To Confront Versus not to Confront: Women's Perception of Sexual Harassment

To define the term “sexual harassment” has proved extremely difficult to achieve. Researchers, lawyers, scholars, and policy makers around the world have not, up to this point, agreed upon a single definition (Pina, Gannon, & Saunders, 2009). In spite of the broad spectrum of approaches for [legally, academically, psychosocially] defining sexual harassment, most researchers contend that it is a psychological experience based on a sexually unwanted, offensive, and threatening behaviour at work (Topa, Morales, & Depolo, 2008).

Several authors have identified three types of sexual harassment: gender harassment, unwanted sexual attention, and sexual coercion (Expósito & Moya, 2005; Gelfand, Fitzgerald, & Drasgow, 1995). Gender harassment (hostile, offensive, intimidating, and degrading verbal and nonverbal behaviour against women) is a type of subtle sexual harassment aimed at deterring women from transgressing male domains rather than being an expression of sexual attraction. As for unwanted sexual attention (verbal and non verbal behaviour, such as persistent nonreciprocal requests for dates, letters, phone calls, deliberate touching, grabbing, sexual advances and propositions, and assault), it is among the most evident types of this behaviour; this behaviour is perceived by the target as unwelcome, unreciprocated, and offensive acts of sexual interest. Sexual coercion (also known as quid pro quo or sexual blackmail) is the most explicit and recognizable type of sexual harassment, where the harasser, a person in power, demands sexual favours from a subordinate worker in exchange for organizational rewards and benefits or threats of reprisal related to job prospects and conditions (e.g., job security and promotion) (Fitzgerald, Drasgow, Hulin, Gelfand, & Magley, 1997). The literature on the incidence of sexual harassment in different countries (O’Connell & Korabik, 2000; Pina et al., 2009; Timmerman & Bajema, 1998) has shown that only a small number of cases of harassment meet the inclusion criteria for coercion, so this research has decided to study gender harassment and unwanted sexual attention, which are the most frequent forms of sexual harassment. This study was undertaken in Spain, a country where there is no official register for the prevalence of sexual harassment at work (Amnesty International, 2011). Nevertheless, the most recent figures revealed a total number of 6,573 sexual offences against women in 2009, of which 330 were cases of sexual harassment (Instituto de la Mujer, 2009).

Though both men and women may be exposed to sexual harassment, the literature on harassment is consistent in reporting that an overwhelming number of victims are women, and harassers are men (Ménard, Hall, Phung, Ghebrial, & Martin, 2003; Ménard, Shoss, & Pincus, 2010; Pryor, 1995). Thus, one out of every two to three women have experienced some type of sexual harassment or have been subjected to unwanted sexual behaviour (European Commission, 1998; Pina et al., 2009). The magnitude of the social problem underscores the need for research focusing on this type of gender violence. However, the data on incidence rates may fail to provide an accurate description of reality, given the lack of consensus and the confusion of terminology, which only serve to further confound the definition of harassment.

The strategies most frequently used by women to cope with harassment range from avoiding or ignoring the harasser to reporting the offence (Cortina & Wasti, 2005; Fitzgerald, Gelfand, & Drasgow, 1995). Unfortunately, none of these strategies has proven to be clearly effective in combating harassment at work, nor in raising the confidence of workers (i.e., potential victims) regarding their expectations towards their current employers. Studies have shown that women who report incidents of harassment are often threatened with reprisals for reporting the incident or making it public (Bergman, Langhout, Palmieri, Cortina, & Fitzgerald, 2002). A further strategy employed by women in coping with sexual harassment is confronting the harasser. Some studies have found that active confrontation benefited victims by empowering women, and by helping them to expose social inequality (Kaiser & Miller, 2004).

Though there may be potential benefits in confronting the harasser, few women appear to opt for this strategy. A few studies on women’s beliefs about whether to confront or not sexual harassment found the response that it was driven by the fear of the harasser’s possible reactions. These reactions may involve threats of reprisal related to job prospects and conditions (e.g., job security and promotion) or on a social dimension (e.g., ostracism) and negative appraisal (Saunders & Senn, 2009; Shelton & Stewart, 2004). The tendency to respond negatively to any woman who attempt to draw limits as to the behaviours of men, particularly if these infringe traditional gender roles, is enshrined and perpetuated by the sexist ideology. A good example is a study where women who challenged traditional gender roles and undermined male authority were found to be negatively evaluated by men (Herrera, Expósito, & Moya, 2012).

