The Behavioral Ecology of Sex Tourism: The Consequences of Skewed Sex Ratios

Introduction

As of 2019, the world population is short of 87 million women in the reproductive age, and this gap will widen to 117 million in 2050 (United Nations 2019). This shortage is the consequence of a higher female fetal mortality rate (Orzack et al. 2015), as well as sex-selective abortions of females in parts of the world (Zeng and Hesketh 2016). The result is a surplus of men in the reproductive-age generations, labeled male-skewed operational sex ratio (OSR). The local ratio of men to women can vary substantially from country to country, and region to region. For example, countries such as China and India or regions in former East Germany (Leibert 2016) have significantly more men than women in the reproductive age. By 2006, China had a surplus of 40 million single men (Chan, Eric, and Chan 2006). Going beyond mere demographical studies, more recent research documents that skewed sex ratios influence various psychological processes and behaviors of humans. This includes mating behavior and family life (Hesketh and Zhu 2006) but also aggravates various problems in society, such as inter-male aggression and crime (Hudson and Den Boer 2002), or sexual violence against women (Diamond-Smith and Rudolph 2018).

Research on animal (Emlen and Oring 1977) and human behavior (Buss and Schmitt 1993; Guttentag and Secord 1983) shows that skewed sex ratios increase competition for mates among members of the more prevalent sex. In response, the sex facing greater competition engages, often subconsciously, in adaptive strategies and behaviors that help to counter the unfavorable state. These behaviors are referred to as ecological plasticities (Sng et al. 2018). Integrating a biological perspective into tourism, this research suggests that sex tourism is such an ecological plasticity, fueled by a local lack of potential female mating partners. This contention derives from animal behavior research that refers to this coping mechanism as natal group dispersal (Pusey 1987). Similar to the young male lion that ventures into new territory to find a mating partner it struggles to find in its natal group, it is suggested that men, when faced with an unfavorable OSR (meaning more men than women) show a higher tendency to seek a mating partner abroad, consequently indicating more positive predispositions toward sex tourism. Hence, this study sets out to explain the phenomenon of sex tourism as a behavioral adaptation to a skewed OSR in the tourist’s home country.

This study draws on both archival data and experimentally gathered data to test the effects of skewed local sex ratios on sex tourism attitudes. As for archival data, the study draws on statistical census data on operational sex ratios in different regions in conjunction with corresponding local Google search-engine results on sex tourism. This correlational study examines a relationship between local sex ratios and people’s interest in sex tourism. Then, in study 2, this effect is investigated in an experimental examination that manipulates sex ratio, thereby providing empirical evidence for a causal relationship of skewed sex ratios on sex tourism attitudes.

The present research generates relevant implications for academics, tourism managers, policy-makers and tourists themselves. While the majority of research on sex tourism focuses on describing the phenomenon and its facets (e.g., Oppermann 1999), this study asks why sex tourism exists in the first place and shows that it is rooted in and fueled by hitherto concealed social ecologies. Specifically, this study is the first to show that the ecological dimension of sex ratio impacts predispositions toward sex tourism. Importantly, this effect is rooted in the environment the individual tourist is living in, hence providing the first examination of how the ecology of an individual drives sex tourism intentions. By doing so, this study offers a novel and creative enhancement to research that often understands sex tourism as a phenomenon driven by individual differences, and not environmental factors. As for managers and policy makers, understanding that the problem of sex tourism is potentially fueled by demographics in the source country adds a novel and overlooked layer to understanding and also countering sex tourism.

Sex Tourism as Ecological Plasticity

Researchers have devoted significant efforts to the study of sex tourism over the recent three decades. The majority of these efforts focuses on understanding and conceptualizing the phenomenon of sex tourism, thereby commonly using case studies that are analyzed qualitatively. For example, previous studies have investigated sex tourism in Thailand in the event of AIDS (Cohen 1988), the nature of sex tourism in Kenya (Omondi and Ryan 2017), or the experience of sex tourists and their relationship with locals in the Caribbean (Herold, Garcia, and DeMoya 2001; Weichselbaumer 2012). These studies, and many others, have contributed to our understanding of sex tourism and conceptualized it in a myriad of ways, including typologies and continua pertaining to geographical regions, relationships, economic and emotional facets, and gendered aspects of sex tourism (e.g., Oppermann 1999). This literature is well-founded and developed, and the present research is not intended to add to this research on the nature of sex tourism but it seeks to contribute a novel layer to the sex tourism literature by revealing an important, yet concealed, root of sex tourism. By doing so, the present research goes beyond existing efforts that suggest sexual desire, novelty-seeking, or emotional longing as motives of sex tourism (e.g., Ying and Wen 2019).

