Introduction

“Slut-shaming” is the act of humiliating a woman based on presumed sexual behavior and appearance, regardless of whether or not she is sexually active [33]; the term “slut” implies that the target lacks value and morality [6]. However, historically, it was a term used primarily against lower-class women as a tool for shaming [3]. In the eighteenth century, the word was not only used by men to denigrate female servants, but by upper-class women as well [3]. In this context, it was not uncommon for lower-class women working in the homes of the upper-class to be called “sluts” when they were sexually assaulted by male employers. Upper-class women’s use of the word signified their allegiance to their socioeconomic status (SES) over their gender. These dynamics set the stage for this word to be used by women as a pejorative term against other women [3].

Although, in some contexts, men may be disparaged for their sexual behavior (e.g., be referred to as a “man-whore” or “man-slut”), women receive the bulk of “slut-shaming” [3, 6, 37]. “Slut-shaming” as a means of social and sexual control is something girls begin to face as early as middle school [34–36]. Women and girls receive mixed signals about their sexuality beginning early in life; they must be innocent yet sexual, which puts them in a bind regarding their own behaviors and may impact their judgment of peers [1, 14, 34, 36, 37]. One way in which young women navigate these complex social messages is through a woman-on-woman control of female sexuality in the form of “slut-shaming” [3].

Intrasexual competition is a primary theory for understanding why women would “slut-shame” other women [4, 10, 40, 41]. “Slut-shaming” may reduce the “value” of a woman with whom another woman is competing with for a partner. Because the behavior of “good” women is thought to be consistent with gender norms [18], when a woman is called a “slut,” it is an advertisement that she is not a “good” woman and is, therefore, worth less to a potential partner [41]. “Slut-shaming” may be precipitated by female jealousy, in that women want the attention the sexually promiscuous woman receives from men [4]. According to this view, the driving force behind this same-sex aggression is that women attempt to bolster their own desirability by lowering the desirability of other women. This may be enhanced if they feel they will not attract a partner if other women are more ‘available’ [4].

“Slut-shaming” may also result from an internalization of stereotyped norms of gender and a desire to punish women who violate these norms. Women may be shamed for not conforming to feminine norms [10, 24, 33], especially those of both modesty and fidelity, two constructs that signify the valuing of women’s monogamy and sexual abstention [27]. Further, sex is expected to be tied to love or emotion for women, and casual sexual activity suggests there is not an emotional connection driving the action [39]. To the extent that these norms are internalized, other women may believe that “slut-shamed” women deserve it, despite the fact that the normalization of “slut-shaming” puts them at risk for being targeted next.

In sum, “slut-shaming” may be motivated by sexual jealousy and a desire to denigrate a sexual competition. If that is the case, it should be associated with the emotion of “jealousy.” Slut-shaming may also be motivated by a desire to monitor social norms and judge women who violate them. If that is the case, then “slut-shaming” would be motivated by a sense of judgment to a greater extent than a sense of jealousy.

Acceptability of “Slut-Shaming”

Little is known about perceptions of those targeted as a “slut” by others as well as perceptions of those who shame women’s presumed sexual behavior. While “slut-shaming” is common [2], it is considered unkind, and those who engage in such behavior may be perceived negatively by others. One prior study looking at perceptions of an online “slut-shaming” incident illustrated that participants viewed “slut-shamers” negatively, particularly if the target of the humiliation was a woman [31]. Female participants also indicated wanting more social distance from the “shamer” than did male participants, indicating that women may be more judgmental about “slut-shaming.” However, the extent to which one views “slut-shaming” in a positive or a negative light may depend on characteristics of the woman being shamed. Thus, one aim of the current study was to understand what impacts the acceptability of “slut-shaming.”

Dressing in a provocative manner may precipitate “slut-shaming” [31]. Women’s clothing may act as a marker of sexual availability [20], and women dressed in “sexy” attire are often judged harshly [29]. Vaillancourt and Sharma [41] found that a confederate dressed in provocative attire spurred women to lash out with indirect aggression; participants rated a provocatively dressed confederate in their study as “bitchy” and indicated they would not want their boyfriends to spend time with her. We hypothesized that a “slut” who was pictured wearing provocative attire would be judged more negatively than if she was dressed conservatively. Given that wearing provocative clothing may be seen to justify “slut-shaming,” we also hypothesized that a “slut-shamer” would be judged less harshly, and the “slut-shaming” would be seen as more justified, if the woman being shamed were dressed provocatively as opposed to conservatively.

