Abstract
Anthropomorphism refers to the assignment of human-like traits and characteristics to non-human objects. In the tourism sector, destinations can be anthropomorphized. In this investigation, we explore potential travelers’ power as a factor that predicts to whom destination anthropomorphism appeals. Because anthropomorphism facilitates individuals’ perceived control, which individuals low on power lack but want to regain, we hypothesize in the current research that destination anthropomorphism would appeal more to travelers low (vs. high) on power. The higher perceived control manifests in travel behavior in the form of fewer perceived risks while traveling. We observe that powerless (vs. powerful) participants perceive greater control (Experiment 2) and fewer risks while traveling (Experiment 3) when the destination is anthropomorphized, explaining their higher travel intentions (Experiments 1–3). Our results suggest that whether or not to use anthropomorphism in destination advertising needs to consider potential travelers’ power.
“Paris is like a woman who, when attacked, will surrender her body in order to save her heart and soul.”—Bernard Dadié
In 2004, the city council of Amsterdam, Netherlands, installed a sign, “I Amsterdam,” to celebrate the city’s citizens in all their diversity. This sign gave human attribution to the city and was a successful example of city rebranding through a short and powerful slogan. Ten years later, the city council removed the sign due to a significant surge in tourism, with over 6,000 tourists taking selfie pictures each day (Hitti 2018). This highlights just one of the many benefits of destination anthropomorphism. Indeed, many destinations are anthropomorphized, with places, cities, and even countries instilled with a personality and/or an identity (Brown and Campelo 2014; Hultman et al. 2015). For instance, people refer to cities or countries as “she” rather than “it”; people also say they are visiting “Mother Russia,” “Mother Bulgaria,” or the “motherland.” Likewise, tourism agencies use anthropomorphized characters to promote destinations, such as Australia’s Ballarat Wildlife Park using a “wombassador”. In these examples, destination marketers convey or promote their destination using human-like features, even though countries, cities, and attractions are non-human objects. This body of work builds upon prior research focusing on destination brand love (Amaro, Barroco, and Antunes 2020), destination image (Tse and Tung 2022), place attachment (Prayag and Ryan 2012), destination loyalty (Stylos and Bellou 2019), and destination personality (Bekk, Spörrle, and Kruse 2016), all treating a destination as a “person” in some way.
Prior research has examined whether destination anthropomorphism is beneficial in attracting potential tourists. Kwak et al. (2020), most notably, found that anthropomorphizing culturally-distant destinations reduces travelers’ travel intentions to those destinations, with cultural distance defined in terms of how similar or different the destination culture is to one’s own (e.g., a Japanese tourist would consider France to be culturally-distant). The authors thus identified cultural distance as a potential moderator for the effects of destination advertising on travel intentions. Letheren, Martin, and Jin (2017) found that consumers’ anthropomorphic tendency and destination personification increase travel intentions due to positive emotions and destination attitude. Finally, Han, Wang, and Li (2020) explored the anthropomorphism of tourism brands, whereas Murphy, Gretzel, and Pesonen (2019) investigated the anthropomorphism of robot services in the hospitality and tourism sectors.
But what’s missing from the existing literature is: to whom does destination anthropomorphism appeal? That is, rather than when or under what circumstances might imbuing a city with human-like features be appealing, there may be some segments of potential travelers that may be more or less receptive to destination anthropomorphism. Our research question thus speaks to segmentation—that is, segmenting the market and identify the target market that destination anthropomorphism appeals to. Indeed, not all tourists are the same but are rather heterogeneous (Dolnicar 2020). One traveler can have very distinct tastes and preferences compared to another. There have been segmentations of the market for lodging (Brey et al. 2007), cultural tourism (McKercher et al. 2002), and sustainable tourism (Dolnicar and Leisch 2008). But, to our knowledge, there has been little examination of market segmentation as it relates to destination anthropomorphism, motivating our current effort. The primary efforts, though, have focused on consumers high versus low on “anthropomorphic tendency.” For example, personifying destination marketing communication improves destination attitude and arouses positive emotions, thus increasing travel intentions, among tourists with high levels of anthropomorphic tendency (Letheren, Martin, and Jin 2017). Similarly, Han, Wang, and Li (2020) showed that tourists who are high on entity theory are more likely to book for a travel agency trip that anthropomorphizes their brand. Similarly, consumers with higher anthropomorphic tendencies showed favorable reactions toward AI-curved trip advisors (Martin et al. 2020). But, it seems worthwhile to examine other individual factors or traits that would lead some travelers, while not others, to be receptive to destination anthropomorphism.
The current research aims to address this gap and answer the question by exploring the role of travelers’ psychological power (Magee and Galinsky 2008) in their reactions toward destination anthropomorphism. Psychological power, or simply power, is perhaps one of the key differences between human beings. Some people have more or less power than others (e.g., boss vs. employee; teacher vs. student). Power is also important in tourism, as tourism has often been described as the “hedonistic face of neo-colonialism” (Crick 1989, 322), with power asymmetries evident in interactions between tourists, tourism operators, and local community members (Bramwell and Meyer 2007; Wearing, Wearing, and McDonald 2010). Because power is so pervasive in tourism and influences tourism decisions, its possibility in affecting how tourists respond toward destination anthropomorphism captured our attention. To the best of our knowledge, this is the first study examining the impact of power on destination anthropomorphism. We hypothesize that destination anthropomorphism increases travel intentions primarily among potential travelers who are low (vs. high) on power. Our posited explanation is that travelers, who are low on power, perceive control while traveling in an anthropomorphized destination, manifested as perceived lower risks when traveling, thus increasing travel intentions. We test our hypotheses in three laboratory experiments, allowing us to determine causation.
Consequently, our work extends an understanding of travelers’ receptivity to destination advertising that includes an anthropomorphized component (Kwak et al. 2020), expanding the understanding of whom destination anthropomorphism appeals to (or not). We conceptualize power at the individual level (Wong, Newton, and Newton 2014) in contrast to the majority of work involving power in the tourism literature that has explored it in terms of asymmetry in interactions between tourists, tourism operators, and local community members (Bramwell and Meyer 2007; Ryan 2002; Wearing, Wearing, and McDonald 2010). We will expand on such conceptualization of power in tourism later on. Our work also has implications in practice for destination advertisers by allowing them to predict who, or which segment of potential travelers, destination anthropomorphism could appeal to. We discuss all of these contributions later on.
