A Study on the Experience Economy Examining a Robot Service in the Restaurant Industry Based on Demographic Characteristics

Abstract

The experience economy of service robots in the restaurant industry was researched in the present study. The dimensions of the experience economy, which include esthetics, entertainment, education, and escapism, were assessed based on customer demographics. Additionally, this study explored how the four sub-dimensions of the experience economy affects word-of-mouth intentions. The study analyzed 292 surveys. Significant differences in the experience economy were found in terms of gender, education, and monthly income, according to the statistical results. Lastly, the results of the regression analysis revealed that the experience economy’s sub-dimensions have a positive effect on word-of-mouth intentions. This is the first study to examine the experience economy of robot services, and thus, it provides theoretical contributions and practical suggestions for the restaurant industry.

1. Introduction

Service robots in the hospitality industry are regarded as one of the most innovative technologies in the Industry 4.0 [1,2]. Their implementation in the hospitality industry, which includes hotels and restaurants, has become widespread, and their significance has been widely acknowledged [3,4]. Modern Restaurant Management [5] reported numerous cases wherein service robots were used in the restaurant industry. For instance, GuGu Asian Table in Pennsylvania employs service robots in order to deliver menus to the tables. This provides an entertaining experience for the customers, because they watch the robots serve their meals. La Duni and Ari Korean BBQ in Texas also utilize service robots in order to clear dishes from the tables. The adoption of robot services was advanced by the COVID-19 pandemic as customers sought to minimize human contact [2]. Moreover, the cost of using robots has decreased by more than half over the past three decades, whereas labor wages have more than doubled [6,7]. Some experts have raised concerns about unemployment issues, although the implementation of service robots can bring about positive benefits, which include unique service experiences and increased productivity [2]. Robots offer these advantages, and they help reduce workplace injuries by replacing robots with human staff [3]. As a result, the global market for service robots in the restaurant industry is projected to increase by 16 percent and reach USD 320 million in 2028 [8].

The current study’s purpose is to investigate the significance of the experience economy in the context of robotic restaurants. Pine and Gilmore [9] developed the experience economy theory, which has been widely used in order to predict tourist/customer behavior and understand their characteristics in various studies [10,11,12]. Oh, Fiore, and Jeong [13] developed the measurement items of the experience economy using a mixed-method approach, which included qualitative and quantitative studies, in the hospitality and tourism fields. Hwang et al. [3] applied the experience economy factors to the context of service experience at restaurants, and they demonstrated their internal consistency, validity, and reliability. The four sub-dimensions of the experience economy exist, which include esthetics, entertainment, education, and escapism, and they were proposed by Pine and Gilmore [9]. These sub-dimensions are separated into categories, in accordance with the level of customer participation and the level of emotional connection to the situations. The customers are attracted to experiential elements, which include emotions and amusement, therefore, it is crucial to provide them with unique experiences in order to enhance their behavioral intentions and loyalty [3,14,15]. This also means that restaurant customers consider various experiences to be as important as the quality of food menus; however, studies concerning the experience economy, in a restaurant context, have been insufficient. Thus, this research study focused on the experience economy in the area of robot services at restaurants. More specifically, the study investigated the differences in the experience economy, with regard to the customers’ demographic characteristics. It is crucial to consider the customers’ demographic characteristics for an efficient marketing performance [16,17]. Moreover, the customers can influence the purchasing behavior of others by exchanging information about the services they experienced [18]; therefore, this study focused on the experience economy and word-of-mouth relationships.

Studies concerning the experience economy in this context are still insufficient despite the trend in the rise of robot services in the hospitality sector. In particular, there are no studies concerning the experience economy in the robot service context that focus on customer demographics and word-of-mouth intentions. Thus, this research provides an examination in order to fill the research gap and contribute to the hospitality sector. The primary research issues to be addressed include the items listed below.

(1)

To identify the differences in the experience economy in accordance with demographic factors.

(2)

To demonstrate the experience economy’s influence on word-of-mouth intentions.

The current study used SPSS software to conduct a principal component analysis, as well as a t-test, one-way analysis of variance (ANOVA), and multiple linear regression analysis. The results of this study present theoretical contributions for the hospitality sector and crucial information for restaurants that use service robots.

