Abstract
Deviant tourist behavior is destructive to both the tourist experience and tourism development. Punishment and reward are two regulation approaches that are widely used to reduce deviant tourist behavior. However, few scholars have considered why punishment and reward are occasionally ineffective. Based on reinforcement sensitivity theory, this article explored the effects of punishment and reward on reducing deviant tourist behavior intention. Following the rules of a quasi-experiment design, a scenario-based survey was conducted to test hypotheses. The results showed that both punishment and reward have negative effects on deviant tourist behavior intention. However, groups with high sensitivity and groups with low sensitivity to punishment or reward exhibited significant disparities. In addition, the awareness of the impact of tourism was found to mediate the effects of punishment and reward on reducing deviant tourist behavior intention. This study contributes to a better understanding of this phenomenon for scholars and practitioners.
Keywords
Introduction
Deviant tourist behavior (DTB) is a deliberate or unconscious practice undertaken by tourists in a tourism-related context that results in damage or loss of property (Li and Chen 2017). Deviant tourist behavior is a common and universal phenomenon since tourists often ignore daily norms and regulations (Turner and Ash 1975). Behaviors such as littering, graffiti, and climbing sculptures or trees to take photos during a journey all belong to deviant tourist behavior. Regulations are widely used to reduce deviant tourist behavior among which punishment is a typical approach (e.g., Akers et al. 1979; Arvey and Ivancevich 1980). In addition to punishment, an increasing number of scenic spots have begun to use rewards to prevent deviant tourist behavior. For example, some scenic areas offer free admission tickets to tourists who collect trash and some scenic areas reward tourists who report others’ deviant tourist behavior. However, no existing literature have tested the effects of punishment and reward on deviant tourist behavior (e.g., Hughes et al. 2008; Li and Chen 2017; Thirumaran 2013; Uriely, Ram, and Malach-Pines 2011).
Reinforcement theory (Ferster and Skinner 1957) suggests that punishment may reduce undesirable behaviors (e.g., deviant tourist behavior), whereas rewards may promote favorable behaviors (e.g., obeying regulations and norms during a journey). Therefore, conventional wisdom seems to suggest that punishment and reward may reduce individuals’ unfavorable behaviors. However, the effectiveness of these methods in reducing deviant tourist behavior remains unclear for two reasons. First, tourists are not always well informed about punishment and reward regulations (e.g., Cole 2007). Second, some tourists engage in deviant tourist behavior even though they know these behaviors are not allowed because the high mobility of tourists enables them to avoid punishment easily. Third, reward does not always meet its goals (e.g., Ganzaroli, De Noni, and van Baalen 2017; Hwang and Mattila 2017). Consequently, both punishment and reward have limits in terms of their ability to reduce deviant tourist behavior.
Therefore, it remains unclear to scholars and practitioners how punishment and reward affect the reduction of deviant tourist behavior intention and the actual behavior. Do punishment and reward have similar effects on different tourists? These issues require further investigation since people with different personalities may react differently to the same stimuli (e.g., punishment or reward). Answers to these questions will help to reveal the mechanisms and limits of punishment and reward in reducing deviant tourist behavior, thereby contributing to a better understanding of deviant tourist behavior and providing guidelines for practitioners to manage this behavior. Scholars have found that behavioral intention is the most accurate prediction of behavior (e.g., Chang 1998; Fishbein and Ajzen 1975). Since deviant tourist behavior could bring improper gains to the tourists who engage in it, deviant tourist behavior intention could be highly correlated with actual deviant behavior. Therefore, based on reinforcement sensitivity theory (RST), this article explores the effects of punishment and reward on deviant tourist behavior intention. Investigating the effects of punishment and reward on deviant tourist behavior intention is of great importance to the academic understanding of this phenomenon as well as to the development of the tourism economy.
Theory Background and Hypotheses
The reinforcement sensitivity theory of personality is a neuropsychological theory that explains neurologic differences between people in response to reinforcing and punishing stimuli (Gray 1970). According to RST, different types of brain systems underlie different types of behaviors. The behavioral inhibition system (BIS) reflects people’s sensitivities to and avoidance of punishment (Gray 1991), whereas the behavioral approach system (BAS) reflects people’s sensitivities to reward and underlies their drive to seek rewards (Pickering and Gray 1999).
The BIS and BAS may both increase attention and arousal (Corr 2002; De Pascalis et al. 2004; Heponiemi et al. 2003). According to RST, the BIS is reflected in individual’s sensitivity to punishment, whereas the BAS is reflected in sensitivity to reward (Torrubia et al. 2001). Sensitivity to punishment and reward are associated with behavior regulation (McNaughton and Corr 2004).
People with different sensitivity to punishment and reward react differently to a given situation. Sensitivity to reward is positively associated with positive affect (Hundt et al. 2013) and related to job satisfaction and other measures of well-being (van der Linden et al. 2007). For example, people with higher sensitivity to reward have stronger positive responses to music (Loxton et al. 2016). Sensitivity to reward would positively affect people’s voice in the workplace (Schreurs et al. 2015) and their moral judgment (Moore, Stevens, and Conway 2011). On the contrary, people with sensitivity to punishment react more effectively to loss-framed message (Kaye, White, and Lewis 2013). Sensitivity to punishment is related to individuals’ tendencies to avoid threats (Krupic, Krizanic, and Corr 2016).