The sexist ideology asserts men and women are different and each should take their corresponding place in society; it also establishes the behaviour appropriate for men and women and what is expected of each in a relationship (McHugh & Frieze, 1997). Thus, the sexist ideology is a variable influencing challenges to the established order between men and women, and what is considered to be acceptable and normal in traditional relationships. This system of beliefs or ‘legitimizing ideologies’ serves to justify and perpetuate gender inequality that has prevailed throughout history (O’Brien & Major, 2005; Sutton et al., 2008).

Women themselves normally acquiesce to ambivalent sexist beliefs, particularly of the benevolent type, as it enables them to perceive of themselves in a positive light, and it is a subtler and in turn more accepted type of sexism than hostile sexism. Moreover, as such beliefs are often covert, they are not perceived as sexist, which hinders any attempt to combat these beliefs (Barreto & Ellemers, 2005). This highlights the need for furthering our understanding of sexism among women given that a factor thwarting social change among discriminated groups of women was the fact that the members of such groups accepted their inferior status, and consequently justified and perpetuated it (Jost & Kay, 2005).

In general, sexism is associated to attitudes legitimizing violence against women, and would explain the nexus between hostile sexism and blaming the victim (Herrera et al., 2012; Lila, Gracia, & Garcia, 2013). In relation to attitudes towards sexual harassment, “a common aspect to all acts of violence are the numerous myths related to them” (Leidig, 1981, p.199). Myths of sexual harassment, including beliefs such as self-victimization, that women enjoy acts of violence, these acts are only committed by mentally deranged men, or that women exaggerate their reports are common to all women. Acceptance of myths hinders the ability to visualize a problem and this lack of awareness in turn undermines the willingness to assist the victim (Lonsway, Cortina, & Magley, 2008). Sexist myths were shared by both men and women and, on the whole, the more both of them adhere to traditional gender roles, the greater the likelihood certain behaviours of sexual harassment were considered to be acceptable or normal. Furthermore, the more an individual considered sexist behaviour was normal, the greater the likelihood they would deny the negative consequence of their own actions or behaviour (Quinn, 2002).

Thus, the aim of this study was to assess the influence of the type of sexual harassment (gender harassment vs. unwanted sexual attention), and victim response (confrontation vs. no confrontation), on women’s perception of harassment and their beliefs about the possible reactions of the harasser. Moreover, the efficacy of confrontation as a victim coping strategy for combating sexual harassment was analysed. Furthermore, the influence of certain ideological variables, such as ambivalent sexism and the acceptance of myths of sexual harassment, on social perceptions of harassment and the evaluation of women was evaluated.

To achieve these goals, we posited the following hypotheses:

1. Gender harassment would be perceived less as sexual harassment in comparison to unwanted sexual attention.

2. Participants believed the type of victim response (confrontation) would influence the harasser’s evaluation of the victim (i.e., women who explicitly responded to harassment would be negatively evaluated by the harasser).

3. A participant’s ideological variables (sexism, myths of harassment) would influence both the perception of harassment and beliefs about how the harasser would react in response to confrontation.

Participants

The sample consisted of 138 female undergraduates enrolled at different faculties of the University of Granada; age range was 18 to 58 years (M = 24.69, SD = 10.49).

Procedure and Design

The sample was obtained through incidental sampling in different classrooms of several faculties at the University of Granada, Spain. Having obtained informed consent, participants were randomly assigned to one of the experimental conditions, and were given approximately 20 minutes to complete a questionnaire in their habitual classrooms. All participants were assured their information and responses would remain anonymous and confidential. Once all students had completed the questionnaire, they were informed about the objectives of the study. All of the participants freely volunteered to respond to the questionnaire, and were awarded an extra 0.1 towards their final grade as compensation.

A 2 (type of harassment: gender harassment vs. unwanted sexual attention) x 2 (type of victim response: confrontation vs. no confrontation) factorial design was employed.

Instruments

A questionnaire was designed containing all the variables to be measured. The first step was to present a scenario, and participants were instructed to adopt the role of the protagonist of the following story (Ann):

John and Ann are workmates, they don’t have a relationship other than at work, they are not friends, they don’t go out together, nor do they meet up anywhere else than at work, they are only workmates. At work, sometimes they talk about everyday tasks, discuss issues about the working day, whether it’s been productive or not, the future of the business, any noteworthy events, just normal questions one would expect to discuss with a workmate. During the conversation…

The type of sexual harassment was manipulated by providing participants the following information. For gender harassment: “… John tells Ann: You would be better off if you showed off more of your charms (body)...” For unwanted sexual attention: “... John tells Ann while he puts his hand on her thigh: I know of a way we can forget about these problems, and get rid of this stress….”

