A Comparative Study of the Impact of Negative Word of Mouth on Travel Intentions of Chinese and Korean Consumers in Tourism Destinations

Abstract

This article is based on reducing the negative effect of negative word-of-mouth on consumers’ travel intentions. We constructed a model of consumers’ travel intentions to explore the influence of negative word-of-mouth about tourist destinations on consumers’ travel intentions, examine the moderating effect of destination familiarity and the mediating utility of negative emotions, and conduct a comparative analysis between China and South Korea. Based on SPSS and AMOS analysis, we found the following: (1) Negative word-of-mouth has no direct positive effect on travel intention, and negative emotion plays a completely mediating role between negative word-of-mouth and travel intention, whereas high-intensity negative emotion is positively related to travel intention. (2) Travel destination familiarity has a moderating effect on negative word-of-mouth and negative emotions. Consumers with higher familiarity make external attributions and generate high-intensity negative emotions when they receive negative word-of-mouth; consumers with lower familiarity make internal attributions and generate low-intensity negative emotions. To improve consumers’ travel intentions, companies can enhance the relationship between tourist destination brands and tourists, adopt different management strategies for negative word-of-mouth according to the level of familiarity of destinations, and formulate tourism strategies.

1. Introduction

In light of the rapid development of tourism, the upgrading of consumers’ consumption demands, and the increasing competitiveness of the tourism market, understanding the factors affecting consumers’ intentions to travel is critical to improving the competitiveness of enterprises [1]. Due to the promotion of the Internet and the connection between the tourism industry and the Internet, the Internet has become an influential method of obtaining tourism information and making tourism decisions. Nowadays, more and more people share, collect, and exchange travel information via the Internet, which not only makes the public aware of tourist destinations, but also exposes the negative sides of some tourist spots [2]. Negative word-of-mouth far outweighs positive word-of-mouth, and the spread of the Internet amplifies the negative word-of-mouth and may cause tourists to change their travel plans, which is a big blow to local tourism.

Scholars have found, however, that negative word-of-mouth does not always lead to negative evaluations and sometimes may even inspire positive evaluations of brands. Different scholars have explored different perspectives on this phenomenon [3]. Escalas and Bettman (2003) [4] and Wilson et al. (2017) [5] argue that when consumers have high self-brand associations, they deal defensively with negative word-of-mouth, ultimately increasing their intentions to purchase. Furthermore, some scholars have also explained it in terms of consumers’ positive attitudes towards brands (Ahluwalia, 2000) [6], product purchase experience (Sweeney et al., 2014) [7], and brand strength (Ho-Dac et al., 2013) [8,9].

Recently, scholars have focused on the negative effects of negative online word-of-mouth on consumers’ travel intentions. So, is there some condition under which negative online word-of-mouth positively influences consumers’ travel intentions? The purpose of this paper is to empirically investigate the positive effects of negative word-of-mouth on consumers’ travel intentions from the perspectives of destination familiarity and negative sentiment, as well as explore the moderating role of destination familiarity and the mediating role of negative sentiment, and to provide useful suggestions for formulating travel strategies for Chinese and Korean consumers [10].

2. Materials and Methods

2.1. Negative Word-of-Mouth

“Word-of-mouth” refers to what consumers say about goods and services using virtual opinion platforms and other channels. There are two types of word-of-mouth: positive and negative. Tourists’ perception formation and behavior decisions are informed by negative word-of-mouth, which conveys negative consumer perceptions of tourist places and tourist risk information [11].

Several studies concluded that negative online word of mouth can cause some negative effects, including changes in attitudes toward the brand and changes in consumer decisions. However, some scholars believe that negative word of mouth may actually have a positive effect online (Escalas and Bettman, 2005) [12,13,14].

In this paper, we combine previous research to define negative word-of-mouth as consumers’ dissatisfaction with a company’s products, brands, or services leading them to describe their unpleasant experiences in text, pictures, and other online comments [5,15].

