Examining the Asymmetric Effect of Multi-Shopping Tourism Attributes on Overall Shopping Destination Satisfaction

Shopping Tourism Attributes

UNWTO (2014, p. 13) defines shopping tourism as “a contemporary form of tourism fostered by individuals for whom purchasing goods outside of their usual environment is a determining factor in their decision to travel.” This definition is consistent with the shopping tourism literature that shopping is a major travel motivation (M. J. Choi, Heo, and Law 2016b) and is treated as an irresistible attraction (Timothy 2014). This motivation leads tourists to allocate a high portion of the total travel expenditure to shopping activities (Albayrak, Caber, and Cömen 2016); thus, shopping tourism is recognized as a rapidly growing tourism sector (Sharma, Chen, and Luk 2018).

Shopping tourists show different shopping behavioral patterns when traveling than when staying in their home countries (I. A. Wong and Wan 2013). Tourists are likely to exhibit more impulse buying (Thomas and LeTourneur 2001) and thus spend more money on shopping than dining or accommodation because they can access merchandise that are either unique to a shopping destination only (Turner and Reisinger 2001) or cheaper than in their home country (Yeung, Wong, and Ko 2004). Henderson et al. (2011) stated that shopping tourism is facilitated or inhibited by critical attributes. The attributes are broken down into shopping-related (shopping opportunities, shop environments, products, service quality, and customer protection) and destination-related attributes (infrastructure, attractions, transportation, marketing campaign, government action). Shopping-related attributes are extensively reviewed in the extant literature (Albayrak, Caber, and Cömen 2016; Henderson et al. 2011; LeHew and Wesley 2007; Lloyd, Yip, and Luk 2011; Sharma, Chen, and Luk 2018). However, most studies have overlooked destination-specific attributes in examining shopping tourism phenomenon. This study reviews shopping- and destination-specific attributes to assess their asymmetric impact on shopping destination satisfaction.

Underlying Theories of Asymmetric Effect of Attributes on Satisfaction

The performance of multi-attributes determines overall satisfaction (Back 2012; J. S. Lee, Choi, and Chiang 2017; J. S. Lee and Min 2013). The asymmetric effect of attributes on satisfaction occurs when a particular attribute does not induce customer satisfaction even after an investment in improving the attribute performance, whereas another attribute exhibits greater impact on customer satisfaction after an equivalent investment is made in this attribute.

The asymmetric effect of attributes on satisfaction is underlain by prospect theory (Kahneman and Tversky 1979), in which (1) gains and losses are evaluated by comparing them with a reference point, (2) losses are given more weight than gains in assessing value (loss aversion), and (3) the marginal value of gains or losses declines with their size (diminishing sensitivity). In particular, loss aversion explains that people are reluctant to realize losses because they consider losses more serious and larger than equivalent-sized gains (Einhorn and Hogarth 1981). This belief explains the phenomenon in which individuals tend to hold losers more than winners in stock investment. Loss aversion indicates that negative asymmetry exists given that the negative performance of a particular attribute is more sensitive to satisfaction than the corresponding size of its positive performance. Moreover, diminishing sensitivity implies that as an attribute approaches a high or low performance level, satisfaction is not as significantly affected as when the attribute’s performance is at an intermediate level. Loss aversion and diminishing sensitivity founded on prospect theory offer theoretical rationales for the asymmetric nature of attributes (Mittal, Ross, and Baldasare 1998).

The asymmetric effect of attributes on satisfaction is conceptualized by attractive quality theory (Kano 1984) composed of five quality dimensions. Kano (1984) asserted that five quality dimensions influence satisfaction differentially and are classified into “attractive,” “one-dimensional,” “must-be” “indifferent,” and “reverse” qualities.

Attractive qualities concern attributes that travelers do not generally expect. Hence, when tourists experience such qualities, they will likely be delighted and positively surprised, thereby ending up with high satisfaction. However, given that tourists do not expect attractive qualities, they will not be dissatisfied even if these qualities are not provided. That is, attractive qualities are positively asymmetric with satisfaction.

Must-be qualities are the opposite of attractive qualities; that is, they display a negative asymmetric relationship with satisfaction. These qualities are perceived by tourists as basic attributes. Thus, tourists feel disappointed when must-be qualities do not meet their expectations, but they are not satisfied even when these qualities are not provided as they take the qualities for granted.

One-dimensional qualities are defined as positively symmetric with satisfaction. Travelers are satisfied (dissatisfied) when these qualities are (are not) offered. Indifferent qualities literally do not have a relationship to satisfaction or dissatisfaction, regardless of whether they are provided or not. Reverse qualities trigger dissatisfaction if available and cause satisfaction if unavailable.

