Introduction

Research has identified the increased health risk behaviors that young people engage in when holidaying abroad, including substance use, violence, and risky sexual activity (Blay et al., 2010; Hughes et al., 2008, 2009; Rogstad, 2004). Some researchers have called this phenomenon behavioral inversion, whereby certain meanings, morals, or norms of ordinary behavior are changed, held in abeyance or even reversed during holiday periods (Graburn, 1983). This can contribute to the suppression of personal limits and favor excessive use of alcohol and drugs and risky sexual activity (de Oliveira Santos & Paiva, 2007; Lomba, Apóstolo, & Mendes, 2009). Thus, previous research has shown that casual sexual encounters in youth holiday resorts can be commonplace and mediated through substance use (Downing et al., 2011). However, less is known about young people’s experiences of more aggressive sexual encounters during holidays abroad, including sexual harassment (SH) and having sex against their will (SAW).

To date much research on SH and SAW among young people has focused on college students, largely in the U.S. (e.g., Abbey, Zawacki, Buck, Clinton, & McAuslan, 2004). For example, one study found that a fifth of female college students reported unwanted sexual contact in the past six months, compared with 8 % of males (Banyard et al., 2007). Another study found that a third of both female and male undergraduates reported having been coerced into sexual activity, while a third of males and 17 % of females reported having perpetrated sexual coercion themselves (Fair & Vanyur, 2011). Thus, SH and SAW can be perpetrated by, and inflicted on, individuals of both sexes (Cairns, 1994; Emmer-Sommer & Allen, 1999; Equal Rights Advocates, 2010), although females are most commonly the victims and males the perpetrators (Bacharach, Bamberger, & McKinney, 2007; Myhill & Allen, 2002; Testa & Park, 1996).

There are strong relationships between sexual aggression and substance use, particularly alcohol consumption (Muñoz-Rivas, Gámez-Guadix, Graña, & Fernández, 2010). Alcohol use precedes incidents of sexual violence in both offenders and victims (Abbey, Zawacki, Buck, Clinton, & McAuslan, 2001; Martin, 1992). Alcohol intoxication, even at a low level, may disinhibit sexually domineering men, making sexual aggression more likely (Noel, Maisto, Johnson, & Jackson, 2009). Social acceptance of high levels of alcohol use can facilitate male sexual aggression towards women (Bacharach et al., 2007) and in groups of males, social norms and peer pressures to pursue potential sexual partners can also contribute. Men’s alcohol use is cited as a factor in more than half of reported sexual coercion incidents (Abbey et al., 2004). Equally, in some cultures gender norms can mean that female alcohol use is considered by some men to suggest sexual availability (Bellis, Hughes, & Hughes, 2006), while intoxication can also make women appear easy targets for sexual activity or sexual assault (Calafat, Juan, Becoña, & Mantecón, 2008). Alcohol intoxication can also relate to women’s risky sexual intentions directly (Schacht et al., 2011) or indirectly through effects on cognitive appraisals of a situation as potentially “sexual.” Thus, alcohol may increase women’s risk of sexual victimization through reduced risk perception (Davis, Stoner, Norris, George, & Masters, 2009). The often highly sexualized nature of bars and nightclubs, combined with extensive use of alcohol and/or other drugs, can contribute to sexual violence (Abbey et al., 2001; Combs-Lane & Smith, 2002; Finney, 2003). Individuals describe a range of aggression, from verbal threats through physical assault and rape, which they experience and witness while or after drinking in bars (Parks, 2000; Parks, Miller, Collins, & Zetes-Zanatta, 1998).

