SEXUAL HARASSMENT AND CONFIRMATION THEORY: HOW RECIPIENTS PERCEIVE THE DIMENSIONS OF ACCEPTANCE AND CHALLENGE By Allison P. Mazur A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Communication—Master of Arts 2019 ABSTRACT SEXUAL HARASSMENT AND CONFIRMATION THEORY: HOW RECIPIENTS PERCEIVE THE DIMENSIONS OF ACCEPTANCE AND CHALLENGE By Allison P. Mazur With the onslaught of sexual violence scandals involving celebrities and politicians, sexual harassment has recently become a topic of public concern. While much attention is being devoted to victims speaking out about their experiences and holding harassers accountable, limited communication research attention has been devoted to supporting victims of sexual harassment. This manuscript applies Dailey’s (2006) confirmation theory to understand what confirming messages participants in a hypothetical sexual harassment scenario will find most effective, and how the severity of the harassment may affect the reception of confirming messages. Participants (N = 170) in a repeated-measures design were presented with one of three hypothetical sexual harassment scenarios varying in severity and asked to rate how validating, motivating, appropriate, and facilitative of reappraisal they perceived four confirmation messages to be. Results indicated that high acceptance/high challenge messages were most effective across outcomes. Challenge and scenario severity interacted such that low challenge messages were rated more validating in the low severity scenario compared to the moderate severity scenario; low challenge messages were also rated more appropriate in the low severity and high severity scenarios compared to the moderate severity scenario. Participant sex and challenge interacted such that males rated low challenge message more validating, motivating and appropriate than females; whereas females rated high challenge messages as more validating and appropriate than males. Theoretical and pragmatic implications are discussed. ACKNOWLEDGEMENTS To my parents and brother – for always believing in me and supporting me along the way. To my advisor, Mandy – for encouraging my growth as a scholar and giving me countless opportunities throughout the years. Many thanks to Sandi and Clare for helping me conceptualize earlier ideas for this manuscript, and to my committee members, Elizabeth and Becki, who helped develop it further and see it through to completion. Endless gratitude to Sammy, my unofficial mentor, who guided me through the thesis process and gave unwavering emotional, esteem (!), and informational support. To my wonderful support system at MSU – Emily, Theresa, Clint, Petee, Nicole, Emma, Alexa, Hannah, Matt, Nick, Amy – I could not have done it without you. To many other professors and friends – thank you for your kindness at moments when I needed it most. iii TABLE OF CONTENTS LIST OF TABLES ......................................................................................................................... vi LIST OF FIGURES ...................................................................................................................... vii INTRODUCTION .......................................................................................................................... 1 Sexual Harassment ...................................................................................................................... 2 Confirmation Theory .................................................................................................................. 7 Acceptance dimension. ........................................................................................................... 8 Challenge dimension. .............................................................................................................. 8 Interaction of acceptance and challenge dimensions. ............................................................. 9 Message Effectiveness .............................................................................................................. 11 Confirming Messages Applied to Sexual Harassment.............................................................. 12 Scenario severity. .................................................................................................................. 14 Participant sex. ...................................................................................................................... 16 Context for the Present Study ................................................................................................... 17 METHODS ................................................................................................................................... 19 Sample....................................................................................................................................... 19 Scenarios. .............................................................................................................................. 20 Confirmation messages. ........................................................................................................ 21 Measures ................................................................................................................................... 22 Message validation................................................................................................................ 22 Message motivation. ............................................................................................................. 22 Message appropriateness. ..................................................................................................... 23 Facilitation of reappraisal. .................................................................................................... 23 Perceived severity. ................................................................................................................ 24 Emotional upset. ................................................................................................................... 24 Realism. ................................................................................................................................ 25 Familiarity with sexual harassment. ..................................................................................... 25 RESULTS ..................................................................................................................................... 26 Preliminary Analyses ................................................................................................................ 26 Main Analyses .......................................................................................................................... 27 Message validation................................................................................................................ 28 Message motivation. ............................................................................................................. 31 Message appropriateness. ..................................................................................................... 33 Facilitation of reappraisal. .................................................................................................... 35 DISCUSSION ............................................................................................................................... 38 Theoretical Implications ........................................................................................................... 39 Pragmatic Implications ............................................................................................................. 44 Limitations and Future Directions ............................................................................................ 45 iv CONCLUSION ............................................................................................................................. 48 APPENDICES .............................................................................................................................. 49 APPENDIX A: Severity Scenarios ........................................................................................... 50 APPENDIX B: Confirmation Messages by Scenario ............................................................... 51 APPENDIX C: Survey .............................................................................................................. 52 REFERENCES ............................................................................................................................. 69 v LIST OF TABLES Table 1: Descriptive and Inferential Statistics for Preliminary Analyses ..................................... 59 Table 2: Independent Samples t-Tests of Biological Sex on Emotional Upset, Realism, Perceived Severity, and Familiarity with Sexual Harassment ....................................................................... 60 Table 3: Paired Samples t-Tests of Acceptance and Challenge on Validation Ratings................ 61 Table 4: Paired Samples t-Tests of Acceptance and Challenge on Motivation Ratings ............... 62 Table 5: Paired Samples t-Tests of Acceptance and Challenge on Appropriateness Ratings ...... 63 Table 6: Paired Samples t-Tests of Acceptance and Challenge on Facilitation of Reappraisal Ratings .......................................................................................................................................... 64 vi LIST OF FIGURES Figure 1: Interaction of acceptance and challenge on message validation ratings. ...................... 65 Figure 2: Interaction of acceptance and challenge on message motivation ratings. ..................... 66 Figure 3: Interaction of acceptance and challenge on message appropriateness ratings. ............. 67 Figure 4: Interaction of acceptance and challenge on facilitation of reappraisal ratings. ............. 