Attractive quality theory is further explored in the extant literature from the perspective of the three-factor theory, in which attributes are categorized as dissatisfiers, satisfiers, and hybrids (Anderson, Fornell, and Mazvancheryl 2004; Back 2012; Deng 2007; Füller, Matzler, and Faullant 2006; Mikulić and Prebežac 2008; Oliver 1997). For example, Oliver (1997) classified attributes into bivalent satisfiers, monovalent dissatisfiers, and monovalent satisfiers by advocating that attributes influence satisfaction differently. In particular, bivalent satisfiers (known as hybrids), which are contingent on the quality level of attributes, induce satisfaction or dissatisfaction. Monovalent dissatisfiers are regarded as must-be attributes (e.g., store temperature). Thus, individuals feel dissatisfied when these attributes are absent. However, even when these attributes are present, people are not necessarily satisfied because they take these attributes for granted. Unlike monovalent dissatisfiers, monovalent satisfiers are considered value-added and delightful attributes (e.g., staff foreign language proficiency) that cause a high satisfaction level and do not evoke dissatisfaction even if they are not provided because these attributes are generally unexpected.

In the tourism and hospitality literature, three-factor theory is adopted to examine the asymmetric effects of attributes on incentive traveler satisfaction (J. S. Lee, Choi, and Chiang 2017), convention attendee satisfaction (J. S. Lee and Min 2013), ski resort customer satisfaction (Füller, Matzler, and Faullant 2006), and restaurant customer satisfaction (Back 2012). In line with the above-mentioned research, the current study explores the asymmetric effects of attributes on shopping destination satisfaction, based on the following three factors (Back 2012; Füller, Matzler, and Faullant 2006; J. S. Lee, Choi, and Chiang 2017):

Dissatisfiers and frustrators belong to negative asymmetrical attributes (Anderson and Mittal 2000). Dissatisfiers are considered must-be attributes that induce dissatisfaction when they are absent while frustrators are regarded as intense dissatisfiers (Füller and Matzler 2008). These attributes provoke frustration (a high level of dissatisfaction) if they are not supplied. Tourists take dissatisfiers and frustrators for granted; thus, these attributes do not trigger satisfaction even if they are provided (J. S. Lee, Choi, and Chiang 2017).

Hybrids refer to symmetrical attributes (Anderson and Mittal 2000). When these attributes are implemented, travelers are satisfied. If they are not implemented, then travelers are dissatisfied (Back 2012; Füller, Matzler, and Faullant 2006; J. S. Lee, Choi, and Chiang 2017).

Satisfiers and delighters are classified as positive asymmetrical attributes (Anderson and Mittal 2000). They exhibit the opposite nature of dissatisfiers and frustrators (Füller and Matzler 2008). That is, satisfiers are considered value-added attributes that cause satisfaction when they are fulfilled. Delighters are perceived as extreme satisfiers, such that tourists feel delighted when these attributes are provided. Travelers do not generally expect satisfiers and delighters. Thus, they are not likely to be dissatisfied when these attributes are unavailable (J. S. Lee, Choi, and Chiang 2017).

Measurement Development and Data Collection

Data collection

This study focused on Chinese tourists who had visited Korea. The Chinese people have considerable opportunities to accumulate disposable income and travel abroad due to the current spectacular economic growth of China (UNWTO 2014). Chinese citizens were ranked as top spenders on international tourism in 2013, with a total expenditure of US$129 billion (UNWTO 2014). Korea has recently become a popular shopping destination among the Chinese. KTO (2017) reported that China had become Korea’s top inbound market, accounting for 46.79% (n = 8,067,722) of its inbound tourists in 2016. These figures exhibit a sharp increase in ratio and number compared with the values in 2011 (i.e., 22.67%, n = 2,220,196); thus, Korea was deemed an appropriate study site.

To collect data, this study hired Sojump, a professional research software company managed by the Shanghai Information Technology Corporation. Since 2005, Sojump has been widely used by private companies and universities to collect online data and conduct market research in China. An online questionnaire about shopping tourism attributes was developed. The questionnaire asked the respondents to indicate their agreement with each item using a 7-point Likert-type scale, where 1 = strongly disagree and 7 = strongly agree. After data collection, 523 usable samples were obtained for data analysis. Table 1 shows the profiles of the respondents. Males accounted for 46.3% (n = 242) and females accounted for 53.7% (n = 281) of the sample. Approximately 79% (n = 413) of the respondents were in their 20s and 30s. Slightly more than half of the sample, that is, 53.9% (n = 282), were first-time visitors, whereas 46.1% (n = 241) were repeat visitors.

OPEN IN VIEWER

Table 1.

Profiles of Respondents (n = 523).

Profile Category	Frequency	Percentage
Gender
Male	242	46.3
Female	281	53.7
Age, years
20 or below	34	6.5
21–30	186	35.6
31–40	227	43.4
41–50	64	12.2
51–60	10	1.9
61 or above	2	0.4
Education level
High school	36	6.9
Associate degree	80	15.3
Undergraduate degree or above	338	64.6
Postgraduate degree or above	69	13.2
Monthly household income
Less than CNY 15,000	110	21.0
CNY 15,001–30,000	236	45.1
CNY 30,001–45,000	95	18.2
CNY 45,001–60,000	47	9.0
CNY 60,001 or above	35	6.7
Frequency of visit to Korea
Once	282	53.9
Twice	162	31.0
Three times	52	9.9
Four times	13	2.5
Five times or more	14	2.7

Note: 1 yuan is approximately 0.15 US dollars and 0.13 Euros (as of December 2017).