“Slut-shaming” may also still reflect its roots of classism [2, 3]. Socio-economic status is characterized by wealth, education, and occupation [25], and research has shown that the way people are perceived may be impacted by their social standing [11]. We hypothesized that the “slut” would be viewed more negatively if she were of a lower SES, a “slut-shamer” would be judged less harshly if the target appeared to be of lower social status, and the shaming incident would be seen as more justified if it was targeted against a woman of a lower social class. We also hypothesized an interaction between attire and SES, such that a woman who was both dressed provocatively and of a lower SES would be judged more harshly than a high SES, conservatively dressed woman.

Feminist Identity

Further, we were particularly interested in whether feminist identity may serve as a protective mindset against endorsement of “slut-shaming.” Feminist blogs coined the term “slut-shaming” in order to point out and problematize the fact that women tend to be valued and denigrated based on their sexuality [48]. The feminist movement has broadly encouraged women’s sexual freedom outside feminine gender roles [15, 22], an example of which is the reclamation of the word “slut” exemplified by “Slut Walks” [37]. Further, prior research has found a link between women who hold feminist ideology and sexual well-being and subjectivity [38]. We hypothesized that self-identified feminists would be less disapproving of the “slut” than would non-feminists, would disapprove of the “slut-shamer” more than would non-feminists, and would see the “slut-shaming” incident as less justified than non-feminist participants.

Although, as discussed above, women have been calling other women “sluts” for centuries, the rise and widespread adoption of social media has allowed women to be victimized in new ways as they have public or semi-private online profiles [7, 9, 13, 26, 28]. One study reported that 20% of women surveyed had experienced cyber harassment [46]. The Pew Research Center [32] reported that young adults, and, more specifically, young women, are most often the targets of internet harassment, citing 50% of young, female respondents have been called offensive names online. One researcher documented the Twitter hashtag #mencallmethings, with which participants tweeted examples of online harassment they receive from men [28]; this harassment includes death and rape threats, “slut-shaming,” and appearance-related insults [7, 23, 28]. While public, online harassment allows online ‘bystanders’ to know how the perpetrator feels about the target, it is unclear how those who witness such abuse feel about the offender and victim.

Understanding Tone

Finally, we were interested in exploring how our participants interpreted the tone of the “slut-shaming” incident. Given that comments made online cannot express a tone of voice, we were interested in the extent to which our participants saw a “slut-shaming” comment as serious, rather than as a lighthearted jab. We also wished to explore whether, overall, a “slut-shaming” comment would be seen as jealous (indicating that it was seen as being motivated by intersexual competition) or judgmental (indicating it was seen as being motivated by a desire to monitor conformity to gender roles).

Method

Participants

We recruited 186 female college students in the United States to complete our survey. On average, participants were 19.46 years old (SD = 1.64). The majority of our participants self-identified as White/Caucasian (80.2%). Participants also identified as Multiracial (5.3%), Asian/Pacific Islander (4.8%), Latina (3.2%), African American/Black (2.1%), and “other” (2.1%); an additional 2.1% chose not to disclose their ethnicity. The majority of participants also identified as heterosexual/straight (86%); participants also identified as bisexual (8%), homosexual/gay/lesbian (2%), and “other” (4%). Additionally, most participants self-identified as middle class (48%) or upper-middle (35%) class; the rest identified as working class (13%), wealthy (3%), or poor (1%). More than half of our participants (60%) identified as non-feminists while the rest (40%) identified as feminists.

Procedure

Participants over the age of 18 who were enrolled in college were recruited using the psychology department subject pool consisting of students. Students were recruited from general psychology classes. They took the survey in a lab in the psychology building, and were spaced apart from one another such that it would be difficult for participants to see the screens of other students. They received credit for participating. The study was described as aiming to learn more about how people make judgments based on what they see on Facebook. The posted study link, hosted through Surveygizmo.com, took participants to an online informed consent. Upon giving consent, participants were randomly placed into one of the following four conditions: conservative attire/high SES, conservative attire/low SES, provocative attire/high SES, provocative attire/low SES. After viewing the randomly assigned Facebook screenshot, participants answered questions about their perceptions of the people involved based on their interpretations of the conversation. The survey was the same for all participants. After completing the survey, participants were taken to an online debriefing statement.