2. Literature Review

2.1. Robot Services in the Hospitality Sector

Service robots mimic human behavior and provide services to customers via their built-in systems [19]. Entrepreneurs are increasingly adopting robotic concepts in order to address challenges, such as increasing labor costs, shortages in labor, and the impact of the COVID-19 pandemic, in order to enhance service quality and reduce disease transmissions [20,21]. The forecasts indicate that robots may take the place of human hospitality employees on a large scale in the next decade, and several well-known hotel brands, such as Hilton and Sheraton, have already incorporated autonomous robots into their operations [22]. Restaurants are also using robots to serve the customers [23,24]. The use of robot services provides various benefits to restaurants, which include cost savings in labor, increased production efficiency, and injury prevention by replacing robots with human staff [2,3,6]. Modern Restaurant Management [5] reported numerous cases of service robots being used in the restaurant industry, such as GuGu Asian Table in Pennsylvania and La Duni in Texas.

Due to this trend, many scholars have examined robot services from the perspective of customer behavior in the hospitality sector, which includes restaurants. For instance, Hwang et al. [25] investigated travelers’ perceived value of robot services in the airline industry, and they discovered that utilitarian/hedonic value significantly influenced intentions to use robot services. Jung et al. [26] studied the behavior of hotel-goers with regard to the adoption of robot services, and the results revealed that caring for the negative aspects of human services, and the moral obligation to accept technology positively, affects attitudes toward robot services; this, in turn, affects intentions to adopt robot services. Kwak et al. [27] researched the motivations of consumers and the perceived value of robot service restaurants in the restaurant industry. They demonstrated that robot services motivated consumer innovativeness, including hedonic and social behaviors; this plays a crucial role in forming perceived value, which leads to intentions to use robot servicea. Chuah et al. [6] investigated the predictors and outcomes of attitudes toward robotic restaurants. The results of their study indicated that attitudes form a customer’s willingness to use robot services and pay more. Moreover, significantly, they identified predictors of attitudes, which include functional, emotional, social, epistemic, co-creational, and conditional value. Kao and Huang [28] successfully investigated the customers’ adoption behavior by applying the technology acceptance model. They identified that perceived ease of use, and the perceived usefulness of service robots at restaurants, can form favorable attitudes, which may be followed by intentions to use robot services. Soliman et al. [29] studied the impact of health and safety considerations on visitors’ satisfaction and loyalty in the robot service tourism sector and in a hospitality services context. They found that emotional well-being, perceived safety, and perceived green image provide satisfaction, which, in turn, impacts customers’ willingness to use service robots and their loyalty with regard to using these services. Zhang et al. [30] examined customer decision-making in the context of product recommendations given by robots. Their results indicated that the appearance and personification of robots positively affects customers’ purchase behaviors, although, it has a negative impact on the time taken for customers to make decisions and their degree of hesitation. Chi et al. [31] investigated customers’ acceptance of artificially intelligent service robots, and they found that the crucial roles of perceived social influence, hedonic motivation, and anthropomorphism comprise customer acceptance behavior. Chen and VG [32] studied the antecedents and outcomes of users’ experiences with airport service robots, and they demonstrated that service satisfaction forms perceived smartness and word-of-mouth intentions. There are still limited studies on the experience economy of robot services.

2.2. The Experience Economy

Creating experiences that generate willing purchases from customers is more important than product or service innovation [33]. Pine and Gilmore [9] proposed the theory of the experience economy, which focuses on the utility that is obtained from experiences in the context of behavioral economics [9,34]. Businesses should provide customers with memorable experiential offerings that positively influence customer perceptions, evaluations, and behaviors [35,36]. Pine and Gilmore [9] classified the experience economy into four sub-categories based on the level of customer participation, which included active and passive participation, and the level of customer connection to the situations, which included absorption and immersion. First, esthetics is classified as passive participation and immersion, and it concerns how customers interpret the environment around them based on their personal understanding of the world; they become absorbed in appreciating the environment without exerting influence on the environmental situation [9,37]. Second, entertainment is a form of passive participation and absorption, and it is a comprehensive concept that is related to humans’ ability to have fun and their interests [9,10]. In other words, customers enjoy watching and experiencing a situation, which makes it an entertaining experience. Third, education is classified as active participation and absorption, and it aims to satisfy the desire to obtain information or knowledge that was previously unknown or desired to be known [9,32]. It is an active experience, because customers actively participate in a specific situation, which is unlike the entertainment experience. Lastly, escapism is characterized by active participation and immersion, and it responds to the desire of modern individuals to soothe their tired minds in their busy daily lives [9,38]. This concept differs from esthetics, which is a form of passive participation and immersion, because customers actively enjoy an escapism experience.