Hypothesis Development
We first discuss the construct of anthropomorphism and how anthropomorphizing a destination such as a city or country influences perceived control. We then discuss how travelers’ power might moderate their receptivity toward destination anthropomorphism.
Anthropomorphism and Perceived Control
There is no single accepted definition of anthropomorphism. But, almost all scholars investigating the matter tend to agree that it involves the assignment of human-like attributes toward non-human objects (Aggarwal and McGill 2007; Chan 2020; Chi, Gursoy, and Chi 2022). For example, people might see clouds in the shape of a sheep or see the face of Jesus on their toast in the morning. In marketing, firms often anthropomorphize brands and products as a method to appeal to consumers and increase sales; consider “Tony the Tiger” or “Michelin Man” (Herak, Kervyn, and Thomson 2020; Kervyn, Fiske, and Malone 2012; MacInnis and Folkes 2017; Puzakova, Kwak, and Rocereto 2013). Anthropomorphism is common in the tourism sector as well, evident by our opening examples with Amsterdam and Australia. Consider also the Japanese region of Kumamoto Prefecture that has used an anthropomorphized mascot “Kumamon” since 2010 to promote tourism, as this anthropomorphized character conveys important symbols and plays the role of spokesperson. Destinations themselves (not just mascots) are also anthropomorphized (Golossenko, Pillai, and Aroean 2020; Huang, Wong, and Wan 2019; Puzakova and Kwak 2017). Consider personifications of countries. In Australia, New South Wales was once and is known as the “Little Boy from Manly”; meanwhile, the “Finnish Maiden” is the personification of Finland. Table 1 presents a list of select findings regarding anthropomorphism in the tourism and hospitality literature. While destination anthropomorphism is among the many papers regarding anthropomorphism in the tourism context (Kwak et al. 2020), how power impacts the receptivity of destination anthropomorphism has not been explored.
Literature Review (Anthropomorphism).
The psychological literature has offered three potential benefits of anthropomorphism. Put differently, assigning human-like attributes to a non-human entity appears to fulfill at least three psychological needs (Epley, Waytz, and Cacioppo 2007; Epley et al. 2008; Waytz et al. 2010). The first benefit is that anthropomorphism facilitates understanding of non-human objects. Because human beings have a rich repertoire of knowledge about humans, anthropomorphizing non-human entities allows people to better understand non-humans. The second benefit is that anthropomorphism can satisfy belonging needs (Wang 2017). When another inanimate thing seems human, human beings are better able to connect with it; this can explain why socially-excluded people express preference for anthropomorphized products and brands (Chen, Wan, and Levy 2017).
The third and final benefit of anthropomorphism is that it facilitates perceived control (Epley et al. 2008; Kulow, Kramer, and Bentley 2021). This is most relevant to our current research. When objects are anthropomorphized, humans seem to perceive an ability to predict the objects’ behaviors and even potentially predict their thoughts. This benefit expands on the first, which says (as above) that anthropomorphism facilitates understanding. When people can better understand others, they are better able to predict the thoughts and behaviors of those others—a benefit that could potentially apply to non-human entities that are anthropomorphized and thus seen as human. Several findings indicate greater perceived control from anthropomorphism. When participants are denied control over a set of animate marbles, they attributed greater agency to those marbles than those given control of those marbles (Johnson and Barrett 2003). Thus, when people are low on perceived control, they are more likely to anthropomorphize as a means to obtain perceived control. In another study, individuals high on chronic desire for control were more likely to anthropomorphize dogs, compared to those who were low on such a desire (Epley et al. 2008). This is about a desire for control and not about perceived control per se, but since people high on the former want to feel more of the latter, they engage in humanizing non-human objects as a means to achieve such goals.
Ajzen (1991) included “control beliefs” in his Theory of Planned Behavior (TPB). In the TPB, control beliefs can either facilitate or impede a particular behavior or behavior intention. When people are high on control beliefs (i.e., high on perceived control), this increases behavior (or behavior intention); when they have low control beliefs (i.e., low on perceived control), this decreases behavior (or intention). “Control beliefs” have similarly been investigated in tourism contexts. For example, Lam and Hsu (2004, 2006) illustrated that perceived control is important in predicting potential travelers’ intention to visit a tourism destination. Namasivayam and Mount (2006) found that greater perceived control increases satisfaction, via using a panel of hotel customers, with this relationship mediated by perceived fairness.
But, what does “perceived control” mean in a tourism context? Researchers have conceptualized perceived control in a tourism context as an ability to reach a particular destination such as Hong Kong (Lam and Hsu 2004); others have similarly conceptualized it as having enough money and time to travel (Sparks 2007). Still others have conceptualized it broadly (and vaguely) as merely “the hotel helped me feel in control” (Namasivayam and Mount 2006). In the current research, we follow classic work on “control” in the psychology literature that has conceptualized it in terms of perceived risks, in that greater perceived control means decreased risk perceptions (Horswill and McKenna 1999; Nordgren, van der Pligt, and van Harreveld 2007; Strickland, Lewicki, and Katz 1966). That is, when people perceive control over their travel journey, they feel there are fewer risks, dangers, or threats during their travels. In turn, tourism has extensively studied risk perception in travelers’ decision-making (Quintal, Lee, and Soutar 2010). Scholars have studied risk perception through certain personality traits (Lepp and Gibson 2008); others suggested that risk perception is more situational and tourists’ risk-taking varies depending on their destination (Perpiña, Camprubí, and Prats 2019). At the very center of these diverging tourist-based studies on risk is that tourists experience cultural, health, social, political, and general risks in the choosing of the destination. Hence, risk perception is a key factor in considering a place to travel and visit (Perpiña, Camprubí, and Prats 2019; Reichel, Fuchs, and Uriely 2007).