Materials

Screenshots of the Facebook Conversations

Facebook conversations were created by the researchers using fictitious Facebook accounts. Each conversation included a photo, a photo caption, and a single comment. In all cases, the status update and the comment were identical. The status read, “So glad Jessica could join us last night [heart]” with the picture attached, and the comment read, “saw her last night…such a slut.” The posts were set up such that it was clear that the post was on the page of “Jessica,” the woman pictured. Last names, profile photos, and dates were blocked to create the illusion that we were protecting the privacy of real Facebook users. The face of “Jessica” was also blurred, but we did not block her employment and education information pictured next to her timeline.

This information was used to provide participants with information about “Jessica’s” socioeconomic status. The high SES target was an intern at a law firm, studied at a well-known, academically challenging college, and attended a private high school. The low SES target was a waitress at a diner, studied at a community college, and attended a public high school. In both photos, “Jessica” is a White, college-aged woman. Her face is not shown, therefore, age would be estimated by participants based on provided biographical information. The photo either showed “Jessica” dressed provocatively (i.e., in a tank top with her bra partially exposed and her midriff visible) or conservatively (i.e., midriff covered and a short-sleeved shirt that showed no cleavage). This resulted in four possible combinations: conservative attire/high SES, conservative attire/low SES, provocative attire/high SES, provocative attire/low SES (see Figs. 1 and 2 for examples).

Fig. 1
figure 1

Image from the conservative attire/low SES condition

Full size image
Fig. 2
figure 2

Image from the provocative attire/high SES condition

Full size image

Person Perception

Participants evaluated both the “slut” (Jessica) and the “shamer” through a semantic-differential measure of person perception [17]. We used one subscale from this measure to assess general evaluation (e.g., “bad–good”). These 12 items were responded to using a 7-point scale (−3 to 3) such that higher scores were associated with more positive evaluations. The Cronbach’s alpha in the study in which the measure was developed was .77. For the current study, “slut” evaluation was .92, and “shamer” evaluation was .78.

Social Distance

We used a measure developed for a previous study [31] to measure social distance. This 15-item measure was used to assess people’s willingness to participate in common social interactions for college students with varying degrees of closeness (e.g., “Would you study with this person?” or “Would you set this person up on a date with a close friend or family member?”). The participants indicated their willingness to engage in these activities using a 6-point scale ranging from 1 (Definitely Yes) to 6 (Definitely No) such that higher scores indicated a greater desire for social distance and a lower willingness to affiliate. In the study in which the measure was developed, the Cronbach’s alpha for the “slut” target was .95, and it was .96 for the “shamer” target. For the current study, the Cronbach’s alpha for the “slut” target was .94, and it was .96 for the “shamer” target.

Perception of Justification

We were interested in whether the participants’ believed the “slut-shaming” comment was justified in the context shown. Participants were asked the extent to which they believed Ashley’s (the “shamer”) comment was justified on a 6-point scale ranging from 1 (Not At All) to 6 (Very).

“Shamer’s” Tone

We asked the participants to report their perceptions of the “shamer’s” tone in the conversation. We instructed participants to do so because it can be difficult to decipher tone from text-based conversations [45], and we hoped to gain insight into how this interaction may have been perceived differently depending on the characteristics of the “slut.” The participants were asked to rate the extent to which they believed the “shamer” was jealous, judgmental, and serious, and they responded on a 7-point scale ranging from 1 (Not At All) to 7 (Very Much So).

Feminist Identity

Participants responded yes (coded 1) or no (coded 0) to the item, “Do you consider yourself to be a feminist?”

Results

In order to determine how the Facebook conversation was perceived, we ran three 2 × 2 × 2 MANOVAs: one to assess perceptions of the “slut” (on both general evaluation and social distance), one to assess perceptions of the “shamer” (on both general evaluation and social distance), and one to assess the perceived tone of the conversation (assessing whether the tone was perceived as jealous, judgmental, and serious). In order to test whether the comment was viewed as more judgmental or more jealous, we ran a repeated measure ANOVA to determine which of these tone variables were more strongly endorsed. Finally, we ran one 2 × 2 × 2 ANOVA in order to answer the overarching question of whether the participants felt as though the slut-shaming incident was justified. For each analysis, the three independent variables were: “slut” attire, “slut” socioeconomic status, and participant feminist self-identification.