2.2. Negative Emotions

Negative emotions are a dimension of consumer emotions, and according to consumer behavior research, things contrary to consumers’ goals often lead to negative emotions (Menon and Dubé, 2000) [16]. According to the definition of consumer emotions by Westbrook and Oliver (1991) [17] and Menon and Dubé (2000), this article defines negative emotions as a negative psychological state experienced by consumers, such as anxiety, anger, or tension [18,19]. Based on the research theme of this paper, according to the intensity of emotions, and drawing on the research results of Feng Jiao et al. (2012) and Qian Jie (2017), negative emotions are subdivided into high-intensity negative emotions and low-intensity negative emotions: ① High-intensity negative emotions include anger, irritability, and regret; ② Low-intensity negative emotions include frustration, helplessness, and worry [20,21].

2.3. Destination Familiarity

Familiarity has received more attention as an important variable influencing consumer behavior. After Cohen introduced familiarity into tourism research in 1972, many scholars have conducted studies on destination familiarity. The research topics related to familiarity mainly include research on the influence of familiarity on destination image and intention to visit, research on the influence of familiarity on motivation and experience, and research on the influence of familiarity on the attractiveness of tourist destinations. All studies found that familiarity had significant effects on other variables [22]. Milman and Pizam (1995) [23] argue that destination familiarity refers not only to tourists’ visual or psychological impressions of the destination, but also to the relevant experiences that stimulate travel intentions.

So far, studies have usually examined experience familiarity, information familiarity, self-described familiarity, and interpersonal relationships as important components of destination familiarity [24]. Based on previous studies, this study defines destination familiarity as a tourist’s knowledge and subjective perception of a place acquired through various channels such as information and personal experiences, which will influence the tourist’s perception of the place and then impact the tourist’s behavioral intentions. The destination familiarity dimension is divided into the experience familiarity dimension, the information familiarity dimension, and the proximity familiarity dimension.

2.4. Traveling Intention

Intention refers to the state of preparation and willingness before the behavior. Woodside and Lysonski (1989) [25] consider intention to travel as the likelihood that a potential tourist will go to a tourist destination. Ye Li (2008) considers intention to travel as an individual’s willingness to participate in tourism activities based on their subjective consciousness. Liu, Wenjuan (2018) believes that the intention to travel is the desire of potential tourists to travel after being stimulated by external factors and forming the judgment tendency of whether to go to a tourist destination or not. Combining the above studies, this study considers the intention to travel as the behavioral tendency of potential tourists to obtain information about tourism products and services through the Internet and thus form intentions on whether they are willing to carry out tourism activities or not [26].

A review of studies related to intention to travel reveals that the main focus is on the study of its influencing factors. Chen (2007) found that event marketing influenced the intention to travel by affecting tourists’ attitudes. Shen, Suyan and Guo, Jianying (2011) found that perceived behavioral control, past related experiences, and cultural tourism participation all impacted tourists’ intention to travel [27]. In the study of Yue, H. and Ma, A. (2021), it was confirmed that attitude, subjective norms, perceived behavioral control, and Internet word-of-mouth had a significant positive effect on travel intention. This paper will further investigate the effects of negative word-of-mouth on travel intentions.

3. Research Model and Hypotheses

3.1. Subsection

Based on the literature review, this paper constructs a model of the effect of negative word-of-mouth on intention to travel, as shown in Figure 1.

3.2. Research Hypotheses

3.2.1. Negative Word of Mouth and Negative Emotions

Negative word-of-mouth usually comes from unsatisfactory consumer experiences. When consumers are exposed to negative word-of-mouth, they are more likely to be motivated by negative emotions, so negative emotions are part of negative word-of-mouth. They can impact consumer behavior [28].

When negative reviews of a brand appear, consumers with high familiarity will automatically process the information they have about the destination, and when negative word-of-mouth conflicts with this information, it will stimulate feelings such as anger and irritation [29]. In contrast, consumers with low familiarity, due to their lack of knowledge and experience of the destination, will assume that the destination is consistent with what the negative information describes based on the available information, resulting in disappointment, disgust, etc. Therefore, the following hypothesis is proposed:

Hypothesis 1a:

When destination familiarity is high, the stronger/weaker the negative word-of-mouth, the stronger/weaker the high-intensity negative sentiment.

Hypothesis 1b:

When destination familiarity is low, the stronger/weaker the negative word-of-mouth, the stronger/weaker the low-intensity negative sentiment.