The notion of the experience economy is closely linked to hospitality- and tourism-based consumer behavior. Pine and Gilmore [39] stated that individuals seek diverse experiences, such as tourism, cultural events, and unique dining experiences. Scholars, such as Penn [40] and Yeoman and McMahon-Beattie [41], additionally focused on micro trends in order to describe the changes that occur in the experience economy by highlighting its consumer-centric nature. As a result, previous studies on consumer behavior adopted the concept of the experience economy and its four sub-dimensions. For instance, Oh, Fiore, and Jeong [13] developed measures for the four sub-variables of the experience economy (i.e., esthetics, entertainment, education, and escapism) using a mixed-method approach in the hospitality and tourism context. Lai et al. [11] applied these concepts and measures to the tourists’ culinary experiences at destinations, and they discovered that all the sub-dimensions of the experience economy positively influenced perceived value. Hwang et al. [3] also adopted these constructs in the restaurant industry by focusing on the different types of service providers, which included human staff and robots. Their study revealed that all four sub-dimensions of the experience economy significantly influenced attitudes towards both types of service providers. However, the research on the experience economy with regard to robot services is still inadequate, particularly in relation to customer demographics and word-of-mouth intentions.

2.3. Differences in the Experience Economy according to Demographic Characteristics

The current study aimed to examine the differences in the experience economy, in accordance with demographic factors. Thus, the study reviewed previous works which demonstrated demographic differences between customer behaviors, in the robotic restaurant context. Numerous studies examined the differences between the predictors of customer behavior in accordance with demographics, such as gender, age, marital status, education level, and income [42,43,44,45,46]. Han and Chong [42] studied the perception of new technology, in accordance with the degree of customer innovation and technological innovation, in the restaurant self-service context. They found that female customers have lower innovativeness and higher innovation resistance than males. Jang and Lee [43] studied market segmentation based on serving robots’ attributes and evaluations in the context of robotic restaurants. Their results revealed that male customers evaluate serving robots’ animacy, likeability, and safety attributes in a more positive manner than females. It also indicated that a high-age customer group evaluates serving robots’ anthropomorphism, animacy, and safety attributes in a more positive manner than a lower-age customer group. Joo and Hwang [45] investigated the differences between perceived risks of robotic restaurants in the context of robot services, in accordance with the customers’ demographic characteristics. The results of their study revealed that females expressed greater concern about psychological risks compared with males, and older customers were more concerned about privacy risks compared with younger customers. Joo and Hwang [46] also studied differences in service experience, in the field of robotic restaurants, by focusing on the concept of SERVQUAL. The results of their study indicated that SERVQUAL factors showed somewhat significant differences in terms of gender, age, education level, and monthly income level. This means that female customers in their 50s with an associate degree, and who were in the low-income bracket, reacted more strongly to experience economy factors. Demographic characteristics are regarded as crucial factors influencing customer experience, but there is no study on the differences in the experience economy of robotic restaurants based on demographic characteristics.

The current study also reviewed existing works which observed the differences in the experience economy, based on demographic characteristics, in other contexts, in order to achieve the research objectives. Hwang and Park [10] found significant differences between the sub-variables of the experience economy of golf tournaments, based on the demographic characteristics of the spectators. The results of their study indicated that educational factors showed statistically significant differences in terms of monthly average income, entertainment factors showed differences in terms of age, and esthetic factors showed differences in terms of education level and age. Visitors in their 40s with a bachelor’s degree, who were in the middle-income bracket, reacted more strongly to experience economy factors. Kim and Choo [12] also found significant differences between the four sub-dimensions of the experience economy of hotel products based on the demographic characteristics of the customers. The results of their study revealed that married customers reacted more strongly to entertainment experiences compared with unmarried customers, and customers in their 30s and above reacted more strongly to aesthetic experiences compared with customers in their 20s. They also demonstrated that females reacted more strongly to escapism experiences compared with males.

The demographic characteristics of customers have a multifaceted impact on responses to their experience when using specific products or services [47,48]. Similarly, previous works found differences in the experience economy, based on demographic characteristics in the hospitality and tourism context [10,12]. There are no studies on the differences in the experience economy, based on demographic characteristics, with a focus on robotic restaurants, but demographic characteristics were crucial factors influencing customer experience [43,44,45,46]. It can be inferred that there are significant differences in the experience economy of robotic restaurants, in terms of demographic characteristics. This study proposes the following hypothesis, which is based on the discussions above.