Punishment and Reward
According to RST, the BIS is activated by aversive conditioned stimuli, such as punishment, termination of rewards and innate fear (Corr 2002; Heponiemi et al. 2003). The BIS strengthens behavioral inhibition. For example, the BIS is positively related to indecision (Hardie and Wright 2013).
Punishment that results in negative outcomes affect individuals’ outcome expectancy and reduces the possibility that others will perform similar behaviors (Trevino and Ball 1992). Punishment is effective in various social settings. For example, punishment has a positive effect on procedure learning (Corr, Pickering, and Gray 1997). The punishment of perpetrators of unethical behavior conveys the importance of obeying ethical norms in social life (Blau 1964), and the punishment of unethical behaviors increases others’ evaluations of justice (Trevino and Ball 1992). Consequently, punishment reduces the likelihood of deviant behavior (Bhati and Pearce 2016). With respect to tourism, the punishment of perpetrators of deviant tourist behavior may convey the notion that disobeying regulations is not allowed, which may be useful in reducing deviant tourist behavior intention. Thus, hypothesis 1a is constructed as follows:
Hypothesis 1a: Punishment of unethical behavior has a negative effect on deviant tourist behavior intention.
Compared to the BIS, the BAS is activated by stimuli such as reward and the termination of punishment (De Pascalis et al. 2004). The BAS promotes behavioral activation. For example, the BAS is positively correlated with senior-executive performance (Hutchison, Burch, and Boxall 2013), whereas it is negatively related to indecision (Hardie and Wright 2013).
Smillie, Jackson, and Dalgleish (2006) stated that rewards might stimulate the BAS and lead to behavioral activation. Reward systems are widely used in social life and have a strong impact on individuals’ motivation and behavior (Hutter and Diehl 2011; Mattila, Hanks, and Zhang 2013). For example, loyalty reward programs are always used to enhance tourists’ loyalty (Hwang and Mattila 2017). In another research, Chung and Lee (2016) found that anticipated reward positively affected tourists’ geotag satisfaction. As one kind of reward, perceived travel benefits would increase travel behavior (Chen and Petrick 2016). Scholars have found that rewarding ethical behaviors positively affects observers’ subsequent ethical behaviors (Trevino and Youngblood 1990), leading to the occurrence of fewer unethical behaviors. In a current study, Thompson, Curran, and O’Gorman (2017) found that reward has a positive effect on diminishing volunteer tourists’ recalcitrant behavior. Rewarding ethical tourist behavior may also have the effect of promoting ethical behavior intention and reducing deviant tourist behavior intention. Thus, hypothesis 1b is constructed as follows:
Hypothesis 1b: Reward of ethical behavior has a negative effect on deviant tourist behavior intention.
Moderate Effects of Sensitivity to Punishment and Reward
Sensitivity to punishment is positively associated with negative affect and behavioral inhibition. For example, sensitivity to punishment is positively related to work stress (van der Linden, Beckers, and Taris 2007), daily depersonalization and emotional exhaustion (Schreurs et al. 2014), but negatively related to voice (Schreurs et al. 2015).
Scholars have found that sensitivity to punishment positively moderates the relationship between work demands and stress (van der Linden, Beckers, and Taris 2007). When work demands are high, individuals with high sensitivity to punishment express higher levels of stress than those with low sensitivity to punishment. Besides, sensitivity to punishment has a positive moderate effect on the positive relationship between employees’ surface acting and their daily emotional exhaustion and depersonalization (Schreurs et al. 2014). Sensitivity to punishment made people suffer more from negative affect.
Sensitivity to punishment has also been found to have similar moderate effects in people’s unethical behavior. Feil and Hasking (2008) found that punishment sensitivity has a negative moderate effect on the relationship between individual’s emotion-focused coping and their alcohol use. For people with low emotion-focused coping, when having high sensitivity to punishment, the level of their alcohol use to escape unpleasant emotions is much lower than those with low sensitivity to punishment. In another research, sensitivity to punishment was found to have negative moderate effect on the positive relationship between individual’s positive expectancies and marijuana use frequency (Simons and Arens 2007). When one’s sensitivity to punishment is high, his or her positive expectancies would lead to less marijuana use compared to others whose sensitivity to punishment is low.
Existing findings reveal that individual with high sensitivity to punishment is more inclined to avoid negative outcomes. Thus tourists with high sensitivity to punishment are more likely to be affected by punishment regulations than those with low sensitivity to punishment. To avoid being punished, these tourists may have lower levels of deviant tourist behavior intention and be less likely to engage in deviant tourist behavior. Thus, hypothesis 2a is constructed as follows:
Hypothesis 2a: Tourists’ sensitivity to punishment has a positive moderating effect on the negative relationship between punishment and deviant tourist behavior intention. For tourists with high sensitivity to punishment, the negative relationship will be stronger.
