Tattoos are increasingly popular across ages and social classes in today's society. Whereas tattoos were once largely considered a deviant form of expression that was popular among bikers, prisoners, and those in the armed forces (DeMello, Reference DeMello1995; Larsen, Patterson, & Markham, Reference Larsen, Patterson and Markham2014), tattoos are now in the mainstream. In a recent survey of individuals across 18 countries, approximately 38 percent of the respondents had at least one tattoo (Tennent, Reference Tennent2018, May 24). The three most heavily tattooed countries were Italy, Sweden, and the U.S. with 48, 47, and 46 percent of the respondents, respectively, adorned with body art. Moreover, it is estimated that 45 million individuals in the U.S. workforce have at least one tattoo (Schouten, Reference Schouten2016, September 29). As tattoos are increasingly popular, their implications for the workplace cannot be ignored.
Despite the mainstreaming of tattoos, individuals adorned with such body art may be subject to potential prejudice in the workplace. Timming's (Reference Timming2015) interviews with hiring managers revealed largely negative attitudes toward job candidates with tattoos. Furthermore, Timming, Nickson, Re, and Perrett (Reference Timming, Nickson, Re and Perrett2017) found that tattoos on individuals resulted in lower employment suitability ratings for job candidates. Regarding coworker preferences, Miller, McGlashan Nicols, and Eure's (Reference Miller, McGlashan Nicols and Eure2009) results demonstrated that individuals perceive coworkers with tattoos as less acceptable compared to those without. Baumann, Timming, and Gollan (Reference Baumann, Timming and Gollan2016) also illustrated that customers have a predominantly negative attitude toward service providers with tattoos. Notwithstanding the value of these insights, research has yet to examine the impact of being tattooed on discriminatory behavior directed toward employees once on the job.
This study contributes to the existing body of research on prejudice and discrimination against employees bearing tattoos and the body of research on workplace sexual harassment by examining the impact of visible tattoos on unwanted sexual attention and potential moderators of this relationship. Unwanted sexual attention is a form of sexual harassment that is comprised verbal and non-verbal behavior of a sexual nature (Gelfand, Fitzgerald, & Drasgow, Reference Gelfand, Fitzgerald and Drasgow1995), which has been demonstrated to have a variety of negative consequences for employees (Willness, Steel, & Lee, Reference Willness, Steel and Lee2007). To help determine what organizations can do to minimize such attention at work, two aspects of organizational climate – perceived sexual harassment climate and inclusion climate – will be examined as moderators of the visible tattoo-unwanted sexual attention relationship. Aspects of organizational climate are important as climate signals what behaviors are acceptable, mobilizes existing biases, impacts how decisions are made, and shapes motivations (Schneider, Reference Schneider1990; Wilson & Thompson, Reference Wilson and Thompson2001). In a less permissive sexual harassment climate where there are more negative repercussions for harassment (Estrada, Olson, Harbke, & Berggren, Reference Estrada, Olson, Harbke and Berggren2011), visible tattoos may be less strongly related to unwanted sexual attention. Furthermore, in a more favorable inclusion climate that is more respectful of individual differences (Nishii, Reference Nishii2013), the relationship between visible tattoos and unwanted sexual attention could be weaker.
Theoretical Background and Hypotheses
Unwanted sexual attention is a widely recognized dimension of workplace sexual harassment that includes such behavior as repeated unwanted requests for dates, discussion of sexual matters, and inappropriate touching (Gelfand, Fitzgerald, & Drasgow, Reference Gelfand, Fitzgerald and Drasgow1995). Unwanted sexual attention does not involve threats and bribes (Gelfand, Fitzgerald, & Drasgow, Reference Gelfand, Fitzgerald and Drasgow1995) and occurs with greater frequency than sexual coercion (McDonald, Reference McDonald2012). Unwanted sexual attention, along with other forms of sexual harassment, has widely acknowledged adverse effects on employees and organizations, such as decreased job satisfaction, lower organizational commitment, greater work withdrawal, and poorer physical and mental health (Willness, Steel, & Lee, Reference Willness, Steel and Lee2007). Reinforcing these findings, Hogh, Conway, Clausen, Madsen, and Burr (Reference Hogh, Conway, Clausen, Madsen and Burr2016) found that experiencing unwanted sexual attention was followed by an increase in long-term absences due to sickness. Furthermore, Nielsen, Bjørkelo, Notelaers, and Einarsen (Reference Nielsen, Bjørkelo, Notelaers and Einarsen2010) provided evidence that unwanted sexual attention is associated with lower job satisfaction and poorer mental health. While tattoos could be viewed as an innocuous form of self-expression, the present study seeks to demonstrate that adornment with tattoos may also have unintended adverse consequences. Two arguments are put forward below to explain the relationship between visible tattoos on employees and unwanted sexual attention, drawing on automatic stereotype activation and the attention-getting potential of visible tattoos.