68 vii INTRODUCTION Sexual harassment can have detrimental and long-lasting effects for victims, including deflated self-esteem, loss of self-confidence, substance abuse, and increased risk of anxiety, depression, and post-traumatic stress disorders (Fitzgerald, Swan, & Fischer, 1995b; Friborg et al., 2017; Houle, Staff, Mortimer, Uggen, & Blackstone, 2011; McDermut, Haaga, & Kirk, 2000; Rospenda, Richman, & Shannon, 2009). Negative effects for victims of sexual harassment in academia include students feeling less accepted, respected, and safe on campus (Cortina, Swan, Fitzgerald, & Waldo, 1998; Pinchevsky, Magnuson, Augustyn, & Rennison, 2019). Additionally, college students affected by sexual harassment have reported disenchantment with male professors, increased intention of leaving school, lowered self-confidence, poorer academic performance, and more negative perceptions of the university (Banyard et al., 2017; Benson & Thomson, 1982; Mengo & Black, 2016; Reilly, Lott, & Gallogly, 1986). Though the definition of sexual harassment varies across federal law, state laws, and local workplace/school policies, McDonald (2012) forwards one that encompasses many definitions – “conduct [that is] unwanted or unwelcome, and which has the purpose or effect of being intimidating, hostile, degrading, humiliating or offensive” (p. 2). Sexual assault, defined as “any nonconsensual sexual act proscribed by Federal, tribal, or State law, including when the victim lacks capacity to consent” (United States Department of Justice, 2019) can be a form of sexual harassment, but sexual harassment also encompasses less criminalized, but still inappropriate offenses, such as unwanted touching, persistent requests for dates, or sexist jokes. Recently, sexual harassment has become a widely politicized issue with the rise of the #MeToo and Time’s Up movements, social justice movements which encourage sympathy and solidarity among victims of sexual assault and harassment (Rodino-Colocino, 2018). In addition, 1 many prominent politicians and celebrities (e.g. Harvey Weinstein, Matt Lauer, Kevin Spacey) have recently been accused of sexual harassment and/or assault, thus making the issue of sexual harassment salient and timely (Dorris, 2017; Farrow, 2017; Setoodeh & Wagmeister, 2017). The added visibility of the prevalence of sexual harassment has enabled more victims to disclose their experiences. More people disclosing harassment offers more opportunities for informal support providers to lend support. The present study is framed by confirmation theory and examines how individuals in academia respond to confirming messages after reading a scenario which asks them to imagine experiencing sexual harassment (in varying degrees of severity). Confirmation is a specific way one can convey support, in which confirming messages intend to validate the feelings, thoughts, and/or abilities of the conversational partner in order to make them feel valued and worthy of respect (Dailey, 2006). Confirmation theory posits that messages high in both acceptance and challenge are the most sophisticated and ideal messages because they both validate the recipient (acceptance) but also encourage them to adjust their behaviors or cognitions (challenge; Dailey et al., 2010a). Literature pertaining to sexual harassment and confirmation theory will be reviewed here, which will lead to hypotheses and research questions about the effects of the reception of acceptance and challenge confirmation messages on message ratings of validation, motivation, appropriateness, and facilitation of reappraisal, and the extent to which these dimensions interact with harassment severity. Sexual Harassment As described above, the current study uses McDonald’s (2012) definition of sexual harassment – any unwanted conduct that is intended to intimidate, humiliate, or offend an individual. Till (1980) suggested there are five categories of sexual harassment: sexist 2 remarks/behavior, inappropriate and offensive sexual advances, solicitation of sexual activity by promise of rewards, coercion of sexual activity by threat of punishment, and sexual assault. Fitzgerald, Gelfand, and Drasgow (1995a) found that these five categories loaded onto three factors – gender harassment, unwanted sexual advances, and sexual coercion. Sexual harassment typically involves a power imbalance between the harasser and the victim (Morganson & Major, 2014; Scarduzio, Sheff, & Smith, 2018). Sexual harassment is often worse for minority women because it can become racialized sexual harassment where both sexist and racist attitudes are apparent (Buchanan & Ormerod, 2002). Due to the element of power differential inherently involved in sexual harassment, harassment often occurs in the workplace or academic context. An estimated 42-64% of women and 14-19% of men have experienced sexual harassment in the workplace (U.S. Merit Systems Protection Board [USMSPB], 1981, 1988, 1995). As for academia, almost 50% of undergraduate female students reported experiencing some form of sexual harassment (Benson & Thomson, 1982; Fitzgerald et al., 1988; Thakur & Paul, 2017). Additionally, organizational conditions often create barriers to formal reporting such as not making reporting policies and processes transparent, thus facilitating a hostile environment that allows sexual harassment to persist (Fitzgerald et al., 1995a; Thakur & Paul, 2017). Past research has examined how victims cope with sexual harassment (see Fitzgerald, 1990; Magley, 2002; Morganson & Major, 2014; Scarduzio et al., 2018, Ullman & Peter- Hagene, 2014). Fitzgerald (1990) proposed that victims of sexual harassment may cope with the harassment either internally or externally. Internal strategies are ones that attempt to manage cognitions or emotions, such as enduring the harassment, denying it happened, reattribution, or illusory control/self-blame. A common response to sexual harassment, especially if it is less 3 severe, is to do nothing or deny the situation (Fitzgerald et al., 1995b). Reattribution can include reinterpreting the sexual harassment as harmless or a once-off. Illusory control or self-blame is quite common, with 25% of female victims in a sample believing that they were in some way responsible for being sexually harassed (Jensen & Gutek, 1982). Many of these internal coping strategies can deter victims from formally reporting because they inherently minimize or trivialize the harassment. External strategies are focused on taking action to address the problem, such as avoiding the harasser, formally reporting the harassment to an institution or organization, or seeking social support (Fitzgerald et al., 1995b). The most infrequent strategy is to seek institutional or organizational support, and most victims see this as a last resort. Within seeking institutional/organizational support, victims may talk to their supervisor, file a formal complaint, or take legal action. Victims may be worried that formal reporting could lead to public embarrassment or could cause the harassment to become more severe (Fitzgerald et al., 1995b). Many victims also fear retaliation from their harasser or institution/organization, which prevents them from utilizing this external strategy. These fears are not unfounded; 62% of state workers reported retaliation after filing formal complaints of harassment (Avina & O'Donohue, 2002; Binder, Garcia, Johnson, & Fuentes-Afflick, 2018; Loy & Stewart, 1984). These instances of retaliation included lowered job evaluations, denial of a promotion, and even getting transferred or fired from jobs (Bergman, Langhout, Palmieri, Cortina, & Fitzgerald, 2002; Loy & Stewart, 1984). Ullman and Peter-Hagene (2014) further categorized sexual assault coping strategies into either maladaptive or adaptive. Maladaptive coping strategies include denial, disengagement, withdrawing from social situations, or substance use. Adaptive strategies include cognitive 4 restructuring or reappraisal, expressing one’s emotions, or seeking social support. Maladaptive strategies are related to increased post-traumatic stress disorder (PTSD) symptoms, whereas adaptive strategies are better for long term coping and recovery, and lead to decreased PTSD symptoms (Ullman & Peter-Hagene, 2014). Informal support is one of the more commonly used adaptive external coping strategies, perhaps because it is perceived to have fewer repercussions than formal reporting. Biaggio, Brownell, and Watts (1991) found that victims of sexual assault cited fearfulness and stigmatization as reasons for not formally reporting, so they more often turned to informal support providers they expected would be very supportive. The U.S. Merit Systems Protection Board (1981, 1988, 1995) found that 68% of victims of sexual harassment discussed the issue with a coworker, and 60% of victims discussed the harassment with friends or family. These statistics indicate that informal support providers may be a major source of support for many victims of sexual harassment and can help them cope with their situation, so it is imperative that support providers give helpful responses when sexual harassment is disclosed to them. Further, negative social reactions have been compared to a “second injury” for victims, because negative reactions to disclosure reject the recipient’s feelings and experience, and they communicate a lack of support from the sender, typically someone close to the victim (Symonds, 2010; Ullman, 1996). This indifference and apathy to the victim can “reinjure” them and cause further psychological trauma (Symonds, 2010). Negative social reactions include blaming the victim, treating the victim differently, trying to control the victim’s actions/taking control away from the victim, and distracting the victim (Ullman, 1996; Ullman & Peter-Hagene, 2014). These reactions to disclosure of sexual assault have been found to be especially detrimental for the recipient and can lead to self-blame, poorer psychological adjustment, the development of PTSD 5 symptoms, and the use of maladaptive coping strategies (Ullman, 1996; Ullman & Peter-Hagene, 2014). In contrast, positive social reactions have less of a robust or protective effect on recovery and coping, however they still contribute to better psychological adjustment (Ullman, 1996; Ullman & Peter-Hagene, 2014). Positive social reactions include providing emotional support, providing tangible/informational support, telling the victim it was not their fault, validating and believing the victim, and listening or encouraging the victim to talk (Ullman, 1996, 2000; Ullman & Peter-Hagene, 2014). Positive social reactions to disclosure of sexual assault have been linked to a victim’s greater perceived control over their recovery, less PTSD symptoms, and more adaptive coping strategies (Ullman & Peter-Hagene, 2014). Few studies have been conducted to examine the type and quality of support informal providers give to sexual harassment victims. One study that examined informal support and sexual harassment was conducted by Bingham and Battey (2005), who used Barbee’s (1990) social support activation model to identify how professors respond to students experiencing unwanted sexual attention. Participants, who were tenure-track professors at a university, were given a hypothetical scenario of a student disclosing that he/she was experiencing sexual harassment from another instructor. Participants were asked to write what they would say if they were actually in that situation. Responses were then coded for the four types of social support behaviors in Barbee’s (1990) social support activation model. Bingham and Battey (2005) found that all responses included either informational or tangible support (solve behaviors), whereas about a third of responses (36.5%) included either emotional or esteem support (solace behaviors). A few responses (3.1%) included dismiss behaviors, which minimized or dismissed the victim’s problem, and no responses included escape behaviors. While it