3.2.2. The Mediating Role of High/Low-Intensity Negative Emotions

Consumer behavior will change when consumers receive negative word-of-mouth followed by negative emotions. However, the intensity of negative emotions has different effects on consumer behavior [30]. Strizhakova et al. found that the angrier consumers are, the more they respond positively and expressively, rather than avoiding them. It has also been pointed out that consumers are more likely to adopt impulsive buying behavior when they are in a strong negative emotion at the time of purchase (Zhang Yunlai (2009) and Piron (1991)), and disgust, which is the opposite of anger, is an emotion with a higher avoidance motivation and is weaker in intensity than anger. Consumers are more motivated to avoid when they feel disgusted, and their purchase intention is lower. From the above literature, it can be found that emotions of different intensities cause different behaviors. Therefore, the following hypothesis is proposed:

Hypothesis 2a:

High-intensity negative emotion mediates between negative word-of-mouth and intention to travel, and high-intensity negative emotion positively influences intention to travel.

Hypothesis 2b:

Low-intensity negative emotion mediates between negative word-of-mouth and intention to travel, and low-intensity negative emotion has a negative effect on the intention to travel.

3.2.3. Moderating Role of Destination Familiarity

Most of the existing studies on negative word-of-mouth have focused on its negative effects, such as shifts in attitudes toward brands and changes in consumer decisions, which may cause public opinion crises if negative word-of-mouth is not dealt with promptly (Yang and Mai (2010), Fei, Ping-Hua (2010), and Song, Xiaobing (2011)). However, some scholars have also argued that negative word-of-mouth may have a positive impact (Escalas and Bettman (2005) and Cheng et al. (2012)), and Wilson et al. and Zhang Jemei explained the positive impact of negative word-of-mouth on consumers’ purchase intentions using self-threat and defense mechanisms from a personal brand association perspective. Jinxin Zhang and Hai Hu and Hua Huang and Haifan Mao also further enriched the study of negative online word-of-mouth on consumers’ positive behavior under brand perception. In addition, there are also some studies explaining the possible positive effects of negative word-of-mouth on consumer behavior in terms of usefulness and reliability.

Negative word-of-mouth will positively stimulate behavioral intentions, and there is an important moderating variable, that is, destination familiarity. Negative word-of-mouth has less impact on consumers with high destination familiarity than on those with low destination familiarity. When negative reviews occur for a brand with high familiarity, consumers will counteract or mitigate the negative effects of negative reviews based on their knowledge and experience of the brand, and it is more difficult to change brand attitudes. However, when negative reviews appear for unfamiliar brands, consumers’ attitudes are more likely to be affected by the negative reviews due to their lack of knowledge and experience. Similarly, tourists with higher familiarity have a higher intention to travel and are less influenced by negative information in tourism studies. Therefore, the following hypothesis is proposed:

Hypothesis 3:

Destination familiarity moderates the effect of negative word-of-mouth on travel intention.

4. Methodology

This research aims to investigate the influence of negative word-of-mouth on travel intentions and to analyze the mediating utility of negative emotions and the moderating role of destination familiarity. As shown from

Table 1. Descriptive Statistics.

Valid		CHINA (Percent)	KOREA (Percent)
Gender	Male	38	37.6
	Female	62	62.4
Age	20–30	57.8	39.5
	31–40	19.5	27.2
	41–50	7.6	16.7
	51–60	9.9	10.7
	60+	5.1	5.8
Jobs	Information and communication industry	4.2	4.9
	Service industry	24.4	25.7
	Manufacturing industry	7.9	7.5
	Trade and distribution industry	11.6	13.4
	Students	51.8	48.4

, this research topic was investigated through questionnaires from 15 June to 18 July 2021, using Chinese and Korean consumers as the research subjects. After removing invalid questionnaires, 819 valid questionnaires were returned, including 466 in Korea and 353 in China. The questionnaires, shown in

Table 2. Questionnaire summary.