On one front, visible tattoos could lead to increased unwanted sexual attention because tattoos may be characterized as stigma (Larsen, Patterson, & Markham, Reference Larsen, Patterson and Markham2014) and automatically prompt the activation of negative stereotypes (Major & O'Brien, Reference Major and O'Brien2005). Stigmas, of which visible tattoos are an example, evoke automatic activation of negative stereotypes of the stigma bearer's group (Major & O'Brien, Reference Major and O'Brien2005). Activated cultural stereotypes have the potential to affect behavior at both a conscious level and an unconscious level without the stereotype holder's conscious intent (Wheeler & Petty, Reference Wheeler and Petty2001). Since those with tattoos may be stereotyped as more sexually promiscuous and as thrill seekers (Swami & Furnham, Reference Swami and Furnham2007; Wohlrab, Fink, Kappeler, & Brewer, Reference Wohlrab, Fink, Kappeler and Brewer2009), perpetrators could believe that visibly tattooed individuals are not averse to sexual attention in the workplace. They may also perceive tattooed individuals as more willing to break workplace norms (Camacho & Brown, Reference Camacho and Brown2018; Langman, Reference Langman2008). In this regard, perpetrators may view visibly tattooed employees as willing participants in a sexual exchange. Furthermore, perpetrators may feel a sense of greater power vis-a-vis a visibly tattooed employee. Perpetrators may believe that they can more readily get away with harassment because of the visibly tattooed employee's lower perceived status (Halnon & Cohen, Reference Halnon and Cohen2006), since those with tattoos are stereotyped as ‘other’ and less intelligent, honest, and capable (Broussard & Harton, Reference Broussard and Harton2017; Dean, Reference Dean2010; Degelman & Price, Reference Degelman and Price2002). This argument is supported by the power differential explanation of sexual harassment which posits that perpetrators are largely motivated by a desire to reinforce a sense of power and escape negative consequences from the organization for their behavior (Stockdale & Nadler, Reference Stockdale and Nadler2012).
Visible tattoos could also lead to an increased likelihood of unwanted sexual attention by the mere fact that visible tattoos, whether or not they are perceived as a stigma, are an attention-getting adornment of the body. As tattoos are on the body and visible tattoos draw attention, others at work may be likely to stare and engage in conversation about someone's tattoo. In this regard, unwanted sexual attention could potentially be inadvertent. Research by Wohlrab, Fink, Pyritz, Rahlfs, and Kappeler (Reference Wohlrab, Fink, Pyritz, Rahlfs and Kappeler2007) highlights that tattoos are particularly attention getting. In their eye-tracking study, Wohlrab et al. found that people gazed longer at tattoos than other forms of body adornment. Evolutionary psychologists have argued that people are unconsciously attracted to people with tattoos (Koziel, Kretschmer, and Pawlowski, Reference Koziel, Kretschmer and Pawlowski2010). Any commentary regarding a visible tattoo or visual scrutiny, however subtle, may be thought of as inappropriate by an employee as workplace norms restrict such behavior. Such behavior may further be unwanted because interest in a visible tattoo could be interpreted as interest in the body, which is closely associated with sexual interest. Moreover, commentary on a visible tattoo could spill over into conversation of a sexual nature because the tattoo is on the body. Employees with visible tattoos could find themselves facing the dilemma of wanting their tattoos to be noticed, but not wanting their bodies to be associated with sex in a work context. In sum, both because visible tattoos prompt automatic activation of negative stereotypes associated with sex and because they attract extra visual attention, possessing a visible tattoo may lead to increased unwanted sexual attention.
Hypothesis 1
Individuals with visible tattoos are more likely to receive unwanted sexual attention.
Two aspects of organization climate – sexual harassment climate and inclusion climate – could moderate the visible tattoo-unwanted sexual attention relationship. Schneider (Reference Schneider1990) defined organizational climate as the perceptions of employees concerning the practices, procedures, and behaviors that get rewarded and supported in a work setting. Irrespective of its focus, climate encompasses multiple formal and informal aspects of the organizational system. Climate serves as a pervasive mechanism for exercising organizational control explicitly and implicitly. The present study assessed climate as an individual perception rather than as a shared perception across individuals within an organization. Thus, the sexual harassment and inclusion climate constructs are best characterized as perceived climate constructs in this investigation (Jauhari & Singh, Reference Jauhari and Singh2013; Madera, Dawson, & Neal, Reference Madera, Dawson and Neal2013).