From an evolutionary perspective, such suggested motives are proximate because they do not explain why individuals seek sexual desire through sex tourism in the first place, and importantly, why some individuals do and others do not. In consequence, while informative, these proximate motives are rather intuitive and do not dig deep enough to understand the root of sex tourism. Indeed, “to this day, the overwhelming majority of tourism studies has focused on proximal motives to explain tourist behavior” (Kock, Josiassen, and Assaf 2018, p. 180). Enhancing such proximate explanations, the key tenet of evolutionary psychology is that most contemporary behaviors have an ultimate explanation rooted in our ancestral past (Mayr 1961; Tinbergen 1963). Evolutionary psychology maintains that evolutionary challenges, such as surviving and finding a mate, have predisposed our ancestors to develop fundamental mechanisms that helped to address each of these challenges (Griskevicius and Kenrick 2013; Kock, Josiassen, and Assaf 2018). The fundamental motive that this study focuses on is mate acquisition, which plays a crucial role in evolutionary considerations (Kenrick and Griskevicius 2015). Thus, an evolutionary psychologist would not conclude that human beings seek sexual encounters because it is pleasant (a proximate explanation) but concludes that mate acquisition is pleasant (thus reinforcing) because it serves a vital evolutionary purpose (ultimate explanation).

Highlighting the potential that evolutionary psychology bears for tourism, researchers have called for more evolutionary-based investigations (Crouch 2013; Kock, Josiassen, and Assaf 2018). For example, Fennell (2006), being the first to bring an evolutionary lens to tourism, explains tourist–host interactions through evolutionary motives. Specifically, he creatively uses the sociobiological theory of reciprocal altruism to explain that cooperative relationships should not exist in tourism because cooperation is based on long-term and predictable reciprocity, a requirement irreconcilable with short-term movement of tourists. More recent research by Kock, Josiassen, and Assaf (2019) examines xenophobia as an evolutionary remnant; Kim and Seo (2019) examine risk-taking from an evolutionary view; and Nørfelt, Kock, and Josiassen (2019) suggest that the ultimate motive behind travel and tourism itself is not mere pleasure but the evolutionary crucial benefits that derive from venturing out and exploring something new. Drawing on this research, and addressing Kock et al.’s (2018, p. 181) recent notion that “in many cases, the holiday serves either as a mate acquisition or mate retention vehicle, such as . . . sex tourism,” this study seeks to reveal an ultimate driver of sex tourism: the local sex ratio the individual is exposed to at home.

The local sex ratio is a socio-ecological factor, which means that it shapes the thinking, feeling, and behavior of the individual living in that environment (Uskul and Oishi 2020). Linking ecology and evolution, the key tenet of this study is that the ecological factor of sex ratio activates the fundamental evolutionary motive of mate acquisition, thereby resulting in the adaptive behavior of sex tourism. Behavioral ecology (Sng et al. 2018) examines the adoption of different behaviors and traits depending on environmental conditions. This adoption process and resulting behavior is labeled phenotypic plasticity, according to which animals and humans can adapt their behavior, physiology, morphology or development in response to different ecological dimensions (Sng et al. 2018). One ecological dimension of fundamental importance to evolution is the operational sex ratio, defined as the ratio of ready to mate females to males in a given population. The OSR is crucial because it affects the probability of attracting and retaining a mating partner, and therefore having mating success. The importance of mating partner attraction for evolution was first coined by Darwin’s seminal sexual selection theory (1871). While natural selection (often referred to as “survival of the fittest,” attributed to Herbert Spencer) stresses the adaptive functions of survival, sexual selection highlights that evolutionary survival is meaningless if the surviving genes are not transferred to the next generation through sexual reproduction. Originating in animal behavior research (Fisher 1930), the OSR determines the availability of potential mates and the level of competition for those mates (Weir, Grant, and Hutchings 2011).

As the OSR determines mate competition and thereby activates the fundamental evolutionary motive of mate acquisition, it predisposes the exposed individual to exert adaptive behaviors (i.e., showing phenotypic plasticity). While the OSR has weak effects in polygynous species in which one or a few alpha males mate with all available females (such as elephant seals or gorillas), sex ratios have strong effects in socially monogamous mating systems (such as humans). Among mammals, the reproductive success of males is more constrained by the number of mates than the reproductive success of females because males have a higher potential biological reproduction rate than females (i.e., females become pregnant while males’ investment in sexual reproduction can be minimal). This difference is formalized in parental investment theory (Trivers 1972) and implies that male-skewed OSR’s (i.e., more males than females) have particularly strong effects on males’ behavior and mating strategies because males face a higher risk of failing to attract a mating partner (Balshine-Earn 1996). Accordingly, animal behavior research documents that male-skewed OSRs increase males’ level of intrasexual competition and mating effort while female-skewed OSRs decrease males’ competition and mating efforts (Forsgren et al. 2004). For example, a lack of females predisposes male birds to invest more in paternal care (Liker, Freckleton, and Székely 2013). In summary, these considerations indicate that a lack of females predisposes males to increase efforts toward mate acquisition but not the other way round.