Moderating Role of Power
An implication of anthropomorphism having the benefit of fostering perceived control thus is that anthropomorphism should primarily appeal to people who are low on perceived control. Indeed, classic research in psychology suggests that control is a fundamental need of human beings (Burger 1985; Deci and Ryan 2000). Consequently, people low on perceived control would want to regain their control. We suggest, then, that anthropomorphism would have particular appeal for travelers low on perceived control.
We relate this to the literature on psychological power, hypothesizing that (destination) anthropomorphism should appeal more to travelers low on power. Psychological power, or simply power, is defined as perceived asymmetric control over one’s own and another’s resources in interactional relationships (Magee and Galinsky 2008). In other words, control is fundamental to power. When people are low on psychological power, they are also low on (perceived) control, and when they are high on power, they are high on control—at least in terms of control over their own and another’s resources. Psychologically, being powerful has many benefits for and is rather advantageous in many aspects to the powerholder. For example, being powerful facilitates optimism and confidence (Anderson and Galinsky 2006), facilitates goal pursuit (Guinote 2007; Overbeck and Park 2006), and increases the self-esteem of the powerholder (Zimbardo 1974). In brief, being powerful is a pleasant state that in turn motivates the powerholder to maintain the state of power, if not even increase it (Jin, He, and Zhang 2014; Magee and Galinsky 2008), while those who are low on power are pessimistic, less confident, and have less self-esteem.
Importantly, another consequence of being powerless is low perceived control over the environment and one’s life. Indeed, the very definition of power is about perceived control, meaning that, by definition, “powerlessness” refers to low perceived control. Consequently, powerless individuals engage in various actions that help them regain mastery, control, and understanding, given that perceived control is fundamental to human beings (Burger 1985; Deci and Ryan 2000). For example, as a larger choice set (more options) is one way of regaining control, powerless people exhibit a preference for large (vs. small) choice sets (Inesi et al. 2011). Similarly, because status represents not just resources but also control, powerless individuals exhibit the desire to attain status goods, more so than the powerful, holding all else (e.g., income) equivalent (Rucker and Galinsky 2008).
In tourism contexts, “power” has also been investigated. For example, Choi et al. (2017) reported that powerless individuals are likely to post fake online reviews, while Wong, Newton, and Newton (2014) reported that powerful travelers were likely to buy tourism packages more so than powerless travelers. Table 2 presents some select findings on power in the tourism and hospitality literatures. However, the role of power in travelers—in destination anthropomorphism is absent from existing knowledge.
Literature Review (Power).
The Current Research
We now port our discussion about anthropomorphism, perceived control, and the moderating role of power into the tourism context. Formally, in the current research, we would predict that travelers low (vs. high) on power should exhibit more favorable attitudes toward destinations that are anthropomorphized. Such travelers, being low on psychological power, are low on perceived control. A destination that is anthropomorphized should help them regain this sense of perceived control, as it is one benefit of anthropomorphism. Accordingly, destination anthropomorphism should appeal more to low- (vs. high-) power travelers because such destinations foster perceptions of control. In a tourism context, this perceived control should result in lower risk perceptions while traveling. Refer to Figure 1 for a formal diagram of our conceptual model.
The conceptual framework.
Tourism has primarily studied power in the form of power asymmetries in interactions between tourists, operators, and the community (Bramwell and Meyer 2007; Cheong and Miller 2000; Choi, Choi, and Mattila 2019; Fallon 2001; Ford, Wang, and Vestal 2012; Kayat 2002; Lim 2007; Mehraliyev, Choi, and King 2021; Ryan 2002; Sin 2010; Wearing, Wearing, and McDonald 2010). For example, the discourse surrounding volunteer tourism reinforces prevailing hegemonic power structures as it presents Western ideologies as “superior” to ideologies of the host community (McGehee 2012). However, the social and cognitive psychology literature also contends that power is also psychological, meaning that power is a construct residing in people’s knowledge structures (Smith and Galinsky 2010). Consequently, though some people are by virtue of their relationships with others are more powerful (or powerless) to others (e.g., teacher-student, supervisor-employee, leader-follower), power can also be elicited based on psychological priming techniques. For example, exposure to concepts related to power or simply having one remind themselves of times when they felt powerful can increase power states (Chan and Saqib 2018; Dubois, Rucker, and Galinsky 2010; Inesi et al. 2011). While such effects of power cues on power states are temporary and ephemeral, they nonetheless are consistent with the long-term effects of power, meaning that even temporary exposure to power cues changes one’s power state and that they act and think or feel in ways that are structurally powerful (or powerless) as in interactional relationships.
Consequently, we designed three experiments in which we manipulated power in lab-controlled contexts. Using experimental-based research also answer the call for more use of experiments in tourism and hospitality research (Viglia and Dolnicar 2020), because it better allows for a proper determination of causation. To be sure, experiments offer greater internal validity, but it comes at the expense of external validity. Yet, because power is often accompanied by constructs such as “status” or even “wealth,” we chose an experimental approach to better determine the role of power—and not status or wealth or other co-varying constructs—in travelers’ receptance of destination anthropomorphism. Thus, in our General Discussion, we acknowledge limitations of our studies, yet at the same time experiments offer a powerful basis from which to better understand causal impacts.
We designed three experiments to test our hypotheses. In Experiment 1, we test our hypothesis that destination anthropomorphism would be more receptive for travelers with low (vs. high) power state. In Experiment 2, we test if perceived control explains why travelers’ power state moderates destination anthropomorphism receptivity. We also manipulate power in a more practical means that travel managers could adopt outside of the lab, thereby offering some external validity. Finally, in Experiment 3, we assess if lower perceived risks might similarly do so, thus testing the role of “perceived control” in a context more relevant to tourism. Hence, Experiment 1 tests the overarching hypotheses and Experiments 2 and 3 investigate the proposed first and second mediators as stated in Figure 1.