Perception of “Slut”

The first MANOVA assessed perceptions of the “slut;” the dependent variables were social distance and general evaluation. The three-way interaction was not significant. Similarly, two of the two-way interactions were not significant (feminist identity × “slut” SES and feminist identity × “slut” attire). However, there was a significant two-way interaction of “slut” attire and SES (see Fig. 3). See Table 1 for the MANOVA results of perceptions of the “slut.”

Fig. 3
figure 3

Mean values of desired social distance from the “slut”

Full size image
Table 1 MANOVA results for perceptions of “slut” and “shamer”
Full size table

Examination of the univariate ANOVAs for “slut” attire and SES indicated a significant difference for social distance but not for general evaluation (see Table 2 for univariate results, means, and standard deviations). Follow-up tests indicated that participants who were shown an image of a high SES, provocatively dressed target reported wanting more social distance from the “slut” than did those who saw an image of a high SES, conservatively dressed “slut.” This indicates that between two women who were high SES, the provocatively dressed “slut” was less socially desirable than the conservatively dressed “slut.” Attire did not significantly impact desired social distance in the low SES conditions. There were also no significant differences in desired social distance for high and low SES targets in either attire condition.

Table 2 Descriptive statistics and univariate ANOVA results for perceptions of the “slut” and “shamer” and the tone of “shamer’s” comment
Full size table

We also found a significant main effect for social distance based on feminist identity. The univariate results, F(1, 177) = 15.59, p < .001, η 2partial  = .08, indicated that participants who identified as non-feminist reported wanting significantly more social distance from the “slut,” regardless of attire or SES (M = 3.48, SD = 0.86), than did the self-identified feminists (M = 2.94, SD = 0.94). Feminist identity increased women’s comfort with being socially close to a woman who had been called a “slut.”

Perception of “Shamer”

The second MANOVA assessed perceptions of the “shamer” in terms of social distance and general evaluation (see Table 1). Once again, the three-way interaction was not significant. Similarly, none of the two-way interactions were significant. Furthermore, there were no significant main effects based on SES or feminist identity. However, there was a significant main effect for “slut” attire.

Examination of the univariate ANOVAs for the “slut” attire condition on the perception of the “shamer” indicated that there was only a significant main effect for social distance (see Table 2 for univariate results, means, and standard deviations). Follow-up tests indicated participants were willing to be significantly closer to the person who shamed the provocatively dressed target than they were to the person shaming the conservatively dressed target. Our participants were more comfortable with the idea of socializing with someone who “slut-shamed,” when the person being shamed was wearing a provocative outfit.

Assessment of “Shamer” Tone

A third MANOVA was conducted in order to assess perceptions of the tone of the “shamer’s” comment. The three independent variables were, again, “slut” attire, “slut” SES, and participant feminist self-identification. The three dependent variables were perceptions that the tone was jealous, judgmental, and serious. Akin to the first two MANOVAs, the three-way interaction was not significant. Further, two of the two-way interactions were not significant (feminist identity × “slut” attire and feminist identity × “slut” SES). However, there was a significant two-way interaction of “slut” attire and SES (see Fig. 4). Finally, there was no significant main effect for feminist identity. See Table 3 for the MANOVA results of perceptions of the “shamer’s” tone.

Fig. 4
figure 4

Mean values of perceived seriousness of the tone of the “slut-shaming” interaction

Full size image
Table 3 MANOVA results for perceptions of “Shamer” tone and ANOVA results for the perceived justification of “Shamer’s” comment
Full size table

Examination of the univariate analyses indicated that there was a significant interaction for serious tone but not for judgmental or jealous tone (see Table 2 for univariate results, means, and standard deviations). If the participant saw the profile of the provocatively dressed target, her interpretation of the comment as serious did not differ between SES conditions. However, if she was shown the photo of the conservatively dressed target, she thought the tone of the comment was significantly more serious when directed toward the low SES target than toward the high SES target. Results suggest that a “slut-shaming” incident involving a high SES woman may be more likely to be considered light-hearted, or a joke, than it would be if it involved a low SES woman. There was not a significant difference in perceptions of the tone as serious based on attire within the low SES condition. Participants did, however, significantly differ on their interpretation of serious tone based on attire in the high SES condition. If the high SES target was dressed provocatively, the “slut-shaming” comment was considered more serious than if she were dressed conservatively.