In contrast, sensitivity to reward also influences people’s moral judgment and behavior (Moore, Stevens, and Conway 2011). Taylor and Eitle (2015) found that the BAS-drive would negatively moderate the positive relationship between lifetime traumas and criminal activities. In another study, scholars found that sensitivity to reward had a positive moderate effect on the relationship between job insecurity and employee voice (Schreurs et al. 2015). In the high job insecurity situations, employees with high sensitivity to reward engage in much more voice compared to those with low sensitivity to reward.
Kaye, White, and Lewis (2013) found that individuals with high sensitivity to reward would receive more gain-framed message. People with high sensitivity to reward are more likely to be affected by positive incentives and to engage in moral behavior. Tourists with high sensitivity to reward may be more likely to react to rewards than those with low sensitivity to reward. These individuals may be more likely to obey regulations to obtain a reward and have a lower level of deviant tourist behavior intention. Thus, hypothesis 2b is constructed as follows:
Hypothesis 2b: Tourists’ sensitivity to reward has a positive moderating effect on the negative relationship between reward and deviant tourist behavior intention. For tourists with high sensitivity to reward, the negative relationship will be stronger.
Tourism Impact Awareness
Deviant tourist behavior has a negative impact on tourism and can result in damage or loss of property. Research has found that awareness of specific situations or consequences may influence people’s choice of behaviors (Schor 1998; Dickinson et al. 2013). For example, awareness of environmental consequences strongly predicts people’s environmental conservation behaviors (Zhang et al. 2016), and awareness of environmental problems may promote pro-environmental behavior (Bamberg and Moser 2007). Researchers have found that awareness of raw water pollution is helpful in promoting environmental protection (Charrière and Aumond 2016; Story and Forsyth 2008). Similarly, managers’ environmental awareness is positively correlated with corporate environmental behavior (Andreu, Casado-Diaz, and Mattila 2015).
Similar results have been found in tourism research. Awareness of the positive consequences of environmental protection has a positive effect on tourists’ daily environmental conservation behaviors and concern for the environmental protection of tourist sites (Zhang et al. 2016). Scholars have found that sustainability awareness is the basis of sustainable tourism. Teaching sustainable tourism ideas may increase tourists’ awareness and environmental protection behavior (Krider et al. 2010), whereas a lack of awareness negatively influences individuals’ environmental behaviors (Jamal and Stronza 2009). Ignorance and a lack of awareness about environmental problems are factors that underlie tourists’ inappropriate behavior (Mihalic 2016; Wiernik, Ones, and Dilchert 2013).
Tourism impact awareness refers to the extent of tourist’s awareness that tourism has an impact on environment, society and economy. Given that individual’s awareness of specific results could affect his/her behavior. Tourism impact awareness may also influence tourists’ deviant behavior. Tourists with high tourism impact awareness may reduce their deviant tourist behavior intention more than those with lower tourism impact awareness. Punishment and reward are measures that may be used to reduce deviant tourist behavior intention. Both punishment and reward may arouse tourists’ awareness of their impact on tourism, suggesting that punishment and reward have the effect of reducing deviant tourist behavior intention by increasing tourists’ tourism impact awareness. In other words, tourism impact awareness has a mediating effect on the relationship between punishment/reward and deviant tourist behavior intention. Thus, hypothesis 3 is constructed as follows:
Hypothesis 3: Tourists’ tourism impact awareness will mediate the relationship between punishment and reward and deviant tourist behavior intention. Tourists’ tourism impact awareness is positively correlated with punishment and reward but negatively correlated with deviant tourist behavior intention.
Figure 1 presents the framework of our model of the effects of punishment and reward on deviant tourist behavior intention.

The model of the effects of punishment and reward on deviant tourist behavior intention.
Research Method
Procedure
To test the hypotheses, following the rules of a quasi-experiment design, a scenario-based survey was conducted. Based on the process of convenient sampling, the participants were recruited in the Nanjing Confucius Temple Scenic Area, a famous scenic area in China, during the May Day festival of China in 2016. Since it is one of the most popular scenic areas in China, it is therefore easy to find a large number of samples during a festival. The quasi-experiment was conducted at the rest area where the tourists were more likely to join the survey and had enough time to complete the questionnaire patiently. Tourists were invited to participant in the survey and given a sealed envelope with a random scenario setting (punishment/reward/no statement) and questionnaire respectively. Besides individual tourist, as for the group tourists, we only invited one or two tourists from the same party to join the survey to avoid their communication with each other and interfere with the independent answer. The purpose of the survey was stated before it was administered. Each participant was given a souvenir worth approximately ¥3 (approximately $0.45).