is valuable to 6 understand what type of support strategies providers might employ, Bingham and Battey (2005) did not examine whether recipients found this support to be effective in making them feel better, which this study aims to do. The present study utilizes the framework of confirmation theory to operationalize supportive responses to sexual harassment victims. Confirmation is a specific way in which one can convey support, in which confirming messages aim to validate the recipient’s feelings and urge the recipient to explore and develop their thoughts. Confirming messages also attempt to encourage recipients to reach for a greater potential, meaning that the messages urge a change or reappraisal in the recipient’s cognitions, behaviors, and/or abilities. Confirmation Theory Confirmation theory posits that individuals have the need to be validated in order to develop their sense of self and identity (Buber, 1965; Laing, 1961; Watzlawick, Beavin, & Jackson, 1967), and sexual harassment can threaten one’s self-esteem and feelings of self-worth (Gruber & Bjorn, 1986). Confirmation messages exist on a single continuum from very disconfirming to very confirming (Dailey, 2006). Confirming messages intend to validate the feelings, thoughts, and/or abilities of the conversational partner in order to make them feel valued and worthy of respect (Dailey, 2006). Thus, confirming messages show a positive regard for the other. In contrast, disconfirming messages discount or reject the conversational partner’s feelings, thoughts, and/or abilities, which makes them feel inferior and less valued. Therefore, disconfirming messages show a negative regard for the other. While confirmation aims to validate a conversational partner, it also promotes growth by challenging the partner to achieve a greater potential by changing or reappraising their cognitions, behaviors, and/or abilities (Buber, 1965; Dailey et al., 2010a). Therefore, confirmation includes dimensions of both acceptance and 7 challenge (Dailey et al., 2010a). The dimensions of acceptance and challenge can be manifested in both verbal and nonverbal strategies, although this study will focus solely on verbal messages. The following sections will explain and apply the dimensions of acceptance and challenge to sexual harassment in further detail. Acceptance dimension. Acceptance can be defined as warmth, positive regard, and attentiveness toward another (Dailey et al., 2010a). Messages high in acceptance indicate to the recipient that they are cared for, their feelings are valid, and that they are not judged (Dailey et al., 2016). Messages low in acceptance do not remind the recipient that they are valued and cared for and/or they invalidate the recipient’s thoughts or feelings. In this study, acceptance is defined as acknowledging and validating the individual’s cognitions and emotions about sexual harassment, highlighting that the individual is valued and supported regardless of the harassment, and reinforcing that the individual is not being judged or blamed for the harassment. Applied to the context of sexual harassment, an example of a message high in acceptance could be, “Experiencing sexual harassment must be stressful for you, but remember I am always here to talk or listen and never judge you.” Challenge dimension. Challenge can be defined as engaging and pushing another to adjust their attitudes and/or behaviors to achieve their greater potential (Dailey et al., 2010a). In past studies, the dimension of challenge has been conceptualized as messages or behaviors that engage the recipient in competition or confrontation – essentially questioning the recipient’s actions or cognitions, stimulate the recipient’s cognitive or behavioral abilities, or call something about the recipient into question (Dailey, 2008). Messages low in challenge do not encourage the recipient to enact new behaviors or cognitions in order to promote growth as an individual. In this paper, challenge is defined as engaging the individual in conversations about their feelings 8 about sexual harassment, advocating that individuals speak up for themselves, and pushing individuals to resolve their issues with sexual harassment. Applied to the context of sexual harassment, an example of a message high in challenge could be, “Understand that this sexual harassment is not your fault. Keep a formal record of all instances of sexual harassment in case you decide to report it later.” Interaction of acceptance and challenge dimensions. Confirmation theory has been applied to the topics of family communication (Dailey, 2006, 2008; Wilson, Dorrance Hall, Gettings, & Pastor, 2015), athlete-coach communication (Cranmer et al., 2018), and couples’ weight loss management (Arroyo, Segrin, Harwood, & Bonito, 2017; Dailey et al., 2010a; Dailey, Romo, & McCracken, 2010b; Dailey, Romo, & Thompson, 2011). This research has suggested that messages high in both challenge and acceptance produce the most optimal outcomes for recipients, such as meeting diet goals in the context of weight loss management (Cranmer et al., 2018; Dailey et al., 2016; Dailey et al., 2010a). Through these past studies, confirmation high in both dimensions has been found to be positively related to self-esteem and identity strength (Dailey, 2009; Ellis, 2002; Schrodt, Ledbetter, & Ohrt, 2007). Confirmation has also been linked to an individual’s self-efficacy, or one’s sense of control and ability to produce a desired result (Dailey et al., 2010a). In contrast, Dailey (2010) found the effects of acceptance and challenge to sometimes be additive rather than interactive. Daley (2010) found the interaction of acceptance and challenge was not a significant predictor for one’s self-concept if the acceptance and challenge messages were coming from the recipient’s mother or father. Conversely, the interaction between acceptance and challenge was significant for messages coming from siblings. Dailey (2010) suggests this reinforcing effect may be more apparent in peer relationships or friendships and 9 suggests that further research about additive versus interactive effects be conducted. The present study examines confirmation in friendships, so the results may point to an interaction between challenge and acceptance, as was found with siblings. The two dimensions are often crossed on the continuum of low and high to create four combinations of messages: low acceptance/low challenge, low acceptance/high challenge, high acceptance/low challenge, and high acceptance/high challenge. Together, high acceptance and high challenge messages indicate that the recipient is validated but can also strive to improve or develop and reach a higher potential (Dailey et al., 2016). An example of a message high in both acceptance and challenge is, “Experiencing sexual harassment must be stressful for you. Try keeping a formal record of all of the instances of harassment in case you would like to report it later.” Theoretically, messages high in one component but lacking the other are seen as less effective. Messages high in challenge but lacking acceptance may invalidate the recipient, causing them to experience negative emotions, or think the provider is too critical (Cranmer et al., 2018; Dailey et al., 2016). An example of a message high in challenge but low in acceptance is, “You need to formally report the sexual harassment to upper administration.” Messages high in acceptance but lacking challenge may cause recipients to feel like they do not need to improve their behaviors or reach a higher potential (Dailey et al., 2016). An example of a message high in acceptance and low in challenge is, “You know I like you just as you are,” (Dailey et al., 2010a). Messages low in both acceptance and challenge are disconfirming messages. These messages are thought to be the least ideal because they convey apathy or indifference, invalidate, or reject the recipient and do not encourage them to do anything about their situation. An example of a message low in both acceptance and challenge that rejects the recipient is, “I can’t talk about this right now,” (Dailey et al., 2010a). 10 Message Effectiveness Message effectiveness refers to whether a message achieves its goal; in the current study, the goal is conceptualized as offering support. In this study, message effectiveness is conceptualized as whether a message ultimately makes the recipient feel validated in their feelings and experiences, motivated to modify their thoughts or behavior, is appropriate to say in the context of the situation, and facilitates reappraisal about their experience. Confirmation messages have been rated similarly in past studies to ensure that they contained validation (acceptance) components and motivation (challenge) components (Dailey et al., 2010a). The dimension of acceptance conveys to a recipient that they are cared for and not judged, thus acceptance messages should validate a recipient’s thoughts and feelings (Dailey et al., 2010a). The dimension of challenge aims to adjust a recipient’s attitudes and/or behaviors to achieve their greater potential, thus challenge messages should strongly motivate the recipient (Dailey et al., 2010a). While four distinct categories of messages (i.e., low acceptance/low challenge, low acceptance/high challenge, high acceptance/low challenge, and high acceptance/high challenge) were created by researchers in Dailey et al.’s (2010a) study, the participants’ ratings of challenge and acceptance within each message in each category varied. For instance, some low acceptance/high challenge messages were rated as containing components of validation, so challenge messages may lead to validation and acceptance messages may lead to motivation although that is not the intended effect. Not only should messages validate and motivate the recipient, but they should also be appropriate. Sexual harassment is a stigmatized and sensitive topic, so it is important that the recipient perceives the supportive messages as appropriate and not victim-blaming (Freyd, 2017; Ullman & Peter-Hagene, 2014). Additionally, messages should help facilitate reappraisal with 11 the aftermath of sexual harassment. Reappraisal is one of the several adaptive coping strategies that Ullman and Peter-Hagene (2014) discuss. While there are other adaptive coping strategies, reappraisal will be the sole focus for this study. Messages can help the recipient reappraise their thoughts regarding the situation and aid them in understanding that the harassment was not their fault. By facilitating reappraisal, victims may perceive greater control over the situation and experience less self-blame (Ullman & Peter-Hagene, 2014). Confirming Messages Applied to Sexual Harassment The sexual violence disclosure literature suggests that good responses to disclosures include validating the victim’s emotions, experience, and strength (Freyd, 2017; Ullman & Peter- Hagene, 2014). Messages high in acceptance should validate the recipient’s thoughts and feelings, and also be considered appropriate because they intend to show care and compassion and refrain from making judgments. However, messages high in acceptance may or may not motivate a recipient, since they do not contain explicit actions a recipient should engage in but some high acceptance messages in past studies have been rated as containing motivation components (Dailey at al., 2010a). Additionally, messages high in acceptance may or may not facilitate reappraisal, because acceptance messages do not explicitly encourage reappraisal of the situation or one’s cognitions. In concert with past confirmation theory research and the sexual violence literature, the following hypothesis and research question are forwarded: H1: Messages high in acceptance will be rated more (a) validating and (b) appropriate than messages low in acceptance. RQ1: Do ratings of (a) message motivation and (b) facilitation of reappraisal of low and high acceptance messages differ? 