Variable	Definition	Related Research
Negative Word of Mouth (NWOM)	I was impressed by the negative word-of-mouth information I see that the negative word-of-mouth message has many supporters The publisher of negative information on the Internet is familiar with information related to the tourist destination Negative word-of-mouth information on the Internet is very convincing	Bi Jidong (2010); Zhang Jiemei (2019)
High-intensity Negative Emotions (HNE)	I will feel angry I will feel annoyed I will feel irritable	Westbrook & Oliver (1991); Feng Jiao, Lv Yilin, He Qingwen (2012)
Low-intensity Negative Emotions (LNE)	I will be disappointed I will feel disgusted I will feel frustrated
Destination Familiarity (DF)	I am familiar with the tourist destination I can tell the characteristics and signs of the tourist destination I have visited this tourist destination many times My relatives/colleagues live in this tourist destination	Milman & Pizam (1995)
Travel Intention (TI)	I want to travel to this tourist destination I expect to travel to this tourist destination in the future I will often recommend or talk about this tourist destination with my friends and family I will give priority to this destination if I have travel plans	Woodside & Lysonski (1989); Liu Wenjuan (2018)

, were evaluated using a 5-point Likert scale, where “5” means strongly agree, “3” means moderate, and “1” means strongly disagree. In order to increase the reliability of the hypothesis testing, descriptive statistics, factor analysis, and reliability and validity analyses of the variables were conducted using the structural equation model analysis method. Then, based on results of the confirmatory factor analysis of the measurement models using AMOS, the measurement models were divided into centralized and discriminant validity to further enhance the understanding validity of the study.

5. Data Analysis and Results

5.1. Reliability Analysis

The reliability test of the questionnaire is the test of the reliability and credibility of the questionnaire. It is mainly based on the consistency or stability of the results obtained by the test tool and reflects the true degree of the measured data. In this study, we will analyze the method’s reliability by calculating Cronbach’s α coefficient. The α coefficient is between 0.80 and 0.90 (very good), 0.70–0.60 (good), or 0.61–0.65 (acceptable). Cronbach’s α coefficient is directly proportional to the reliability of the measurement content. That is, the larger the coefficient, the greater the reliability of the measurement content. After testing the reliability of the data, the results are shown in

Table 3. The results of the confidence analysis.

Variable	Cronbach’s Alpha (CHINA)	Cronbach’s Alpha (KOREA)	Items
NWOM	0.869	0.89	4
HNE	0.854	0.846	3
LNE	0.834	0.813	3
DF	0.84	0.828	4
TI	0.908	0.902	4

. The Cronbach’s alpha coefficients of the five variables are greater than 0.8, indicating very good reliability [31].

5.2. Validity Analysis

Validity is how the researcher measures how closely the content aligns to the actual situation. The higher the validity, the more accurately the subjects can be measured. In this study, a factor analysis was used to test the validity of the questionnaire. The results of the factor analyses are shown in

Table 4. The validity analysis for the data from China.

	1	2	3	4	5
TI2	0.84	0.134	0.114	0.248	0.156
TI3	0.827	0.18	0.097	0.213	0.158
TI1	0.807	0.133	0.133	0.232	0.139
TI4	0.797	0.18	0.131	0.225	0.173
NWOM3	0.12	0.836	0.157	0.133	0.078
NWOM4	0.191	0.825	0.171	0.11	0.047
NWOM2	0.079	0.808	0.205	0.029	0.095
NWOM1	0.167	0.773	0.178	0.043	0.059
DF4	0.136	0.139	0.828	0.039	0.102
DF2	0.071	0.215	0.801	0.002	0.094
DF1	0.066	0.183	0.789	0.046	0.039
DF3	0.126	0.137	0.754	0.122	0.032
HNE3	0.234	0.076	0.038	0.842	0.108
HNE2	0.23	0.106	0.052	0.835	0.083
HNE1	0.329	0.097	0.118	0.796	0.111
LNE2	0.166	0.067	0.061	0.102	0.867
LNE3	0.112	0.076	0.1	0.035	0.846
LNE1	0.202	0.093	0.074	0.147	0.806
KMO			0.878
Bartlett’s Test of Sphericity Approx.			Chi-Square	3459.494
			df	153
			Sig.	0.000

and

Table 5. The validity analysis for the data from Korea.