With young people’s holidays abroad typically characterized by frequent nightlife participation and intoxication, holiday periods are likely to be key risk periods for sexual aggression. However, sexual aggression is a difficult and sensitive subject to explore, filled with ethical and methodological dilemmas (Finney, 2004). Cross-national studies can raise additional issues, with cultural factors likely affecting what individuals understand by SH or SAW. Fitzgerald et al. (1988) identified three forms of SH: sexual coercion, unwanted sexual attention, and gender harassment. Thus, a wide spectrum of behaviors can be considered as sexual harassment. These can be verbal (such as comments about one’s body, spreading sexual rumors, sexual remarks, dirty jokes or stories), physical (e.g., grabbing, rubbing or touching in a sexual way), or visual (such as the display of pictures of naked people or sex-related objects, or obscene gestures). Whether or not such behaviors are perceived as sexual harassment is likely to vary in different cultures, contexts, individuals, and moments of time. For example, a cross-national study of college students found that those from Brazil were more likely to see as flattering some behaviors that were considered harassing by American students (Pryor et al., 1997). American women perceived several scenarios in the study as more harassing than did American men, but this gender pattern did not emerge in other countries. Timmerman and Bajema (1999) reviewed 74 studies on the incidence of harassment in northwest Europe. Frequencies ranged from 2 to 90 % and Timmerman and Bajema concluded that the methodologies of surveys varied so widely that no firm estimates could be made of the relative frequency of harassment in the various countries; thus, no conclusions could be formed as to whether cultural factors within the countries affected rates of harassment.

Having sex against one’s will (SAW) is also a difficult matter to define objectively. Modern definitions of sexual assault and rape center on the lack of consent for sexual activity by one party, including through force, threat, intimidation, and incapacitation. Yet, having sex against one’s will can also be consensual with, for example, individuals agreeing to unwanted sex to please their partner or because they feel obligated to do so (Impett & Peplau, 2002). Substance use can complicate the matter further, as individuals may classify sex as having been against their will if they are unable to remember whether they consented or not or if they later regret sex that was initially consensual. Despite these issues of definition, few studies to date have examined young people’s experiences of either SAW or SH on holiday. Where SH during holidays has been explored in other studies it has mainly focused on SH of female tourists by local people (e.g., at the beach or in nightclubs) (de Albuquerque & McElroy, 2001; Kozak, 2007) or, in the case of males, harassment by sex trade workers in, for example, Caribbean or Asian countries such as China or Thailand (Cohen, 2003).

The aim of this study was to explore the incidence of sexual harassment and having sex against one’s will in young holidaymakers visiting popular holiday destinations in Southern Europe: Portugal, Spain, Italy, Greece, and Cyprus. Using a sample of 6,502 British and German holidaymakers aged 16-35, we examined reported levels of SH and SAW during the holiday period and identified demographic and holiday-related factors associated with these negative sexual experiences.

Method

Participants

Participants were young (aged 16-35) British and German holidaymakers visiting five traditional Mediterranean holiday destinations. Based on previous studies (Bellis, Hughes, Calafat, Juan & Schnitzer, 2009, Hughes et al., 2009, 2011a), a target sample of 700 individuals of each nationality was set for each location. British and German holidaymakers were chosen as these two nationalities are the most common visitors to all the surveyed locations. The survey took place in the departure areas of airports in each location: Palma de Mallorca in Spain; Faro in Portugal; Venice (Treviso and Marco Polo airports) in Italy; Crete (Heraklion airport) in Greece; Larnaca in Cyprus. The final sample was 6,502 British and German holidaymakers aged 16-35 years.

Measures

An anonymous questionnaire was developed based on established research tools (Hughes et al., 2008). The questionnaire explored: holidaymakers’ characteristics (including age, sex, nationality, self-rated social status, length of stay, and whether they were travelling with a sexual partner); reasons for choosing their holiday destination (“Why did you choose to visit [destination]?: cost, nightlife, culture, weather, work, and visiting family/friends” Tick all that apply); frequency of alcohol use, drunkenness (self-defined), tobacco use, and illicit drug use (cannabis, ecstasy, cocaine, amphetamines, ketamine, and gammahydroxybuterate [GHB]) on holiday; sexual activity on holiday (number and gender of sexual partners, use of condoms); features that participants favored when choosing bars or nightclubs to visit (“When choosing a bar or nightclub to visit, do you favour any of the following factors: a friendly atmosphere, opportunities for sex, places where people get drunk, etc.”); and personal experience of problems on holiday (“During your holidays did you experience any of the following?: Been injured in an accident, been involved in an argument, been involved in physical fight, etc.”). In this study, we focused on two negative sexual experiences: 1) sexual harassment (SH) (“During your stay in [destination], did you personally experience any of the following…been sexually harassed?”); and 2) sex against one’s will (“On any occasion this holiday, did you have sex that you felt was against your will?”). Consequently, both SH and SAW were self-defined. The questionnaire was translated into German following a research meeting to ensure the meaning of all questions was understood across nationalities and tested with native speakers. Ethical approval for the study was obtained from the National and Kapodistrian University of Athens, Greece.