Factor Analysis

EFA was conducted using principal axis factoring with oblique rotation to identify the underlying dimensions of shopping tourism. Two distinct domains/concepts (shopping- and destination-specific attributes) exist over shopping tourism; thus, EFA was run separately for the two domains.

The EFA results show that 3 factors (shopping atmosphere, store service orientation, merchandise) underlay shopping-specific domains, with 66.78% of the variance (Table 2), whereas destination-specific domains comprised 6 dimensions, with 72.43% of the variance (Table 3). The destination-specific domains are affordability, K-pop culture/media (represented by drama, music, and film), safety, accessibility, government promotion, and attraction. Consequently, a nine-factor structure with 45 items underlay shopping tourism.

OPEN IN VIEWER

Table 2.

Results of Exploratory Factor Analysis for Shopping-Specific Attributes.

Factors	Factor Loading	Cronbach’sAlpha
Factor 1: Shopping Atmosphere (eigenvalue: 13.05; % of variance: 48.96)		.96
1. Temperature in the store was comfortable.	.96
2. Store was neat and clean.	.91
3. Overall lighting level in the store environment was appropriate.	.80
4. Noise level in the store was acceptable.	.72
5. Store staff was neat in appearance.	.70
6. Color used in the décor scheme added excitement to the store’s environment.	.66
7. Store was conveniently located in a city center.	.65
8. In-store merchandise information was available.	.63
9. Overall, store layout made getting around easy.	.59
10. Store staff had good service attitude.	.58
11. Store staff had good knowledge of their jobs.	.57
12. Store’s architecture gave it an attractive character.	.56
13. Interior was decorated in an attractive fashion.	.55
14. Overall design was interesting.	.53
15. Store was easily accessible from the hotel.	.50
16. Pre-shopping information was available.	.48
17. Store staff was easy to communicate with in a foreign language.	.45
Factor 2: Store Service Orientation (eigenvalue: 3.15; % of variance: 10.26)		.91
1. Store offered customer-friendly product return/exchange policy.	.75
2. Overseas delivery service was provided.	.74
3. Store offered after-sales services if necessary.	.70
4. Store was competent in handling complaints.	.65
5. Store offered appealing special deals (e.g., discounts, coupons, etc.).	.47
Factor 3: Merchandise (eigenvalue: 2.03; % of variance: 7.56)		.92
1. Merchandise selection varied.	.84
2. Quality of merchandise carried by the store was high.	.82
3. The merchandise that I was looking for was available.	.80
4. Merchandise was reasonably priced.	.77
5. Merchandise offered value for money.	.72
6. Merchandise was authentic.	.64
7. Merchandise was fashionable.	.62

Note: Total variance explained = 66.78; Kaiser–Meyer–Olkin measure of sampling adequacy = .97; Bartlett’s test of sphericity, p < .001.

OPEN IN VIEWER

Table 3.

Results of Exploratory Factor Analysis for Destination-Specific Attributes.

Factors	Factor Loading	Cronbach’sAlpha
Factor 1: Affordability (eigenvalue: 3.33; % of variance: 21.19)		.88
1. Hotel accommodations were affordable.	.83
2. Dining was affordable.	.74
3. Local transportation was affordable.	.71
Factor 2: K-pop Culture/Media (eigenvalue: 2.47; % of variance: 17.26)		.93
1. My shopping in Korea was inspired by Korean celebrities.	−.87
2. My shopping in Korea was motivated by destinations featured in Korean pop culture.	−.84
3. My shopping in Korea was encouraged by products/services featured in K-pop culture.	−.82
4. My shopping in Korea was induced by K-pop culture (drama, music, film).	−.80
Factor 3: Safety (eigenvalue: 1.92; % of variance: 12.03)		.83
1. Korea is a safe and secure destination.	.92
2. Korea is a politically stable destination (e.g., free of coups and strikes).	.65
Factor 4: Accessibility (eigenvalue: 1.43; % of variance: 8.23)		.86
1. Local transport was convenient.	.80
2. Korea was properly accessible.	.47
Factor 5: Government Promotion (eigenvalue: 1.20; % of variance: 7.55)		.85
1. Visitor information/service centers were available in Korea.	.64
2. Korean government–sponsored special promotional campaign (e.g., nationwide shopping festival) was attractive.	.63
3. Korea operated a user-friendly tax refund system (e.g., number of certified tax-free stores, availability of self-service tax refund kiosks at airports, etc.).	.62
Factor 6: Attraction (eigenvalue: 1.08; % of variance: 6.17)		.85
1. Korea had a variety of tourist attractions.	−.55
2. A variety of international cuisines were available in Korea.	−.52

Note: Total variance explained = 72.43; Kaiser–Meyer–Olkin measure of sampling adequacy = .96; Bartlett’s test of sphericity, p < .001.