Sexual harassment climate has consistently been demonstrated to be a significant antecedent of workplace sexual harassment, as proposed in a Fitzgerald, Drasgow, Hulin, Gelfand, and Magley's (Reference Fitzgerald, Drasgow, Hulin, Gelfand and Magley1997) seminal model. Hulin, Fitzgerald, and Drasgow (Reference Hulin, Fitzgerald, Drasgow and Stockdale1997) characterize sexual harassment climate as an environment that is associated with the tolerance of sexual harassment and inaccessible and ineffective harassment remedies. Sexual harassment climate relates to an employee's perceptions of ‘the degree to which an organization is perceived as insensitive to or tolerant of sexual harassment’ (Hulin, Fitzgerald, & Drasgow, Reference Hulin, Fitzgerald, Drasgow and Stockdale1997: 129). A permissive sexual harassment climate is one where individuals perceive that it is normative to engage in sexually harassing behavior, believe it is risky for victims to complain, think that claims will not be taken seriously, and perceive no serious sanctions against perpetrators. Since their meta-analysis demonstrated that sexual harassment climate exhibited significant pervasive effects on sexual harassment, Willness, Steel, and Lee (Reference Willness, Steel and Lee2007) asserted that ‘it is quite clear that organizational climate and the workplace environment are central to understanding the conditions under which harassment is more likely to occur and how victims are affected’ (p. 143). Additional reviews have further validated Willness et al.'s assertion (e.g., Chan, Lam, Chow, & Cheung, Reference Chan, Lam, Chow and Cheung2008; O'Leary-Kelly, Bowes-Sperry, Bates, & Lean, Reference O'Leary-Kelly, Bowes-Sperry, Bates and Lean2009).
A less permissive sexual harassment climate is proposed to attenuate the hypothesized relationship between visible tattoos and unwanted sexual attention. When a climate is less permissive of sexual harassment, potential perpetrators may be less apt to engage with potential targets for fear of negative repercussions. Although sexual harassment climate typically has been examined as a main effect rather than as a moderator, other climate variables have been found to strengthen or attenuate relationships in other workplace research (Hofmann, Morgeson, & Gerras, Reference Hofmann, Morgeson and Gerras2003; Smith-Crowe, Burke, & Landis, Reference Smith-Crowe, Burke and Landis2003). Sexual harassment climate may interact with visible tattoos to reduce unwanted sexual attention because, in a less permissive climate, where claims of harassment are less risky and remedies are more effective, perpetrators are less free to act on their stereotypes about people with tattoos. They may then engage in more self-monitoring. Accordingly, in a less permissive sexual harassment climate, a potential perpetrator may be less likely to perceive an employee with a tattoo as a target and may be more concerned about discussing sexual matters or staring at someone in a sexually suggestive way. Conversely, in an organization with a more permissive sexual harassment climate, a potential perpetrator may be more apt to act on the tendency to behave inappropriately toward those with visible tattoos and feel freer to break workplace norms. Based on these arguments, it is hypothesized:
Hypothesis 2
A less permissive perceived sexual harassment climate will reduce the likelihood that individuals with visible tattoos will receive unwanted sexual attention, and a more permissive perceived sexual harassment climate will increase the likelihood.
Inclusion climate is also hypothesized to moderate the relationship between visible tattoos and unwanted sexual attention. Psychological climate for inclusion, referred to here as inclusion climate, reflects an individual's perception regarding the degree to which he or she feels as though ‘… people of all identities and many styles can be fully themselves while also contributing to the larger collective, as valued and full members’ (Ferdman, Reference Ferdman2017). Nishii (Reference Nishii2013) conceptualized inclusion climate as consisting of three dimensions: foundations of equitable employment practices, inclusion in decision-making, and integration of differences. Foundations of equitable employment practices refer to diversity-specific human resource management practices that are designed to help eliminate bias in recruitment, selection, compensation, and employee development. In turn, inclusion in decision making refers to the degree to which the perspectives of all employees, regardless of their demographic characteristics, are sought, valued, and integrated in organizational functioning. Finally, the integration of differences refers to the degree to which the individual differences of diverse employees are valued and integrated rather than changed or suppressed on the job. Integration of differences reflects an employee's perception of the collective norms and expectations that guide the treatment and valuing of diverse individuals in an organization. For the present study, we focus specifically on integration of differences because visible tattoos continue to be viewed as a stigmatizing art form, and they are inherently unique from person to person.