12 While challenge messages are intended to motivate a recipient, it may not be the case in the context of sexual harassment. Drawing from the literature on sexual violence disclosures, researchers recommend providers avoid minimizing victims’ experiences, inserting judgments or blame, and taking away victims’ autonomy (Freyd, 2017; Ullman & Peter-Hagene, 2014). Messages high in challenge may increase a victim’s self-efficacy and belief that they have agency to change the situation and hold their harasser accountable. However, victims of sexual harassment may feel too helpless to do anything or too disillusioned with the organizational response to take any further action and may still feel unmotivated by the challenge messages (Gutek & Koss, 1993; Holland & Cortina, 2017). Messages high in challenge may or may not validate a recipient, since they do not contain explicit acknowledgement of the recipient’s feelings or thoughts, although Dailey et al. (2010a) found that some high challenge messages were rated as containing components of validation. Additionally, in the context of sexual harassment, challenge messages may be viewed as inappropriate. Challenge messages could be perceived as a threat to autonomy because they push recipients to engage in action-oriented strategies, minimize victims’ experiences by suggesting the victim is not handling the issue properly, or imply blame by telling the victim there are actions they can take to avoid or prevent further harassment. Further, challenge messages may or may not facilitate reappraisal. Recipients may view challenge messages as telling them what to do rather than helping them reappraise the situation or their feelings. Since it is unknown if challenge messages will motivate or validate recipients, be appropriate in the context of sexual harassment disclosure, and facilitate reappraisal, the following research question is forwarded: 13 RQ2: Do ratings of (a) message validation, (b) message motivation, (c) message appropriateness, and (d) facilitation of reappraisal of low and high challenge messages differ? This study aims to further explore whether high levels of both challenge and acceptance create the most effective messages according to the recipient, as would be predicted by confirmation theory. If challenge components are not viewed as motivating, but are instead viewed as inappropriate or critical, high acceptance/high challenge messages will not have their intended effect. Since it is unclear how challenge will operate in the context of sexual harassment, predictions cannot be made about how acceptance and challenge will interact. Thus, the following research question is proposed: RQ3: How will acceptance and challenge interact in regard to (a) message validation, (b) message motivation, (c) message appropriateness, and (d) facilitation of reappraisal ratings? Scenario severity. Fitzgerald et al. (1988; 1995a) created the Sexual Experiences Questionnaire (SEQ) which separates sexual harassment into three increasingly severe levels: gender harassment (e.g., sexist remarks, inappropriate jokes), unwanted sexual attention (e.g., persistent requests for dates, inappropriate sexual advances, uncomfortable touching/stroking/fondling), and sexual coercion (e.g. coercion of sexual activity by threat of punishment, quid pro quo). Sexual harassment episodes that are explicit, repeated, and obviously harassing are thought of as more severe (i.e., the second and third levels of the SEQ), and these episodes are more likely to elicit a formal complaint (Baker, Terpstra, & Larntz, 1990; Fitzgerald et al., 1995a). Sexual harassment episodes that are brief or less explicit (i.e., the first and perhaps second level of the SEQ) are usually ignored or diffused with humor on the part of the recipient (Fitzgerald et al., 14 1995a). Additionally, as sexual harassment becomes more frequent and severe, victims will utilize more coping strategies, including seeking social support (Cortina & Wasti, 2005; O’Leary-Kelly, Bowes-Sperry, Bates, & Lean, 2009). Baker et al. (1990) studied how severity of sexual harassment will influence the victim’s reaction and coping strategies (i.e., whether they would internally or externally report, physically or verbally react, leave their job, ignore it, etc.), and they found that as harassment becomes more severe, more assertive reaction strategies are utilized. However, previous research has not examined what support messages victims might find effective, depending on the severity of the harassment. This study will utilize the three levels of severity forwarded by the SEQ to determine how severity of a sexual harassment scenario influences how recipients perceive confirming messages. The dual-process theory of supportive communication outcomes has been used to examine how individual and situational factors, such as severity of a participant’s problem and participant’s level of emotional upset, will affect how they process a supportive message (Bodie & Burleson, 2008; Bodie et al., 2011; Feng & MacGeorge, 2010; Holmstrom et al., 2015). Feng and MacGeorge (2010) found that message factors in advice became more influential when problems are more severe; meaning that as problem severity increases, recipients will be more motivated to think critically about the advice content. As for emotional support, Bodie and Burleson (2008) theorized that as a problem becomes more severe and an individual becomes more upset about it, the recipient’s desire for emotional support and their motivation to critically process supportive messages will increase. However, these effects drop off if a participant faces an extremely severe problem and is too overwhelmed to process supportive messages. Both of these variables – problem severity and emotional upset – feed into how distressed an individual will feel in a situation, and in turn, an individual will process messages with more scrutiny if 15 they are experiencing a moderate level of distress, but the same effect will not be found if the individual is too distressed or not distressed enough to be motivated to think critically. Seeing as problem severity can have an effect on how messages are processed, severity may moderate the main effects of acceptance and challenge messages on message validation, message motivation, message appropriateness, and facilitation of reappraisal. Since it is unknown how severity will interact with acceptance and challenge messages, the following research questions are proposed: RQ4: How does acceptance interact with scenario severity in regard to the ratings of (a) message validation, (b) message motivation, (c) message appropriateness, and (d) facilitation of reappraisal? RQ5: How does challenge interact with scenario severity in regard to the ratings of (a) message validation, (b) message motivation, (c) message appropriateness, and (d) facilitation of reappraisal? RQ6: How does acceptance and challenge interact with scenario severity in regard to the ratings of (a) message validation, (b) message motivation, (c) message appropriateness, and (d) facilitation of reappraisal? Participant sex. Both female and male participants will be included in this study, although the hypothetical harasser’s sex will be controlled (i.e., male). This manipulation of harasser sex was done purposefully because males are more often harassers than females (USMSPB, 1981, 1988, 1995). While females experience sexual harassment far more often than males, as heavily documented in the literature (Berdahl & Moore, 2006; Terpstra & Baker, 1987; Thakur & Paul, 2017), males also experience sexual harassment (Bingham & Battey, 2005; Lucero, Allen, & Middleton, 2006; Street, Gradus, Stafford, & Kelly, 2007). Therefore, it is useful to collect 16 responses from both male and female participants, although it is important to note that these responses may have inherent differences. Sex differences may be found in the results for a variety of reasons, one being that male participants may find the scenarios less realistic or too far-fetched because they have less experience with sexual harassment. Past research has found that females are more likely than men to perceive ambiguously sexual interactions as sexual harassment (Gutek, Morasch, & Cohen, 1983; Gutek, Nakamura, Gahart, Handschumacher, & Russell, 1980; Terpstra & Baker, 1987). Preliminary analyses will be conducted to examine significant discrepancies due to sex. Context for the Present Study The present study examines college student reactions to messages varying in acceptance and challenge provided by a close friend in response to a sexual harassment episode. Most confirmation research has focused on individuals in highly interdependent relationships (i.e., parent-child, couples in romantic relationships, coach-athlete; Cranmer, Brann, & Weber, 2018; Dailey, 2006; Dailey et al., 2010a), whereas this study examines confirmation within a less interdependent relationship: two close friends. While this relationship is less interdependent than others, a close friend is a central relationship for college students, this study’s sample. College students may be far away from home and their established support networks (i.e., family and friends from their hometown), thus they may turn to their peers and friends at college for support during stressful times. Social support from friends has been found to lessen stressful life events and psychological distress for college students (Rodriguez, Bingham, Myers, Morris, & Cardoza, 2003; Zaleski, Levey-Thors, & Schiaffino, 1998). Therefore, it is appropriate to examine confirmation in close friendships, a main support source for college students. 