Two scenarios were designed to test the effects of punishment and reward. In the punishment scenario, before completing the questionnaire, the participants were asked to read the following statement: “Currently, China is one of the most important tourist markets. However, deviant tourist behavior is still a problem that cannot be ignored. Last April, the National Tourism Administration issued the Regulations of Deviant Tourist Behavior Management. Six kinds of deviant behaviors were listed in the regulations.” The Regulations of Deviant Tourist Behavior Management is the first rule in China that defined the specific deviant tourist behavior and its corresponding penalty. For instance, deviant behaviors such as polluting the environment, vandalizing public facilities, or breaking the customs of destination would be charged to tourists’ personal credit record. By September 2016, 20 tourists who broke the regulations have been punished and announced to the public. Then, the scenario could be effective to convey the purpose of punishment in deviant tourist behavior management.
In the reward scenario, the participants were asked to read the following statement before completing the questionnaire: “Last February, a ‘trash bank’ was opened in Hailuo Valley in Sichuan Province. During a visit to the valley, you can collect a bag of trash and deliver it to the scenic office, and they will record it in your trash bank account. After collecting ten bags, you will have free access to the scenic area for the following five years.” Littering is one kind of common deviant tourist behavior in China. No effective way was found in reducing littering. A “trash bank” is novel and profitable for tourists. Then, the scenario statement could be effective to convey the purpose of rewarding ethical behaviors to reduce deviant tourist behavior.
In addition, a control group was used to test the effects of punishment and reward. In the control group, participants read no statements before completing the questionnaire. Since Participants were randomly shown a scenario statement and questionnaire (punishment, reward, or control). Participants who filled out the corresponding scenario questionnaire belonged to the same group (e.g., those who filled out the reward scenario questionnaire belonged to the reward group). For each group, 200 participants were recruited. SPSS 21.0 was used to analyze the data. Samples with missing data were deleted, and 562 samples with full data were used for data analysis.
Samples
The demographics for each group are presented in Table 1. In the control group, 183 questionnaires were completed. Of the participants in the control group, 98 (53.55%) were male, 75 (40.98%) were 21–30 years of age, 94 (51.37%) were 31–40 years of age, 11 (6.01%) were 41–50 years of age, and 3 (1.64%) were older than 50 years of age. Three (1.64%) of the participants in the control group had no work experience, 11 (6.01%) had 1–2 years, 39 (21.31%) had 3–5 years, 62 (33.88%) had 6–10 years, and 68 (37.16%) had more than 10 years of work experience. In terms of monthly income among the control group participants, 7 (3.83%) earned less than ¥2,000, 65 (35.51%) earned between ¥2,001 and ¥5,000, 76 (41.53%) earned between ¥5,001 and ¥10,000, 31 (16.94%) earned between ¥10,001 and ¥20,000, and 4 (2.19%) earned more than ¥20,000.
Participants Demographics.
In the punishment group, 188 questionnaires were completed. Of the participants in the punishment group, 106 (56.38%) were male, 1 (0.53%) was younger than 20 years, 86 (45.75%) were 21–30 years of age, 96 (51.06%) were 31–40 years of age, 4 (2.13%) were 41–50 years of age, and 1 (0.53%) was older than 50 years of age. Nine (4.79%) of the participants in the punishment group had no work experience, 26 (13.83%) had 1–2 years, 59 (31.38%) had 3–5 years, 64 (34.04%) had 6–10 years, and 30 (15.96%) had more than 10 years of work experience. In terms of monthly income among the punishment group participants, 14 (7.45%) earned less than ¥2,000, 59 (31.38%) earned between ¥2001 and ¥5,000, 80 (42.55%) earned between ¥5,001 and ¥10,000, 21 (11.17%) earned between ¥10,001 and ¥20,000, and 14 (7.45%) earned more than ¥20,000.
In the reward group, 191 questionnaires were completed. Of the participants in the reward group, 90 (47.12%) were male, 1 participant (0.52%) was under 20 years of age, 82 (42.93%) were 21–30 years of age, 100 (52.36%) were 31–40 years of age, and 8 (4.19%) were 41–50 years of age. Three (1.57%) of the participants had no work experience, 21 (10.99%) had 1–2 years, 59 (30.89%) had 3–5 years, 67 (35.08%) had 6–10 years, and 41 (21.47%) had more than 10 years of work experience. In terms of monthly income among the reward group participants, 2 (1.05%) earned less than ¥2,000, 61 (31.94%) earned between ¥2,001 and ¥5,000, 85 (44.50%) earned between ¥5,001 and ¥10,000, 35 (18.32%) earned between ¥10,001 and ¥20,000, and 8 (4.19%) earned more than ¥20,000. As showed in Table 1, the percentages of demographic characteristics concerning gender, age, work experience, and monthly income of the three groups do not show great dissimilarity, with the difference generally no more than 10%. Demographic similarity among the three groups indicates the samples are applicable for experimental analysis.
Variables
Tourism impact awareness
The participants’ tourism impact awareness was measured using seven items from Ap and Crompton’s (1998) tourism impact scale. These items reflect the negative impact of tourism on the local environment, society, and economy (e.g., Tourism will increase litter; Tourism will lead to a phony folk culture). The respondents indicated the extent of their agreement with each item from 1 (strongly disagree) to 7 (strongly agree). Reliability was calculated using the data collected in the research. Through Cronbach’s alpha test, the reliability was 0.93. The tourism impact awareness score was calculated based on the mean value of the items.