Due to the participants’ differing interpretation of the “shamer’s” tone as serious, we re-ran our MANOVAs assessing perceptions of the “slut” and “shamer” to see if including participants’ perceptions of the tone as serious impacted our results. The results remained largely unchanged, with the exception of the general evaluation of the “shamer.” With serious tone as a covariate, the general evaluation of the “shamer” based on “slut” attire met traditional standards for statistical significance, F(1, 175) = 4.75, p = .03, η 2partial  = .03. The pattern of means is the same as our original analyses indicated. With serious tone as a significant covariant, the “shamer” received the lowest evaluation when she commented on the photo of the conservatively dressed, high SES “slut” and the highest evaluation when she commented on the photo of the provocatively dressed, high SES “slut.”

In order to determine whether, overall, participants viewed “slut-shaming” as more motivated by a sense of jealousy (indicating sexual competition) or by judgment (indicating a concern with norm violation), we ran a repeated measures ANOVA specifically looking at jealous and judgmental tones. Overall, judgmental tone (M = 5.62, SD = 0.90) was endorsed to a higher degree than was jealous tone (M = 4.20, SD = 1.30), F(1, 186) = 203.37, p < .001, η 2partial  = .52, although the mean levels of endorsement for both tones was above the midpoint of the scale.

The final analysis assessed whether the “shamer’s” comment was justified. We ran a 2 × 2 × 2 ANOVA with the same independent variables of “slut” attire, “slut” SES, and participant feminist self-identification (see Table 3). The three-way interaction was not significant. Further, two of the two-way interactions were not significant (“slut” attire × “slut” SES and “slut” SES × feminist identity). The third two-way interaction of “slut” attire and feminist identity was significant (see Fig. 5).

Fig. 5
figure 5

Mean values of participants’ belief the “slut-shaming” comment was justified

Full size image

Follow-up analyses indicated no significant difference between feminists and non-feminists in their assessment of whether the comment was justified within the conservative attire condition, F(1, 78) = 0.08, p = .77, η 2partial  = .001. However, if the target was dressed provocatively, non-feminists thought the “shamer’s” comment was significantly more justified (M = 1.85, SD = 1.09) than did feminists (M = 1.38, SD = 0.59), F(1, 103) = 6.3, p = .01, η 2partial  = .06. Within the group of self-identified feminists, there was no significant difference regarding perception of justification based on attire, F(1, 73) = 0.73, p = .40, η 2partial  = .01. Within the group of non-feminists, the comment about the provocatively dressed target was perceived as more justified (M = 1.85, SD = 1.09) than was the comment aimed toward the conservatively dressed target (M = 1.22, SD = 0.47), F(1, 108) = 12.96, p < .001, η 2partial  = .11. Overall, the consistent low scores regarding justification indicated the low acceptability of the incident.

Discussion

The purpose of this research was to build upon prior research of perceptions of “sluts” and “slut-shamers” on social media by varying the socioeconomic status and attire of our target “slut” and examining the role of feminist identity of the participant. This study focused on how female participants reacted when presented with one woman “slut-shaming” another woman. Our hypotheses regarding social class were not supported, but our hypotheses regarding dress and participants’ feminist identity were supported.

We hypothesized that a “slut” with low SES/provocative attire would be judged more harshly than her high SES/conservative attire counterpart. Unexpectedly, our interaction illustrated that participants wanted more social distance from the high SES, provocatively dressed target, whereas we predicted the low SES target would be evaluated more negatively. This finding may be related to previous research on intrasexual competition [40], such that participants may feel like the wealthy “slut” is more competition because of her social status. Further, expectations are different for high SES individuals than for low SES individuals [11, 30, 43]. The low SES, provocatively dressed “slut” may be perceived as behaving in line with what our participants expect from a low SES woman, whereas the high SES, provocatively dressed target is expected to behave with more ‘class’ [8]. This relationship between dress and class has been supported in prior research, in one study a sexily dressed CEO was seen as having less social competence compared to a sexily dressed assistant [47], and in another study, participants rated a sexily dressed manager as less competent than a manager wearing business casual clothing [19]. The results of both studies suggest that women of higher social status are expected to appear more modest than women of lower social status.

We hypothesized that our participants would perceive the “slut-shamer” more negatively if she shamed a high SES/conservatively dressed “slut” than if she shamed a low SES/provocatively dressed “slut.” Consistent with prior research and our hypothesis regarding perceptions of the “slut-shamer,” the “shamer” was generally perceived negatively [31]. Women wanted the most social distance from the “shamer” who shamed the conservatively dressed target, likely because she was shaming someone who was not as openly violating feminine social norms in terms of her clothing [19].