Hypothesis 1 (H1).

There are significant differences in the experience economy in terms of demographic characteristics.

2.4. The Effect of the Experience Economy on Word-of-Mouth Intentions

The present study also suggests a causal relationship between the experience economy and word-of-mouth intentions. Although previous studies examined the outcomes of the experience economy and predictors of word-of-mouth intentions in the restaurant context, there are no studies on the direct effect of the experience economy on word-of-mouth intentions (see

Table 1. Previous studies on the outcomes of the experience economy and predictors of word-of-mouth intentions in the restaurant context.

Authors	Aim/Context	Main Constructs	Analysis	Main Conclusions
Hwang et al. [3]	Customer experience based on the types of service providers	The experience economy, brand attitude, brand loyalty, and types of employees (robots vs. humans)	Structural equation modeling	(Robots) Education, entertainment, escapism → Brand attitude (Humans) Entertainment, esthetics, escapism → Brand attitude
Lai et al. [11]	Tourists’ destination culinary experience	The experience economy, perceived value, satisfaction, WOM	Structural equation modeling	Education, esthetics entertainment, escapism → Functional value Esthetics, entertainment, escapism → Emotional value
Joo and Hwang [46]	Service experience in robotic restaurants	SERVQUAL, WOM intentions, and demographic factors	Multiple linear regression	Tangibles → WOMI Empathy → WOMI Assurance → WOMI
Babin et al. [49]	Restaurant experiences triggering WOM	Perceived values, customer satisfaction, and WOM intentions	Structural equation modeling	Utilitarian value → WOMI Hedonic value → WOMI
Jeong and Jang [50]		Service quality, food quality, atmosphere, price fairness, and WOM motives	Structural equation modeling	Service quality → WOM motives Atmosphere → WOM motives
Han and Ryu [51]	Driving WOM in full-service restaurants	Service encounter performance, customer satisfaction, WOM intentions	Structural equation modeling	Service encounter performance → WOMI Customer satisfaction → WOMI

Note: The arrow ‘→’ means a causal positive effect; WOM = word-of-mouth; WOMI = word-of-mouth intentions.

). Thus, the study reviewed extant works examining the causal relationship between restaurant customers’ service experiences and word-of-mouth intentions. Numerous studies attempted to investigate the predictors of word-of-mouth intentions among restaurant customers [46,49,50,51] (see Table 1). For instance, Babin et al. [49] explained the impacts of utilitarian value, hedonic value, and customer satisfaction on the word-of-mouth intentions of restaurant patrons. Jeong and Jang [50] studied restaurant experiences by activating positive word-of-mouth motivations. They found that customers’ satisfactory experiences with service providers, and a superior atmosphere, triggered word-of-mouth. Han and Ryu [51] identified that service encounter performance, customer satisfaction, and commitment are significant positive predictors of word-of-mouth intentions in the context of full-service restaurants. Joo and Hwang [46] investigated how SERVQUAL affects word-of-mouth intentions in the context of robotic restaurants. Their study indicated that intangible emotions, such as empathy and assurance, positively affect word-of-mouth intentions. It can be inferred that the customers’ positive responses to service experiences play a crucial role when forming their word-of-mouth intentions.

The existing works above noted that customers’ positive responses to service experiences play a crucial role in forming their word-of-mouth intentions [52,53,54,55]. Recent studies on restaurant customer behavior also support the causal relationship between service experience and word-of-mouth intentions [49,50,51]. Thus, it can be hypothesized that the experience economy of robotic restaurants affects customers’ word-of-mouth intentions. The present study hypothesized causal relationships between the four sub-dimensions of the experience economy and customers’ word-of-mouth intentions, which are based on the discussions above.

Hypothesis 2 (H2).

Esthetics positively and significantly affect word-of-mouth intentions.

Hypothesis 3 (H3).

Entertainment positively and significantly affects word-of-mouth intentions.

Hypothesis 4 (H4).

Education positively and significantly affects word-of-mouth intentions.

Hypothesis 5 (H5).

Escapism positively and significantly affects word-of-mouth intentions.

2.5. Proposed Conceptual Model

The conceptual model, which is illustrated in Figure 1, is presented in this study in accordance with the proposed hypotheses.