Sensitivity to punishment and reward
The participants’ sensitivity to punishment and reward was measured using Torrubia et al.’s (2001) Sensitivity to Punishment and Sensitivity to Reward Questionnaire, which consists of 48 items. The participants replied with yes or no responses to the items (e.g., Are you often afraid of new or unexpected situations? Do you often do things to be praised?). The sensitivity to punishment and reward scores were calculated based on the sum of the number of yes responses.
Deviant tourist behavior intention
Three items from Li and Chen’s (2017) deviant tourist behavior scale were used to measure deviant tourist behavior intention. Littering, feeding animals in the zoo, and climbing on a sculpture to take a photograph from the scale were selected as measures of deviant tourist behavior intention for two reasons. First, these behaviors occur frequently during travel; thus, the participants might have personal experience with these behaviors. Second, these behaviors are not overly destructive; thus, the participants might be less likely to hide their real intention to gain social approval. Serious deviant tourist behavior (e.g., carving on a historical relic, stealing something for a souvenir) may be more readily viewed as unacceptable by society. Participants may simply choose strongly disagree when responding to such behaviors, rendering these behaviors unsuitable for the purpose of this research. Therefore, the three kinds of behaviors selected for this research were not only representative but also appropriate for measuring the participants’ deviant tourist behavior intention. The participants were asked to rate the extent to which they would engage in such behaviors (e.g., I would litter during the journey for convenience) on a scale of 1 (strongly disagree) to 7 (strongly agree). Reliability was calculated through Cronbach’s alpha test. The reliability was 0.87. Deviant tourist behavior intention score was calculated based on the mean value of the items.
Control variables
The participants’ gender, age, work experience, and monthly income were collected for use as control variables. Social desirability was also designated as a control variable to measure the respondents’ need for social approval, which could affect the results. It was measured using Reynolds’s (1982) short form of the Marlowe–Crowne Social Desirability Scale. The reliability was calculated using the collected data through Cronbach’s alpha test (reliability was 0.88). The social desirability score was calculated based on the mean value of the items.
The classical scales used in this research have been used by many scholars and proved to be effective in unethical behavior research (e.g., Choi and Sirakaya 2005; Davis et al. 2007). Reliability and validity of the scale have been proved in prior research. All the scales were translated into Chinese using Brislin’s (1980) multistage translation/back-translation procedure.
Results
Scenario manipulation assessment
The participants’ tourism impact awareness was used to test whether the scenario was effective. According to the results of independent-sample T test, the participants’ tourism impact awareness was significantly higher in the punishment group (mean=3.91, SD=1.22) (F=5.64, p<0.05; t=−5.39, p<0.01) than in the control group (mean=3.29, SD=1.01). In addition, tourism impact awareness was also significantly higher (F=9.42, p<0.01; t=−3.15, p<0.01) in the reward group (mean=3.67, SD=1.29) than in the control group. This indicates that the scenarios used in this research were effective.
Sensitivity to punishment and reward grouping
To test the hypotheses, the participants were sorted into high and low groups according to their sensitivity to punishment and reward. Those whose sensitivity to punishment was higher than the mean were categorized into the high sensitivity to punishment group. The remaining participants were categorized into the low sensitivity to punishment group. The categorization of the participants into groups based on sensitivity to reward followed the same procedure. Regarding the participants’ sensitivity to punishment, according to the results of independent-sample T test, the high group (mean=17.88, SD=3.54, n=297) was significantly higher (F=1.00, p>0.05; t=33.92, p<0.01) than the low group (mean=7.42, SD=3.76, n=265). In terms of the participants’ sensitivity to reward, according to the results of the independent-sample T test, the high group (mean=17.60, SD=3.09, n=337) was significantly higher (F=0.82, p>0.05; t=31.78, p<0.01) than the low group (mean=8.61, SD=3.56, n=225). These findings indicate that the high and low grouping was effective.
The means, standard deviations, and correlations of the variables in different groups are shown in Table 2. Univariate general linear models were used to test the hypotheses. The effects of punishment and reward were tested separately.
Means and Standard Deviations of Correlations of Variables in Different Groups.
Note: *p<0.05, **p<0.01, ***p<0.001.
Effects of punishment
To test the effects of punishment on deviant tourist behavior intention, the control group was viewed as the low punishment context and the punishment group was viewed as the high punishment context. Table 3 presents the between-subjects effects of punishment and sensitivity to punishment on tourism impact awareness and deviant tourist behavior intention.
Tests of Between-Subjects Effects of Punishment and Sensitivity to Punishment on Tourism Impact Awareness and Deviant Tourist Behavior Intention.
Note: *p < 0.05, **p < 0.01, ***p < 0.001.
As shown in step 1 of Table 3, punishment has a main effect on deviant tourist behavior intention (F=9.10, p<0.01). According to the results of independent-sample T tests, the deviant tourist behavior intention of the punishment group (mean=3.32, SD=1.16) was significantly lower (F=0.02, p>0.05; t=4.44, p<0.001) than that of the control group (mean=3.85, SD=1.14). This finding indicates that punishment has a negative effect on deviant tourist behavior intention. Hypothesis 1a is thus supported.