17 In addition, most confirmation research examines confirmation across interactions and relationships, rather than a one-shot response to a discrete, upsetting event, such as sexual harassment (Cranmer et al., 2018; Dailey, 2010; Dailey et al., 2010a). I argue that there is value in expanding confirmation theory to this context. As mentioned previously, the sexual harassment coping strategies proposed by Fitzgerald (1990) include both internal and external strategies. Internal coping strategies may be aided by the acceptance dimension of confirming messages, as internal strategies are used to manage emotions and cognitions relating to the harassment and the acceptance dimension aims to validate a recipient’s emotions and experiences. In parallel, external coping strategies may be facilitated by the challenge dimension of confirming messages. External coping strategies are focused on problem-solving and taking action, and the challenge dimension encourages recipients to enact certain behaviors and resolve their issues in order to reach a greater potential. To summarize, the aims of the current study are twofold. The first aim is to test the boundaries of confirmation theory. It has been previously thought that messages that were high in both dimensions of acceptance and challenge were of the best quality. However, sexual harassment is a sensitive topic and perhaps some challenge messages may not be appropriate in this context. The second aim is to understand if the severity of sexual harassment will affect how effective participants find challenge and acceptance messages following disclosure of harassment. More severe sexual harassment may produce different responses to the messages than less severe sexual harassment. 18 METHODS Before completing the online survey, participants were informed that the survey focused on sexual harassment and supportive messages and that their responses are anonymous to the researchers. Participants were informed that their participation is completely voluntary and at any time they can leave a question unanswered or terminate their participation with no penalty. After completing the informed consent, participants were directed to the survey. Participants were randomly presented with one of three hypothetical scenarios and were instructed to imagine that they experienced sexual harassment from a male professor. The three scenarios varied in severity of sexual harassment, as identified by the SEQ (Fitzgerald et al., 1995a). After participants read the scenario, they were presented with four messages (i.e., low acceptance/low challenge, low acceptance/high challenge, high acceptance/low challenge, and high acceptance/high challenge) in a randomized order. Participants were instructed to imagine that they told a close friend what happened in the scenario and the close friend responded with these messages. Participants were asked to rate each of the four messages on scales measuring message validation, message motivation, message appropriateness, and facilitation of reappraisal. Participants then completed a manipulation check of scenario severity, emotional upset, and realism. Participants also completed the Sexual Experiences Questionnaire – Short Form (SEQ-s; Fitzgerald et al., 1995a) to gauge familiarity with sexual harassment. Lastly, participants completed demographic information. Sample Participants (N = 170) were recruited through two methods: through communication courses at Michigan State University and via snowball sampling on the author’s social media sites. Participants who were recruited through communication courses were compensated for 19 their time through course credit at the discretion of the course instructor, and participants recruited through snowball sampling completed the survey voluntarily. Participants had to be at least 18 years old and a current undergraduate/graduate student or graduated from college within the past year to be eligible to take the survey. Data from participants who did not complete the majority of the survey were excluded. The sample was 62.9% female and 37.1% male. The average age was 21.61 years old (SD = 2.16), ranging from 18 to 30 years old. Participants primarily identified as White/Caucasian (68.2%), followed by Asian/Pacific Islander (19.4%), Black/African American (5.9%), multiple races (2.9%), Hispanic/Latino (1.2%), “other” (1.2%), and Native American/American Indian (0.6%). 89.4% of the sample were current or past students at Michigan State University. The majority of participants were fourth-year seniors (36.5%), followed by juniors (22.9%), graduate students (11.2%), sophomores (10.2%), recent graduates (8.2%), in their fifth year or later of college (7.6%), and freshmen (2.4%). Domestic students (78.2%) comprised the majority of the sample. Communication majors made up 45.3% of the sample. 25.3% of the sample reported they had been sexually harassed, and of those that had been sexually harassed 93% were female. Scenarios. The scenarios were adapted from Baker et al. (1990) to reflect an academic setting rather than a workplace setting to be more relatable for college students (see Appendix A for scenarios). Terpstra and Baker (1987) based the scenarios on sexual harassment incidents reported in Fair Employment Practices Guidelines (1978-1982). These three scenarios were chosen because they reflect the three distinct levels of the SEQ: gender harassment, unwanted sexual attention, and sexual coercion. In the scenarios, Professor X was used as the name for the male sexual harassment perpetrator, and participants were instructed to imagine themselves as 20 the harassment victim. Hypothetical scenarios have been used in past sexual violence studies and have been found to elicit varying perceptions of different forms of sexual violence, such as stranger vs. acquaintance rape and less severe vs. highly severe sexual harassment (Baker et al., 1990; Check & Malamuth, 1983). The scenarios contain a similar word count to control for length, as to not conflate the results. Confirmation messages. After reading one of the scenarios, participants were randomly presented with four pre-constructed scenario-specific messages. Challenge and acceptance were crossed to create four styles of messages: low acceptance/low challenge, low acceptance/high challenge, high acceptance/low challenge, and high acceptance/high challenge. Low acceptance/low challenge messages were created similarly to Wilson et al. (2015), in that they do not contain an acceptance or challenge component, but they suggest the recipient focus on something else. For a message to be considered high in challenge, it has to encourage the recipient to reflect on the situation and reappraise it. The low acceptance/low challenge message includes a vague statement about focusing on something else, but this message does not push for reflection or reappraisal. For a message to be considered high in acceptance, it has to convey warmth, attentiveness, validation, and no judgment. The low acceptance/low challenge message rejects the recipient instead of conveying warmth, attentiveness, and validation. High acceptance/high challenge messages contain both components from high acceptance/low challenge and low acceptance/high challenge messages to better isolate the effect of the combination of high acceptance/high challenge rather than creating entirely different messages (Dailey et al., 2010a). Messages were modeled off of messages from past confirmation theory studies (e.g., Dailey et al., 2010a; Wilson et al., 2015) to incorporate applicable components for recipients experiencing sexual harassment. Participants were asked to rate each message on 21 scales measuring if a message is validating, motivating, appropriate in the context of the conversation, and facilitates reappraisal. See Appendix B for the messages. Measures Unless stated otherwise, all of the following measures rate items on a 7-point Likert-type scale, ranging from 1 = “strongly disagree” to 7 = “strongly agree.” Message validation. The validation measure assessed if a message validated the recipient’s feelings and experience. Participants were instructed to respond to the items as if a close, same-sex friend offered them the message after they told them about the sexual harassment scenario. Validation was measured through five items created by Dailey et al. (2010a): “This message shows he/she cares about me,” “This message is judgmental” (reverse coded), “This message is conveyed in a friendly voice,” “This message accepts my feelings,” and “This message is emotionally cold” (reverse coded). These items were averaged to create a composite validation score for each message. Reliability for low acceptance/low challenge messages was  = .75, M = 2.79, SD = 1.02; for low acceptance/high challenge messages  = .88, M = 5.43, SD = 1.22; for high acceptance/low challenge messages  = .86, M = 5.21, SD = 1.11; and for high acceptance/high challenge messages  = .87, M = 5.90, SD = 1.08. Message motivation. The motivation measure evaluated if a message motivated the recipient to modify their thoughts or behavior. Participants were instructed to respond to the items as if a same-sex friend offered them the message after they told them about the sexual harassment scenario. Motivation was measured through four items adapted from Dailey (2008) and Dailey et al. (2010a), including “This message presents new information, ideas, or options about the scenario I read,” “This message pushes me to advocate for myself,” “This message helps me channel my negative emotions about the scenario I read into more positive actions,” 22 and “This message encourages me to explore different ideas.” These items were averaged to create a composite motivation score for each message. Reliability for low acceptance/low challenge messages was  = .85, M = 2.58, SD = 1.27; for low acceptance/high challenge messages  = .88, M = 4.51, SD = 1.37; for high acceptance/low challenge messages  = .82, M = 3.36, SD = 1.17; and for high acceptance/high challenge messages  = .86, M = 4.80, SD = 1.36. Message appropriateness. The appropriateness measure gauged whether recipients found the messages appropriate in the context of the situation. Participants were instructed to respond to the items as if a same-sex friend offered them the message after they told them about the sexual harassment scenario. Appropriateness was measured through four items adapted from Canary and Spitzberg’s (1987) Specific Appropriateness subscale. These items include “This message includes some things that should not have been said” (reverse coded), “Some of the things said in this message are embarrassing to me” (reverse coded), “All of the things in this message are in good taste as far as I’m concerned,” and “At least one of the remarks in the message is rude” (reverse coded). These items were averaged to create a composite appropriateness score for each message. Reliability for low acceptance/low challenge messages was  = .74, M = 3.16, SD = 1.18; for low acceptance/high challenge messages  = .87, M = 5.50, SD = 1.23; for high acceptance/low challenge messages  = .85, M = 5.22, SD = 1.21; and for high acceptance/high challenge messages  = .90, M = 5.65, SD = 1.26. Facilitation of reappraisal. The facilitation of reappraisal measure