3. Methodology

3.1. Measurement Items

The measurement scales, which were suggested by prior studies, were employed in order to develop the questionnaire. The experience economy was divided into four sub-categories, which included esthetics, entertainment, education, and escapism, and this study used 12 items that were cited from Hosany and Witham [37] and Hwang and Lee [56]. Three measurement scales, which were borrowed from Hennig-Thurau, Gwinner, and Gremler [57], were used in order to analyze word-of-mouth intentions. A seven-point Likert’s scale was utilized in order to evaluate the responses, which ranged from (1) strongly disagree to (7) strongly agree.

3.2. Data Collection

In order to collect data, this study employed Embrain, which is the largest survey firm in South Korea. A data collection company employed 10 data investigators who surveyed patrons of a South Korean restaurant brand, specifically, the M brand, at a robot restaurant. The M brand is a garlic-themed Italian restaurant, and the brand pioneeringly adopted service robots in restaurants in South Korea. The brand’s stores are predominantly located in the bustling areas of Seoul, primarily targeting couples in their 20s and 30s, and families in their 40s and 50s. Their robots can deliver food to four tables at once with their vertical trays, they have speakers to communicate in Korean and English languages, and they feature an LCD screen.The survey was conducted from 1 p.m. to 4 p.m. during the day and from 7 p.m. to 10 p.m. in the evening over a period of four weeks. Following the guidelines of the Korea Centers for Disease Control and Prevention (KCDCP), one-on-one interviews were conducted on the streets of Korea during the COVID-19 pandemic. The interviewers positioned themselves at the entrance of a restaurant and they approached individuals who had finished their meals, inviting them to participate in the survey. Only those who agreed were included in the survey. Prior to commencing the survey, the purpose of the study was thoroughly explained to the respondents. Upon completion of the survey, each respondent received a token of appreciation in the form of a gift valued at around USD 5. Eight questionnaires were discarded out of the 300 surveys that were collected as they were multivariate outliers. Thus, the final analysis was based on 292 surveys from the robot restaurant.

4. Data Analysis

4.1. Profile of the Rspondents

Table 2. Respondent profiles (n = 292).

Variable	n	Percentage
Gender
Male	125	42.8
Female	167	57.2
Age
20s	61	20.9
30s	106	36.3
40s	77	26.4
Over 50s	48	16.4
Education Level
High school diploma	8	2.7
Associate’s degree	25	8.6
Bachelor’s degree	242	82.9
Graduate degree	17	5.8
Marital Status
Single	127	43.5
Married	164	56.2
Others (divorced and widow/widower)	1	0.3
Monthly income (USD)
More than 7001	75	25.7
6001~7000	68	23.3
5001~6000	82	28.1
4001~5000	46	15.8
3001~4000	12	4.1
Less than 3000	9	3.1

displays the participants’ demographic characteristics, which include gender, age, education level, marital status, and monthly income. The majority of the respondents were female, which accounted for 57.2% of the sample (n = 167), and 42.8% were male (n = 125). The age of the respondents was primarily concentrated between 20 and 39 years old. Furthermore, 26.4% (n = 77) of respondents were between 20 and 29 years of age, and 36.3% (n = 106) were between 30 and 39 years of age. The majority held a bachelor’s degree, which comprised 82.9% of the sample (n = 242). Moreover, 56.2% of the respondents were married (n = 164), and 43.5% were single (n = 127). Regarding monthly income, 28.1% of the participants earned between USD 5001 and USD 6000 per month (n = 82).

4.2. Principal Component Analysis

Table 3. Results of the principal component analysis for the cognitive drivers.

Variables (Mean and Standard Deviation)	Factor Loading	Eigen Value	Explained Variance	Cronbach’s α
Esthetics (5.66 and 0.89)		2.734	22.783	0.877
The appearance of the robotic server was good.	0.823
The robotic server was attractive.	0.812
The robot server looked good.	0.800
Entertainment (5.72 and 0.80)		2.614	21.782	0.931
This robotic server was entertaining.	0.799
This robotic server was fun.	0.793
This robotic server kept me amused.	0.773
Education (5.36 and 0.96)		2.562	21.347	0.919
This robotic server provided a real learning experience.	0.866
This robotic server stimulated my curiosity to learn new things.	0.798
This robotic server made me more knowledgeable.	0.751
Escapism (5.61 and 0.80)		2.301	19.176	0.901
I felt like I was in a different place while using this robot server at this restaurant.	0.809
I felt I was in a different world while using this robot server.	0.724
I completely escaped from my daily routine while the robotic server offered me its services at this restaurant.	0.711

Note: KMO measure of sampling adequacy = 0.915, Bartlett’s test of sphericity p < 0.001, and total explained variance = 85.088%.