In addition, the interaction of punishment and sensitivity to punishment had a significant effect on deviant tourist behavior intention. This finding suggests that tourists’ sensitivity to punishment has a moderating effect on the relationship between punishment and deviant tourist behavior intention. Figure 2 shows the estimated marginal means of deviant tourist behavior intention for the punishment and control groups. The covariates appearing in the model were evaluated at the following values: gender=1.45, age=2.62, work experience=3.70, income=2.79, social desirability=4.32. As shown in Figure 2, for the control group, there was no significant difference between the participants with high and low sensitivity to punishment in terms of deviant tourist behavior intention. However, for the punishment group, the participants with high sensitivity to punishment exhibited significantly lower levels of deviant tourist behavior intention than did the participants with low sensitivity to punishment. This finding indicates that sensitivity to punishment has a positive moderating effect on the negative relationship between punishment and deviant tourist behavior intention. Hypothesis 2a is thus supported.

Estimated marginal means of deviant tourist behavior intention for the punishment and control groups.
Two steps were taken to test the mediating effect of tourism impact awareness. As shown in step 2 of Table 3, punishment has a significant main effect (F=14.30, p<0.001) on tourism impact awareness. According to the results of independent-sample T tests, the tourism impact awareness of the punishment group (mean=3.91, SD=1.22) was significantly higher (F=5.64, p<0.01; t=−5.38, p<0.001) than that of the control group (mean=3.29, SD=1.00). This finding suggests that tourism impact awareness is positively correlated with punishment. In step 3, tourism impact awareness was added as a covariate in the univariate general linear model of deviant tourist behavior intention. As shown in step 3 of Table 3, tourism impact awareness has a main effect on deviant tourist behavior intention (F=318.98, p<0.001). However, the effect of punishment on deviant tourist behavior intention was insignificant in contrast to the results for step 1. In addition, tourism impact awareness was negatively correlated with deviant tourist behavior intention. Thus, tourism impact awareness has a mediating effect on the relationship between punishment and deviant tourist behavior intention. Hypothesis 3 is therefore partially supported.
Effects of reward
To test the effects of reward on deviant tourist behavior intention, the control group was viewed as the low reward context and the reward group was viewed as the high reward context. Table 4 presents the between-subjects effects of reward and sensitivity to reward on tourism impact awareness and deviant tourist behavior intention.
Tests of Between-Subjects Effects of Reward and Sensitivity to Reward on Tourism Impact Awareness and Deviant Tourist Behavior Intention.
Note: *p<0.05, **p<0.01, ***p<0.001.
As shown in step 1 of Table 4, reward has a main effect on deviant tourist behavior intention (F=6.32, p<0.05). According to the results of independent-sample T tests, the deviant tourist behavior intention of the reward group (mean=3.53, SD=1.16) was significantly lower (F=0.17, p>0.05; t=2.69, p<0.01) than that of the control group. This finding indicates that reward has a negative effect on deviant tourist behavior intention. Hypothesis 1b is thus supported.
In addition, the interaction of reward and sensitivity to reward has a significant effect on deviant tourist behavior intention (F=4.57, p<0.05). This finding indicates that sensitivity to reward has a moderating effect on the relationship between reward and deviant tourist behavior intention. Figure 3 shows the estimated marginal means of deviant tourist behavior intention for the reward and control groups. Covariates appearing in the model were evaluated at the following values: gender=1.50, age=2.64, work experience=3.81, income=2.86, social desirability=4.25. As shown in Figure 3, for tourists with low sensitivity to reward, there was no significant difference between the control and reward groups in terms of deviant tourist behavior intention. However, for tourists with high sensitivity to reward, those in the reward group exhibited significantly lower levels of deviant tourist behavior intention than did the tourists in the control group. This finding indicates that sensitivity to reward has a positive moderating effect on the negative relationship between reward and deviant tourist behavior intention. Hypothesis 2b is thus supported.

Estimated marginal means of deviant tourist behavior intention for the reward and control groups.
Two steps were also taken to test the mediating effect of tourism impact awareness. As shown in step 2 of Table 4, reward has a significant main effect (F=7.08, p<0.01) on tourism impact awareness. According to the results of independent-sample T tests, the tourism impact awareness of the reward group (mean=3.67, SD=1.29) was significantly higher (F=9.42, p<0.01; t=−3.14, p<0.01) than that of the control group. This finding indicates that tourism impact awareness is positively correlated with reward. In step 3 of Table 4, tourism impact awareness was added as a covariate in the univariate general linear model of deviant tourist behavior intention. As shown in step 3, tourism impact awareness has a main effect on deviant tourist behavior intention (F=145.89, p<0.001). However, the effect of reward on deviant tourist behavior intention was insignificant in contrast to the results for step 1. Thus, tourism impact awareness has a mediating effect on the relationship between reward and deviant tourist behavior intention. Hypothesis 3 is thus supported.