Experiment 1
The objective was to test our main premise that destination anthropomorphism appeals more to low- (vs. high-) power travelers, expressed as greater travel intentions. We relied on a 2 (destination: non-anthropomorphized vs. anthropomorphized) ×2 (power: low vs. high) between-participants design. Destination anthropomorphism was the first factor, with half of the participants seeing an ad for an anthropomorphized destination or not. Power was the second factor, with half of the participants randomly receiving the powerless or powerful experimental treatment. Here, we hypothesized that destination anthropomorphism would appeal more to powerless (vs. powerful) participants. Among powerful participants, there should be no difference in travel intentions whether the destination was anthropomorphized or not. We also expected a main effect of power since prior work suggests that powerful people are action-oriented (Anderson and Galinsky 2006); power should increase higher travel intentions regardless of destination anthropomorphism, consequently.
Participants
For this study, we recruited 338 young adult foreign travelers in partnership with a local hostel in Melbourne’s Central Business District. These participants indicated on a pre-screener that they had not visited Bali, Indonesia, before. In Australia, Bali is a popular destination for young adults and couples to travel to, similar to how American college students visit Mexico or the Caribbean for spring break. As we randomly assigned participants to one of our conditions, we only measured age and gender, assuming that all other demographic and/or psychometric variables would be randomized and balanced across conditions. Participants completed this study in the university’s behavioral lab.
Procedure
We firstly randomly assigned participants to either the powerless or powerful condition. Following existing work (Galinsky, Gruenfeld, and Magee 2003), participants were asked to recall a recent time when they felt either powerless or powerful; they had to write down what happened. Participants spent about 2–3 minutes on this writing task. Instructions in the powerless condition were: Please recall a particular incident in which someone else had power over you. By power, we mean a situation in which someone had control over your ability to get something you wanted, or was in a position to evaluate you. Please describe this situation in which you did not have power—what happened, how you felt, etc.
Meanwhile, instructions in the powerful condition were as follows: Please recall a particular incident in which you had power over another individual or individuals. By power, we mean a situation in which you controlled the ability of another person or persons to get something they wanted, or were in a position to evaluate those individuals. Please describe this situation in which you had power—what happened, how you felt, etc.
We then presented participants with an ad for Bali that was either anthropomorphized or not, assigned randomly (see Appendix 1). The anthropomorphized advertisement described Bali as a person and used personal instead of non-personal pronouns; the stimuli was adapted from Kwak et al. (2020). Participants indicated their intentions to visit Bali over the winter break (June–July) on three 7-point scales: “I would choose Bali for my holiday”;“Bali seems like a place I would travel for my winter break”; and “I would definitely consider Bali as a holiday spot over the winter break” (1 = Strongly Disagree, 7 = Strongly Agree). There was an attention check item that asked participants to select 2 on the 7-point scale; we excluded people who failed the attention check to ensure the quality of our data.
We finally included a power manipulation check by asking participants “How powerful do you feel at the moment?” (1 = Not at All, 7 = Very Powerful). This is a one-item manipulation check consistent with research on power (Smith et al. 2008). We also included a manipulation check for anthropomorphism by asking “How much do you consider Bali to be a destination with human-like features?” (1 = Not at All, 7 = Very Much). Participants were debriefed, dismissed, then thanked for their time. They received a small monetary compensation in exchange for their time.
Results
Data cleaning
In total, 46 failed our attention check. Our final sample size had a mean age of 19.32 years old (SD = 1.22) and there were 149 men, 143 women, and 1 non-binary participants. A sensitivity analysis revealed that a sample size of 292 could test an effect size of f = 0.19 at the alpha = 0.05 and power = 0.80 levels with df = 3 and groups = 4.
Power manipulation check
A 2 × 2 ANOVA revealed only a main effect of power manipulation (M = 4.00, SD = 1.57) versus powerless (M = 3.61, SD = 1.54) on felt power, F(1, 288) = 5.54, p = .02, d = 0.27. There were no main effect of destination anthropomorphism, F(1, 288) = 1.30, p = .25, and no interaction that crossed the two variables, F(1, 288) = 0.20, p = .65. Hence, our manipulation was successful in affecting participants’ psychological power.
Anthropomorphism manipulation check
Our anthropomorphism manipulation was also successful. A 2 × 2 ANOVA revealed only a main effect of destination anthropomorphism (M = 5.74, SD = 1.14) versus none (M = 5.03, SD = 1.78), F(1, 288) = 16.45, p < .001, d = 0.47. There were no main effect of power, F(1, 288) < 0.001, p = .98, and no interaction crossing the two independent variables, F(1, 288) = 0.62, p = .43.
Travel intentions
We averaged the three statements to produce a single index (α = 0.94), with higher scores indicating higher travel intentions. A 2 × 2 ANOVA on this index revealed no main effect of anthropomorphism, F(1, 288) = 2.53, p = .11. But, there was a main effect of power, F(1, 288) = 4.98, p = .02, d = 0.26, with powerful participants more likely to visit (M = 5.69, SD = 1.13) than powerless ones (M = 4.38, SD = 1.40). Critically, the interaction was significant, F(1, 288) = 13.11, p < .001, d = 0.42. For powerless participants, anthropomorphism increased intentions (M = 5.74, SD = 1.11) compared to no anthropomorphism (M = 4.99, SD = 1.58), F(1, 139) = 10.99, p = .001, d = 0.56. But, for powerful participants, anthropomorphism did not influence intentions (M = 5.54, SD = 1.17) versus no anthropomorphism (M = 5.84, SD = 1.07), F(1, 149) = 2.60, p = .10 (see Figure 2).
Study 1: Travel intentions as function of power and destination anthropomorphism.
Discussion
We find that destination anthropomorphism only increases travel intentions for participants low (vs. high) on power. This is consistent with our thesis. For high-power travelers, whether or not Bali was did not influence their travel intentions, consistent with our hypotheses. But, we did not include measures to understand why low- (but not high-) power travelers express higher intentions to travel for a destination endowed with human-like features. Consequently, we turned to Experiments 2 and 3 to examine perceived control and risks.
Experiment 2
The goal of Experiment 2 was to show evidence that destination anthropomorphism appeals primarily to low-power travelers since powerless individuals are low on perceived control, which destination anthropomorphism facilitates. Here, we measured perceived control, and tested our hypotheses in the form of moderated mediation.