In organizations characterized by high levels of integration of differences, employees may feel as though they are able to display their true selves and enact core identities without fear of reprisal from fellow employees because norms and expectations dictate that all employees will be treated with respect and dignity. Furthermore, when an organizational climate is characterized by the acceptance and integration of the differences in others, there is less likelihood that interpersonal interactions will be negatively affected by perceptions of status differentials (DiTomaso, Post, & Parks-Yancy, Reference DiTomaso, Post and Parks-Yancy2007; Nishii, Reference Nishii2013), as such status differentials are less likely to develop (Shore, Cleveland, & Sanchez, Reference Shore, Cleveland and Sanchez2018). A favorable inclusion climate may reduce the effect of visible tattoos on unwanted sexual attention because individuals are less likely to negatively stereotype tattoos and those adorned with them in an environment that is appreciative of individual differences. Others may also be inclined to perceive visible tattoos in a positive light, as a signal of individuality and creativity to be valued (Dickson, Dukes, Smith, & Strapko, Reference Dickson, Dukes, Smith and Strapko2015). Furthermore, in a favorable inclusion climate, there is reinforcement for treating individuals with dignity and respect, whereby a potential perpetrator may be less apt to harass even when motivated to do so. Thus, a favorable perceived inclusion climate could reduce the likelihood that individuals with visible tattoos will receive unwanted sexual attention.
Hypothesis 3
A more favorable perceived inclusion climate will reduce the likelihood that individuals with visible tattoos will receive unwanted sexual attention, and a less favorable perceived inclusion climate will increase the likelihood.
Method
Sample and procedure
A sample of 417 working individuals in the U.S. was recruited from Qualtrics business panels to complete an online survey to provide data for the focal constructs. Specifically, the sample was comprised non-management and non-supervisory individuals employed in retail and restaurant settings. Non-management and non-supervisory employees were selected for this research because such individuals are more likely to be subject to harassment given that they have less power in the organizational hierarchy. Individuals employed in restaurant and retail settings were targeted as these industries have been estimated to account for the highest incidences of unwanted sexual attention (Frye, Reference Frye2017, November 20). The survey was framed and marketed to potential participants as a survey about individuals' experiences at work with no mention of tattoos and sexual harassment to help eliminate selection bias. The participants had been employed with their current employers for 3.93 years (SD = 5.08). Their average age was 33.52 years old (SD = 14.41). Fifty-five percent of the respondents were female, and 76 percent were Caucasian.
Measures
Unwanted sexual attention
Six items from the unwanted sexual attention dimension of Fitzgerald, Gelfand, and Drasgow's (Reference Fitzgerald, Gelfand and Drasgow1995) revised Sexual Experiences Questionnaire were used for the present study. Sample items included: Someone at work attempted to discuss sex with you when you didn't want to and Someone at work gave you unwanted sexual attention. A seventh item (Someone at work attempted to stroke or fondle you) was not used because it was thought the item was too intrusive. The respondents were instructed to indicate how frequently each behavior occurs on the job using a five-point scale with anchors of 1 = never and 5 = almost daily. The internal consistency reliability estimate for the scale was .92.
Although unwanted sexual attention was measured as an interval variable in the survey, the variable was dichotomized for the analyses. The scale scores ranged from 1.00 to 5.00, but the distribution was extremely right, or positively, skewed (skewness = 2.45, M = 1.40, Mdn = 1.00). Fifty-seven percent of the sample reported no incidences of unwanted sexual attention, a score of 1.00. This low base rate of harassment is consistent with previous research (Ilies, Hauserman, Schwochau, & Stibal, Reference Ilies, Hauserman, Schwochau and Stibal2003). The variable was dichotomized for purposes of the analyses because typical transformations recommended for right skewness failed to normalize the distribution or make it symmetrical. For example, after a natural log transformation, the data were still highly skewed (skewness = 1.64). For the dichotomization, the variable was recoded 0 for those who reported no experience of unwanted sexual attention and 1 for those who reported any experience of unwanted sexual attention. After this recoding, the unwanted sexual attention scale can be interpreted as whether or not unwanted sexual attention was received.
Tattoos
To measure tattoo visibility, the participants reported whether they have tattoos that are visible to others at work. In addition, the participants reported whether they have tattoos of any kind, whether visible or non-visible. This variable was included in the analysis to isolate the effect of having a visible tattoo on unwanted sexual attention from having a tattoo in general. For both variables, responses were coded 1 = yes and 0 = no. Sixty-four percent of the participants had a tattoo (n = 268); whereas thirty-eight percent (n = 160) of the participants had a tattoo that was visible to others in the workplace.