Impact Range Performance Analysis (IRPA) and Impact Asymmetry Analysis (IAA)

The current study adopted IRPA and IAA to assess the asymmetric impact of an attribute on satisfaction (Mikulić and Prebežac 2008). Penalty–reward contrast analysis (PRCA) was undertaken using multiple regression analysis and dummy variables to conduct IRPA and IAA. The lowest “attribute performance score” (APS) was input as 1 to create the first set of dummy variables, that is, the penalty index (PI). That is, if an attribute performance is 1, then it will be coded as 1. When an attribute performance is rated as 2, 3, 4, 5, 6, or 7, then it will be coded as 0. The opposite procedure is used to generate the second set of dummy variables, that is, the reward index (RI). The highest APS (i.e., 7) was input as 1. Then, a low attribute performance (rated as 1, 2, 3, 4, 5, or 6) was input as 0. RI and PI were created through regression analysis using dummy variables. RI refers to attributes that are positively associated with satisfaction, whereas PI signifies attributes that are negatively associated with satisfaction (Table 4). RI and the absolute value of PI were summed to produce the value of an attribute’s “range of impact on satisfaction” (RIS), after which the equations below (Mikulić and Prebežac 2008) were adopted to generate the “satisfaction-generating potential” (SGP) and the “dissatisfaction-generating potential” (DGP). SGP and DGP were consequently used to determine “impact asymmetry” (IA).

where RI = reward index for attribute i, PI = penalty index for attribute i, RIS_i =│PI│+ ri = range of impact on satisfaction, and SGP_i + DGP_i = 1.

OPEN IN VIEWER

Table 4.

Results of IRPA and IAA.

Incentive Travel Attributes	RI	PI	RIS	SGP	DGP	IA	AR
Shopping Atmosphere: Store Staff
Store staff was neat in appearance.	0.18	−0.39	0.57	0.32	0.68	−0.37	Dissatisfier
Store staff had good service attitude.	0.50	−0.45	0.95	0.53	0.47	0.05	Hybrid
Store staff had good knowledge of their jobs.	0.29	−0.6	0.89	0.33	0.67	−0.35	Dissatisfier
Store staff was easy to communicate with in a foreign language.	0.13	−0.04	0.18	0.76	0.24	0.51	Delighter
Shopping Atmosphere: Servicescape
Temperature in the store was comfortable.	0.06	−0.16	0.22	0.29	0.71	−0.43	Frustrator
Store was neat and clean.	0.34	−0.48	0.82	0.42	0.58	−0.17	Dissatisfier
Overall lighting level in the store environment was appropriate.	0.32	0.005	0.32	0.98	0.02	0.97	Delighter
Noise level in the store was acceptable.	0.21	−0.12	0.33	0.63	0.37	0.27	Delighter
Color used in the décor scheme added excitement to the store.	0.68	−0.40	1.08	0.63	0.37	0.26	Delighter
Overall, store layout made getting around easy.	0.18	−0.58	0.75	0.23	0.77	−0.53	Frustrator
Store’s architecture gave it an attractive character.	0.26	−0.12	0.38	0.69	0.31	0.38	Delighter
Interior was decorated in an attractive fashion.	0.39	−0.31	0.70	0.56	0.44	0.11	Delighter
Overall design was interesting.	0.37	−0.25	0.62	0.59	0.41	0.19	Delighter
Shopping Atmosphere: Store Accessibility