Participants
For this study, we also recruited 292 young adult foreign travelers in partnership with a local hostel in Melbourne’s Central Business District. They had indicated on a pre-screener that they had not visited Uluru in the Northern Territories before. Uluru is a landmark of cultural significance to Indigenous Australians (Chan 2019); and a popular place to visit among both locals and foreign tourists. We ensured that the participants differed from those who took part in Experiment 1.
Procedure
The procedure was identical to Experiment 1, except that the destination was Uluru, not Bali (see Appendix 2). Moreover, all questions were responded on 9-point instead of 7-point scales. And before demographic questions, we measured perceived control by asking participants to indicate their agreement to three items: “I would like to feel more in control”;“Control is something that I lack right now”; and “I feel like I have a lot of control right now” (1 = Strongly Disagree, 9 = Strongly Agree; first two items reverse-scored). We included the attention check as Experiment 1, embedded within the perceived control items.
Results
Data cleaning
In total, 30 failed our attention check. Our final sample size had a mean age of 20.39 years old (SD = 1.93) and there were 116 men, 144 women, and 2 non-binary participants. A sensitivity analysis revealed that a sample size of 262 could test an effect size of f = 0.20 at the alpha = 0.05 and power = 0.80 levels with df = 3 and groups = 4.
Power manipulation check
Our power manipulation was successful. A 2 × 2 ANOVA revealed a main effect of our powerful (M = 4.61, SD = 1.46) versus powerless manipulation (M = 4.08, SD = 1.36), F(1, 258) = 8.39, p = .004, d = 0.36. There were no main effect of anthropomorphism, F(1, 258) = 0.59, p = .44, and no interaction, F(1, 258) = 0.12, p = .72.
Anthropomorphism manipulation check
Our anthropomorphism manipulation was successful. A 2 × 2 ANOVA revealed only a main effect of destination anthropomorphism (M = 4.11, SD = 1.44) versus none (M = 3.43, SD = 1.56), F(1, 258) = 12.58, p < .001, d = 0.44. There were no main effect of power, F(1, 258) = 0.09, p = .75, and no interaction, F(1, 258) = 0.14, p = .70.
Travel intentions
We averaged the three statements to produce a single index (α = 0.94). A 2 × 2 ANOVA revealed a main effect of destination anthropomorphism, F(1, 258) = 10.30, p = .001, d = 0.39, with anthropomorphism increasing travel intentions (M = 5.84, SD = 1.94) compared to none (M = 4.98, SD = 2.32). There was also a main effect of power, F(1, 258) = 4.04, p = .04, d = 0.25, with powerful participants more likely to visit (M = 5.68, SD = 2.09) than powerless ones (M = 4.97, SD = 2.27). Importantly, the two-way interaction was significant, F(1, 258) = 8.42, p = .004, d = 0.36. Among the powerless, anthropomorphism increased travel intentions (M = 5.97, SD = 2.36) compared to no anthropomorphism (M = 4.34, SD = 1.98), F(1, 126) = 17.66, p < .001, d = 0.74. Among the powerful, who generally scored higher, anthropomorphism did not increase travel intentions (M = 5.73, SD = 1.50) compared to no anthropomorphism of Uluru (M = 5.65, SD = 2.46), F(1, 132) = 0.05, p = .82 (see Figure 3).
Study 2: Travel intentions as function of power and destination anthropomorphism.
Perceived control
We averaged the three items to produce a single index (α = 0.91), with higher scores indicating higher perceptions of control. A 2 × 2 ANOVA revealed only a main effect of power, F(1, 258) = 5.86, p = .01, d = 0.30, with powerless participants lower on perceived control (M = 4.21, SD = 1.80) than powerful participants (M = 4.77, SD = 1.75). There was no main effect of anthropomorphism, F(1, 258) = 0.19, p = .65, and no significant two-way interaction, F(1, 258) = 0.15, p = .69.
Moderated mediation analysis
We lastly conducted a moderated mediation analysis to determine if anthropomorphism increases intentions to travel to Uluru only for those in the powerless (vs. powerful) condition since powerless participants were low on perceived control. Conceptually, our model called for a mediated moderation analysis (where the moderation effect of power is mediated by perceived control). Please refer to Muller, Judd, and Yzerbyt (2005) for a discussion about the conceptual distinction that exists between mediated moderation (moderation that is mediated) and moderated mediation (mediation that is moderated). However, we were using Preacher and Hayes’ PROCESS protocols for SPSS, and there is no PROCESS template for mediated moderation, but only one for moderated mediation. Preacher and Hayes argued that, statistically, mediated moderation is equivalent to moderated mediation, and the latter is more “interesting” and “meaningful” than the former (Hayes 2021). Consequently, we tested a moderated mediation effect with Model 15 (Preacher and Hayes 2008), with power as the independent variable, travel intentions as the dependent variable, perceived control as the mediating variable, and destination anthropomorphism as the moderating variable. The results revealed a main effect of only power on perceived control, so Model 15 but not Model 8 was appropriate. When Uluru, NT, was anthropomorphized, powerlessness increased intentions to travel through perceived control (B = 0.20, SE = 0.10, 95% CI 0.03, 0.45). When Uluru was not anthropomorphized, powerlessness did not affect participants’ travel intentions through perceived control (B = −0.02, SE = 0.06, 95% CI −0.17, 0.09).
Discussion
These findings indicate that destination anthropomorphism increases travel intentions primarily along powerless (vs. powerful) travelers. Therefore, we replicate the critical main hypotheses this time using a similar participant pool but employing a different destination, thus demonstrating generalizability of our predictions somewhat. Also, and most critically, we find evidence that this effect arises likely because low-power travelers are low on perceived control. Because this is a fundamental need that they want to restore, and because anthropomorphized destinations facilitate perceived control, this explains why destination anthropomorphism appeals more to powerless than powerful potential travelers.