Homo sapiens are monogamous, and a skewed OSR should therefore shape human behavior in significant ways. Indeed, research on the relationship between sex ratio and human behavior mirror findings in the animal literature. Similar to animals, men face more difficulties in finding a female romantic partner when populations become more men-biased. This shortage of females predisposes them to engage, often subconsciously, in adaptive behaviors to increase their reproductive fitness. US regions with male-skewed sex ratios have lower divorce rates and higher paternal investment (Pedersen 1991), as well as a larger share of young adult men who are already married (Kruger and Schlemmer 2009). These findings are supported by research (Moss and Maner 2016) documenting that men become more monogamous and seek more long-term relationships in environments with less women, thus shifting to the sociosexual orientation typically favored by women. Yet, in societies that are characterized by male-skewed sex ratios, violence increases (Raffield, Price, and Collard 2017; Weir, Grant, and Hutchings 2011) and men become more aggressive toward the same sex (Moss and Maner 2016). Further, the sex ratio also shapes important behaviors that are less obviously linked to mating behavior. When women are scarce, men become more risk-taking (Ackerman, Maner, and Carpenter 2016), seek immediate gains more often (Griskevicius et al. 2012), and corroborating the theory of competitive altruism, donate more to social causes (Tognetti et al. 2016). A lack of women also causes parents to save more for their sons in order to increase their relative marriage attractiveness (Wei and Zhang 2011). These studies document that a male-skewed OSR shapes men’s behavior in important ways, often without the conscious awareness of the actor. This study thus suggests that another important, yet neglected, phenomenon driven by a male-skewed sex ratio is sex tourism.

It is hypothesized that sex tourism is a way for males to, subconsciously, cope with a local male-skewed sex ratio that they are exposed to in their home region. While some men, when confronted with a difficulty to find a mating partner due to an unfavorable sex ratio, turn toward adaptive behaviors that increase their competitiveness on the local mating market (such as increasing spending on luxury products; Griskevicius et al. 2012), others may engage in natal dispersal and thus seek a potential mate in a more distant mating market (i.e., the sex tourism destination). While existing research has examined a number of adaptive behavioral responses of men toward an unfavorable sex ratio, sex tourism as a contemporary form of natal dispersal has been neglected in human behavior research, although prevalent among animals. The present study follows up on this observation, thereby contributing to both existing tourism and psychology research.

The behavior of males leaving their own group and territory is referred to as male-biased natal dispersal. Animal research documents that male-biased natal dispersal is the norm among mammals, including primates (Johnson 1988). In contrast, most females stay in their natal group, referred to as female philopatry (Greenwood 1980). Natal dispersal reduces the risk of inbreeding (Perrin and Mazalov 1999) and increases the probability of mating for males who are not successful in finding a mate in their own local group. Thus, it is more prevalent in groups that are characterized by a male-skewed sex ratio (Drickamer and Vessey 1973). In humans, we find anecdotal evidence that a male-skewed sex ratio predisposes men to find a mating partner across borders. For example, foreign single men, who face shortages of women in their home country, search for wives in Vietnam and are willing to pay high bride prices to marry them. Yet, no study has academically investigated whether a contemporary form of natal dispersal may also exist among humans.

Given that a male-skewed OSR increases the degree of intrasexual competition for mates, this research predicts that such local OSRs predispose men to hold more favorable attitudes toward sex tourism, which is suggested to be a contemporary manifestation of the natal group dispersal phenomenon (Pusey 1987). This is because sex tourism is an adaptive compensatory behavioral strategy to cope with a male-skewed OSR that the man is facing at home. More specifically, by traveling to other countries in order to receive sex in exchange for financial resources, the man subconsciously compensates the high intrasexual mating competition at home with his pecuniary strength at the destination. Thus, the man “regains” sexual fitness through pecuniary strength at the destination, regardless of the destination’s sex ratio. Accordingly, lacking sexual fitness is best to be compensated for in poor destinations because the man’s pecuniary strength is leveraged to give him access to sexual partners, regardless of sex ratios. Thus, it is no coincidence that sex tourism mostly happens in developing countries. Importantly, sex tourism is a means for pursuing both short- and long-term romantic relationships as existing research documents (e.g., Weichselbaumer 2012; Ying and Wen 2019). As a male-skewed OSR triggers both short- and long-term mating acquisition motives among men (Moss and Maner 2016), it is expected that sex tourism addresses both of these motives. In conclusion, one would expect to observe this more positive predisposition toward sex tourism both in real-world examinations and in the laboratory, given that the OSR is skewed toward men.

Study 1: Sex Ratio Effects on Interest In Sex Tourism

The first examination of the link between sex ratios and sex tourism is a correlational study, using field data on OSR’s and interest in sex tourism. The study obtaining census data (Geis and Orth 2017) on the local operational sex ratios in the 16 German states. As for measuring interest in sex tourism, data on its prevalence is almost nonexistent because of its clandestine nature. This study therefore developed a new approach to approximate the interest in sex tourism by drawing on a large database of online search activity. Drawing on online databases is suitable when exploring sex tourism because information search and sex travel preparation often happens online, as well as sex tourism marketing and contact making before the actual travel (Hawke and Raphael 2016).