Procedure

Experienced researchers were recruited in each study site and were trained to implement the surveys. They were instructed to approach all individuals that appeared to be aged 16-35 years and travelling without children or older relatives, who were waiting to check in for flights bound for the UK or Germany. Data collection took place between 10 July and 30 August 2009, covering peak summer holiday periods. Overall, 11,417 individuals were approached and asked if they had time to complete a short survey. Of these 35.3 % (n = 4,026) declined before being provided with any details of the survey or the topics it explored. Those stating they had time were provided with an explanation of the survey, assured of its anonymity and confidentiality, and asked if they would be willing to participate. At this stage, compliance was 92.5 % (6,834 out of 7,391). Those agreeing to participate were handed a questionnaire, clip-board, pen, and envelope and asked to self-complete the questionnaire and seal it in the envelope for collection by researchers. Researchers were instructed to stand a discrete distance away to permit the questionnaire to be completed in private yet allow participants to seek assistance if required.

Data Analyses

Completed questionnaires were returned to the UK and entered into a dedicated database using SPSS v15. At this point, 332 questionnaires were excluded due to participants being outside the target age, nationality or for questionnaires being incomplete or defaced. The final sample was 6,502. Target samples were achieved in all locations with the exception of German holidaymakers in Crete and Portugal (see Table 1). Analyses used χ2, with logistic regression used to identify factors independently associated with having being sexually harassed and having experienced sex against their will during the holiday. Logistic regression analysis used a stepwise backward conditional model in which all variables identified as significant in the χ2 analysis were entered into the initial model, with those that were not significantly (p < .05) related to the dependent variable being removed from the model and subsequently the model recalculated.

Table 1 Demographic characteristics
Full size table

Results

Samples varied between locations by gender, age, and nationality (see Table 1). A greater proportion of those surveyed in Italy were female and participants were younger in Mallorca and Crete than in other locations. There were also differences between destinations in relation to holidaying with a partner, length of stay, self-defined financial status, and reasons for choosing destination. For example, nightlife was a major reason for destination choice in those visiting Mallorca, Crete, and Cyprus, but not those visiting Portugal or Italy.

Two-thirds (68 %) of participants had gone on holiday without a sexual partner. More than half of all holidaymakers reported having had sex on holiday (53 %), including 31 % of those who arrived without a sexual partner. Overall, 8.6 % reported having been sexually harassed during the holiday. χ2 analysis examined relationships between SH and SAW and 13 variables measuring participants’ demographics and holiday characteristics (Table 2). All but three variables (length of stay on holiday and choosing the destination based on weather and family/friends) had significant relationships with SH. Thus, reported SH was higher in holidaymakers who were British, younger (aged 16-19 years), and travelling without a sexual partner, while both females and gay or bisexual males reported higher levels than heterosexual males (Table 2). Visitors to Mallorca or Crete were most likely to report SH, as were those who had chosen their destination due to nightlife. Those who had chosen their destination based on cost, culture or weather were less likely to report SH. The mean length of stay of individuals was 8.9 days. Overall, 1.5 % of holidaymakers reported having had sex against their will during their holiday. Only five of the 13 variables examined in χ2 analysis were significantly related to SAW at the p < .05 level (Table 2). Again, SAW was most common in visitors to Mallorca or Crete and in younger holidaymakers. Gay or bisexual males reported significantly higher levels of SAW than either females or heterosexual males.

Table 2 Proportion of participants reporting having been sexually harassed on holiday and having had sex against their will, by demographics and holiday characteristics
Full size table

Table 3 shows the associations identified in χ2 analysis between reported SH or SAW and nine variables measuring participants’ substance use on holiday and nightlife choices. All variables had significant associations with SH and all but one (tobacco use on holiday) with SAW. Thus, frequent drunkenness (on at least half of the days of the stay) was associated with both SH and SAW, while smoking tobacco was associated with SH only. Use of any form of illicit drug on holiday was strongly associated with both SH and SAW. Both outcomes were also associated with individuals being attracted to bars and nightclubs that have opportunities for sex (i.e., meeting sexual partners) and where people get drunk. Participants who reported being attracted to bars that have friendly atmospheres were less likely to report either SH or SAW.