Perceived sexual harassment climate
Estrada et al.’s (Reference Estrada, Olson, Harbke and Berggren2011) nine-item scale was used to measure sexual harassment climate. Sample items included: A sexual harassment complaint would not be taken seriously by my employer and At my workplace, individuals who sexually harass others get away with it. The items were accompanied by a five-point response scale with anchors of 1 = strongly disagree and 5 = strongly agree. The internal consistency reliability estimate for the scale was .89.
Perceived inclusion climate
Nishii's (Reference Nishii2013) six-item measure was used to assess inclusion climate. Sample items included: Employees in my workplace are valued for who they are as people, not just for the jobs that they fill and My workplace is characterized by a non-threatening environment where people can reveal their true selves. The items were accompanied by a five-point response scale with anchors of 1 = strongly disagree and 5 = strongly agree. The internal consistency reliability estimate for the scale was .89.
Control variables
Four control variables were used in the analyses: tenure, age, ethnicity, and gender. Tenure, measured in years with one's current employer, was included because employees may be more likely to leave a job when confronted with unwanted sexual attention. Although decades old research failed to find a relationship between job tenure and harassment (Gruber & Bjorn, Reference Gruber and Bjorn1982; Stedham & Mitchell, Reference Stedham and Mitchell1998), the relationship is more plausible in the current environment of the #Me Too Movement. The three demographic variables were included because they may be related to both adornments with tattoos and the dependent variable. Previous research has found that younger individuals, non-Caucasians, and males are more likely to bear tattoos (Heywood et al., Reference Heywood, Patrick, Smith, Simpson, Pitts, Richters and Shelley2012; Laumann & Derick, Reference Laumann and Derick2006). These groups are also more likely to rebel, and rebelliousness is one of the most frequently cited motives for acquiring a tattoo (Camacho & Brown, Reference Camacho and Brown2018). Moreover, younger individuals, non-Caucasians, and females have been found to be the most likely victims of unwanted sexual attention (Kearl, Reference Kearl2018; Leskinen, Cortina, & Kabat, Reference Leskinen, Cortina and Kabat2011; O'Connell & Korabik, Reference O'Connell and Korabik2000). These groups tend to be less powerful, and harassment increases as target power declines. Gender was coded 1 = female and 0 = male. Ethnicity was coded 1 = Caucasian and 0 = for other ethnicities.
Confirmatory factor analysis
In order to assess the factor structure and discriminant validity of the two climate variables, we conducted confirmatory factor analysis in AMOS 24. Following recent research (e.g., Cooper, Kong, & Crossley, Reference Cooper, Kong and Crossley2018), we constructed item parcels for use as indicators of the latent climate constructs (Landis, Beal, & Tesluk, Reference Landis, Beal and Tesluk2000; Little, Cunningham, Shahar, & Widaman, Reference Little, Cunningham, Shahar and Widaman2002). The nine perceived sexual harassment climate items were randomly distributed into three parcels, and the six perceived inclusion climate items were also randomly distributed into three parcels (Matsunaga, Reference Matsunaga2008). Results indicated that the hypothesized two-factor measurement model fit the data well, (χ2 = 21.03, p < .05, df = 8, CFI = .99, TLI = .98, RMSEA = .06, SRMR = .03). A single-factor measurement model demonstrated significantly poorer model fit (χ2 = 299.76, df = 9 [Δχ2 = 278.73, df = 1, p < .01], CFI = .80, TLI = .70, RMSEA = .28, SRMR = .11). These results support the discriminant validity of the two climate constructs.
Analytic strategy
Binary logistic regression was used to assess the impact of the independent variables on unwanted sexual attention. The independent variables were entered into the regression in three blocks, resulting in three models. Model 1 included the control variables. Model 2 included the addition of main effects for the tattoo variables, perceived sexual harassment climate, and perceived inclusion climate. Finally, Model 3 added the visible tattoo × perceived sexual harassment climate and visible tattoo × perceived inclusion climate interaction terms. The coefficient for the visible tattoo in Model 2 was used to test Hypothesis 1. The coefficients for the interaction terms in Model 3 were used to test Hypotheses 2 and 3.