Data for a measure of interest in sex tourism was obtained from the Google Trends database. Search engine databases such as Google Trends allow researchers to track the volume of specific search terms over time and divided by region to generate insights about phenomena, such as financial market behavior (Preis, Moat, and Stanley 2013). As search engines have become a primary tool for tourists’ holiday planning (Fesenmaier et al. 2011), tourism researchers make increasing use of this rich data to predict tourist arrivals (e.g., Dergiades, Mavragani, and Pan 2018; X. Li and Law 2020). An “interest in sex tourism index” was then calculated based on the relative frequency of a pool of search terms people use on Google. This index averages the relative frequency (quantified by the number of searches relative to inhabitants in the respective IP-determined area) of sex tourism–related search terms such as “sex holiday” and “sex Thailand.” The search terms were identified through a three-step methodology: content analysis of two online forums dealing with sex tourism, interviews with four individuals recruited in those forums, and Google’s autocomplete function to identify related search terms. After scrutinizing the pool of search terms for redundancy, face and content validity, as well as data availability, this methodology yielded six search terms from which the index measure was calculated.

Study 2: The Effect of Experimentally Manipulated Sex Ratios on Sex Tourism Rationalization and Intent

Study 2 manipulated local sex ratios and examined their effects on the two sex tourism–related dependent variables sex tourism rationalization (STR) and sex tourism intent. While STR is designed to capture an individual’s justification and acceptance of sex tourism as a social phenomenon, sex tourism intent captures an individual’s personal attitude toward engagement in sex tourism while on holiday. In addition, this study examined whether age, education, relationship status, or political opinion shapes the main effect. By doing so, this study overcomes the shortcomings of study 1 and controls for potential confounders in the census data. Consistent with the author’s prediction and correlational findings from study 1, male-skewed OSRs should relate to a higher rationalization of and intention to engage in sex tourism. Hence, seeking a mating partner outside the disadvantageous ecology of too many men at home is an adaptive ecological plasticity of the individual male to reconquer reproductive fitness.

Method

Participants

Participants were recruited in the United States, using the crowdsourcing panel Mechanical Turk (MTurk). Crowdsourcing online panels offer access to a large number of comparably diverse respondents and are considered to be of at least equal data quality to street intercepts and student samples (Goodman and Paolacci 2017). Twenty-four participants were discarded because they failed the intentional manipulation check (explained below), and, following Ackerman, Maner, and Carpenter (2016), an additional 68 were discarded because they indicated a homo- or bisexual orientation, resulting in a sample size of 533.

Design and procedure

The author developed a 3 (male-skewed vs. female-skewed vs. neutral control) × 2 (male vs. female participants) between-subjects design that participants were presented with on a computer screen. The design consisted of two parts: In part 1, we participants’ sex ratio perception were manipulated, and in part 2, participants’ sex tourism rationalization and sex tourism intent were captured, as well as various demographics.

Sex ratio manipulation

In part 1, participants viewed three series of photos of people that varied in operational sex ratio (Ackerman, Maner, and Carpenter 2016; see supplemental material). Specifically, participants were shown three arrays of 18 photos each that contained headshots of either a female or male between the ages of 18 and 30. In order to minimize suspicion, the first part was introduced as a memory task by instructing the participants to count the number of females and males in each array, and to remember that number. Participants were told that the photos show students from a local (state) university campus. Post-study interviews with 11 participants in a pre-study revealed no guessing of hypotheses or suspicion. Also in the pre-study, it was verified, in support of Ackerman, Maner, and Carpenter (2016), that the photos of females and males were rated as similarly attractive.

In the male-skewed OSR condition, 12, 13, and 14 of the 18 photos showed a male; similarly, in the female-skewed condition, 12, 13, and 14 photos showed a female. In line with the cover story that has introduced part 1 as a memory game, participants viewed each array for 1 second, and were then asked on the next page to indicate how many females and males appeared in each array. Then, participants viewed the same three arrays again for 15 seconds, allegedly to allow them to check the accuracy of their memory. Then, participants were again asked to indicate the number of females and males in each array. Asking participants to indicate the number of females and males again did not only fortify the sex ratio awareness but also served as a manipulation check. Specifically, 97% counted accurately after the second viewing round. In the neutral control condition that was conducted as a post hoc test, participants viewed either an array of neutral nature photos (e.g., a field, a leaf, or clouds, among others) or were not shown any photos (thus dropping part 1). Through these two controls, it was scrutinized whether the exposure to photos of people or exposure to photos in general exerted any effect. After having finished part 1, participants continued with part 2, which included the dependent, control, and demographic variables. Part 2 consisted of dependent variables in order to analyze whether the sex ratio indeed influences predispositions that participants harbor toward sex tourism.

