intentions. The items on the questionnaire for measuring service fairness were based on the literature review. Since this study incorporated two new services fairness dimensions: seller-to-seller fairness and customer-tocustomer fairness, a focus group study was employed. There were eight people participated in the focus group who provided their restaurant experiences that were related to service fairness issues. The results of the focus group approach helped to generate the items for measuring these two new service fairness dimensions. A total of 16 items were developed which represent the proposed five service fairness dimensions. They were measured based on a 7-point, Likert-type scale ranging from strongly disagree to strongly agree.

The four dependent variables included in the study are overall fairness, customer satisfaction, overall quality, and behavioral intentions. One item was used to measure overall fairness: “Overall, I was treated fairly.” It was measured on a 7-point, Likert-type scale ranging from strongly disagree to strongly agree. To measure customer satisfaction, two items were used: “Overall, this restaurant visit was very satisfactory,” and “My overall experience left me feeling quite happy.”The items were also measured on a 7-point, Likert-type scale ranging from strongly disagree to strongly agree. Respondents were asked one question regarding their overall perceived quality: “The overall quality of the restaurant is…” The question was measured on a 7-point, Likert-type scale ranging from “much worse than expected to much better than expected. Two questions regarding behavioral intentions were used: “Based on my experience today, I would recommend this restaurant to my friends or relatives,” and “Based on this visit, if I were choosing between this particular restaurant and another similar type of restaurant, I would choose this restaurant again.” These two items were measured on a 7-point, Likert-type scale ranging from strongly disagree to strongly agree.

In the last section, respondents’ age and gender were asked. Consultation with the managers of the restaurant that cooperated in this study and a pilot study were conducted to arrive at the final versions of all the questions. The pilot study was conducted at the restaurant that participated in this study. Questionnaires were filled out by ten restaurant patrons after they finished their dinners. Respondents were asked to indicate whether any part of the questionnaire was difficult to understand. The size of the print was also tested to determine if it was too small for older people to read. The suggestions from respondents were incorporated into the final revision of the questionnaire.

Data collection

An on-site survey was conducted at a table-service restaurant in the northeastern United States. A table-service restaurant was preferred to a fast food restaurant or restaurant which primarily offers a buffet because more interactions between customers and restaurant employees were desired for this study. The on-site survey was conducted over a seven-day period, and it was conducted at breakfast, brunch, lunch, and dinner meal periods. The questionnaires were given to restaurant customers after they finished their main courses. The patrons were asked to participate in the survey when they were waiting to pay. Each respondent was offered a coupon, which could be redeemed for free ice cream as an incentive for participating in the survey. A total of 397 useable questionnaires were collected.

Descriptive analysis

The characteristics of the sample are presented in Table 1. Among 397 participants, the largest age group (38.2%) was between 21 and 34 years of age. More than half of the people surveyed were under 34. Only 6.5% of participants were above 55. This indicates that a large portion of customers of the restaurants were young. About half of them were males, and the other half of them were females.

Table 1: Sample Characteristics of Study Respondents (N = 397).

The means and standard deviations of all service fairness items are presented in Table 2. Note that the three items that represent the customer-to-customer dimension were worded in a negative way; thus, after converting the means, the scores of these items are between five and six. Therefore, mean scores show that most respondents rated five or six on all of the items.

Table 2: Mean Score and Standard Deviations of Items for Service Fairness Items.

Reliability tests were used to test the five service fairness dimensions. The Cronbach alpha reliability coefficients were .90, .96, .92, .95, and .91 for procedural fairness, interactional fairness, distributive fairness, customer-to-customer fairness, and seller-to-seller fairness respectively. These high alpha values indicate that these measurement items are reliable, and there is good internal consistency among the items. Subsequently, the measurement items were averaged to form a composite score that represent each service fairness construct. Those composite scores were used for the next analyses.

Several multiple regression analyses were conducted to investigate the influence of the five service fairness dimensions on the four dependent variables. Table 3 presents the results of the relationship between the five dimensions and overall fairness. The regression equation was significant. Among the five service fairness dimensions, interactional fairness, distributive fairness, and seller-to-seller fairness had significant impacts on overall fairness. Distributive fairness had the strongest influence on respondents’ overall fairness. The multicollinearity of the five fairness dimensions was examined. All variance inflation factor values were less than 10, and all tolerance values were less than 1. Thus, evidence of multicollinearity among the fairness constructs was not found.