	1	2	3	4	5
TI2	0.846	0.201	0.136	0.194	0.112
TI3	0.829	0.141	0.117	0.189	0.154
TI4	0.806	0.135	0.146	0.206	0.151
TI1	0.799	0.175	0.149	0.206	0.121
NWOM4	0.162	0.882	0.054	0.121	0.029
NWOM1	0.179	0.85	0.085	0.126	0.019
NWOM3	0.088	0.847	0.07	0.037	0.002
NWOM2	0.145	0.804	0.096	0.094	0.103
DF4	0.14	0.096	0.828	0.058	0.082
DF2	0.093	0.039	0.823	0.045	0.077
DF1	0.051	0.073	0.793	0.053	0.035
DF3	0.166	0.075	0.741	0.081	0.046
HNE3	0.206	0.09	0.076	0.842	0.099
HNE2	0.2	0.129	0.05	0.827	0.099
HNE1	0.281	0.134	0.119	0.803	0.153
LNE2	0.124	0.018	0.068	0.107	0.868
LNE3	0.099	0.061	0.092	0.071	0.835
LNE1	0.186	0.046	0.059	0.134	0.794
KMO			0.871
Bartlett’s Test of Sphericity Approx.			Chi-Square	4423.229
			df	153
			Sig.	0.000

. The factor extraction of the factor rotation matrix is consistent with the factor attribution theory assumption of each measurement item. The aggravation value of each variable exceeds 0.5, indicating that the scale has good construct validity. Therefore, it can be roughly determined that the overall validity of the scale is good.

5.3. Hypothesis Test

In Baumgartner and Homburg’s study, it was demonstrated that SEM is appropriate for data analysis only when the sample size is more than five times the estimated parameters. This model has 18 estimated parameters. The data collected from China and South Korea in this study are from 353 and 466 questionnaires, respectively, which is greater than the required sample size. Consequently, the sample data meet the requirements to test structural equation models.

The variables and assumptions of this study refer to relatively mature scales and scientific research results at home and abroad, so this paper must carry out confirmatory factor analysis. Through the calculations of AMOS 23.0 software, the revised structural equation model fit index is shown in

Table 6. Fitness index.

Items	CMIN/DF	NFI	TLI	CFI	RMSEA	GFI	AGFI
Ideal value	>1, <3	>0.9	>0.9	>0.9	<0.08	>0.8	>0.8
China	1.64	0.958	0.979	0.983	0.043	0.956	0.936
Korea	2.072	0.959	0.973	0.978	0.048	0.955	0.936

. The estimated values of the seven indexes are within the range of the fitting standard, indicating that the model has a good fit with the sample data. This model can effectively explain the behavior of the users in the march.

The structural equation model analysis was performed using AMOS 23.0 software to validate the proposed research hypotheses. After analyzing the sample data using the maximum likelihood estimation method (ML), the fit indices of the model were obtained and the standardized regression coefficients (path coefficients) and significance of each path were calculated. The parameter estimation results for their structural equation models are shown in Figure 2 and Figure 3.

The paths can be analyzed for significance using AMOS software. According to the results of the path test, all the paths passed the test, and the path effects were more significant. The results of the structural Equation-AMOS Model Path Analysis are shown in

Table 7. The structural Equation-AMOS Model Path Analysis Results for the data from China.

			Estimate	S.E.	C.R.	p
HIGH	←	NWOM	0.349	0.061	5.762	***
LOW	←	NWOM	0.25	0.055	4.52	***
Intention	←	LOW	0.3	0.053	5.705	***
Intention	←	HIGH	0.532	0.051	10.336	***

Notes: *** p < 0.001.

and

Table 8. The structural Equation-AMOS Model Path Analysis Results for the data from Korea.

			Estimate	S.E.	C.R.	p
HIGH	←	NWOM	0.393	0.054	7.251	***
LOW	←	NWOM	0.158	0.05	3.149	**
INTENTION	←	LOW	0.232	0.047	4.976	***
INTENTION	←	HIGH	0.481	0.045	10.655	***

Notes: ** p < 0.01; *** p < 0.001.

.

5.4. Mediation Analysis

This paper further investigates the mediating effect of negative emotions between negative word-of-mouth and travel intention. The mediation effect was tested by the Bootstrapping confidence interval method. According to the test criteria, if the confidence interval does not contain 0 at a 95% confidence interval, the mediation path exists. The results of the tests are shown in

Table 9. The results of the Intermediary Effectiveness Test for the data from China.