Experiment 3
Experiment 2 findings are useful. But, they are limited in two ways. The first and most important limitation is that we measured “perceived control” generally and vaguely in that experiment. In a tourism context, and derived from classic literature linking control to risk perceptions, we would predict that destination anthropomorphism appeals to powerless travelers by reducing their perceived risks while traveling to the destination. Thus, in Experiment 3, we measured perceptions of risk, expecting that it would explain why power moderates the receptivity of destination anthropomorphism. The second limitation is minor conceptually but major practically. We so far manipulated power in laboratory-controlled contexts, allowing for internal validity. However, in actual practice, travel managers and marketers do not ask potential travelers to recall a time in their lives in which they felt powerful. In Experiment 3, we as such manipulated power in a way that would be feasible and realistic for tourism operators, managers, and marketers.
Participants
For Experiment 3, we recruited 350 Americans from the Qualtrics online panel. We employed such a panel in order to generalize our findings beyond young adults, whom we sampled in Experiments 1 and 2. Our participants indicated on a pre-screener that they had not visited Hong Kong, China, or Taiwan. Participants did not speak Chinese (neither Mandarin nor Cantonese dialect). Participants completed the study online.
Procedure
The procedure was largely identical to the prior studies except for four major changes. First, the destination that was anthropomorphized or not was Hong Kong (see Appendix 3), which on a pre-screener participant indicated that they had never visited before. Second, here, participants were not seeking accommodation via Airbnb, but rather a local Hong Kong-based alternative. Third, in both Experiments 1 and 2, we did not explicitly instruct participants to consider their trip (to Bali or Uluru) either as a single traveler, with a significant other, or in a group. In Experiment 3, we asked participants to imagine that they were traveling by themselves. This allowed for greater precision and less variance.
The fourth change was in our manipulation of power. Research in the marketing and advertising literatures suggest that merely a subtle exposure to the word “power” within advertising contexts may be enough to prime and increase psychological power (Rucker, Dubois, and Galinsky 2011). For example, Wang, Minton, and Zhang (2020) documented that the tagline “You’re powerful!” was enough to influence power. This is possible as power is a psychological construct (Smith and Galinsky 2010), and therefore based on psychological priming principles, exposure to the word “power” was enough to trigger associated felt power. Thus, we here manipulated power using a similar “You’re powerful!” tagline directly in the Hong Kong ad (see Appendix 3). Participants indicated their responses on 9-point scales.
Finally, we measured participants’ perceived risks with traveling to Hong Kong. They indicated their responses to three items: “I would feel safe while traveling in Hong Kong”; “Hong Kong seems like a risky place to visit”; and “The risks while traveling in Hong Kong are low” (1 = Strongly Disagree, 9 = Strongly Agree; second item reverse-scored). Participants, further, completed the same power manipulation check as before. Participants received monetary compensation for their time.
Results
Data cleaning
In total, 50 failed our attention check. Our final sample size had a mean age of 39.92 years old (SD = 13.46) and there were 135 men, 164 women, and 1 non-binary participants. A sensitivity analysis revealed that a sample size of 300 could test an effect size of f = 0.19 at the alpha = 0.05 and power = 0.80 levels with df = 3 and groups = 4.
Power manipulation check
Our power manipulation was successful. A 2 × 2 ANOVA revealed only a main effect of our powerful (M = 5.90, SD = 2.14) versus powerless manipulation (M = 4.47, SD = 2.17), F(1, 296) = 32.03, p < .001, d = 0.65. There was no main effect of destination anthropomorphism, F(1, 296) = 0.002, p = .96, and no interaction, F(1, 296) = 1.83, p = .17.
Anthropomorphism manipulation check
Our anthropomorphism manipulation was successful. A 2 × 2 ANOVA revealed only a main effect of destination anthropomorphism (M = 4.85, SD = 2.73) versus none (M = 4.40, SD = 2.57), F(1, 296) = 5.31, p = .02, d = 0.26. There were no main effect of power, F(1, 296) = 0.12, p = .72, and no interaction, F(1, 296) = 0.17, p = .67.
Travel intentions
We averaged the three items to produce a single index (α = 0.92). A 2 × 2 ANOVA revealed a main effect of destination anthropomorphism, F(1, 296) = 12.89, p < .001, d = 0.41, with anthropomorphism increasing travel intentions (M = 6.05, SD = 2.09) compared to none (M = 5.05, SD = 2.34). There was also a main effect of power, F(1, 296) = 41.12, p < .001, d = 0.74, with powerful participants more likely to visit (M = 6.36, SD = 1.92) than powerless ones (M = 4.76, SD = 2.31). Importantly, the interaction was significant, F(1, 296) = 5.56, p = .01, d = 0.27. Among the powerless, anthropomorphism increased travel intentions (M = 5.52, SD = 2.18) compared to no anthropomorphism (M = 4.09, SD = 2.22), F(1, 147) = 15.61, p < .001, d = 0.65. Among the powerful, who generally scored higher, anthropomorphism did not increase travel intentions (M = 6.49, SD = 1.92) relative to none (M = 6.20, SD = 1.93), F(1, 149) = 0.87, p = .35 (see Figure 4).
Study 3: Travel intentions as function of power and destination anthropomorphism.
Perceived risks
We averaged the three items to produce a single index (α = 0.94), with higher scores thus indicating fewer perceived risks (a safer travel experience in Hong Kong). A 2 × 2 ANOVA revealed a main effect of anthropomorphism, F(1, 296) = 4.99, p = .02, d = 0.25, with destination anthropomorphism increasing scores (M = 5.91, SD = 2.37) compared to no anthropomorphism (M = 5.21, SD = 2.56). There was also a main effect of power, F(1, 296) = 7.28, p = .007, d = 0.31, with powerful participants scoring higher (M = 5.97, SD = 2.39) than powerless ones (M = 5.17, SD = 2.52). Importantly, the interaction was significant, F(1, 296) = 6.95, p < .01, d = 0.30. Among the powerless, anthropomorphism increased scores (M = 5.90, SD = 2.45) in comparison to no anthropomorphism of Hong Kong (M = 4.53, SD = 2.43), F(1, 147) = 11.66, p = .001, d = 0.56. For the powerful, who generally scored higher, anthropomorphism did not increase scores (M = 5.92, SD = 2.31) in comparison to Hong Kong not anthropomorphized (M = 6.03, SD = 2.50), F(1, 149) = 0.08, p = .77 (see Figure 5).