Store was conveniently located in a city center.	0.20	−0.18	0.39	0.52	0.48	0.05	Hybrid
In-store merchandise information was available.	0.32	−0.77	1.09	0.29	0.71	−0.41	Frustrator
Store was easily accessible from the hotel.	0.35	−0.3	0.65	0.54	0.46	0.09	Hybrid
Pre-shopping information was available.	0.15	−0.31	0.47	0.33	0.67	−0.34	Dissatisfier
Store Service Orientation
Overseas delivery service was provided.	0.08	−0.25	0.33	0.25	0.75	−0.50	Frustrator
Store offered customer-friendly product return/exchange policy.	0.14	−0.51	0.64	0.21	0.79	−0.58	Frustrator
Store offered after-sales services if necessary.	0.17	−0.38	0.55	0.31	0.69	−0.39	Dissatisfier
Store was competent in handling complaints.	0.09	−0.21	0.30	0.30	0.70	−0.40	Dissatisfier
Store offered appealing special deals (e.g., discounts, coupons, etc.).	0.32	−0.58	0.90	0.36	0.64	−0.28	Dissatisfier
Merchandise
Merchandise selection was varied.	0.24	−0.35	0.59	0.41	0.59	−0.18	Dissatisfier
The merchandise that I was looking for was available.	0.19	−0.30	0.49	0.38	0.62	−0.24	Dissatisfier
Merchandise was reasonably priced.	0.02	−0.22	0.24	0.08	0.92	−0.83	Frustrator
Merchandise offered value for money.	0.15	−0.03	0.18	0.82	0.18	0.64	Delighter
Quality of merchandise carried by the store was high.	0.19	−0.16	0.35	0.54	0.46	0.09	Hybrid
Merchandise was fashionable.	0.20	−0.39	0.59	0.34	0.66	−0.32	Dissatisfier
Merchandise was authentic.	0.16	−0.33	0.49	0.33	0.67	−0.33	Dissatisfier
Affordability
Dining was affordable.	0.19	−0.48	0.67	0.29	0.71	−0.42	Frustrator
Hotel accommodations were affordable.	0.23	−0.35	0.58	0.40	0.60	−0.20	Dissatisfier
Local transportation was affordable.	0.23	−0.76	0.99	0.24	0.76	−0.53	Frustrator
K-pop Culture
My shopping in Korea was induced by K-pop culture.	0.23	−0.46	0.69	0.33	0.67	−0.34	Dissatisfier
My shopping in Korea was inspired by Korean celebrities.	0.15	−0.20	0.35	0.43	0.57	−0.15	Dissatisfier
My shopping in Korea was motivated by destinations featured in K-pop culture.	0.21	−0.66	0.87	0.24	0.76	−0.51	Frustrator
My shopping in Korea was encouraged by products/services featured in K-pop culture.	0.24	−0.37	0.61	0.40	0.60	−0.21	Dissatisfier
Safety
Korea is a politically stable destination (e.g., free of coups and strikes).	0.28	−0.50	0.78	0.36	0.64	−0.29	Dissatisfier
Korea is a safe and secure destination.	0.49	−1.22	1.71	0.28	0.72	−0.43	Frustrator
Accessibility
Korea was properly accessible.	0.51	−0.61	1.12	0.46	0.54	−0.08	Hybrid
Local transport was convenient.	0.36	−0.83	1.19	0.30	0.70	−0.40	Dissatisfier
Government Promotion
Korea operated a user-friendly tax refund system.	0.19	−0.75	0.93	0.20	0.80	−0.60	Frustrator
Visitor information/service centers were available in Korea.	0.31	−0.10	0.41	0.75	0.25	0.50	Delighter
Korean government–sponsored special promotional campaign was attractive.	0.50	−0.35	0.85	0.59	0.41	0.18	Satisfier
Attraction
Korea had a variety of tourist attractions.	0.46	−0.72	1.18	0.39	0.61	−0.22	Dissatisfier
A variety of international cuisines were available in Korea.	0.40	−0.67	1.08	0.37	0.63	−0.25	Dissatisfier