The type of victim response was manipulated by providing participants the following information. For confrontation: “In view of the situation, Ann responds: Listen John, your behaviour is entirely unacceptable, what you are doing is called sexual harassment, so I would appreciate it if you would make sure it never happened again.” For no confrontation: “In view of the situation, Ann feels distressed, but decides to do or say nothing.”

Thereafter, the following scales were administered:

Perception of sexual harassment was measured using the item: “Do you think John’s behaviour constitutes sexual harassment?” with a 7-point Likert type response format ranging from 1 (it is not sexual harassment) to 7 (it is sexual harassment).

Women’s beliefs about how men would evaluate the victim were measured using the items: “If you were Ann, to what extent would you match John´s ideal partner?” with a 7-point response format ranging from 1 (not at all) to 7 (very much); “(Remember, you are Ann) John believes I behaved like this just to be provocative”, measured on a 7-point response format ranging from 1 (totally disagree) to 7 (totally agree); and “(Remember, you are Ann) John thinks I am impertinent”, rated on a 7-point Likert type response format ranging from 1 (totally disagree) to 7 (totally agree).

Importance ascribed to the incident of harassment, described by the item “If you were Ann and it was the first time it happened to you, would you take no notice, and think it wasn’t important?”, rated on a 7-point Likert type response format ranging from 1 (totally disagree) to 7 (totally agree).

Four items were used as manipulation checks: a) type of harassment, “To what extent did John suggest to Ann she would be better off if she showed more of her charms?”, measured using a 7-point Likert type scale ranging from 1 (he didn’t suggest it at all) to 7 (he clearly suggests it), “To what extent did John deliberately touch Ann while he suggested a way of forgetting their problems?”, measured using a 7-point Likert type scale ranging from 1 (he didn’t do it at all) to 7 (he did it deliberately); and b) victim response, confrontation, “To what extent did Ann respond assertively to John’s behaviour, that is, clearly tell him his behaviour was entirely unacceptable, and that he should ensure it never happened again?”, rated on a 7-point Likert type scale ranging from 1 (she didn´t respond assertively) to 7 (she responded assertively), “To what extent does Ann not respond to John’s behaviour?”, measured using a 7-point Likert type scale ranging from 1 (no response at all) to 7 (responded fully).

Gender stereotyping: a scale consisting of 26 items for measuring the masculine perception of women (instrumental) and the feminine perception (expressivity) designed and validated by Expósito (1997) was used. In this study, only the instrumental scale was used given that the objective was to obtain masculine perception. Examples of instrumental items were: ambitious, independent, self-confident, individualist, leadership qualities, strong. Participants were told to imagine they were Ann, and were instructed to indicate the degree to which they thought each item was applicable by John to the protagonist of the story (herself) using a 7-point Likert type response format ranging from 1 (not at all) to 7 (very much). The coefficient for participants on the instrumentality subscale was .82.

The Illinois Sexual Harassment Myth Acceptance (ISHMA; Lonsway et al., 2008; adapted to the Spanish context by Expósito, Herrera, Valor-Segura, Herrera, & Lozano, 2014). The scale consisted of 20 items scored on a 7-point Likert type scale (1 = strongly disagree, 7 = strongly agree). High scores were indicative of greater acceptance of myths of sexual harassment. The scale consisted of four dimensions: invention/exaggeration (alpha, .63), hidden motives (alpha, .80), natural heterosexuality (alpha, .81), and woman’s responsibility (alpha, .72). The alpha coefficient for the total scale was .85. In this study, only global scores were analysed as the primary objective was to obtain an overview of myths rather than a detailed analysis of each specific dimension.

The Ambivalent Sexism Inventory (ASI; Expósito, Moya, & Glick, 1998) consisted of 22 items rated using a 6-point response format ranging from 0 = totally disagree to 5 = totally agree. High scores revealed more sexist attitudes. Half of the items were related to hostile sexism (HS) (women get easily offended, women always exaggerate the problems they have at work…), and the other half were related to benevolent sexism (BS) (women are bestowed with a purity that few men possess…). The Cronbach alpha coefficient for hostile subscale was .88, and for the benevolent subscale .89. The alpha coefficient for the total scale was .92.