The perceived tone of the “shamer’s” remark was important in this context as it can be difficult to determine the tone of a comment online [45]. Our participants’ responses indicated they perceived the tone as most serious when directed at the high SES, provocatively dressed target. Perhaps because she is perceived as violating social norms of her class by dressing provocatively, they may have felt she was acting out of line [19, 47]. There was no difference in perceived seriousness of the “shamer’s” tone for low SES targets, which may be explained by participants’ lowered expectations based on SES [11]. “Slut-shaming” was perceived as both jealous and judgmental by our participants, as indicated by rating both tones above the midpoint for the scale. They may have believed that the “shamer” was jealous because she was in competition for a partner [40]; however, participants perceived the tone as more judgmental than jealous [19] indicating a greater concern with social norm violation than sexual jealousy.

Further, we believed that feminist identity might protect some women from endorsing “slut-shaming” behavior regardless of the target’s SES or attire. Our hypotheses about participants’ feminist identity being related to more negative evaluation of the “slut-shamer” and more positive evaluations of the “slut” in comparison with non-feminist participants were generally supported. Feminist participants reporting wanting less social distance from the “slut” than did non-feminist participants. Further, while participants generally viewed “slut-shaming” as unacceptable, feminists indicated that the “shamer’s” comment was less justified than did non-feminist participants when the target was dressed provocatively. Feminists found “slut-shaming” equally unjustified regardless of attire; whereas non-feminists indicated that it was significantly more justifiable to “slut-shaming” a woman wearing provocative attire. When comparing feminists’ to non-feminists’ perception of justification for only the provocatively dressed “slut,” non-feminist women found the “shaming” to be justified significantly more than the feminist-identified women. Thus, self-identification as a feminist does meaningfully shape how one views an online incident of “slut-shaming.” These findings reflected the belief represented in a great deal of feminist theory [5, 22, 42, 44] and research [12] that women should be allowed to enjoy sex and not be judged based on their sexuality.

It is important to understand this study within the context in which it was conducted. Our participants were required to be undergraduates in college at the time of the survey; therefore, we may not have fully tapped into the “slut-shaming” culture since casual sex is not unusual in college (see [16], for a review of ‘hookup culture’). Furthermore, we used a fictitious Facebook account rather than real-life confederates. Making the situation more personal by using confederates may result in stronger feelings, positive or negative, regarding “slut-shaming.” We cannot determine whether the difference between self-identified feminists and non-feminists willingness to associate with the target was truly impacted by our experiment without control conditions in which there is no “slut-shaming” comment. It could be that feminists are more willing to socially affiliate with provocatively dressed women whether or not she has been “slut-shamed.” Researchers who wish to continue investigating this issue should consider the benefits of including a control condition. Additional research could also provide meaningful information about this phenomenon by examining how other demographic factors of a target, such as race, weight, and sexual orientation, may influence the acceptability of “slut-shaming,” as our target was white, thin, and blonde. Participants’ racial, economic, gender, and sexual background may also play a role in how they interpret a “slut-shaming” incident, and a more representative sample may help us understand these relationships better. Future researchers may wish to investigate why women may see “slut-shaming” as either judgmental or motivated by jealousy by including a condition that shows sexually suggestive commentary from the target.

Further, while we know that many women face harassment online [7, 32, 46], there is a dearth of research regarding specific types of online harassment and how they relate to interpersonal bullying. It is important to understand how much harassment is related to “slut-shaming,” be it based on clothing, actual sexual activity, or perceived sexual activity. Women who receive rude comments on their public social media sites may have different emotional responses depending on content or theme of the comments (e.g., calling a woman “stupid” may evoke a different emotional response than calling her a “slut” due to the history, prevalence, and meaning of the term “slut”). It is also important to note that feminists may differ from non-feminists in the way they respond to this harassment due to possible experience with online provocation [21, 28]. Given this, future research would benefit from exploring experiences of and responses to online harassment among both feminist and non-feminist identified women.

Furthermore, feminist work regarding online harassment has centered on man-on-woman abuse with no analysis of woman-on-woman online harassment. This may be because of the level of violence shown against women on- and offline by men [7], and therefore, harassment from men is seen as more threatening. However, our research suggests that feminist self-identification is protective against acceptance of woman-on-woman harassment. Feminist activists may wish to turn additional theoretical attention to the harm caused by women harassing other women online and the ways in which online “slut-shaming” creates an environment where women are more focused on policing each other than on working together to create a safer online world.