Path	Intermediary Effect	BootSE	95% Confidence Interval CI		Ratio of Total Effect
			Lower	Upper
TOTAL	0.1524	0.0308	0.097	0.218	44%
HNE	0.1076	0.0263	0.0608	0.1643	31%
LNE	0.0448	0.0129	0.0217	0.0727	13%
HNE-LNE	0.0629	0.0278	0.0112	0.1209	-

and

Table 10. The results of the Intermediary Effectiveness Test for the data from Korea.

Path	Intermediary Effect	BootSE	95% Confidence Interval CI		Ratio of Total Effect
			Lower	Upper
TOTAL	0.1355	0.0242	0.0893	0.1857	36%
HNE	0.1097	0.0217	0.0685	0.1548	30%
LNE	0.0258	0.0097	0.0094	0.0473	7%
HNE-LNE	0.0839	0.0233	0.0406	0.1326	-

, which show that negative emotion mediates between negative word-of-mouth and willingness to travel. Therefore, we can judge that hypothesis 2a and hypothesis 2b are not valid.

5.5. Moderating Effect Analysis

To explore the influence of destination familiarity on consumers’ travel willingness under negative word of mouth, this article further analyzes the difference in the influence of consumers with different levels of destination familiarity on their travel intentions when facing negative word of mouth, that is, the moderating effect test. In this paper, a Model 59 in SPSS macro prepared by Hayes (2012) was used to test the moderating effect, as shown in

Table 11. An analysis of the moderating effect of familiarity.

	CHINA
	HNE			TI
	Coeff.	SE	T	Coeff	SE	T
Constant	−0.0554	0.0434	−1.2784	−0.0595	0.0372	−1.5996
NWOM	0.2239	0.0560	4.0009 ***	0.2901	0.0480	6.0403 ***
DF	0.1142	0.0499	2.2868 ***	0.1617	0.0429	3.7720 ***
NWOM × DF	0.1812	0.0542	3.3438 ***	0.1946	0.0465	4.1841 ***
R-sq	0.1068			0.2116
F	13.9145 ***			31.2250 ***
	KOREA
	HNE			TI
	Coeff.	SE	T	Coeff	SE	T
Constant	−0.0317	0.0338	−0.9374	−0.0381	0.0282	−1.3513
NWOM	0.2216	0.0486	4.5637 ***	0.2534	0.0404	6.2692 ***
DF	0.1454	0.0393	3.6993 ***	0.2054	0.0327	6.2789 ***
NWOM × DF	0.2495	0.0532	4.6873 ***	0.2993	0.0443	6.7572 ***
R-sq	0.1516			0.2806
F	27.5115 ***			60.0808 ***

Notes: *** p < 0.001.

. The analysis of Chinese data showed that the interaction term between negative online word-of-mouth and familiarity showed a significant relationship (β = 0.1812, p < 0.01), and Korean data also showed a significant relationship (β = 0.2495, p < 0.01), which indicates that there is a positive moderating effect of familiarity between negative word-of-mouth and highly negative emotions, suggesting that familiarity plays a positive moderating role between negative word-of-mouth and travel intentions. Therefore, hypothesis 1a and hypothesis 3 hold. In contrast, the moderating effect of familiarity was not supported under the low negative sentiment, and hypothesis 1b did not hold.

To reflect the moderating effect more intuitively, this study used the simple slope method for testing, the core idea of which is to add or subtract one standard deviation from the mean and perform a regression analysis at both high and low levels. The effect of negative word-of-mouth on the negative emotions at different familiarity levels was analyzed first, and the results are shown in

Table 12. The impact of negative word-of-mouth on travel willingness under different destination familiarity levels.

	DF	Effect	BootSE	BootLLCI	BootULCI
CHINA	eff1(M + 1 SD)	0.0148	0.0209	−0.0229	0.0605
	eff2 (M)	0.0971	0.0313	0.0384	0.1618
	eff3(M − 1 SD)	0.2424	0.0647	0.1248	0.3779
KOREA	eff1(M + 1 SD)	0.0014	0.0185	−0.0389	0.0358
	eff2 (M)	0.0766	0.0210	0.0365	0.1187
	eff3(M − 1SD)	0.2082	0.0465	0.1246	0.3068

. When familiarity is high, there is a positive moderating effect of familiarity between negative word-of-mouth and high negative emotion. As shown in Figure 4 and Figure 5, there is a strong positive effect of negative word-of-mouth on an increased negative effect at high familiarity with a steeper slope.