displays the principal component analysis results that were used to examine the sub-dimensions of the experience economy. The analysis revealed that all four dimensions were unidimensional, as they had eigenvalues higher than 1.0. The Kaiser–Meyer–Olkin (KMO) value was 0.915, which indicated the validity of the model, and Bartlett’s test of sphericity was statistically significant at p < 0.001. Furthermore, all factor loadings were greater than 0.711. The factor model accounted for 85.088% of the variance, and the first domain explained 22.783%, the second domain explained 21.782%, the third domain explained 21.347%, and the fourth domain explained 19.176%. Each domain demonstrated suitable internal consistency, which was proven by the Cronbach’s alpha values exceeding 0.70 in each domain [58]. The four domains were identified as esthetics, entertainment, education, and escapism.

Table 4. Results of the principal component analysis for behavioral intentions.

Variables (Mean and Standard Deviation)	Factor Loading	Eigen Value	Explained Variance	Cronbach’s α
Word-of-mouth intentions (5.52 and 0.86)		2.634	87.794	0.930
I am likely to encourage others to use this restaurant.	0.846
I am likely to say positive things about this restaurant to others.	0.933
I am likely to recommend this restaurant to others.	0.932

KMO measure of sampling adequacy = 0.763, Bartlett’s test of sphericity p < 0.001.

presents the results of the principal component analysis, which were used to confirm the dimension concerning word-of-mouth intentions. The eigenvalue for intention to use was found to be greater than 1.0, which indicated the suitability of the analysis. The Kaiser–Meyer–Olkin (KMO) value was 0.763, which verified the appropriateness of the PCA, and Bartlett’s test of sphericity was statistically significant at p < 0.001. The model accounted for a total variance of 87.794%, and all factor loadings exceeded 0.846. The model additionally demonstrated internal consistency, which was indicated by the Cronbach’s alpha value exceeding 0.70 [58].

4.3. The t-Test and One-Way Analysis of Variance (ANOVA)

The t-tests and the one-way ANOVA were performed in order to measure the differences between demographic characteristics in the experience economy. The results are illustrated in

Table 5. Results of the t-test and the one-way ANOVA: The experience economy and the respondents’ demographic factors.

Gender	Male			Female			t-Value	p-Value
Esthetics	5.52			5.77			2.337	0.020 *
Entertainment	5.57			5.83			2.789	0.006 *
Education	5.19			5.49			2.693	0.008 *
Escapism	5.42			5.75			3.499	0.001 *
Age	20s		30s	40s	50s		F-value	p-value
Esthetics	5.81		5.59	5.62	5.72		0.959	0.413
Entertainment	5.83		5.63	5.67	5.85		1.377	0.250
Education	5.36		5.25	5.33	5.66		2.062	0.105
Escapism	5.73		5.55	5.53	5.68		1.031	0.379
Education level	High school diploma		Associate degree	Bachelor’s degree	Graduate degree		F-value	p-value
Esthetics	6.04		5.89	5.63	5.66		1.105	0.347
Entertainment	6.20		6.18	5.65	5.74		4.415	0.005 *
Education	6.25		5.81	5.31	5.07		4.955	0.002 *
Escapism	6.08		3.00	5.57	5.31		3.915	0.009 *
Marital status	Single		Married		Others		F-value	p-value
Esthetics	5.71		5.62		7.00		1.412	0.245
Entertainment	5.71		5.72		7.00		1.263	0.284
Education	5.30		5.40		7.00		1.786	0.170
Escapism	5.64		5.57		7.00		1.783	0.170
Monthly income level	Less than 3000	3001~4000	4001~5000	5001~6000	6001~7000	More than 7001	F-value	p-value
Esthetics	6.11	6.33	5.93	5.42	5.56	5.70	4.174	0.001 *
Entertainment	6.55	6.25	5.78	5.53	5.74	5.68	4.097	0.001 *
Education	5.85	5.83	5.42	5.19	5.46	5.30	1.771	0.119
Escapism	6.33	6.02	5.73	5.53	5.55	5.50	2.872	0.015 *

Notes: * p < 0.05.

. The results of the t-tests indicated that differences with regard to gender were significant in the four sub-dimensions of experience economy. In addition, the results of the one-way ANOVA indicated partial differences in education level, which included entertainment, education, and escapism, and monthly income level, which included esthetics, entertainment, and escapism. However, there is no statistical difference in age and marital status, which differs from our expectation.