Results
Table 1 presents the descriptive statistics and correlations among the study variables. Table 2 presents the results of the logistic regression analyses predicting unwanted sexual attention. The significant χ2 for each model indicates that at least one predictor in each model was nonzero. The Nagelkerke R 2s for Models 1, 2, and 3 are .14, .33, and .36, respectively. The χ2Δs were significant both from Model 1 to Model 2 (73.73, p < .01) and from Model 2 to Model 3 (10.17, p < .01), indicating that the contribution of each additional set of variables accounted for a statistically significant increase in explained variance. The main entries in Table 2 are logistic coefficients, standard errors, and exponentiated logistic coefficients, exp(b). For the exponentiated logistic coefficients, values exceeding 1 indicate an increased effect on unwanted sexual attention, while values below 1 indicate a decreased effect. Exp(b) may be interpreted as the proportional change in the odds of experiencing unwanted sexual attention per unit change in the independent variable. Of the control variables, age and gender were significant predictors in the analyses. As the age of the respondents increased, the odds of receiving unwanted sexual attention declined (Model 1: b = −.05, p < .01; Model 2: b = −.06, p < .01; Model 3: b = −.06, p < .01). With respect to gender, the odds of females experiencing unwanted sexual attention were greater than for males (Model 1: b = .47, p < .05; Model 2: b = .44, p < .05; Model 3: b = .52, p < .05).
Table 1. Descriptive statistics and correlations among variables
Note. n = 417. Ethnicity: 1 = Caucasian and 0 = other. Gender: 1 = female and 0 = male. Correlations with unwanted sexual attention, ethnicity, gender, tattoo, and visible tattoo are point biserial. Significance levels reflect two-tailed tests.
*p < .05 **p < .01.
Table 2. Logistic regression predicting unwanted sexual attention
Note. n = 417. Ethnicity: 1 = Caucasian and 0 = other. Gender: 1 = female and 0 = male. Significance levels reflect two-tailed tests.
*p < .05 **p < .01.
Hypothesis 1, which proposed that individuals with a visible tattoo would be more likely to receive unwanted sexual attention, was supported. The logistic coefficient of .88 for visible tattoos in Model 2 was significant at the .01 level. Holding constant the other variables in the model, possessing a visible tattoo increased the odds of experiencing unwanted sexual attention on average by a multiplicative factor of 2.42.
Hypothesis 2, which proposed that a less permissive perceived sexual harassment climate would decrease the likelihood that individuals with visible tattoos would receive unwanted sexual attention and that a more permissive climate would increase the likelihood, was not supported. The interaction between visible tattoos and perceived sexual harassment climate was non-significant (b = .18, p > .05) in Model 3.
Hypothesis 3, which proposed that a more favorable perceived inclusion climate would decrease the likelihood that individuals with visible tattoos would receive unwanted sexual attention and that a less favorable perceived inclusion climate would increase it, was supported. The logistic coefficient of −.77 for the visible tattoo × perceived an inclusion climate interaction term in Model 3 was significant at the .01 level. Holding constant the other variables in the model, a favorable inclusion climate decreased the odds of experiencing unwanted sexual attention for employees with visible tattoos on average by a multiplicative factor of .46. This interaction is presented graphically in Figure 1. As illustrated in the left plot, when employees did not have visible tattoos, inclusion climate had almost no effect on unwanted sexual attention. However, as illustrated in the right plot, when employees did have visible tattoos, inclusion climate decreased unwanted sexual attention.
Figure 1. Interaction between visible tattoo status and perceived inclusion climate predicting unwanted sexual attention
Perceived inclusion climate was found to significantly interact with visible tattoos in predicting unwanted sexual attention, but not perceived sexual harassment climate. One reason for the non-significant sexual harassment climate × visible tattoo interaction may have been the moderate correlation between the two climate variables; whereby the shared variance may have suppressed the effect of sexual harassment climate. To further explore this possibility, a post hoc analysis was conducted in which Model 3 was re-estimated separately for each climate variable. The variance that sexual harassment climate formerly shared with inclusion climate could then be attributed to only sexual harassment climate. See Table 3, where the results for sexual harassment climate are presented on the left side of the table and the results for inclusion climate are presented on the right side. The interaction between inclusion climate and visible tattoos was again significant (b = −.88, p < .01). Although the logistic coefficient for the interaction between sexual harassment climate and visible tattoos was larger than it was in the primary analysis, the interaction was still non-significant (b = .56, p > .05).
Table 3. Post hoc logistic regression predicting unwanted sexual attention with moderators in isolation
Note. n = 417. Ethnicity: 1 = Caucasian and 0 = other. Gender: 1 = female and 0 = male. Significance levels reflect two-tailed tests.
*p < .05 **p < .01.