Table 3 Proportion of participants who reported sexual harassment and sex against their will by holiday substance use and bar preferences
Full size table

To identify factors independently related to SH and SAW on holiday, logistic regression analysis was conducted using all variables identified as significant in χ2 analysis (Tables 2, 3). Twenty predictor variables were entered into the model for SH, of which ten emerged as significant. Thus, analysis confirmed the increased risk of SH in visitors to Mallorca and Crete, females and gay or bisexual males, younger participants, and British holidaymakers. Odds of SH were also higher in those with a self-reported medium high/high income and lower in those who chose the destination due to cost. Overall, alcohol use on holiday was related to SH. Among drinkers, odds of SH increased with frequency of drunkenness. However, non-drinkers reported risks of SH at higher levels than drinkers who never got drunk on holiday. Those who were attracted to bars where people get drunk or where there were opportunities for sex were at increased risk. SH was also strongly associated with cocaine use on holiday.

For SAW, of 13 predictor variables tested four had independent relationships with SAW. Gay and bisexual males were almost seven times more likely to report SAW on holiday (Table 4). The only other factors independently associated with SAW were cannabis use on holiday and being attracted to bars where there were opportunities for sex. Odds of SAW were significantly lower among visitors to Cyprus.

Table 4 Adjusted odds ratios for reporting sexual harassment and having sex against their will on holiday
Full size table

Discussion

A growing body of research is identifying the high levels of substance use young people engage in during holidays abroad and the negative consequences they experience (Bellis et al., 2009; Elliot et al., 1998; Hughes et al., 2009). However, this study is the first to identify unwanted sexual experiences in young tourists across different destinations and nationalities. Our study found that over 1 in 12 young British or German tourists experienced sexual harassment while holidaying in Southern European destinations. Despite the average holiday being only 8.9 days, 1.5 % of all participants reported having had sex against their will during their holiday. Whether such sex involved physical force, coercion or alcohol/drug-induced stupor preventing individuals from making rational decisions was not measured by the survey. However, with millions of young people holidaying in such destinations each year, findings suggest that a significant number of young holidaymakers will experience negative sexual experiences during their time abroad. The impact of such experiences can range from mild annoyance up to the more serious, health-damaging impacts of sexual assault, including physical injury, sexual health problems, and long-term psychological harm.

We found wide variations in negative sexual experiences between both nationalities and destinations. British holidaymakers were more likely than their German counterparts to report sexual harassment, as were those who holidayed in Mallorca and Crete. Both of these destinations attract young people specifically due to their reputation for nightlife and see high levels of alcohol use and drunkenness in visiting tourists (particularly British tourists; see Hughes et al., 2011a). Associations between sexual harassment and visiting either Mallorca or Crete remained after other factors had been controlled for, while those between choosing a destination due to its nightlife and sexual harassment did not. However, sexual harassment was significantly associated with alcohol use, with odds greatest in those who frequently got drunk. Those who preferred to visit bars where drunkenness occurs were also at increased risk (Tables 3, 4) with both factors being strongly interrelated (e.g., 41.4 % of those who got drunk were attracted to bars where drunkenness occurs compared to 5.4 % of drinkers who never got drunk, p < .001). Youth-focused bars in destinations, such as Mallorca and Crete, frequently offer cheap drink promotions and entertainment focused around excessive alcohol use and sexually suggestive activities. Such activities and participation in them by tourists may create perceptions of sexual availability that promote unwanted (as well as initially wanted) sexual attention. Our results add to evidence that bar characteristics can influence behavior (Hughes et al., 2011b). However, holiday locations can operate with little public health focus as underlying pressures are for the operation of an economy based on nightlife and the consumption of excessive alcohol. The long-term implications on the health of individuals are rarely considered given that these are likely to fall on nations other than those hosting the resort.

Our survey did not identify whether sexual harassment was perpetrated by other holidaymakers, by tourism staff or by local populations. However, media reports suggest that both can occur, with drunken female tourists being seen as easy targets for both predatory tourists and resident males. Such predatory behavior has also been reported in domestic situations (e.g., Kelly, Lovett, & Regan, 2005). Thus, while much research links sexual aggression to the alcohol consumption of the male aggressor, the use of alcohol by the victim can also be a major factor. For example, in a study on aggressive (sexual and non-sexual) incidents reported by women as victims, half reported that their behavior was changed by the alcohol they had consumed immediately before or at the time of the aggression (Parks, 2000). Consequently, there is a need to ensure that young holidaymakers are aware of the links between substance use and unwanted sexual attention when abroad. However, working with bars that trade on intoxicating nights should also be a priority to prevent the creation of norms and expectations (e.g., of sexual availability) in holiday resorts that facilitate sexual aggression.