4.4. Multiple Linear Regression Analysis

The study employed a regression analysis in order to examine the relationships between esthetics, entertainment, education, escapism, and word-of-mouth intentions; this supported Hypotheses 2, 3, 4, and 5. The results, which are presented in

Table 6. Results of the regression analysis: The effect of the experience economy on word-of-mouth intentions.

Independent			Dependent	Beta	t-Value	Hypothesis
H2	Esthetics	→	Word-of-mouth intentions	0.149	2.533 *	Supported
H3	Entertainment	→		0.312	5.074 *	Supported
H4	Education	→		0.104	1.976 *	Supported
H5	Escapism	→		0.309	4.870 *	Supported

Notes: * p < 0.05, F-value = 106.366, and Adjusted R² = 0.592.

, revealed that esthetics positively impacted word-of-mouth intentions (β = 0.149, t = 2.533, and p < 0.05), which was hypothesized in H2. Entertainment had a similarly significantly positive effect on word-of-mouth intentions (β = 0.312, t = 5.074, and p < 0.05), which confirmed H3. Education also positively affects word-of-mouth intentions (β = 0.104, t = 1.976, and p < 0.05), which is aligned with H4. Finally, the relationship between escapism and word-of-mouth intentions was found to be positive (β = 0.309, t = 4.870, and p < 0.05), which supported H5.

5. Discussion and Conclusions

5.1. Theoretical Implications

The current study preliminarily identified differences in the experience economy based on the customers’ demographic characteristics in the area of robot services. The study focused on restaurants that use service robots and it applied the four sub-dimensions of the experience economy that are proposed by Pine and Gilmore [9]. Previous studies on golf tournaments and hotel products accentuated the significance of demographics in the area of the experience economy [10,12], but there is a limited amount of research on the experience economy of robot services. Thus, the study conducted a differences analysis in order to assess differences based on demographic characteristics in the experience economy. The results of the analysis indicated that there are complete or partly significant differences in the experience economy, based on demographics. Female customers responded more positively to robot service experiences than male customers, which is in line with the study by Kim and Choo [12]. On the other hand, customers with lower education levels and lower income levels responded more positively than higher income groups, which is contrary to the study by Hwang and Park [10]. Golf is a leisure activity that is predominantly enjoyed by the upper class, therefore, this may be the reason for these outcomes. The upper class is likely to prefer full-service fine dining with professional staff as opposed to robots in the restaurant context. Thus, this may be the reason why customers with lower education levels and lower income levels responded more positively to robot services than higher educational and income groups.

However, there were no statistically significant differences in the experience economy based on age and marital status, which was contrary to the proposed hypotheses. Hwang and Park [10] found that the respondents in their 30s to 50s responded more positively to experiencing golf tournaments compared with the respondents in their 20s. This outcome could be attributed to the insufficient experience and understanding of golf among respondents in their 20s. It was assumed that similar responses would be observed across different age groups in the context of robot services due to ease of use, even without sufficient experience or understanding; this contrasts with the leisure context. Furthermore, Kim and Choo [12] identified that married customers responded more positively to experiencing hotel products compared with single customers. Married customers may prioritize certain aspects, such as a comfortable atmosphere and family-friendly services when they select a hotel; this is because they might be traveling with their spouse and/or children. On the other hand, single customers who are planning vacations may have different priorities, such as meeting new people and having new experiences outside of the hotels, thus resulting in married customers responding more positively to experiencing hotel products compared with singles. There might be observed differences when experiencing child-friendly restaurants based on marital status or accompanying children, but the data were collected from the customers of casual restaurant brands using service robots. As a result, it is likely that no significant differences with regard to using robot services at restaurants and marital status were found in this study.

Moreover, the present study demonstrated the causal relationship between the four sub-dimensions of the experience economy and word-of-mouth intentions. Numerous studies support the causal relationship between service experience and word-of-mouth [52,53,54,55], therefore, this study hypothesized the causal relationships between the four sub-dimensions of the experience economy and word-of-mouth intentions in the context of robot services. The results indicated that all four sub-dimensions of the experience economy have statistically significant effects on word-of-mouth intentions. A previous study by Hwang et al. [3] investigated the experience economy in the restaurant service context, and they found a causal relationship between the experience economy and attitudes toward restaurants. The current study proved that the experience economy directly influences customer behavior, which differs from their study. The result of the relationship between the experience economy and word-of-mouth intentions also can be supported by existing tourism studies (e.g., [59,60]). Tourists’ experiences regarding functional, satisfactory, and emotional elements can positively boost the future prediction of tourism development [59]. Gajić et al. [60] empirically proved that high-quality tourist service experiences directly form tourists’ loyalty to the destination. Consequently, this study presents a theoretical extension of the hospitality sector by identifying this relationship.