A second post hoc analysis was performed to ascertain whether there were gender differences in the visible tattoo-unwanted sexual attention relationship. To do so, gender was interacted with the focal variables in the primary logistic analysis. This analysis was conducted in two stages, resulting in two models. (The control variables and main effects for the focal variables were not included in separate models because the coefficients would have been identical to those in the primary analysis.) Model 1 included the control variables, main effects, and five two-way interactions, including three new gender interaction terms: gender × visible tattoo, gender × perceived sexual harassment climate, and gender × perceived inclusion climate. Model 2 included the addition of two three-way interaction terms: gender × visible tattoo × perceived sexual harassment climate and gender × visible tattoo × perceived inclusion climate. The results are presented in Table 4. In Model 1, none of the gender interaction terms were significant, but the visible tattoo × inclusion climate interaction term continued to be significant (b = −.90, p < .01). In Model 2, none of the gender interaction terms were significant, and the visible tattoo × perceived inclusion climate interaction term was also non-significant. These results thus indicate that gender did not significantly interact with having a visible tattoo in predicting unwanted sexual attention. Further, the inclusion of the gender interactions may obscure other relationships.
Table 4. Post hoc logistic regression predicting unwanted sexual attention with gender interactions
Note. n = 417. Ethnicity: 1 = Caucasian and 0 = other. Gender: 1 = female and 0 = male. Significance levels reflect two-tailed tests.
*p < .05 **p < .01.
Discussion
Tattoos remain somewhat controversial in today's workplace, and unwanted sexual attention persists. This study has demonstrated that there is a link between these two pressing issues, and it has contributed to both the literature on tattoos in the workplace and the literature on sexual harassment. This study has expanded the body of research on tattoos in the workplace by demonstrating that the effects of tattoos extend to actual adverse treatment on the job in the form of unwanted sexual attention. At the same time, this study has contributed to the sexual harassment literature by demonstrating both that tattoos, a characteristic of employees that is not explicitly related to gender or sex, may lead to increased unwanted sexual attention and that inclusion climate may reduce such attention.
As hypothesized, this study demonstrated the positive relationship between visible tattoos on employees and unwanted sexual attention, a relationship that has not been explicitly considered and empirically tested heretofore. Previous workplace tattoo research has primarily addressed others' perceptions of individuals with tattoos, such as views on employment suitability, coworker preferences, and customer preferences (e.g., Baumann, Timming, & Gollan, Reference Baumann, Timming and Gollan2016; Miller, McGlashan Nicols, & Eure, Reference Miller, McGlashan Nicols and Eure2009; Timming, Reference Timming2015; Timming et al., Reference Timming, Nickson, Re and Perrett2017). Extending such research, the present study has further demonstrated that tattoos are not an innocuous form of self-expression and that individuals with visible tattoos may be subject to more maltreatment once on the job. This relationship held similarly to both females and males. Notwithstanding the validity of the findings herein, future research would be valuable that examines why the visible tattoo-unwanted sexual attention relationship may manifest itself, drawing on the automatic stereotype activation and attention-getting arguments. That is, research is warranted that examines whether unwanted sexual attention associated with visible tattoos is more a function of negative stereotypes or the attention paid to visible tattoos.
The present research highlights that specific aspects of organizational climate may moderate the visible tattoo-unwanted sexual attention relationship. The significant effect of the interaction of perceived inclusion climate with visible tattoos highlights that the appreciation of individual differences is important in curbing unwanted sexual attention stemming from visible tattoos. In such an accepting climate, individuals may be less apt to activate negative stereotypes about those with tattoos, view individuals with tattoos more positively, and be more restrained in making suggestive comments about physical appearance. Furthermore, in such a climate supportive of individual differences, those adorned with visible tattoos may be less likely to construe comments related to tattoos as unwanted sexual attention, since individuals may have greater trust in coworkers in such an environment.
Counter to expectations, perceived sexual harassment climate did not moderate the visible tattoo-unwanted sexual attention relationship. In both the primary and post hoc analyses, the interaction term with visible tattoos was in the hypothesized direction, but was non-significant. One possible explanation for the non-significant effect is that sexual harassment climate is conceived of and measured as a pervasive context for harassment regardless of individual attributes. Consequently, sexual harassment climate may be too broad a construct to reflect unwanted sexual attention as a function of individual differences such as visible tattoos. Another explanation is that sexual harassment climate may not be perceived as germane to governing behavior concerning attention paid to tattoos if the initiator does not view such attention as sexual. As discussed beforehand, unwanted sexual attention may be inadvertent due to the attention-getting nature of visible tattoos. If tattoo-related attention is believed by an initiator to be non-sexual, penalties and consequences associated with sexual harassment are arguably irrelevant to self-monitoring. Even though sexual harassment climate was not found to be a significant moderator, it should be highlighted that it did have a significant main effect on unwanted sexual attention, consistent with previous research (Willness, Steel, and Lee, Reference Willness, Steel and Lee2007).