Typically, both research and the media focus on sexual harassment and assault perpetrated by males towards females. While our study found that women in general reported higher levels of sexual harassment than males, gay and bisexual males reported similar levels of harassment to females. Women were no more likely to report having had sex against their will than heterosexual males. However, gay or bisexual men were seven times more likely than either heterosexual males or women of any sexuality to report having had sex against their will. Young gay men can face a range of pressures to fit into gay subcultures and particularly high levels of pressure to have sex (Valentine & Skelton, 2003). Although little work has been undertaken on this in tourist resorts, many European youth holiday destinations feature fashionable gay and bisexual resorts and nightlife scenes which, combined with high levels of substance use and disinhibition, are likely to increase these pressures in young (and particularly naive) gay tourists. Higher levels of sexually transmitted infections in gay populations (Loosier & Dittus, 2010; Xu, Sternberg, & Markowitz, 2010) increase the health risks of both wanted and unwanted sexual activity in such environments. Work in gay communities has shown how STIs (e.g., syphilis) can be spread through international social networks (Clark, Cook, Syed, Ashton, & Bellis, 2001). Thus, gay holidaymaker populations should be seen as a key group for targeting sexual health promotion work in order to reduce the risks of STI transmission including HIV.

Limitations

While our samples were intended to be broadly representative of holidaymakers visiting each destination, recruitment was opportunistic, conducted on a convenience basis and focused solely on holidaymakers using air transport. Despite this, we attained a high compliance among those who were informed of the nature of the survey, reducing the potential for participant self-selection to affect findings. As is usual with surveys covering issues of a sensitive nature, our study may have been affected by compliance and either under-reporting or exaggeration of unwanted sexual experiences on holiday, as well as issues such as substance use. To minimize this, we used an established methodology that included measures to ensure that participants were informed of the purpose of the study, assured of its confidentiality, and provided with a clearly anonymous mechanism of participation. Despite this, our quantitative design prevented us from validating exactly what each participant understood by sexual harassment and having sex against their will. It is possible that the differences between nationalities in reported experience of sexual harassment, for example, may represent cultural variation in what constitutes sexual harassment. We were also unable to gauge the severity of incidents in which participants had sex against their will or any circumstances surrounding reported incidents. Future research could use more detailed questions presenting holidaymakers with specific scenarios that could constitute SH or SAW and asking both whether participants have experienced these scenarios and whether they consider them to be SH or SAW. Whilst this level of detail was not possible within our study, our findings do provide the first data on reported levels of SH and SAW in holidaymakers and identify SH in particular as a common experience. Our findings should help inform future research in this area that incorporates more descriptive and qualitative techniques to help understand the nature of sexual aggression experienced by holidaymakers and how understanding of sexual harassment and sex against one’s will varies between cultures.

Conclusion

During what is a relatively short period (on average less than 9 days), nearly one in 10 British or German holidaymakers to Southern European resorts were sexually harassed and 1.5 % report having had sex against their will. As well as the potential for emotional distress, the consequences of such unwanted sexual activity may include harm to mental and physical health as well as STIs and unwanted pregnancy. While our study could not identify the lasting effects on victims, it could identify the resorts in which young holidaymakers were most likely to experience sexual harassment and having sex against the will and risk factors for such harms which can be used to inform interventions. In particular, sexual harassment has a strong association with drunkenness and drug (cocaine) use while gay and bisexual holidaymakers were at significantly increased risk of involvement in sex against their will. Those travelling abroad can suppress their propensity to be both cautious in their patterns of consumption and sexual exploration. However, the fact that young individuals adopt such an approach should not just be seen as an opportunity to exploit them but also to protect their health. Any sexual health information and interventions are likely to have relatively few effects if the considerable assets of an international tourist destination are aimed at encouraging intoxication and providing a setting permissive to sexual harassment. Consequently, measures to protect sexual health should focus not only on the behavior of tourists but also on the environment they holiday in and the way it is marketed to them at home.