assessed whether the recipients found the messages effective in helping them reappraise their situation. Facilitation of reappraisal was measured through five items adapted from Jones and Wirtz’s (2006) reappraisal scale. Participants were asked to respond to the following items after receiving their friend’s 23 message: “My friend made me think about the events I described during the conversation,” “I feel that I ought to re-evaluate the event now after the conversation,” “I don’t really see the event in a different light after the conversation (reverse coded),” “Talking with my friend about the event helped me get my mind off it,” and “I understand the situation better now that I talked about it with my friend.” The items were averaged to create a composite reappraisal score for each message. Reliabilities increased when excluding the item “I don’t really see the event in a different light after the conversation (reverse coded).” Reliability for low acceptance/low challenge messages was  = .76, M = 2.89, SD = 1.21; for low acceptance/high challenge messages  = .75, M = 4.53, SD = 1.11; for high acceptance/low challenge messages  = .67, M = 3.84, SD = 1.02; and for high acceptance/high challenge messages  = .66, M = 4.82, SD = 1.07. Perceived severity. Four items adapted from MacGeorge, Graves, Feng, Gillihan, and Burleson (2004) were used to assess participants’ perceptions of the scenario severity. These items include, “The scenario described a significant problem,” “The scenario described a trivial problem” (reverse coded), “This scenario described a major problem,” and “The scenario described an important problem.” The items were averaged to create a composite severity score to be used as a manipulation check. Reliability was  = .71, M = 5.41, SD = 1.09. Emotional upset. Four items adapted from Bodie et al. (2011) were used to assess participant’s degree of emotional upset. These items include, “The scenario I read was distressing,” “The scenario I read upset me,” “The scenario I read troubled me,” and “The scenario I read bothered me.” The items were averaged to create a composite emotional upset score to be used as a manipulation check. Reliability was  = .95, M = 4.70, SD = 1.57. 24 Realism. Three items were used as a manipulation check to assess how realistic participants perceive the scenarios to be. These items were created by Shebib et al. (2019) and adapted from Burleson, Holmstrom, and Gilstrap (2005) and include, “The scenario I read was realistic,” “The scenario I read was believable,” and “The scenario I read commonly occurs.” The items were averaged to create a composite realism score to be used as a manipulation check. Reliability was  = .88, M = 5.33, SD = 1.17. Familiarity with sexual harassment. Participants’ familiarity with sexual harassment was assessed using the SEQ-s (Fitzgerald et al., 1995a), which is a shortened form of the original SEQ. The SEQ-s has been used with a college sample in prior studies (Cortina et al., 1998). The SEQ-s consists of eight items that tap the three categories of sexual harassment: gender harassment (3 items), unwanted sexual attention (3 items), and sexual coercion (1 item). Participants were instructed to indicate whether they have experienced any of the behaviors from a male professor or instructor during their time at their university on a 5-point Likert scale, ranging from 1 = “never” to 5 = “always.” The gender harassment items include, “Crude sexual remarks,” “Sexist comments,” and “Offensive remarks.” The unwanted sexual attention items include, “Unwanted touching,” “Repeated requests for dates,” and “Repeated requests for sexual favors.” The sexual coercion item is “Implicit or explicit demands for sexual favors to avoid negative consequences.” Lastly, participants were asked “Have you ever been sexually harassed?” to understand how they label their experiences. This is a dichotomous yes/no item instead of a 5-point Likert scale item, and it is used as a criterion item to correlate with the other three subscales. The words sexual harassment do not appear until the end of the measure to avoid demand characteristics. All items except the criterion item were averaged to create one composite score. Reliability across all items was  = .91, M = 1.25, SD = .50. 25 Preliminary Analyses RESULTS Independent samples t-tests were performed to see if there were any significant differences between the two participant samples (MSU communication courses and snowball sampling on the author’s social media sites) on ratings of realism, perceived severity, emotional upset, and familiarity with sexual harassment. No significant differences emerged, so the samples were collapsed, and all cases were analyzed together. Since severity was manipulated between each scenario, a one-way analysis of variance (ANOVA) was performed to see if participants’ perceived severity ratings differed between each scenario severity condition. The ANOVA revealed significant differences between perceived severity ratings for different severity conditions, F(2, 167) = 12.65, p < .001, partial η2 = 0.13. A post-hoc Bonferroni test revealed a couple of significant differences. The moderate severity condition (M = 5.56, SD = .96) was perceived significantly more severe than the low severity condition (M = 4.86, SD = 1.15), p = .001, 95% CI [0.23, 1.16]. The high severity condition (M = 5.81, SD = .96) was also rated significantly more severe than the low severity condition (M = 4.86, SD = 1.15), p < .001, 95% CI [0.47, 1.42]. The high severity condition (M = 5.81, SD = .96) and moderate severity condition (M = 5.56, SD = .96) were not significantly different in perceived severity ratings, p = .57, 95% CI [-0.21, 0.71]. Three one-sample t-tests (one for each scenario) were conducted to determine if participants found the scenarios emotionally upsetting. Ratings in the moderate and high severity condition were significantly above the midpoint on a 7-point scale, indicating that they found the scenarios emotionally upsetting. The ratings for the low severity condition were not significantly different from the midpoint. Additional one-sample t-tests were run for each scenario to 26 determine if realism ratings were statistically above the midpoint. All t-tests indicated that realism ratings were significantly above the midpoint on a 7-point scale. Means, standard deviations, p-values, and confidence intervals for each scenario are reported in Table 1. Independent samples t-tests were conducted to see if there were significant differences due to participant sex between ratings of emotional upset, realism, perceived severity, and familiarity with sexual harassment. Females’ scores were significantly higher than males’ on emotional upset (M = 4.97, SD = 1.51; M = 4.25, SD = 1.57), realism (M = 5.50, SD = 1.04; M = 5.03, SD = 1.33), and perceived severity (M = 5.60, SD = 1.04; M = 5.01, SD = 1.12). Females’ (M = 1.25, SD = .45) scores of familiarity with sexual harassment did not significantly differ from males’ (M = 1.24, SD = .58). Inferential statistics for these independent samples t-tests can be found in Table 2. Because significant differences between males and females emerged for emotional upset, realism, and perceived severity, biological sex was added as an independent variable in the main analyses. Main Analyses A series of four 3 (scenario severity) x 2 (acceptance) x 2 (challenge) x 2 (biological sex) repeated-measures ANOVAs were conducted to test all hypotheses and research questions for the main effects of acceptance and challenge, and interaction effects between scenario severity, biological sex, acceptance, and challenge on message validation, motivation, appropriateness, and facilitation of reappraisal ratings. Scenario severity, the first independent variable, is a 3- level categorical variable (low, moderate, or high) and a between-subjects factor. The second independent variable is biological sex, which is a 2-level categorical variable (male or female) and a between-subjects factor. The third independent variable is the level of acceptance in each message, and this is a 2-level categorical variable (low or high) and within-subjects factor, as 27 each participant received messages with both low and high levels of acceptance. The final independent variable is the level of challenge in each message, and this is a 2-level categorical variable (low or high) and within-subjects factor as participants received messages with both low and high levels of challenge. The dependent variables are (a) message validation, (b), message motivation, (c) message appropriateness, and (d) facilitation of reappraisal where each participant rates how validating, motivating, appropriate, and facilitative of reappraisal they perceive each of the four messages they received to be. Results of all analyses are reported below, organized by the dependent variable. Message validation. For H1a, the results of the 3 (scenario severity) x 2 (acceptance) x 2 (challenge) x 2 (biological sex) repeated-measures ANOVA on the dependent variable of message validation revealed a significant main effect for acceptance on validation ratings, F(1, 164) = 302.26, p < .001, partial 2 = 0.65. Participants rated high acceptance messages (M = 5.54, SD = 0.93) as more validating than low acceptance messages (M = 4.10, SD = 0.59), thus the data are consistent with H1a. RQ2a explored the main effects of challenge on validation ratings, and the results revealed a significant main effect, F(1, 164) = 282.64, p < .001, partial 2 = 0.63. Participants rated high challenge messages (M = 5.60, SD = 1.01) as significantly more validating than low challenge messages (M = 4.03, SD = 0.65). For RQ3a, the results revealed a significant interaction between acceptance and challenge on validation ratings, F(1, 164) = 138.83, p < .001, partial 2 = 0.46. To see where differences existed, six paired samples t-tests were conducted (low acceptance/low challenge vs. high acceptance/low challenge, low acceptance/low challenge vs. low acceptance/high challenge, low acceptance/low challenge vs. high acceptance/high challenge, low acceptance/high challenge vs. high acceptance/high challenge, high acceptance/low challenge vs. low acceptance/high 28 challenge, and high acceptance/low challenge vs. high acceptance/high challenge). All paired samples t-tests were significant. High acceptance/high challenge messages (M = 5.91, SD = 1.08) were rated significantly more validating than low acceptance/low challenge messages (M = 2.79, SD = 1.02), low acceptance/high challenge messages (M = 5.43, SD = 1.22), and high acceptance/low challenge messages (M = 5.21, SD = 1.11). High acceptance/low challenge messages (M = 5.21, SD = 1.11) were also rated significantly more validating than low acceptance/low challenge messages (M = 2.79, SD = 1.02). Low acceptance/high challenge messages (M = 5.43, SD = 1.22) were rated significantly more validating than high acceptance/low challenge messages (M = 5.21, SD = 1.11) and low acceptance/low challenge messages (M = 2.79, SD = 1.02). However, the difference in validation ratings between high and low acceptance messages was greater in the low challenge message condition than in the high challenge message condition (see Figure 1). When both challenge and acceptance are low, message validation ratings are particularly low. The high challenge component appears to be more critical than the high acceptance component for validation since the low acceptance/high challenge messages were rated significantly more validating than high acceptance/low challenge messages. Inferential statistics for these paired samples t-tests are included in Table 3. For RQ4a, the interaction between acceptance and scenario severity on validation ratings was insignificant, F(2, 164) = 1.24, p = 0.29, partial 2 = 0.02. For RQ5a, the interaction between challenge and scenario severity on validation ratings was significant, F(2, 164) = 5.90, p = 0.003, partial 2 = 0.07. To see where differences existed, two one-way ANOVAs were conducted, one for the low challenge messages and one for the high challenge messages. The first ANOVA showed that the effect of scenario severity on message validation ratings was significant for low challenge messages, F(2, 167) = 8.58, p < .001, partial 29 η2 = 0.09. A post hoc Games-Howell test revealed that the only significant difference between groups was found between low severity (M = 4.25, SD = .51) and moderate severity scenarios (M = 3.77, SD = .66), with participants who read the low severity scenario rating low challenge messages as more validating than participants who read the moderate severity scenario, p < .001, 95% CI [0.22, 0.74]. The second ANOVA showed that the effect of scenario severity on high challenge messages was significant, F(2, 167) = 3.27, p = .04, partial η2 = 0.04. A post-hoc Bonferroni test revealed no significant differences between groups, although the difference between the low severity and high severity scenarios was approaching significance (p = .053), with high challenge messages rated as more validating in the high severity scenario (M = 5.85, SD = .92) compared to the low severity scenario (M = 5.39, SD = 1.02). For RQ6a, the three-way interaction between acceptance, challenge, and scenario severity on validation ratings was not significant, F(2, 164) = 0.45, p = .64, partial 2 = .01. Additionally, there was an interaction between biological sex and challenge on validation ratings, F(1, 164) = 12.97, p < .001, partial 2 = .07. To parse out these differences, two independent samples t-tests were performed, one for low challenge messages and one for high challenge messages. Significant results emerged for low challenge messages, t(168) = 2.54, p = .01, 95% CI [0.06, 0.46], Cohen’s d = .39, r = .19. Males (M = 4.16, SD = .66) rated low challenge messages as more validating than females (M = 3.90, SD = .63). Significant results also emerged for high challenge messages, t(168) = -3.25, p = .001, 95% CI [-0.82, -0.20], Cohen’s d = .50, r = .24. Females (M = 5.86, SD = 0.98) rated high challenge messages as more validating than males (M = 5.35, SD = 1.00). There were no other significant two-, three-, or four-way interactions for message validation ratings. 30 Message motivation. For RQ1a, the results of the 3 (scenario severity) x 2 (acceptance) x 2 (challenge) x 2 (biological sex) repeated-measures ANOVA on the dependent variable of message motivation revealed a significant main effect for acceptance, F(1, 164) = 60.11, p < .001, partial 2 = 0.27. Participants rated high acceptance messages (M = 4.08, SD = 0.98) as more motivating than low acceptance messages (M = 3.55, SD = 0.84). RQ2b explored the main effects of challenge on motivation ratings, and the results revealed a significant main effect, F(1, 164) = 160.49, p < .001, partial 2 = 0.50. Participants rated high challenge messages (M = 4.62, SD = 1.25) as more motivating than low challenge messages (M = 3.02, SD = 1.03). For RQ3b, the results revealed a significant interaction between acceptance and challenge on motivation ratings, F(1, 164) = 10.26, p = .002, partial 2 = 0.06. To see where differences existed, six paired samples t-tests were conducted (low acceptance/low challenge vs. high acceptance/low challenge, low acceptance/low challenge vs. low acceptance/high challenge, low acceptance/low challenge vs. high acceptance/high challenge, low acceptance/high challenge vs. high acceptance/high challenge, high acceptance/low challenge vs. low acceptance/high challenge, and high acceptance/low challenge vs. high acceptance/high challenge). All paired samples t-tests were significant. High acceptance/high challenge messages (M = 4.80, SD = 1.36) were rated significantly more motivating than low acceptance/low challenge messages (M = 2.58, SD = 1.27), low acceptance/high challenge messages (M = 4.51, SD = 1.37), and high acceptance/low challenge messages (M = 3.36, SD = 1.17). High acceptance/low challenge messages (M = 3.36, SD = 1.17) were also rated significantly more motivating than low acceptance/low challenge messages (M = 2.58, SD = 1.27). Low acceptance/high challenge messages (M = 4.51, SD = 1.37) were rated significantly better than low acceptance/low challenge messages (M = 2.58, SD = 1.27) and high acceptance/low challenge messages (M = 31 3.36, SD = 1.17), indicating that the high challenge component plays a more important role in motivation ratings than the high acceptance component. The difference in motivation ratings between low and high acceptance messages was greater in the low challenge condition than the high challenge condition (see Figure 2). Messages that are low in both challenge and acceptance receive particularly low motivation ratings. Inferential statistics for these paired samples t-tests are reported in Table 4. For RQ4b, the results did not reveal a significant interaction between acceptance and scenario severity on motivation ratings, F(2, 164) = 0.20, p = 0.82, partial 2 = 0.002. For RQ5b, there was not a significant interaction between challenge and scenario severity on motivation ratings, F(2, 164) = 2.26, p = 0.11, partial 2 = 0.03. For RQ6b, the results did not reveal a significant interaction between acceptance, challenge, and scenario severity on motivation ratings, F(2, 164) = 0.45, p = 0.64, partial 2 = 0.01. Additionally, the results revealed a significant interaction between biological sex and challenge, F(1, 164) = 4.29, p = 0.04, partial 2 = 0.03. To see where differences existed, two independent samples t-tests were conducted, one for low challenge messages and one for high challenge messages. Significant differences appeared for low challenge messages, t(168) = 2.75, p = .007, 95% CI [0.12, 0.76], Cohen’s d = .42, r = .21. Males (M = 3.25, SD = 1.01) rated low challenge messages as more motivating than females (M = 2.80, SD = 1.00). The results were not significant for high challenge messages, t(148.82) = -0.80, p = .42, 95% CI [-0.53, 0.22], Cohen’s d = 0.13, r = .07; males (M = 4.56, SD = 1.11) and females (M = 4.71, SD = 1.32) did not significantly differ on how they rated high challenge messages. There were no other significant two-, three-, or four-way interactions for message motivation ratings. 32 Message appropriateness. For H1b, the results of the 3 (scenario severity) x 2 (acceptance) x 2 (challenge) x 2 (biological sex) repeated-measures ANOVA indicated a significant main effect for acceptance on appropriateness ratings, F(1, 164) = 168.28, p < .001, partial 2 = 0.51. Participants rated high acceptance messages (M = 5.42, SD = 1.11) as more appropriate than low acceptance messages (M = 4.31, SD = 0.70); thus, the data are consistent with H1b. RQ2c explored the main effects of challenge on appropriateness ratings, and the results revealed a significant main effect, F(1, 164) = 173.62, p < .001, partial 2 = 0.51. Participants rated high challenge messages (M = 5.49, SD = 1.16) as more appropriate than low challenge messages (M = 4.23, SD = 0.78). For RQ3c, there was a significant interaction between acceptance and challenge on appropriateness ratings, F(1, 164) = 147.29, p < .001, partial 2 = 0.47. To see where differences existed, six paired samples t-tests were conducted (low acceptance/low challenge vs. high acceptance/low challenge, low acceptance/low challenge vs. low acceptance/high challenge, low acceptance/low challenge vs. high acceptance/high challenge, low acceptance/high challenge vs. high acceptance/high challenge, high acceptance/low challenge vs. low acceptance/high challenge, and high acceptance/low challenge vs. high acceptance/high challenge). All paired samples t-tests were significant. High acceptance/high challenge messages (M = 5.65, SD = 1.26) were rated significantly more appropriate than low acceptance/low challenge messages (M = 3.16, SD = 1.18), low acceptance/high challenge messages (M = 5.50, SD = 1.23), and high acceptance/low challenge messages (M = 5.22, SD = 1.21). High acceptance/low challenge messages (M = 5.22, SD = 1.21) were also rated significantly more appropriate than low acceptance/low challenge (M = 3.16, SD = 1.18). Low acceptance/high challenge messages (M = 5.50, SD = 1.23) were rated significantly more appropriate than low acceptance/low challenge 33 messages (M = 3.16, SD = 1.18) and high acceptance/low challenge messages (M = 5.22, SD = 1.21), meaning that the high challenge component seems to be the driving force behind the appropriateness ratings. For appropriateness ratings, the difference between low and high acceptance messages was larger in the low challenge condition compared to the high challenge condition (see Figure 3). Appropriateness ratings are especially low when both challenge and acceptance are low. Inferential statistics for these paired samples t-tests are reported in Table 5. For RQ4c, the results did not reveal a significant interaction between acceptance and scenario severity on appropriateness ratings, F(2, 164) = 1.74, p = 0.18, partial 2 = 0.02. For RQ5c, there was a significant interaction between challenge and scenario severity on appropriateness ratings, F(2, 164) = 3.63, p = 0.03, partial 2 = 0.04. To further parse out differences, two one-way ANOVAs were conducted, one for low challenge messages and one for high challenge messages. The first ANOVA showed that the effect of scenario severity on message appropriateness ratings was significant for low challenge messages, F(2, 167) = 8.84, p < .001, partial η2 = 0.10. A post hoc Bonferroni test revealed that there was a significant difference between the low severity (M = 4.46, SD = .62) and moderate severity scenarios (M = 3.88, SD = .79), with participants rating low challenge messages as more appropriate in the low severity condition, p < .001, 95% CI [0.24, 0.92]. There was another significant difference between the moderate severity (M = 3.88, SD = .79) and high severity scenario (M = 4.25, SD = .81), with participants rating low challenge messages more appropriate in the high severity condition, p = .03, 95% CI [-0.71, -0.03]. The second ANOVA showed that the effect of scenario severity on message appropriateness ratings was not significant for high challenge messages, F(2, 167) = 1.35, p = .26, partial η2 = 0.02. 