Study 3: Perceived risks as function of power and destination anthropomorphism.
Moderated mediation analysis
Lastly, we tested and ran a moderated mediation analysis to determine if destination anthropomorphism increases intentions to travel via perceived risks only among the powerless (vs. powerful). We utilized Model 8 of the Preacher and Hayes (2008) bootstrapping protocols, with power as the independent variable, Hong Kong travel intentions as the dependent variable, perceived risks as the mediating variable, and destination anthropomorphism as a moderating variable. Conceptually, low-power participants’ perceived risks while traveling in Hong Kong was presumed to be influenced by destination anthropomorphism, with downstream consequences for travel intentions, thus Model 8 was appropriate, unlike Model 15 for Experiment 2. When the destination was anthropomorphized, powerlessness increased intentions to travel via perceived risks (B = 0.15, SE = 0.09, 95% CI 0.002, 0.37). But, when the destination Hong Kong was unanthropomorphized, powerlessness did not influence travel intentions via risk perceptions (B = 0.001, SE = 0.04, 95% CI −0.09, 0.09).
Discussion
These findings indicate that destination anthropomorphism reduces low-power travelers’ perceived risks while traveling. This is in-line with both the importance of perceived risks in tourism decision-making (Lepp and Gibson 2008; Perpiña, Camprubí, and Prats 2019; Quintal, Lee, and Soutar 2010) and how perceived control reduces these risks (Horswill and McKenna 1999; Nordgren, van der Pligt, and van Harreveld 2007; Strickland, Lewicki, and Katz 1966). Therefore, the results of this study indicate that destination anthropomorphism may be one means to reduce the perceived risks while traveling; however, the appeal of such perceived risks is primarily for low-power travelers.
General Discussion
Across three experiments, we find that power can moderate the receptivity toward destination anthropomorphism. Specifically, destination anthropomorphism increases travel intentions primarily for low-power rather than high-power travelers. This is because powerlessness reduces perceived control, perceived control is something fundamental to human beings and needs to be regained, and anthropomorphism facilitates perceived control. Within the tourism context, then, for low-power travelers, destination anthropomorphism increases perceived control and reduces perceived risks while traveling. We find support for the main hypotheses in Experiment 1. In Experiment 2, the effect arises because of perceived control, which we then demonstrate in Experiment 3 in the tourism context in the form of risk perceptions. We recruited Australian and American participants and we used travel destinations including Bali, Uluru, and Hong Kong to generalize our predictions.
To be sure, we are not the first to examine destination anthropomorphism. Our work is situated amongst literature on the receptivity of destination anthropomorphism (Han, Wang, and Li 2020; Letheren, Martin, and Jin 2017; Martin et al. 2020). We approach this question differently, focusing on travelers’ power. Power is, of course, studied in tourism contexts (Bramwell and Meyer 2007; Cheong and Miller 2000; Choi, Choi, and Mattila 2019; Fallon 2001; Ford, Wang, and Vestal 2012; Kayat 2002; Lim 2007; Mehraliyev, Choi, and King 2021; Ryan 2002; Sin 2010; Wearing, Wearing, and McDonald 2010), yet there is very limited work in this area by experimentally manipulating power (for some exceptions, see Wong, Newton, and Newton 2014). The benefit to doing so is the ability to derive conclusions about cause and effect. Because we manipulated power in our studies, we are able to deduce that power has a causal moderating impact on travel intentions (Experiments 1–3), perceived control (Experiment 2), and perceived risk (Experiment 3). This stands in contrast to the dominant study of power in tourism that has considered power asymmetries in interactions; but since power is a psychological construct (Magee and Galinsky 2008), it is also possible to experimentally “induce” power to study its effects, separate from potential confounds.
Therefore, our use of experiments to test our hypotheses, especially in a study on power, is a major contribution of our work. In doing so, we also study the role of “perceived control” within tourism contexts manifested as perceived risks as they relate to destinations that become anthropomorphized. This stands in contrast to the largely-vague study of “perceived control” in tourism contexts (Lam and Hsu 2004; Sparks 2007). We do not mean to suggest that prior definitions of perceived control are improper. In fact, there are likely many ways that perceived control could manifest itself in tourism (and other) contexts. But, we stick close toward the original literature in psychology connecting “perceived control” (“control beliefs”) to “perceived risks,” merely applied in a tourism context. Lastly, our findings indicate a benefit of anthropomorphizing destinations. We do so in particular by paying close attention to established findings about the benefits of anthropomorphism, again rooted in the psychology literature (Epley, Waytz, and Cacioppo 2007; Epley et al. 2008; Waytz et al. 2010). According to this literature, imbuing non-human entities with human-like qualities offers three benefits—the primary one relevant to us being greater perceived control. We apply this to a tourism context, to explain why destination anthropomorphism is appealing and to whom.
Aside from theoretical contributions, there are also practical implications of our work. Specifically, one dominant finding of our experiments is that travelers in a low-power state are more interested in travel to anthropomorphized destinations than travelers with a high-power state. This suggests that travel marketers may wish to target their ads with anthropomorphized destinations appropriately. As we stated earlier, some individuals are low or high on power by virtue of their status or position (e.g., teacher-student). But, because power is a psychological construct (Magee and Galinsky 2008), it is also possible to induce power. Experiment 3 offers a method to do so that should also be practical to travel managers. Other ways to do so are possible. For example, prior research has indicated that power can be affected via temperature and ambient scent (Madzharov, Block, and Morrin 2015), with cooler scents reducing felt power. As such, travel agencies may use cooler colors like peppermint and eucalyptus to induce low-power states when selling anthropomorphized destinations. In contrast, they can employ warmer scents like vanilla and cinnamon to induce a high-power state. In this vein, some colors like purple and gold denote power and control over the environment via authority (Aslam 2006).