Note: IRPA = impact range performance analysis; IAA = impact asymmetry analysis; RI = reward index; PI = penalty index; RIS = range of impact on satisfaction; SGP = satisfaction-generating potential; DGP = dissatisfaction-generating potential; IA = impact asymmetry; AR = asymmetry range.

IA was used as a threshold to identify attributes as dissatisfiers, hybrids, or satisfiers (Mikulić and Prebežac 2008). That is, if SGP is greater than DGP, then an attribute is deemed to induce more satisfaction than dissatisfaction and is regarded as a satisfier. Conversely, if DGP is larger than SGP, then an attribute is considered to cause more dissatisfaction than satisfaction and is identified as a dissatisfier. If the difference between SGP and DGP is arithmetically slight, then an attribute is classified as a hybrid because of its equivalent impacts on satisfaction and dissatisfaction. The current study adopted the following cutoff point, as proposed by Mikulić and Prebežac (2008), to identify attributes as frustrators (extreme dissatisfiers, IA < −0.4), dissatisfiers (−0.4 ≤ IA < −0.1), hybrids (−0.1 ≤ IA ≤ 0.1), satisfiers (0.1 < IA ≤ 0.4), and delighters (extreme satisfiers, IA > 0.4).

This study assessed the asymmetric effects of shopping- and destination-specific attributes on shopping destination satisfaction based on the aforementioned mechanism (Table 4). Shopping atmosphere, store service orientation, and merchandise underlay shopping-specific domains, in which shopping atmosphere was dominant, with 17 attributes. Shopping atmosphere comprised many attributes that were clearly reflective of three subdimensions (store staff, servicescape, and store accessibility). The asymmetric effect of shopping atmosphere was presented using the three subdimensions to clarify its dynamics. Store staff-related attributes were shown to be dissatisfiers, hybrids, and delighters. Staff appearance and job knowledge were classified as dissatisfiers, whereas service attitude and foreign language communication were found to be a hybrid and a delighter, respectively. Servicescape attributes were either negative or positive asymmetric attributes. Store temperature and layout (frustrators) and cleanliness (a dissatisfier) fell into negative asymmetry, whereas delighters or positive asymmetry appeared in the remaining attributes (lighting, noise, décor color, architecture, interior, overall design). Store accessibility attributes indicate accessibility to physical location and store information. Physical location and accessibility (hybrids) were found to be symmetrically related to shopping satisfaction, whereas merchandise information and pre-shopping information were a frustrator and a dissatisfier, respectively. Store service orientation attributes were all characterized by negative asymmetry. Delivery service and return/exchange policy fell under frustrators. After-sales services, complaint handling, and special deals were classified under dissatisfiers. Merchandise attributes were also strongly related to negative asymmetry, except for merchandise quality (a hybrid) and value for money (a delighter). Merchandise variety, availability, fashionable style, and authenticity were found to be dissatisfiers, whereas merchandise price was a frustrator. Accessibility to Korea was considered a hybrid attribute, whereas the convenience of local transport turned out to be a dissatisfier attribute. In government promotion, the tax refund system was deemed as a frustrator, visitor information/service center as a delighter, and special promotional campaign as a satisfier.

On the basis of the IA and RIS values presented in Table 4, Figure 1 is designed to illustrate the relative position of attributes by simultaneously judging the asymmetric impact (IA, Y-axis) and the range of impact (RIS, X-axis) of attributes on shopping satisfaction. RIS was divided into low, medium, and high values using the RIS mean of attributes. A detailed picture of the asymmetric nature of attributes on shopping satisfaction was presented through the simultaneous interpretation of IA and RIS. For example, even if the RIS value of Attributes 5 and 7 in shopping atmosphere-servicescape slightly differed, the two attributes were found to exhibit an extremely opposite asymmetric nature (i.e., frustrator vs. delighter). The simultaneous illustration of the IA and RIS of attributes enables concerned government bodies or industry practitioners to identify which attributes should be prioritized over others in managing shopping destination satisfaction, which is discussed in the practical implication section.

OPEN IN VIEWER

Figure 1.

IRPA grid.

Theoretical Implications

This study examines the asymmetric effect of shopping- and destination-specific attributes on shopping destination satisfaction from the perspective of Chinese tourists under the dimensions of shopping atmosphere, store service orientation, merchandise, affordability, K-pop culture, safety, accessibility, government promotion, and attraction. Jin, Moscardo, and Murphy (2017) encouraged further research on Chinese tourists in the shopping tourism literature given that they are considered the world’s biggest spenders in international tourism. The findings of this study contribute to the extant shopping tourism literature from the perspective of Chinese tourists in several ways.

First, the existing shopping literature measures shopping satisfaction in a limited manner that uses either a mean value of an attribute or overall satisfaction. For example, the perception and expectation of each attribute were compared with each other using a mean score (Heung and Cheng 2000; Yeung, Wong, and Ko 2004) to identify a satisfaction level. If the perception of an attribute is greater than its expectation, then the attribute is defined as satisfied. Otherwise, the attribute is considered dissatisfied. I. A. Wong and Wan (2013) operationalized shopping satisfaction as overall satisfaction and adopted a symmetric, linear relationship to identify the antecedents and consequences of shopping satisfaction. That is, the previous literature relies on the mean value of attributes and the linear relationship of overall satisfaction with shopping dimensions while not considering the asymmetric impact of shopping attributes on shopping satisfaction.

Asymmetric relationships afford researchers with a more detailed picture of the responses of attributes to satisfaction that symmetrical relationships cannot show when they assess the dynamic impact of attributes on satisfaction. Symmetric linear relationships enable researchers to test whether relationships are statistically significant either in a positive or negative manner. If relationships are statistically insignificant, then no relationship is interpreted over the hypothesized paths. By contrast, asymmetric relationships built on IRPA and IAA allow researchers to look into negative asymmetric (frustrators, dissatisfiers), positive asymmetric (satisfiers, delighters), or symmetric (hybrids) impacts of attributes on shopping destination satisfaction. The identification of the dynamic impacts of attributes assists concerned industry practitioners or government bodies in prioritizing attributes for the strategic management of shopping tourism, which is further discussed under practical implications.

Second, the previous shopping tourism literature (M. Choi, Law, and Heo 2016; Heung and Cheng 2000; I. A. Wong and Wan 2013; Yeung, Wong, and Ko 2004) has overlooked destination-related attributes in examining shopping tourism, although the Globe Shopper Index recognizes the critical role of destination attributes in shopping destination competitiveness. The current study identifies attributes of destination dimensions (affordability, K-pop culture, safety, accessibility, government promotion, and attraction) within shopping tourism and assesses the asymmetric impacts of these attributes on shopping satisfaction. Among the destination dimensions, K-pop culture and government promotion, which were developed from the qualitative research process, are considered unique to shopping tourism, whereas the remaining dimensions are deemed under basic tourism domains.