Measuring sex tourism rationalization

Surprisingly, no valid scale exists in the extant literature that measures tourists’ attitude toward sex tourism. Following existing research, an attitude is an individual’s overall evaluative representation of a psychological entity (Kock, Josiassen, and Assaf 2016), and thus characterized by the imputation of some degree of goodness or badness (Eagly and Chaiken 1993). Thus, an attitude captures cognitive, affective, or conative evaluations about an entity that an individual uses to derive behavior toward that entity. While a recent study measures tourists’ motivational reasons to engage in sex tourism (Ying and Wen 2019), such as sexual desire, this approach does not measure a personal attitude toward sex tourism. For example, a tourist may agree that his sexual desire can cause sex tourism yet does not evaluate sex tourism positively. Thus, this existing scale does not capture an attitude toward sex tourism as such but rather the proximate reasons for why people may engage in sex tourism. Also, the scale may be potentially subject to social desirability and response bias because of the highly sensitive nature of the topic. This is particularly true for respondents from cultures in which sex tourism and prostitution is not condoned. This study therefore turns now to developing a scale that captures tourists’ positive beliefs about, and thus a rationalizing attitude toward, sex tourism. This approach is advantageous because capturing an attitude toward sex tourism as an opinion instead of asking whether the individual would personally engage in sex tourism can significantly lower social desirability. Importantly, such a measure is a valid approximation of own behavior (i.e., someone who condones sex tourism is also more likely to personally engage in it), and therefore commonly used in seminal research that deals with sensitive sociopsychological phenomena (Cuddy, Fiske, and Glick 2007). As will be outlined later in detail, this cognitive rationalizing attitude toward sex tourism is complemented by capturing people’s conative element toward sex tourism, that is, their intention to engage in sex tourism at the holiday destination.

In order to arrive at an item pool for a person’s attitude toward sex tourism, the author consulted both online sources and existing research. First, we consulted the platforms reddit.com and debate.org, and opinions and arguments that rationalize or justify sex tourism were extracted. On these two platforms, a diverse set of users discusses topics in anonymity, including sex tourism, making it suitable to generate unbiased (i.e., not socially desirable) insights into how rationalization of sex tourism is articulated among laypersons. Second, the existing, albeit very scarce, literature on sex tourism rationalization (Bender and Furman 2004; Evans, Forsyth, and Wooddell 2000) was reviewed to generate additional items. This procedure yielded 10 items, which were included in the questionnaire.

The scale’s reliability and convergent validity was scrutinized by following established scale development procedures on the collected data set. First, an exploratory factor analysis was conducted that validated the existence of one factor through a parallel analysis (Horn 1965), thereby indicating unidimensionality of the developed measure. Then, four criteria were employed to establish the final scale: (1) items with factor loadings or item-to-total correlations of less than 0.5 were deleted; (2) redundant items were deleted, indicated by very high (>0.9) inter-item correlations; (3) items were sequentially deleted to probe whether its deletion would increase the composite reliability; and (4) an iterated χ²-difference test procedure (Kock, Josiassen, and Assaf 2019) was conducted. However, even if a statistical criterion indicated the deletion of an item, the item was retained if its deletion would had impaired the conceptual integrity and content validity of the scale (Wieland, Kock, and Josiassen 2018). This procedure left six items in the final STR scale; all items and their parameters are shown in Table 1.

OPEN IN VIEWER

Table 1.

Parameters of the Sex Tourism Rationalization (STR) Scale.

Constructs/Items	Item Mean	Item Loadings (CFA)	Corrected Item-to-Total Correlation	Scale Parameters
1. Sex tourism helps the local economy.	3.67	0.89	0.83
2. Sex tourism is a valid way of bringing wealth to a poor country.	3.35	0.88	0.83
3. Sex tourism is a win-win for both the engaging tourist and the locals.	3.29	0.89	0.85
4. Many tourist destinations would be worse off without sex tourism.	3.26	0.80	0.77
5. I would not think worse of a friend if I got to know that (s)he buys sex on holiday.	3.53	0.72	0.71
6. Sex tourism can help people who cannot find a partner at home.	3.58	0.80	0.78
Composite reliability (CR)				0.95
Average variance extracted (AVE)				0.75
Measurement model comparative fit index (CFI)				0.975

Note: The items are scored on a 7-point ordinal scale (1 = strongly disagree; 7 = strongly agree).

Other dependent and demographical variables

In addition to participants’ rationalization of sex tourism, they were asked to indicate the likelihood that they would pay for sexual encounters when holidaying abroad. This measure captures both planned and situational sex tourism intent, and is thus more encompassing than asking respondents merely about the deliberate planning of trips with the main purpose of sex tourism. While measuring actual interest in sex tourism (as quantified in study 1) or actual engagement in sex tourism is not feasible within the chosen data collection design, measuring intent serves as a proxy and allows to generate insights beyond the STR variable. In order to mitigate social desirability, the variable was framed by withholding the label “sex tourism” and measured with a one-item seven-point ordinal scale: “Imagine you were single in a foreign destination and encounter a physically attractive person of the opposite sex. It is clear to you that this person expects financial resources in return for her/his affection (which involves sexual affection). How likely would it be that you engage sexually with this person in exchange for monetary rewards (money, gifts)?” It was decided to measure the variable with a single item because it captures a unidimensional intention (and not a complex latent variable), thereby contributing to a more parsimonious questionnaire design (Dolnicar 2018).

In addition, a range of demographical and control variables was captured. First, gender, sexual orientation, and relationship status were measured. These three variables are crucial to include because they are likely to impact how the sex ratio influences dependent variables (Moss and Maner 2016). Sex ratio effects should be particularly present among heterosexual men, but not homosexual ones or females. As indicated earlier, only heterosexual participants were included in the sample. Further, the relationship status should impact the predispositions of single men more because they should be more inclined to and concerned with finding a mating partner. In addition, participants were asked for their ethnicity, political orientation, education, and age. Demographics of the sample, as well as the neutral control condition of the post hoc, are presented in Table 2.