Of course, because some people are lower or higher on power because of their position within society or an organization, our work suggests that destination anthropomorphism can consider potential travelers’ power in that society or community. Consider how, for example, high-power people are typically higher on social class (Dubois, Rucker, and Galinsky 2010). Therefore, travel managers can target lower social class travelers in their advertisements for anthropomorphized destinations. But, if tourism marketers choose to use slogans as we did in Experiment 3, there are other ways to do so besides what we employed in that experiment. For example, potentially, managers can impact travelers’ decisions by manipulating their sense of power through messages like “Are you powerless? Visit Prague, The Mother of Cities!”
As we showed in Experiment 3, a benefit of destination anthropomorphism (for low-power travelers) is fewer perceived risks. Because travel, especially international travel, is inherently associated with high levels of perceived risk (Reisinger and Mavondo 2006), it thus is imperative to understand how tourism managers and marketers can reduce perceived risks. Indeed, greater perceived risks while traveling reduces travel intentions and satisfaction. In our work, we show that destination anthropomorphism may be one possibility. This only involves changing the ad copy, without, for example, using direct statements to do so, which consumers may find not believable or psychologically react against. The (negative) impact of (greater) perceived risks while traveling is especially important given COVID-19, which has heightened perceived risks of travel (Khatib et al. 2020). The SARS pandemic in 2003, for example, reduced tourist entry into South Korea from 5.3 million in 2002 to 4.8 million in 2003 (Korea Tourism Organization 2004). Because the COVID-19 pandemic is still ongoing, it is still unclear the extent to which the coronavirus has decreased travel because of greater travel risks. But, it is likely uncontroversial to state that it has and will have a sizeable impact upon both the tourism and hospitality sectors. This only heightens the need to find ways to reduce such perceived risks, in order to improve willingness to travel.
That said, there are limitations to this research. First, an immediate limitation is that, in both Experiments 2 and 3, we relied on moderated mediation analyses (Models 15 and 8), rather than mediated moderation, because there is no PROCESS model for the latter (Hayes 2021; Preacher and Hayes 2008). Although, statistically, both produce equivalent co-efficients, other ways to investigate if perceived control and/or perceived risks mediate(s) the moderating effect of power would be beneficial. Second, Zhong et al. (2006) reported differences in how cultures understand and represent power. We recruited participants from different cultures, but both Australia and America are arguably more similar than they are different. It would be worthwhile to examine if our hypotheses would arise in, say, Eastern cultures.
Third, further research might wish to hone-in exactly on what perceived control and perceived risks “mean” when traveling. Even in Experiment 3 where we measure perceived risks, we do so generally by asking questions about perceived safety. But, is it physical safety? Healthy safety? Again, it seems unlikely that people would perceive control over the police response and crime rates in a foreign destination. But, perhaps they perceive they are able to protect themselves better, and not necessarily because they perceived fewer safety or security hazards while traveling. Either, though, would still be consistent with the idea of greater perceived control and fewer risks when traveling. Fourth, we recruited participants who had never visited our target destinations before, but they likely know about them. It would be interesting to see if familiarity would moderate our key effects. Following Kwak et al. (2020), who reported cultural distance as a moderator, perhaps our posited effect might only arise among an unfamiliar (or unvisited) destination, but not with ones that are familiar (or visited).
Another limitation comes from our use of experiments. As mentioned, there are many benefits to using experimental research especially in tourism (Viglia and Dolnicar 2020). The primary benefit is that it allows us to determine causation. But, it only does so to an extent. For example, in Experiments 2 and 3, while our data suggest that power has a causal effect on intentions to travel in Uluru and Hong Kong, and that it also has a causal effect on perceived control and perceived risks, respectively, the causal role(s) of perceived control and/or perceived risks on travel intentions cannot be confirmed. Conceptually, both constructs are widely established as determinants of travel intentions and many other travel outcomes, but in our specific case, their causal direction can only be implied and not confirmed. We note that we could have used “reverse mediation” techniques by swapping the roles of the mediator and the dependent variable. For example, we could have tested if low-power increases perceived control because of higher travel intentions toward anthropomorphized destinations. But, conceptually, this seems unlikely (travel intentions should be an effect, not predictor, of perceived control). Statistically, Lemmer and Gollwitzer (2017) argued that statistical reverse mediation techniques are not foolproof. Thus, the benefits of our work by using experimental technique are strong, although limitations still do exist. Another limitation of course is lack of external validity. But, in our case, we were more interested in the conceptual effect of power on travel intentions depending on whether a destination is anthropomorphized or not and we shed light on theoretical roles that power, perceived control, and perceived risks play.
In sum, we explore and find moderating impact of travelers’ psychological power in travelers’ receptivity toward destination anthropomorphism. Our work adds to literature in destination anthropomorphism and to the role of power in travel by conceptualizing power at the psychological, individual level. Our work motivates further research in these areas, permitting travel managers, travel marketers, and travel scholars to understand their interplay in the touristic arenas.
Footnotes
Appendix 1
Stimuli for Experiment 1 (Bali)
[Anthropomorphized Condition]
I am Bali, Indonesia.
Come visit and explore my heritage!
[Non-Anthropomorphized Condition]
This is Bali, Indonesia.
Come visit and explore the city’s heritage!
Appendix 2
Stimuli for Experiment 2 (Uluru)
[Anthropomorphized Condition]
I am Uluru, the Indigenous landmark.
Come visit and check out my long history!
[Non-Anthropomorphized Condition]
This is Uluru, the Indigenous landmark.
Come visit and check out the landmark’s long history!
Appendix 3
Stimuli for Experiment 3 (Hong Kong)
[Anthropomorphized/Powerful Condition]
Don’t you feel powerful right now? Travel!
I am Hong Kong, Asia’s World City.
Come visit and check out my storied culture!
[Non-Anthropomorphized/Powerless Condition]
Don’t you feel powerless right now? Travel!
This is Hong Kong, Asia’s World City.
Come visit and check out the city’s storied culture!
Acknowledgements
The authors would like to acknowledge Xinran Lehto for comments on an earlier version of the paper.
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) received no financial support for the research, authorship, and/or publication of this article.