6. Discussion and Conclusions

This paper focuses on three variables—negative word of mouth, negative emotions, and destination familiarity—and explores their influence on the travel intentions of Chinese and Korean tourists. Despite the differences in national income and education patterns, the same geographical attributes, such as cultural homogeneity, result in similar outcomes. Since China and Korea belong to the same Asian circle and are located in relatively similar cultural circles, the differences that exist in this study are not significant. The following conclusions were obtained by analyzing data from both countries: (1) Negative Internet word of mouth has no direct positive influence on travel intention. Through the moderating effect and mediation effect tests, it is found that negative Internet word of mouth positively affects travel intention through high-intensity negative emotions. (2) When the destination is more familiar, negative word of mouth is positively correlated with high-intensity negative emotions. (3) Destination familiarity moderates the effect of negative word-of-mouth on travel intention.

First, negative word-of-mouth about tourist destinations will directly affect the negative emotions of travelers, thus affecting the willingness of information receivers to travel. High-intensity negative emotions play a completely mediating role in negative word-of-mouth and intention to travel. The studies of Strizhakova et al. (2012) and Piron (1991) confirm this view. Secondly, this article mainly studies familiarity, an important moderating variable. When consumers with higher familiarity receive negative word-of-mouth, consumers will make more external attributions. For example, there was inaccurate news on the Internet about the damage to the ecological reserve in China’s Guangdong Province, and after investigation, the information was found to be false. When netizens learned that the negative word of mouth was false, they generated high-intensity negative emotions, such as anger, and thus did not affect their intentions to travel to it. This is consistent with the research views of Escalas and Bettman (2005) and Cheng et al. (2012). In conclusion, negative word-of-mouth does not directly affect consumers’ willingness to travel, but rather generates negative emotions of different intensities through the moderating effect of destination familiarity, and positively influences willingness to travel through high-intensity negative emotions. Previous research mainly focused on the negative impact of negative word-of-mouth on travel intentions. This article expands the research on negative word-of-mouth, high/low negative sentiment, and travel intentions, and considers two cognitive pathways of consumers.

At present, more and more people learn about tourism information through the Internet, and negative word-of-mouth may cause tourists to change their travel plans, which has caused a huge blow to the development of local tourism. So, solving the adverse effects of negative word-of-mouth and stimulating consumers’ travel intentions will affect the development of enterprises. Through the research of this article, the following management enlightenment can be obtained:

• Actively disseminate positive information about the tourist place and promptly eliminate negative information in the network. Enterprises should form their characteristics of tourist places to enhance consumers’ familiarity with them.
• Different responses are taken depending on the level of consumer familiarity with the destination. Consumers with high familiarity will make self-judgments of negative online information, which will not impact the corporate image and will even stimulate consumers’ desire to protect the destination and can reverse the negative online image to a certain extent. For consumers with low familiarity, enterprises should try their best to reverse the negative online image and transform the negative word-of-mouth; otherwise, there will be a public opinion crisis.
• Companies also need to pay attention to the negative emotions of consumers. Companies can use negative emotions to design publicity plans when conducting online publicity, which will make it easier for consumers to resonate.

The purpose of this paper is to study the mediating role of negative emotions, but this study only classifies negative emotions into high- and low-intensity and does not break down negative emotions. China and Korea are representative countries in Asia, and because of the author’s limited energy, only samples from China and Korea were selected for analysis in this paper. In addition, most of the questionnaire samples are drawn from student populations, and the distribution of occupations and educational backgrounds is quite uneven. The high proportion of students in the sample will make the model biased. The relatively high percentage of students in this study (around 50%) may bias the results of the study due to the uniqueness of the student population. The samples will be equalized by SMOTE (Synthetic Minority Oversampling Technique) in subsequent research. Future studies need to expand the sample to cover a wider range of occupational and educational backgrounds.