OPEN IN VIEWER

Table 2.

Sample Characteristics of Study 2.

Sample	Study 2	Study 2 Post-hoc
Sample size	342	191
Age (%)
18–25 years	13.2	14.1
26–33 years	38.9	37.2
34–41 years	22.8	22.0
42–49 years	11.4	11.0
50 and above	13.7	15.7
Gender (%)
Female	48.0	38.2
Male	52.0	61.8
Education (%)
Finished a master’s degree or higher	14.1	19.9
Finished a bachelor’s degree	50.3	52.9
Enrolled at university	14.9	11.0
Finished secondary school	18.7	14.1
Finished primary school	2.0	2.1
Ethnicity (%)
White	74.8	79.6
Black or African American	8.5	11.0
Asian	7.6	4.7
Hispanic or Latino	7.0	4.7
Other	2.1	0
Relationship status (%)
Single	33.3
In a relationship (<3 years)	12.9
In a relationship (>3 years)	14.0
Married	39.8

In addition, a high level of data quality was ensured by including a variety of variables to control for different forms of bias: satisficing, common method bias (CMB), and social desirability. First, questionnaire-based data collections may be subject to biasing response behavior. Specifically, a few respondents may engage in satisficing (Barber, Barnes, and Carlson 2013). The use of instructional manipulation check (IMC) questions can effectively detect and alleviate such biasing response behavior (Paas, Dolnicar, and Karlsson 2018). The author therefore included one IMC question (“Please select agree as the answer here”; seven-point ordinal scale) in the questionnaire. Twenty-two respondents were dropped because they failed the IMC question. In order to deter biasing response behavior, participants were informed, before they accessed the questionnaire, that IMCs were used. Second, some biasing response behavior may be fueled by common method questionnaires, resulting in CMB (Podsakoff, MacKenzie, and Podsakoff 2012). This study has been a priori designed to alleviate CMB as it obtains the predictor variable (i.e., the experimentally manipulated sex ratio) and the criterion variables from different sources. Yet, more than one variable is collected through ordinal scales; hence, a theoretically unrelated marker variable was included (Lindell and Whitney 2001), measured on an ordinal seven-point scale: “How much do you like spicy food?” This marker variable did not significantly correlate with any of the other variables, hence indicating that CMB did not impair the data quality of this study.

Third, because of the sensitive nature of the topic, the study controlled for respondents’ tendency to provide socially desirable answers by including three items from Crowne and Marlowe’s (1960) social desirability bias scale. The included items are “I have never intensely disliked anyone”; “Before voting I thoroughly investigate the qualifications of all the candidates”; “When I don’t know something I don’t at all mind admitting it.” All variables were correlated with the social desirability items but no significant results were obtained, hence indicating that a social desirability bias was unlikely to impair the results. In addition to this post hoc testing, social desirability was mitigated a priori by ensuring participants that the study is anonymous and void of right and wrong answers.

Results

The data were analyzed according to the developed 3 (male-skewed vs. female-skewed vs. neutral control condition) × 2 (male vs. female participants) between-subjects design, and an omnibus analysis of variance (ANOVA) on the two dependent variables STR (composite reliability: 0.95; average variance extracted: 0.75) and sex tourism intent was executed. Further, discriminant validity was achieved between all constructs, indicated through the Fornell–Larcker and heterotrait–monotrait ratio heuristic, as well as through a nonsignificant χ²-difference test between the unconstrained and constrained measurement model. Then, a general linear-model ANOVA was executed to predict variance in participants’ STR through the experimental condition (male-skewed vs. female-skewed OSR). A significant main effect of the OSR, F(1, 342) = 12.537 (p < .001) was found, indicating that participants’ rationalization of sex tourism was significantly higher in the male-skewed OSR condition (mean OSR_male-skewed = 3.76, SE = 0.13, mean OSR_{female-skewed} = 3.15, SE = 0.12). It was followed up on this positive finding and tested for OSR effects within female and male participants in a second ANOVA. For females, the OSR had no effect on STR: mean OSR_male-skewed = 2.74, SE = 0.16, mean OSR_{female-skewed} = 2.67, SE = 0.15, F(1, 163) = 0.119, p = .730. Males, however, displayed a higher STR when they were subconsciously primed with an unfavorable male-skewed OSR: mean OSR_male-skewed = 4.69, SE = 0.13, mean OSR_{female-skewed} = 3.61, SE = 0.16, F(1, 177) = 26.988, p < .000. Conducting the post hoc control, there was no difference between the neutral control condition and the female-skewed OSR, neither for men, mean OSR_{female-skewed} = 3.61, SE = 0.16, mean neutral control = 3.78, SE = 0.15, F(1, 209) = 0.578, p = .448; nor for women, mean OSR_{female-skewed} = 2.67, SE = 0.15, mean neutral control = 3.00, SE = 0.20, F(1, 158) = 1.811, p = .180). This result documents that the mere exposure to young females and males does not account for variance in the dependent variables. Also, there was no difference between the neutral control condition with and without photos, documenting that the mere exposure to photos does account for variance in the dependent variables. In sum and in line with the study’s main prediction, and as depicted in Figure 1, the results of study 2 show that a male-skewed OSR led men to rationalize sex tourism more.