Table 3: Relationships of the Five Service Fairness Dimensions to Overall Fairness

Multiple regression analysis was used to determine the influence of the five fairness dimensions on customer satisfaction judgment. Two items that represent customer satisfaction were highly correlated and were averaged to form a composite score in the analysis. Table 4 shows that procedural fairness, interactional fairness, distributive fairness, and seller-to-seller fairness were positively related to customer satisfaction. Distributive fairness had the strongest influence on respondents’ overall satisfaction. The multicollinearity of the five fairness dimensions was examined. Evidence of multicollinearity among the fairness constructs was not found.

Table 4: Relationships of the Five Service Fairness Dimensions to Customer Satisfaction

The next multiple regression analysis measured the influence of the five fairness dimensions on overall quality. Table 5 shows that all five fairness constructs made significant contribution to overall quality judgment and the interactional fairness is the strongest predictor. The results also reveal that multicollinearity of the five fairness dimensions was not found.

Table 5: Relationships of the Five Service Fairness Dimensions to Overall Quality.

The last regression analysis was employed to test the relationship between the five fairness dimensions and respondents’ behavioral intentions. Two items that represent behavioral intentions were highly correlated and were averaged to form a combined score in the analysis. Table 6 shows that distributive fairness and seller-to-seller fairness significantly influenced behavioral intentions. No multicollinearity was found among the five fairness dimensions in this regression equation.

Table 6: Relationships of the Five Service Fairness Dimensions to Behavioral Intentions.

The results also show that the five service fairness dimensions have different degrees of impact on overall fairness, customer satisfaction, overall quality, and behavioral intentions. Distributive fairness and seller-to-seller fairness have significant impacts on all four dependent variables. Interactional fairness had a significant link with overall fairness, customer satisfaction, and overall quality. There is no significant relationship between interactional fairness and behavioral intentions. Procedural fairness had significant impacts on customer satisfaction and overall quality. It did not contribute significantly to overall fairness judgment and behavioral intentions. The relatively new customer-to-customer fairness had significant influence only on overall quality. When comparing its predictive power with other fairness dimensions, customer-to-customer fairness is relatively weaker than the others.

In summary, quality of food, correctness of orders, value customers received, and how the restaurant performed compared to other similar restaurants were the most critical factors for customers of this restaurant. Although all five service fairness constructs had different degrees of impact on overall fairness, customer satisfaction, overall quality, and behavioral intentions, distributive fairness and seller-to- seller fairness are the two critical factors that determine whether or not the participants will recommend or return to the restaurant in the future.

Bowen et al. [4] suggested that there is a psychological contract between a business and its customers that concerns what each party gives and receives in the relationship. Customers expect a company to fulfill its obligations of delivering services that they pay for. Service fairness is the base of forming and maintaining enduring relationship with customers. It is imperative for restaurants to ensure that customers are treated fairly.

This study found that there is a significant link between distributive fairness and overall fairness, customer satisfaction, overall quality, and behavioral intentions. Food is the most basic product that a restaurant offers. Generally, it is the reason why people visit a restaurant in the first place. Failure to provide correct order of food or good quality of food at reasonable price is very likely to trigger perceptions of unfairness.

This study also reveals that seller-to-seller fairness had a significant influence on different kinds of evaluations. This suggests that consumers’ comparisons of different restaurants can have a substantial effect on their evaluations of a restaurant. This finding indicates that restaurants need to pay close attention to what competitors offer and their level of quality.

In addition, restaurants need to ensure that the design of its service delivery facility and process are efficient so that every customer can be served promptly, and waiting time for seating or food and beverages can be minimized. Employee training is also a key issue in terms of interactional fairness so that customers feel they are welcomed and that their needs are well taken care of.

An important result of this study is that the two new fairness dimensions: seller-to-seller and customer-to customer fairness were found to have significant impacts on customers’ restaurant experience. This finding supports previous suggestions [7,8] that these two dimensions need to be included when studying service fairness. Since this is the first study that includes both of these two dimensions, more research is needed to examine if the same effects can also be found.

Furthermore, this study is exploratory in nature; thus, repeat testing of the five service fairness constructs is required. Future work needs to continue testing of the measurement items that represent the two new dimensions: customer-to-customer dimension and seller-to-seller dimension. For instance, the seller-to-seller dimension was measured in terms of quality of food and friendliness of staff. Future research may ask consumers to compare different restaurants for other issues, such as price and atmosphere. Furthermore, researchers [8] also suggested that there may be other unknown dimensions that need to be explored.