Discussion
According to the findings, punishment is an effective way to reduce deviant tourist behavior intention. When punishment regulations were used, the tourists who participated in this study exhibited significantly lower levels of deviant tourist behavior intention. Punishment was more effective in reducing deviant tourist behavior intention among tourists whose sensitivity to punishment was higher; however, punishment was not effective in reducing deviant tourist behavior intention among tourists with low sensitivity to punishment. This finding indicates that punishment is not always a reliable approach to deviant tourist behavior management. More strategies are needed to effectively manage deviant tourist behavior.
Reward is also an effective way to reduce deviant tourist behavior intention. When reward was provided, the tourists exhibited significantly lower levels of deviant tourist behavior intention. However, similar to punishment, reward also has limitations in terms of reducing deviant tourist behavior intention. Rewards were more effective for tourists with high sensitivity to reward than for those with low sensitivity to reward.
The tourists’ sensitivity to punishment had no significant effect on deviant tourist behavior intention when reward was used, and sensitivity to reward had no significant effect on deviant tourist behavior intention when punishment was used. This finding indicates that the effects of tourists’ sensitivity to punishment and reward on reducing deviant tourist behavior intention are independent. For tourists with high sensitivity to punishment, punishment regulations may be a more effective means of reducing their deviant tourist behavior intention. However, for tourists with high sensitivity to reward, rewards may be more effective in reducing their deviant tourist behavior intention. Thus, punishment and reward are complementary rather than redundant strategies in deviant tourist behavior management. It is necessary to use both punishment and reward in deviant tourist behavior management because both strategies have limitations in terms of their effects.
Both punishment and reward may lead to higher tourism impact awareness. According to the findings, when considering tourism impact awareness in the model, the effect of punishment or reward has no main effect on deviant tourist behavior intention. This finding indicates that tourism impact awareness is not only a consequence of punishment and reward but also plays a mediating role in deviant tourist behavior management. Increasing tourism impact awareness may ultimately reduce deviant tourist behavior intention.
Furthermore, tourists’ sensitivity to punishment has a moderating effect on the relationship between punishment and tourism impact awareness (see step 2 of Table 3; the interaction of punishment and sensitivity to punishment is significant). In addition, tourists’ sensitivity to reward has a moderating effect on the relationship between reward and tourism impact awareness (see step 2 of Table 4; the interaction of reward and sensitivity to reward is significant). Punishment is more effective in arousing tourism impact awareness among tourists with high sensitivity to punishment, whereas reward is more effective in arousing tourism impact awareness among tourists with high sensitivity to reward. This finding provides supplementary evidence that tourism impact awareness is a mediator in the relationship between punishment/reward and deviant tourist behavior intention.
Conclusion
Based on reinforcement sensitivity theory, this article explored the effects of punishment and reward regulations on deviant tourist behavior intention. By introducing BIS and BAS personality differences into the model, the study found that both punishment and reward had limitations in terms of their effects. Tourists’ sensitivity to punishment and sensitivity to reward had independent effects on reducing deviant tourist behavior intention. Since intention is an effective indicator of behavior, the findings provide a theoretical model to explain why punishment and reward are not always effective in reducing deviant tourist behavior. This research contributes to a better understanding of this important but neglected phenomenon and highlights the importance of adopting effective strategies to reduce destructive behavior.
Theoretical Contributions
This research explored the mechanisms of punishment and reward in deviant tourist behavior management. Several kinds of prevalent and representative deviant tourist behavior were chosen in this study. However, findings of this study could be extended to other kinds of deviant tourist behavior. For example, serious deviant behaviors such as carving on a cultural relic or gambling not just break the tourism regulations but also break the laws. Tourists would be punished by laws for conducting such kinds of serious deviant tourist behavior. The effects of law on serious deviant tourist behavior is equivalent to the punishment scenario used in this study. Therefore, the findings could be generalized to a large extend.
Although the findings do not seem to be unexpected, such judgment is more of a post hoc justification. Few scholars had inspected the phenomenon, the mechanisms underlying the antecedents and outcomes of deviant tourist behavior requires a number of exploratory studies. In many developing countries, regulations are not sound. The rapid development of global tourism has brought lots of adverse impacts. Of which, deviant tourist behavior is one cause. Findings of this study provide one explanation of individuals’ different reactions to punishments and rewards toward deviant tourist behavior intention based on RST. This study aims at attracting more scholars’ attention on this destructive phenomenon. It contributes to a better understanding of the phenomenon in several ways.
First, the study developed and tested a theoretical model of the use of punishment and reward to reduce deviant tourist behavior intention. Deviant tourist behavior is a common and destructive phenomenon that has received little scholarly attention. The model provides an explanation of the mechanisms of punishment and reward in deviant tourist behavior management. The model shows how and when deviant tourist behavior intention may be reduced via punishment and reward, which contributes to a better understanding of deviant tourist behavior and its relationship with approaches to the management of deviant tourist behavior.