In addition to accessibility and government promotion, most destination attributes are found to be must-be attributes (frustrators or dissatisfiers). Among accessibility attributes, destination accessibility is found to be a hybrid, whereas local transport is considered a frustrator. Among government promotion attributes, the tax refund system falls under dissatisfiers, government-sponsored special promotional campaign under satisfiers, and visitor information/service center under delighters. Thus, active government promotion of shopping tourism enhances shopping satisfaction by offering value-added experience, whereas tourists recognize the tax refund system as a basic attribute.

Another noteworthy implication arises from K-pop culture because this dimension is neither discussed nor tested empirically in the extant shopping tourism literature. The current study operationalizes K-pop culture as K-pop celebrities (dissatisfier), products/services (dissatisfier), and destinations (frustrator) featured in K-pop media (frustrator) and K-pop-induced shopping (dissatisfier). All K-pop culture attributes are viewed as must-be attributes; K-pop culture is believed to be fundamental to Chinese shopping tourists as K-pop is very popular among the Chinese. The powerful role of K-pop in shopping tourism can be construed from film-induced tourism. Film tourism advocates that media affect tourist motivations and visit patterns, along with destination image (Croy 2010; Laing and Crouch 2009), in which individuals develop their personalized memory and associate symbolic meanings, emotions, and attitudes with celebrities and places featured in dramas or films (H. Kim and Richardson 2003; S. Lee, Scott, and Kim 2008). Media audience is motivated to purchase memorabilia and products related to figures featured in media (Ferguson 1992) given the strong emotional bond that develops from media. The theoretical underpinning of film tourism supports the asymmetric effect of K-pop culture on shopping tourism.

Third, the method used in this study offers advantages over qualitative and descriptive methods for assessing the asymmetric impact used in the previous literature. For example, Cadotte and Turgeon (1988) identified the attributes of triggering compliments and complaints from restaurants and hotels based on the critical incident technique (CIT). They examined the relative frequency of compliments- and complaints-causing attributes and typologized attributes into satisfiers (causing satisfaction/compliments), dissatisfiers (triggering dissatisfaction/complaints), criticals (leading to satisfaction and dissatisfaction), and neutrals (not significantly causing satisfaction nor dissatisfaction). Similarly, Johnston (1995) analyzed anecdotes incurred in bank services and identified service attributes as dissatisfiers, satisfiers, and criticals. Their studies (Cadotte and Turgeon 1988; Johnston 1995) shed light on the asymmetric effect of attributes on satisfaction in the service literature but still suffer from the limitations of CIT. For example, the analysis of complaints and compliments suggests that only extreme anecdotes of satisfaction and dissatisfaction experiences are considered. Only a few attributes are likely to fall under the categories of satisfiers and dissatisfiers given the possibility that particular attributes are always dominant in favorable anecdotes (compliments), whereas other attributes frequently appear in service failures (complaints). Therefore, researchers are not allowed to examine unobserved attributes in critical incidents. Consequently, only a handful of attributes with asymmetric impact are identified. This limitation explains why Johnston (1995) discovered only two attributes as satisfiers and two attributes as dissatisfiers among banking service attributes. Moreover, CIT asks survey participants to recall their perceptions of critical anecdotes that they experienced before. The time lapse can cause participants to interpret anecdotes differently from their original interpretation of the incidents because of other events that followed (Johnston 1995).

The current study examines how asymmetric shopping tourism attributes affect shopping destination satisfaction. Thus, this study identifies not only shopping-specific attributes but also destination-based attributes that the previous shopping literature has overlooked using IRPA and IAA. Asymmetric relationships between attributes and shopping satisfaction offer an expanded view of the dynamic effects of attributes, whereas destination-specific attributes allow researchers to examine the unexplored roles of government promotion and K-pop media in shopping tourism. The current findings present theoretical implications that have not been discussed and explored in the existing shopping tourism literature, thereby adding value to the extant literature.

Practical Implications

An insight into the asymmetric effects of shopping tourism attributes allows concerned retailers/government bodies to prioritize such attributes over others for the strategic management of shopping tourism for Chinese tourists. Attributes are prioritized as illustrated in Figures 2 and 3 based on the categorization of their asymmetric impact. This study recommends that an in-depth management of must-be attributes (frustrators, dissatisfiers) and hybrid attributes is fundamental to the success of shopping tourism given that these attributes result in dissatisfaction when they are not properly managed. If the quality of these attributes meets the expectation of shopping tourists, then tactical focus should be placed on value-added attributes (satisfiers, delighters) to make shopping tourists delighted.

OPEN IN VIEWER

Figure 2.

Prioritizing shopping-specific attributes for shopping destination satisfaction.

OPEN IN VIEWER

Figure 3.

Prioritizing destination-specific attributes for shopping destination satisfaction.