As discussed in fairness/justice literature, there is ongoing debate on the interrelationship among the fairness dimensions [27]. Although the interrelationship was not the focus of this study, it needs to be researched in order to better understand various fairness constructs.

Variables that influence perceived service fairness need to be examined. For instance, Huppertz, Arenson, & Evans [16] included social distance between consumers and sellers in their study. They found that in an inequitable situation, customers who had higher shopping frequency with sellers had lower fairness ratings than consumers who had less shopping experience with the same business. They concluded that the closer the relationship between buyers and sellers, the more likely that the consumers will perceive unfairness. Other researchers [14,15] also found familiarity with a service provider affects consumers’ equity perceptions. However, those earlier studies did not measure the effects of familiarity or the length of buyer-seller relationship on any specific fairness dimensions. One may assume that loyal customers are more sensitive to certain fairness principles. For instance, repeat customers may have higher expectations in terms of friendliness of staff or fulfilling special requests due to the closer relationship they have with a service provider. Thus, more work in this area is guaranteed to investigate how the exchange relationship moderates consumers’ evaluations of different fairness dimensions.

This study used four dependent variables: overall fairness, customer satisfaction, service quality, and behavioral intentions to test the five fairness dimensions. Other dependent variables can be considered when examining fairness dimensions. For instance, Folger and Konovsky [28] compared the effects of distributive justice and procedural justice on employees’ reactions to pay raises. They found that procedural justice had a significant impact on employees’ organizational committee and trust in supervisor. Not much research has investigated the relationship between the fairness issue, trust, and commitment in a business. Future marketing research is needed to examine how different fairness constructs affect those different variables.

An important result of this study is that the two new fairness dimensions: seller-to-seller and customer-to customer fairness were found to have significant impacts on customers’ restaurant experience. This finding supports previous suggestions [7,8] that these two dimensions need to be included when studying service fairness. Since this is the first study that includes both of these two dimensions, more research is needed to examine if the same effects can also be found.

Furthermore, this study is exploratory in nature; thus, repeat testing of the five service fairness constructs is required. Future work needs to continue testing of the measurement items that represent the two new dimensions: customer-to-customer dimension and seller-to-seller dimension. For instance, the seller-to-seller dimension was measured in terms of quality of food and friendliness of staff. Future research may ask consumers to compare different restaurants for other issues, such as price and atmosphere. Furthermore, researchers [8] also suggested that there may be other unknown dimensions that need to be explored.

As discussed in fairness/justice literature, there is ongoing debate on the interrelationship among the fairness dimensions [27]. Although the interrelationship was not the focus of this study, it needs to be researched in order to better understand various fairness constructs.

Variables that influence perceived service fairness need to be examined. For instance, Huppertz, Arenson, & Evans [16] included social distance between consumers and sellers in their study. They found that in an inequitable situation, customers who had higher shopping frequency with sellers had lower fairness ratings than consumers who had less shopping experience with the same business. They concluded that the closer the relationship between buyers and sellers, the more likely that the consumers will perceive unfairness. Other researchers [14,15] also found familiarity with a service provider affects consumers’ equity perceptions. However, those earlier studies did not measure the effects of familiarity or the length of buyer-seller relationship on any specific fairness dimensions. One may assume that loyal customers are more sensitive to certain fairness principles. For instance, repeat customers may have higher expectations in terms of friendliness of staff or fulfilling special requests due to the closer relationship they have with a service provider. Thus, more work in this area is guaranteed to investigate how the exchange relationship moderates consumers’ evaluations of different fairness dimensions.

This study used four dependent variables: overall fairness, customer satisfaction, service quality, and behavioral intentions to test the five fairness dimensions. Other dependent variables can be considered when examining fairness dimensions. For instance, Folger and Konovsky [28] compared the effects of distributive justice and procedural justice on employees’ reactions to pay raises. They found that procedural justice had a significant impact on employees’ organizational committee and trust in supervisor. Not much research has investigated the relationship between the fairness issue, trust, and commitment in a business. Future marketing research is needed to examine how different fairness constructs affect those different variables.

While this study made important contributions towards understanding the service fairness dimensions, there are limitations. The surveys were conducted at a mid-priced, table service restaurant in a small northeastern college town. Due to the location, type of the restaurant, and the demographics of its customers, the results of this study may be not generalized to all restaurants. It is expected that the five service fairness dimensions may play different roles for different types of restaurants in different markets.