OPEN IN VIEWER

Figure 1.

Sex tourism rationalization as a function of female- and male-skewed operational sex ratios.

Further, another general linear-model ANOVA was conducted, with sex tourism intention as the dependent variable. The result substantiated the previous finding as males showed a significantly higher intention to engage in sex tourism when exposed to a male-skewed OSR compared to exposure to a female-skewed OSR, mean OSR_male-skewed = 3.81, SE = 0.21, mean OSR_{female-skewed} = 2.64, SE = 0.20, F(1, 177) = 16.596, p < .000. For females, the OSR had no significant effect on sex tourism intentions, mean OSR_male-skewed = 1.91, SE = 0.18, mean OSR_{female-skewed} = 1.64, SE = 0.16, F(1, 163) = 1.305, p = .255. The result is depicted in Figure 2. These experimental findings provide strong support for the author’s evolution-based prediction that a male-skewed OSR causes men, but not women, to hold more positive predispositions toward sex tourism. The findings are consistent with the correlational result in study 1, which documented that male-skewed OSRs correlate with higher search activity on sex tourism. While the secondary search data in study 1 does not reveal the gender of the searcher or relationship status, study 2 indicates that the observed effects are driven by men and caused by a male-skewed OSR.

OPEN IN VIEWER

Figure 2.

Sex tourism intent as a function of female- and male-skewed operational sex ratios.

Going beyond the main predictions, the analysis also yielded a significant main effect of participant sex on STR, mean male = 4.13, SE = 0.11, mean female = 2.70, SE = 0.11, F(1, 341) = 82.50, p < .001, indicating that men hold more positive attitudes toward sex tourism than women regardless of the manipulation. This finding supports parental-investment theory and is in line with existing research documenting that men tend to be more sexually unrestricted (Buss and Schmitt 1993; Moss and Maner 2016). No differential effects on the dependent measures were observed with age as covariate. This finding is in line with evolutionary research because male fertility does not drop significantly in age, hence supporting the effectiveness of the OSR manipulation also among older men. As for relationship status, no differences were observed between men who are single, short-term or long-term liaised, or married (even after controlling for the manipulated scenario). This result implies that men, regardless of their relationship status, subconsciously react to unfavorable sex ratios, thus highlighting the important and persistent effect of this mechanism. Furthermore, this finding again corroborates the parental investment theory tenet that men are more sexually unrestricted. In addition, no significant differences were observed for any of the control variables, that is, ethnicity, political orientation and education. This finding highlights that attitudes toward sex tourism are insensitive to demographical variables (except gender), thus, further increasing the relevance of the study’s identification of sex ratio as a determinant of sex tourism.

Conclusion

The present study reveals a novel and overlooked relationship between a local lack of women and men’s predispositions toward sex tourism. Building on biology and psychology research, it was proposed, and found, that a lack of women causes men to indicate more favorable attitudes toward sex tourism. Study 1 operationalized online search behavior for sex tourism as a proxy for sex tourism interest and correlated it with the ratio of men to women in the state in which the respective search behavior was tracked. The finding of this correlational study indicates that a local sex ratio may indeed relate to a higher interest in sex tourism. Considering the substantial correlation coefficient, the moderate significance may likely be the result of low power because of the small sample size (Wasserstein and Lazar 2016). Addressing the inevitable shortcomings of a field study design, study 2 developed and tested an experimental setup that manipulated local sex ratios. In line with this study’s conceptual reasoning, it was found that male-skewed sex ratios led men to exert more positive attitudes toward sex tourism. Specifically, it was found that men who were primed with a male-skewed sex ratio rationalized sex tourism more and had a higher intention to personally engage in sex tourism. In line with the study’s prediction, women were insensitive to varying sex ratios. This finding was obtained regardless of men’s age, relationship status, political opinion, ethnicity, or education; hence, it is deemed to be a robust, deep-rooted evolutionary mechanism at work.

This study is the first that documents a link between people’s social environment, specifically the ratio of reproductive-age women to men, and men’s predispositions toward sex tourism. Enhancing an even broader picture, this study is the first that conceptually develops and empirically documents a sensitivity of tourists toward their own social ecologies, and subsequent impact on their behavior. By considering the social ecology of an individual tourist as a determinant of behavior, instead of individual traits or attitudes, this study goes beyond current research mindsets in tourism and adds a novel approach to the examination of tourist behavior and decision making, thus adressing recent calls for discipline-bridging research (Kock, Assaf, and Tsionas 2020). As such, this study can pave the way for new original research that seeks to understand how the ecologies in which a tourist resides at home may shape her or his travel behavior abroad. Thus, the author calls for more research that sets out to examine touristic ecological plasticities, that is, behaviors in the context of tourism that are responses to ecological determinants. Specifically, this ecological approach lends itself to the examination on the coronavirus pandemic: An important ecological determinant impacted by the pandemic is the one of pathogen threat. This study calls for future research that examines the pathogen ecology both in the tourist’s home country and the destination as a key determinant of tourist behavior. Existing research in biology provides an exciting starting point for such an endeavor (e.g., Ackerman, Tybur, and Mortensen 2018).