5.2. Practical Suggestions

First, entertainment is the strongest predictor with regard to forming word-of-mouth intentions. The managers can insert elements of fun and humor into the dialogue systems of robot services. For instance, when a customer orders spicy food, the service robot can be programmed to say “Choose our signature spicy menu! We’ll show you a taste that’s as fiery as it gets. The fire extinguisher is on the beverage menu." This creates a fun experience for the customers as the service robots can make witty conversation.

Second, escapism also positively affects word-of-mouth intentions. It can plan services that reflect a certain worldview in order to provide customers with escapism-based experiences. For example, this study’s data were collected from customers of the M brand, which is known for being a garlic-themed Italian restaurant. Service robots can make it feel as though customers are entering a different world, as they are programmed to say phrases such as “Welcome to Garlic World! Our menus are dominated by garlic. Let’s go mad for garlic today.” These types of remarks can also be considered as entertainment elements, which have previously been described.

Third, esthetics play an important role in the formation of the customers’ word-of-mouth intentions. Previous works emphasized the robots’ anthropomorphism/personification in the hospitality sector [30,31]. On the other hand, the brand examined by the current study used service robots that did not look humanoid. The managers should also consider the esthetic appearance of service robots; for instance, they could wear human staff uniforms. They can design the robots’ appearance by taking into consideration the brand’s unique colors, logos, and uniform styles. Moreover, robots can present esthetic experiences during the serving process by displaying videos that capture the natural environment of the origin of the main ingredients that are used in the menu.

Fourth, education positively influences word-of-mouth intentions. Service robots can provide customers with various educational experiences via dialogue or video displays. Customers can call robots instead of the chef or human staff when they need explanations about the menu before ordering. The robot can provide explanations about the menu via the configured dialogue system, and it can display information concerning the main ingredients or cooking methods on the screen. The robot can also provide information concerning the order or manner of eating, pairing menus, and other educational experiences for the customers when serving the menu.

Lastly, it is important to consider the demographic characteristics of the target customers when planning or promoting a robotic restaurant. The results imply that lower-income groups may be a more suitable target customer compared with upper-class groups. Utilizing robot services in casual restaurants is therefore more effective than it is in fine dining restaurants. The results also suggest that females may be more suitable target customers than males, therefore, restaurants with menus that cater to female customers may benefit from utilizing robot services. Furthermore, maximizing promotional effects by implementing indirect advertising in dramas that women frequently watch can also be effective.

5.3. Limitations and Future Research

First, the findings are somewhat difficult to generalize, because the samples in this study were from South Korea only. Further research should consider incorporating cultural dimensions as a moderator, because differences in individuals’ cognition and behavior based on cultural differences may occur [61]. The study also collected data from selected research panels who visited the specific brand, therefore, the results are somewhat difficult to generalize. Furthermore, the study collected samples that were collected in accordance with age group in order to examine the differences in the experience economy based on demographic characteristics. Future research could consider studying generational differences (e.g., generation X/Y/Z and baby boomers) rather than age groups. Second, the study relied on data that were collected from a single survey, which could potentially cause a common method bias. The future research should therefore employ methods that reduce the likelihood of this issue while collecting data [62]. Third, this study only adopted word-of-mouth intentions as an outcome of the experience economy. It is recommended that future studies consider other constructs of customer loyalty and mediators between the experience economy and loyalty. For instance, subsequent studies can develop a comprehensive framework that includes the effect of the experience economy on perceived value and attitude, the effects of two outcomes on brand attachment, and the effects of brand attachment on revisit and word-of-mouth intentions. The framework should be tested by structural equation modeling rather than regression analysis. The moderating role of demographics, such as gender and age, could also be studied further. Fourth, it also only focused on the customers’ demographic characteristics, therefore, an investigation that considers behavioral characteristics is recommended, such as dining-out frequency, average check amount, and type of companion. In addition, there might be differences in the experience economy based on whether customers have experienced robot services or not. It can be tested with a t-test or multiple-group analysis. Lastly, this study presents results that are based on the stated preferences that are reported by the customers themselves. The relationship between revealed preferences, which reflect the factors the customers perceive as being important and actual behavioral outcomes, need to be investigated.