Effects of the Type of Sexual Harassment, Victim Response, and Ideological Variables on the Perception of Sexual Harassment, and the Evaluation of Women

All the experimental manipulations were effective. In order to assess how women perceived harassment and women’s beliefs about how they would be evaluated by men according to the type of sexual harassment (gender harassment vs. unwanted sexual attention) (Hypothesis 1); type of victim response to harassment (confrontation vs. no confrontation) (Hypothesis 2); and the role of sexist ideology and the acceptance of myths of sexual harassment (Hypothesis 3) several hierarchical regression analyses were performed with the following dependent variables: 1) instrumentality, 2) fits the image of the ideal partner, 3) evaluation of women’s behaviour as provocative, 4) evaluation of women as impertinent, 5) the item: “If it were the first time it happened to you, would you take no notice, and think it wasn’t important?”, and 6) perception of behaviour as sexual harassment.

The same procedure was applied for each analysis. In Step 1, the type of sexual harassment: gender harassment (value 0) vs. unwanted sexual attention (value 1); type of victim response: no confrontation (value 0) vs. confrontation (value 1); and the participants’ centred scores in HS, BS, and ISHMA were analysed. Step 2 involved second order interactions between the experimental manipulations and the ideological measures. The results obtained for the first three dependent variables are shown in Table 1, and the results for the remaining three dependent variables are shown in Table 2.

Table 1

Type of Sexual Harassment, Victim Response, and Ideological Variables as Predictors of the Evaluation of Women

Table 2

Type of Sexual Harassment, Victim Response, and Ideological Variables as Predictors of the Evaluation of Women, and the Perception of Harassment

As shown in Table 1 and Table 2, more participants in the unwanted sexual attention condition perceived behaviour as sexual harassment than participants in the gender harassment condition ( = .173, p = .05), which substantiated Hypothesis 1.

Most participants believed women who confronted harassment (vs. those who did not), would be attributed more instrumental traits by men ( = .531, p < .001), perceived as impertinent ( = .334, p < .001), and women would take no notice of the incident, and think it was not important (= .27, p < .01), which substantiated Hypothesis 2.

Moreover, the results indicated that in terms of unwanted sexual attention, women believed they fitted better men’s ideal image of a woman ( = .221, p < .05), and that men would perceive them as provocative ( = .218, p < .05). In addition, an interaction was observed for the type of harassment x type of victim response in the variable “if it were the first time it happened to you, would you take no notice, and think it wasn’t important?”, that measured the importance women assigned to certain behaviours of harassment (= -.24, p < .01). Though no significant differences were observed for unwanted sexual attention ( = .04, ns), for gender harassment participants gave less importance to this behaviour if the victim had confronted the harasser vs. no confrontation ( = .53, p < .01). Furthermore, a significant interaction type of harassment x BS ( = .22, p = .05) was found in the variable that evaluated women as impertinent, for unwanted sexual attention participants with high BS scores tended to believe the women would be perceived by men as impertinent ( = .08), and the opposite effect was observed for gender harassment ( = -.19). Participants scoring high on acceptance of myths of sexual harassment tended not to take any notice or think harassment was important ( = .372, p < .001).

Finally, to examine the relationship between the dependent variables and the ideological measures, correlational analyses were performed (see Table 3).

Table 3

Mean Scores (and Standard Deviations). Correlation between Variables

These analyses revealed that participants believed that men who attributed women instrumental traits (r = .29, p < .01) would also perceived of them as impertinent, and the participant would not think the incident of harassment was important (r = .22, p < .05). In comparison, participants perceived behaviour as sexual harassment when they thought men saw the woman as provocative (r = .20, p < .05) and the woman gave little importance to the incident of harassment (r = -.38, p < .01). Most participants scoring high on acceptance of myths of sexual harassment believed the harasser would attribute the woman more instrumental traits (r = .19, p < .05) and the participant gave little importance to the incident of harassment (r = .29, p < .01). Similarly, participants scoring high on hostile sexism, scored high on acceptance of myths of sexual harassment (r = .65, p < .01), and gave little importance to harassment (r = -.17, p < .05). All of ideological measures were correlated with one another.

The aim of this study was to explore how women, potential victims, think men will react to women who confront sexual harassment. The hypotheses predicted the type of harassment, type of victim response (confrontation), and participant’s ideology would drive women’s beliefs about how harasser may react.