Second, based on reinforcement sensitivity theory, the study explored the limitations of punishment and reward in deviant tourist behavior management. Using RST, the study introduced tourists’ personality into deviant tourist behavior management. The findings explain why regulations are not always effective in preventing misbehavior. Tourism is a special environment in which tourists can easily evade punishment after breaking rules. In addition, tourists are more likely to forego rewards since engaging in the behaviors required to obtain rewards may reduce their valuable tourism time. Individuals’ sensitivity to punishment and reward help to explain these phenomena. This finding also calls for scholarly attention to the reconsideration of prior empirical conclusions.
Third, the research introduced the process of a quasi-experiment method. Experiments and quasi-experiments are widely used in both natural and social science. They are utilized to explore the causal relationship between variables. The study followed the standard process of a quasi-experiment design and data analysis. Introducing the process of an experiment research may provide an effective tool for future tourism research.
Finally, the study introduced tourism impact awareness into the management of deviant tourist behavior. As Turner and Ash (1975) noted, tourists may ignore daily norms during tourism, and a lack of awareness of the impact of tourism may lead to deviant tourist behavior. Tourism impact awareness plays a bridging role in the use of punishment and reward to reduce deviant tourist behavior intention. It helps to build a connection between tourists’ reaction (reducing deviant tourist behavior intention) and reinforcement (punishment/reward).
Practical Contributions
Scholars found that tourists’ self-consciousness could not guarantee their ethical behavior (Dolnicar, Knezevic Cvelbar, and Grün 2017). They suggested that external management tools be used to change tourists’ behavior. Findings of this research provide feasible guidelines for practitioners to assist in reducing destructive behavior intention.
First, practitioners should use both punishments and rewards to reduce deviant tourist behavior intention. Although conventional wisdom suggests that the use of either punishment or reward is helpful in reducing deviant tourist behavior, based on RST, this study determined that punishment is not an effective approach for tourists with low sensitivity to punishment. In addition, reward cannot achieve its goal of reducing deviant tourist behavior among tourists with low sensitivity to reward. Consequently, a combination of punishment, reward, and other approaches may be effective in reducing deviant tourist behavior.
Second, increasing tourists’ tourism impact awareness provides another approach to reduce deviant tourist behavior intention. Governments and practitioners should pay more attention to tourist education (e.g., responsible tourism, the harmfulness of deviant tourist behavior) since these regulations are currently not well publicized and tourists are not well informed about them (Cole 2007). As scholars have found, tourists’ lack of awareness of regulations may lead to undesirable behaviors (Bhati and Pearce 2016). This type of education may increase tourists’ awareness of the impact of tourism. As the findings showed, tourism impact awareness is negatively correlated with deviant tourist behavior intention. Therefore, increasing tourists’ tourism impact awareness may be helpful in reducing deviant tourist behavior.
Limitations and Future Research Directions
Although this article explored the effects of punishment and reward on deviant tourist behavior intention, there are still several limitations to this research. First, the study tested only the effects of punishment and reward; it did not test the interactive effect of punishment and reward. In a current research about online reviews, Tanford and Kim (2019) found that travelers prefer to avoid negative experience instead of seeking reward. Future research could explore how punishment and reward affect deviant tourist behavior when both approaches are used or other factors are considered, since tourists’ behavior is affected by various situation-specific factors (Line, Hanks, and Miao 2018). Second, the findings do not explain how punishment and reward may affect deviant tourist behavior intention in the long term. Future research could use time series data to explore the change of deviant tourist behavior intention or tourism impact awareness after punishment or reward regulations are issued. In addition, several questions remain unsolved on this topic and require additional attention from scholars. For example, given that tourists can easily avoid punishment because of their high mobility, how could deviant tourist behavior intention be affected if punishment or reward regulations are not executed? Besides, samples used in this research are from China. Although deviant tourist behavior is a worldwide phenomenon, tourists with different cultures may react differently to the same reward or punishment. For example, Hwang and Mattila (2017) found that cultural backgrounds affected travelers’ reactions to loyalty reward programs. This issue could also be explored in the future.
Supplemental Material
Scales_used_in_the_research-_online_supp-2018-8-31 – Supplemental material for Do Regulations Always Work? The Moderate Effects of Reinforcement Sensitivity on Deviant Tourist Behavior Intention
Supplemental material, Scales_used_in_the_research-_online_supp-2018-8-31 for Do Regulations Always Work? The Moderate Effects of Reinforcement Sensitivity on Deviant Tourist Behavior Intention by Tao Li and Yun Chen in Journal of Travel Research
Footnotes
Acknowledgements
The authors would like to thank the editor and the anonymous reviewers for their constructive suggestions. The authors also express their gratitude for the participants in this study.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Grant of Humanities and Social Sciences Research Projects of Ministry of Education of China (No. 18YJC630076), Philosophy and Social Science Research Project of Jiangsu Provincial Department of Education (No. 2015SJB401), PhD Programs Foundation of Jiangsu Normal University (No. 16XWR033; 17XLW008), National Natural Science Foundation of China (No. 41571131) and National Social Science Foundation of China (No. 14BGL085).
Supplemental Material
Supplemental material for this article is available online.
Author Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