It is important to emphasize that tourism and travel is an exceptional consumption vehicle in this context because it allows the tourist to travel between different ecologies with different determinants, thereby coping with different ecological pressures. In the current study, the first determinant is the unfavorable sex ratio at home, which results in lower reproductive fitness for the individual. This state is compensated through the second determinant, a favorable financial fitness of the individual, that materializes in sex tourism and consequently caters for a reconstitution of reproductive fitness. While this study has focused on the ecological dimension of sex ratio to investigate an arguably pressing and important phenomenon, various other ecological dimensions (such as financial or cultural scarcity, spatial density, or unpredictability to name a few) have the potential to advance our understanding too. While this research draws on evolutionary psychology, it is important to emphasize that most behaviors include both an evolutionary and cultural explanation, thus highlighting that nature and nurture should play complementary roles in research endeavors.

The study also contributes directly to existing research on sex tourism. First, this study contributes an empirical field study and experimental proof to a literature stream that is heavily relying on qualitative methods. While these often well executed studies have significantly enhanced our understanding of the nature of sex tourism, the current study empirically reveals what can cause this pressing problem, thereby going beyond descriptives. Understanding the root of a problem is important because it allows to examine how it can potentially be alleviated, if not solved. This is particularly true if the root is as prevalent and pervasive as the local sex ratio. Also, while the current study takes its point of departure in the phenomenon of sex tourism, future studies should extend the identified sex ratio effect to noncommercial sex in tourism settings. Specifically, there is no conceptual argument against the contention that also noncommercial sex, for example between tourists, is a means to cope with an unfavorable sex ratio at home. Given that noncommercial sex also plays a significant role in both tourism research and tourists’ experiences (Berdychevsky, Gibson, and Poria 2013; Uriely, Ram, and Malach-Pines 2011), the author calls for studies to enhance this research stream. In a related vein, future studies are welcome that investigate how those men who are financially constrained and thus cannot travel may react to an unfavorable OSR. Potential ecological plasticities may be investments in physical attractiveness, buying counterfeit status symbols, increased porn consumption, or increased aggression toward other men. Further, this study did not include homosexual respondents, yet future research could examine whether the sex ratio effects also hold for those tourists. Also, it is acknowledged that there could be other reasons for engaging in sex tourism. For example, men who perceive themselves to lack appeal for women (e.g., viewing themselves as physically unattractive) may exhibit a higher baseline tendency to engage in sex tourism. Future research is invited to test such other sources of sex tourism that may also potentially function as contingencies.

Further, this research constitutes a significant enhancement to existing studies on sex ratio effects, particularly because most conducted studies are correlational in nature and experimental evidence is still scarce (Moss and Maner 2016). By examining sex tourism as an ecological plasticity of skewed sex ratios, this study combines it with the biological concept of natal group dispersal. This approach can motivate future research to understand not only tourism but all human mobilities through sex ratios. For example, the decision of a young man to study in either the local town or abroad may be subconsciously shaped by perceived sex ratios that could have materialized at school. Further, this study also sheds new light on existing sex ratio research. Specifically, studies find that men become more monogamous (i.e., sexually restricted) when women are scarce (Schmitt 2005). Interestingly, this study suggests the contrary as it documents that another way to cope with a scarcity of women is to seek a mating partner abroad, hence implying that men can become indeed more polygynious. Similarly, existing studies have often documented a shift toward the relationship preferences of the scarcer sex (i.e., women; Moss and Maner 2016). In the light of the findings herein, the present study documents that this is not necessarily the case because a scarcity of women can also lead to the evasive male behavior of sex tourism. Indeed, this consequence does not imply a shift toward the relationship preferences of women but quite the contrary, only shifting the male-dominated relationship preferences to another local ecology in which not women but money is the scarce resource. In addition, while the present study has examined the effects of different sex ratios, future research could test for the effect of the absolute number of male and females. While this study held the number of people constant in the conditions to control for such an effect, it would be interesting to examine a potential interaction effect. A methodological limitation of the current study pertains to its use of data from two different countries for the two studies. Yet, and in accordance with the broad consensus in evolutionary psychology, adaptive psychological mechanisms like the examined sex ratio effect are deep-rooted in human nature and thus universal across the human species. While it is unlikely that results have been impaired by the chosen setup, future research on replicating the results in other countries is welcome.

This study has various practical implications for policy makers. First and foremost, the current study identifies a root of sex tourism, and thus how altering that root may also help alleviating the problem of sex tourism. Importantly, while a sex ratio is an objectively quantifiable variable, a lack of women is first and foremost perceptual. In consequence, policy makers could attempt to alter this perception through publicly visible communication of balanced sex ratios. However, creators of media content should be involved as well because perception and expectation is often altered by what is consumed. Thus, communicating balanced sex ratios through movies, series, or content on social media may also help to alleviate problems. While conceptual development and empirical proof for such strategies is lacking, the current study helps to spur such welcomed endeavors.

Supplemental Material