As for the typologies of harassment to be found in the literature, subtle forms of sexual harassment are considered to be those that are not easily recognizable as discriminatory acts (Baker, Terpstra, & Larntz, 1990; Herrera, Herrera, & Expósito, 2014; Jones & Remland, 1992; Saunders & Senn, 2009). The results of this study agree with the findings of these authors in that, in general, perception of sexual harassment was lower in gender harassment than in unwanted sexual attention. The fact that gender harassment was not perceived as such, or even trivialised, in comparison to other more explicit and direct forms (unwanted sexual attention), may lead to this type of behaviour becoming normalised in relationships between men and women, the implications of which transcend both occupational and social settings. Thus, victims exposed to sexual harassment at work who do not dispose of strategies for detecting and coping with harassment feel vulnerable and inept. This instils a sense of helplessness in potential victims that is conveyed in the responses of participants in relation to both types of harassment. Faced with a situation of unwanted sexual attention, most participants thought the woman would perceive herself as coming closer to fitting a man’s image of an ideal partner, and the woman would see herself as being seen by John as provocative, regardless of the type of victim response to the incident of harassment. This highlights the influence of myths in the perception and evaluation of events (Lonsway et al., 2008), given that myths modulated the image women had of themselves, and what was expected of them. Similar results have been reported for other types of violence against women, which underscores that attitudes to gender violence are crucial for understanding reactions and behaviours towards victims and harassers (Gracia, García, & Lila, 2009).

In relation to victim response, the results show that participants believed women who confronted their harasser would be evaluated negatively by men and would be attributed more instrumental traits than women who did not confront the harasser, regardless of the type of sexual harassment.

As for the evolution of feminine stereotyping, most democratic nations have progressively evolved into perceiving women as competent - at first sight this would appear to be a positive step towards gender equality. Notwithstanding, the data reveals that women who took the initiative or challenged traditional gender roles were chastised for lacking the femininity that is expected of a woman (Rudman & Glick, 1999).

Most of the participants in this study believed that women who confronted their harasser would be perceived by men as impertinent in comparison to women who did not. Previous studies have reported similar results confirming the influence of traditional gender attitudes on the evaluation of victims of violence against women (Herrera et al., 2014; Valor-Segura, Expósito, & Moya, 2008; Yamawaki, 2007). Moreover, participants indicated they would overlook and take no notice of an incident of harassment, probably because they considered the confrontation response was sufficiently explicit and clear and they had done what was necessary to deter the harasser, particularly in response to more subtle forms of gender harassment.

Furthermore, effects of sexist ideology on perception of harassment were found. Women scoring high on acceptance of myths of sexual harassment gave less importance to incidents of harassment. Several studies have observed how myths of violence against women, e.g., on sexual assault, serve to deny, trivialize, and justify patriarchal violence against women (Gerger, Kley, Bohner, & Siebler, 2007; Herrera et al., 2014; Herrera, Herrera, & Expósito, 2016). These myths include beliefs and behaviours that blame the victim, minimize the psychological impact on victims, and justify the behaviour of the harasser (Lonsway et al., 2008). All of these myths are designed to raise harassment tolerance levels; consequently, they have negative repercussions on victims and hinder the recovery process (Campbell, Wasco, Ahrens, Sefl, & Barnes, 2001), and reduce the chances victims will report incidents of harassment.

The results have shown that there is no fully efficacious strategy for coping with sexual harassment, and that women’s beliefs about the negative consequences and reactions deter many women from confronting the harasser and/or reporting the incident (Cala, Trigo, & Saavedra, 2016; Shelton & Stewart, 2004).

The results of this study have important implications for sexual harassment prevention and/or education programs designed to help women to understand the construct of sexual harassment, and to develop strategies for detecting and coping with harassment. Finally, studies on perceptions of harassment and victim response contribute in providing data for the design and implementation of evidence-based social policy on sexual harassment and other types of violence against women. Throughout society and from all quarters of government victims are encouraged to report offences to the police, but those who dare to do so face a barrage of obstacles with negative repercussions on their lives. The results of the influence of ideological variables highlight the importance of raising social awareness, and in developing effective tools and strategies for detecting and coping with harassment.

It should be noted that this study has several limitations which can most certainly be rectified in the future. The study assessed how women think men will react to women who confront sexual harassment, using a story and questionnaires. Though the literature on the usefulness of these types of studies is extensive, it nonetheless entails certain limitations. Due to the impossibility of recreating real-life situations, the participants in this study may have to fully evaluate and react as they would in real life. Thus, the size and the occupation of the sample used, as all participants are university students from Spain, is a limitation, so future studies should try to work with more heterogeneous samples which would allow the possible influence of cultural values, age, and sociodemographic and economic status to be analyzed.

Bearing in mind the limitations of this study, the results shed some light on the understanding of how women think that men react to victims of harassment when they challenge such behaviour and on understanding some of the main obstacles hindering women from reporting harassment.

This underscores the need for further research to advance our understanding of this phenomenon and to provide accurate definitions representing real-life incidents of sexual harassment that will help victims to identify and report incidents of harassment.