Practical implications
With respect to practical implications, at least two options could be considered. One strategy for limiting unwanted sexual attention based on visible tattoos is to foster a more favorable inclusion climate, where individual differences are valued and appreciated rather than suppressed. Such a climate is beneficial for organizations not only because it may curb unwanted sexual attention, but may more generally foster favorable workplace attitudes among employees (Dwertmann, Nishii, & Van Knippenberg, Reference Dwertmann, Nishii and Van Knippenberg2016). Instituting formal diversity programs and ensuring the existence of supportive and diverse leadership may be viable strategies to help create an inclusive climate (Herdman & McMillan-Capehart, Reference Herdman and McMillan-Capehart2010; Shore, Cleveland, & Sanchez, Reference Shore, Cleveland and Sanchez2018). A second strategy to limit unwanted sexual attention is to have a non-visible tattoo policy, where employees are required to cover any potentially visible tattoos while on the job. At the same time, a disadvantage of implementing such a policy is that it runs counter to promoting an inclusive climate. It could be construed as constraining employees' freedom of expression and may prove to be an obstacle to recruiting employees in tight labor market conditions. Moreover, such a policy could be interpreted as attributing blame for harassment to the targets, which is likely not a message that organizations would seek to convey. In this regard, non-visible tattoo policies should be carefully considered. The results herein should not be interpreted as blaming the victim. Rather, these results should be interpreted as highlighting the susceptibility of those with visible tattoos to unwanted sexual attention and the need for organizations to ensure that appropriate preventive and corrective measures are in place to minimize its occurrence.
Limitations
The results of this study should be interpreted in the context of its limitations. One limitation is that the data on visible tattoos and unwanted sexual attention were obtained from a sample of restaurant and retail employees. While this sample was deemed appropriate for this area of inquiry given the relatively high rates of unwanted sexual attention in restaurant and retail environments (Frye, Reference Frye2017, November 20), research would be valuable with employees in other industries. For example, research would be valuable that examines the effects of visible tattoos on unwanted sexual attention in professional occupations (e.g., law and medicine) and progressive high-tech environments (e.g., social media and information technology). Another limitation is that the data were obtained from employees only in the U.S. Since interpretations associated with tattoos could vary in different cultural contexts, along with workplace norms regarding sexually oriented comments, research on visible tattoos and unwanted sexual attention would be fruitful beyond the U.S. and other Western nations. A final limitation is that the data on visible tattoos and unwanted sexual attention were obtained at one point in time. Although the direction of causality is in little doubt, further research would be valuable that obtains data collected at multiple points in time to more fully substantiate cause-and-effect relationships.
Future research
Beyond those suggested previously, several additional opportunities for future research are worth pursuing. One opportunity is focusing on different types of visible tattoos in relationship to unwanted sexual attention. As tattoos vary in content, size, and location on the body, it would be of interest to evaluate the extent to which factors such as these moderate the visible tattoo-unwanted sexual attention relationship. Such research would involve focusing on an entire sample of tattooed individuals with different tattoo characteristics, unlike the present study that focused on differences between tattooed and non-tattooed individuals. Another opportunity is analyzing the effects of visible tattoos on female versus male employees. Although no significant differences were found in the present study, other factors could be considered in the visible tattoo-gender-unwanted sexual attention relationship, such as the gender composition of the workplace and employee sexual orientation. Future research would also be valuable that analyzes whether the visible tattoo-unwanted sexual attention relationship differs by the perpetrator. For example, research could assess whether effects are similar across supervisors, coworkers, and customer perpetrators. On a positive front, research could also examine the relationship between tattoos and favorable workplace outcomes. To the extent that those with tattoos are perceived as creative and outgoing, such individuals could be targeted for innovative projects and assignments.
Conclusion
Tattoos are increasingly prevalent. Individuals obtain tattoos for a wide variety of reasons, but certainly one of them is not to receive unwanted sexual attention in the workplace. The extant body of research has demonstrated that tattoo stigmatization persists in organizations, and the present study has signaled that visible tattoos may have yet another unintended negative consequence for individuals. The current research was not intended as a definitive test of the potential impact of visible tattoos on workplace unwanted sexual attention, but rather a first step toward examining this issue. It is hoped that the findings from this study are thought provoking and will stimulate further research in this area.