Figure 2 illustrates that the attributes of staff, merchandise, store accessibility, and store service orientation are strongly present across frustrators, dissatisfiers, or hybrids under shopping domains, whereas servicescape attributes are dominant among delighters. Retailers who target shopping tourists should regularly monitor whether must-be attributes (frustrators and dissatisfiers) cater to the expectations of shopping tourists. If these attributes fail to meet tourists’ anticipation, then their shopping experience is likely to be ruined (although high-quality servicescape attributes are provided) because these attributes are taken for granted.

In destination domains (Figure 3), frustrators, dissatisfiers, or hybrids are represented by affordability, K-pop culture, safety, accessibility, and attractions, whereas government promotion triggers delight and provides value-added experience to shopping tourists. Affordability, safety, accessibility, and attractions are well recognized as elementary attributes in the tourism literature, whereas pop culture is not known as a must-be attribute. In this study, K-pop media are found to be a fundamental attribute in shopping tourism for Chinese tourists. Korean government bodies that are concerned with shopping tourism should strategically utilize K-pop media and celebrities in promoting products/services and destinations to Chinese tourists. Simultaneously, government efforts are geared toward the design of shopping festivals/campaigns and tourist facilities to enhance the value-added experience of Chinese shopping tourists.

The asymmetric categorization and prioritization of shopping tourism attributes may vary with shopping destinations and the ethnicities of shopping tourists. Concerned retailers/government bodies are recommended to conduct individual surveys to identify the asymmetric classification and prioritization of shopping tourism attributes that are unique to a particular shopping destination or ethnic group. Such classification and prioritization allow practitioners to understand the relative importance and priority of attributes from the perspective of target shopping tourists, thereby leading to the strategic management of shopping tourism.

Another practical implication is derived from the simultaneous comparison of RIS and IA in Figure 1, wherein the asymmetric impact of each attribute is categorized with a corresponding RIS. The simultaneous observation of the RIS and IA of each attribute enables concerned practitioners to identify attributes that fall under dissatisfiers or frustrators with high RIS. Particular attention should be given to attributes that will have a tremendous impact on dissatisfaction (if not addressed adequately) because these attributes have a high RIS level, which indicates an attribute’s range of impact on satisfaction (Mikulić and Prebežac 2008). As shown in Figure 1, dissatisfiers or frustrators with high RIS at the shopping level are (1) staff knowledge of their job, (2) store cleanliness, (3) store layout, (4) availability of in-store merchandise information, (5) special deals, (6) merchandise variety, and (7) fashionable merchandise. Counterpart attributes at the destination level are (1) local transportation, (2) shopping induced by destinations featured in K-pop, (3) tax refund system, and (4) safety and security.

This study categorizes shopping tourism attributes as must-be, hybrid, and value-added ones and prioritizes them thereafter. The classification and prioritization of the attributes provide Korean Tourism Organization (KTO) in particular with an insight into the strategic management of shopping tourism for Chinese tourists, on whom Korea heavily depends. For example, to remain as an attractive shopping tourism destination to Chinese tourists, KTO can assess the performance of shopping tourism attributes identified in this study through periodic surveys. If the survey results indicate poor performance of value-added attributes when KTO aims to make shopping tourists delighted by offering extra value, KTO can make a strategic decision on more investment in value-added attributes to enhance their quality and thus shopping destination satisfaction. In case the low performance of must-be attributes is identified through the survey, KTO can consider an efficient investment to improve the quality of must-be attributes that just meets but does not significantly exceed shopping tourist expectation. The substantial investment in must-be attributes is not necessary because they do not induce satisfaction even if their quality is exceeding (as tourists take them for granted).

Furthermore, KTO could extend its “Korea Quality” certified tourism service scheme to retail shops (it is currently limited to accommodation and restaurants), based on the classification and prioritization of shopping-specific attributes identified in this study. Certification can be obtained after a document review and an on-site evaluation and is valid for three years once a retail shop is accredited. For an on-site evaluation, KTO can prioritize attributes by assigning a different weight to them and assess their performance accordingly. This certification would establish overall quality control over Korea Quality–accredited retail shops and ensure that tourists feel secure about their shopping experience. Shopping tourism attributes possess differing nature in responding to satisfaction. When DMOs identify the attributes and have a good understanding of their different sensitivity to satisfaction, the DMOs are better able to monitor and assess the performance of their destination, thereby ensuring shopping destination satisfaction in a strategic and effective manner.

Limitations and Future Research

This study has a number of limitations. The findings of this research may not be generalized to other ethnic shopping tourists given that this study is based on the perspective of Mainland Chinese shopping tourists. To minimize this limitation, a cross-cultural study will be conducted to identify the distinct perceptions of shopping tourism attributes. In addition, this study adopts cross-sectional data with self-report measures. Although carryover effects can be prevented with cross-sectional data, cross-sectional research can account for an event(s) only during a particular time slot. This limitation poses a bias, that is, a cross-sectional study will generate varying findings when a different time slot is selected (Bland 2001). Future research can explore the moderating role of K-pop in shopping tourism considering that all K-pop-related attributes are found to significantly impact shopping tourist satisfaction as must-be attributes.