34 For RQ6c, the interaction between acceptance, challenge, and scenario severity was not significant, F(2, 164) = 0.11, p = 0.89, partial 2 = 0.001. There was also a significant interaction between challenge and biological sex, F(1, 164) = 18.69, p < .001, partial 2 = 0.10. To see where differences existed, two independent samples t- tests were conducted, one for low challenge messages and one for high challenge messages. Significant differences appeared for low challenge messages, t(168) = 2.05, p = .04, 95% CI [0.01, 0.50], Cohen’s d = .32, r = .16. Males (M = 4.35, SD = .73) rated low challenge messages as more appropriate than females (M = 4.09, SD = .80). Significant differences also appeared for high challenge messages, t(168) = -3.68, p < .001, 95% CI [-1.00, -0.30], Cohen’s d = .57, r = .27. Females (M = 5.82, SD = 1.10) rated high challenge messages as more appropriate than males (M = 5.16, SD = 1.16). There were no other significant two-, three-, or four-way interactions for message appropriateness ratings. Facilitation of reappraisal. For RQ1b, the results of the 3 (scenario severity) x 2 (acceptance) x 2 (challenge) x 2 (biological sex) repeated-measures ANOVA revealed a significant main effect for acceptance on facilitation of reappraisal ratings, F(1, 164) = 76.14, p < .001, partial 2 = 0.32. Participants rated high acceptance messages (M = 4.35, SD = 0.85) as more facilitative of reappraisal than low acceptance messages (M = 3.73, SD = 0.76). RQ2d explored the main effects of challenge on facilitation of reappraisal ratings, and the results revealed a significant main effect, F(1, 164) = 160.29, p < .001, partial 2 = 0.49. Participants rated high challenge messages (M = 4.67, SD = 0.96) as more facilitative of reappraisal than low challenge messages (M = 3.41, SD = 0.89). 35 For RQ3d, there was a significant interaction between acceptance and challenge on facilitation of reappraisal ratings, F(1, 164) = 26.25, p < .001, partial 2 = 0.14. To further parse out differences, six paired samples t-tests were conducted (low acceptance/low challenge vs. high acceptance/low challenge, low acceptance/low challenge vs. low acceptance/high challenge, low acceptance/low challenge vs. high acceptance/high challenge, low acceptance/high challenge vs. high acceptance/high challenge, high acceptance/low challenge vs. low acceptance/high challenge, and high acceptance/low challenge vs. high acceptance/high challenge). All paired samples t-tests were significant. High acceptance/high challenge messages (M = 4.82, SD = 1.07) were rated significantly more facilitative of reappraisal than low acceptance/low challenge messages (M = 2.89, SD = 1.21), low acceptance/high challenge messages (M = 4.53, SD = 1.11), and high acceptance/low challenge messages (M = 3.84, SD = 1.02). Also, high acceptance/low challenge messages (M = 3.84, SD = 1.02) were rated significantly more facilitative of reappraisal than low acceptance/low challenge messages (M = 2.89, SD = 1.21). Low acceptance/high challenge messages (M = 4.53, SD = 1.11) were significantly more facilitative of reappraisal than low acceptance/low challenge messages (M = 2.89, SD = 1.21) and high acceptance/low challenge messages (M = 3.84, SD = 1.02), indicating that the high challenge component is key to higher facilitation of reappraisal ratings, more so than the acceptance component. For facilitation of reappraisal ratings, the difference between low and high acceptance was greater for the low challenge messages than the high challenge messages (see Figure 4). Messages low in both acceptance and challenge were exceptionally low in facilitation of reappraisal ratings. Inferential statistics for the paired samples t-tests are reported in Table 6. 36 For RQ4d and RQ5d, scenario severity did not significantly interact with acceptance, F(2, 164) = 0.79, p = 0.45, partial 2 = 0.01, or challenge, F(2, 164) = 1.51, p = 0.22, partial 2 = 0.02. For RQ6d, there was not a significant interaction between acceptance, challenge, and scenario severity, F(2, 164) = 0.46, p = 0.63, partial 2 = 0.006. There were no other significant two-, three-, or four-way interactions for facilitation of reappraisal ratings. 37 DISCUSSION The present study was conducted to investigate if levels of acceptance and challenge impacted the effectiveness of confirmation messages following a hypothetical episode of sexual harassment, and if scenario severity had any effect on how messages were rated. Participants (N = 170) were randomly assigned to read one of three hypothetical sexual harassment scenarios that differ in level of severity and asked to rate how validating, motivating, appropriate, and facilitative of reappraisal they perceived four scenario-specific messages varying in levels of acceptance and challenge to be. There were significant main effects for acceptance, with high acceptance messages rated more validating, motivating, appropriate, and facilitative in reappraisal than low acceptance messages. Additionally, there were significant main effects for challenge, with high challenge messages rated more validating, motivating, appropriate, and facilitative in reappraisal than low challenge messages. Across all four dependent variables, the main effects created a significant interaction, with high acceptance/high challenge messages rated most validating, motivating, appropriate, and facilitative in reappraisal, followed by low acceptance/high challenge messages, high acceptance/low challenge messages, and low acceptance/low challenge messages. As for the effect of scenario severity, there were no significant interactions between acceptance and scenario severity in terms of validation, motivation, appropriateness, or facilitation of reappraisal ratings. There were significant interactions between scenario severity and challenge on the validation and appropriateness ratings; low challenge messages were rated more validating in the low severity condition compared to the moderate severity condition, and there was a curvilinear effect for appropriateness ratings with low challenge messages rated more appropriate in the low severity and high severity conditions compared to the moderate severity conditions. There were also 38 significant interactions between participants’ biological sex and challenge messages, with males rating low challenge messages as more validating, motivating, and appropriate than females, and females rating high challenge messages as more validating and appropriate than males. Theoretical and pragmatic implications are discussed in the next section. Theoretical Implications The present study’s data are in line with much of the past confirmation theory research. Theoretically, messages high in both acceptance and challenge are found to have the best outcomes because they convey both warmth and validation to the recipient, but also push them to create new cognitions or behavior regarding their situation (Cranmer et al., 2018; Dailey et al., 2016; Dailey et al., 2010a). These outcomes were also found with the current data; high acceptance/high challenge messages were consistently rated the most validating, motivating, appropriate, and facilitative in reappraisal. Interestingly, low acceptance/high challenge messages were consistently rated the second-most validating, motivating, appropriate, and facilitative in reappraisal. Dailey et al. (2010a) found this same result, with low acceptance/high challenge messages being the second most effective message to participants trying to engage in weight management behaviors, after high acceptance/high challenge messages. The operationalization of message effectiveness in both studies may be one explanation for why low acceptance/high challenge messages are rated better than other messages. In the present study, a message is deemed more effective than others if it validates the recipient, motivates the recipient, is appropriate for the context of the situation, and facilitates reappraisal more than other messages. High challenge messages were written with the sexual violence context in mind and were intended to be helpful rather than harmful, so they should also be rated highly for being appropriate, although this was not certain because high 39 challenge messages could come across as face threatening or stripping the victim of their autonomy. High challenge messages were found very appropriate in the current study, indicating that engaging a recipient in reappraisal is viewed as an appropriate tactic. High challenge messages were also written with the intent of helping the recipient change their initial thoughts regarding the situation (i.e., believing they were at fault for the sexual harassment from their professor), so they should be rated highly in that they help facilitate reappraisal. High acceptance messages did not push for explicit reappraisal in the present study, so they may have seemed less appropriate as well because they did not give the recipient any guidance for their situation. In Dailey et al. (2010a), message effectiveness was operationalized as a message motivating a recipient to make or engage in healthy diet/exercise decisions. Thus, she found that challenge messages are better suited for getting a recipient to enact a behavior, therefore they should be more effective than acceptance messages. The way message effectiveness was operationalized in both studies leans toward a message being effective if a recipient follows through with an action, which the acceptance component does not actively do. This may be why low acceptance/high challenge messages are seen as more effective – because the message ratings are not tapping the role of the acceptance component. However, acceptance interacted with challenge to create the highest-rated messages – high acceptance/high challenge messages. While challenge seems to play a greater role in eliciting behaviors or creating new cognitions, acceptance is also important and may have a greater impact when examining different dependent variables. Just as Dailey et al. (2010a) found that acceptance was critical to developing an internal motivation to follow through with healthy diet/exercise behaviors, acceptance may be important in other ways for sexual harassment disclosure (such as relational satisfaction between the support provider and recipient), since the 40 results are showing that high acceptance/high challenge messages are rated better for all of the dependent variables than low acceptance/high challenge messages. Future studies should continue to examine this effect and explore the role of acceptance in this context further. Three scenarios were created for the present study to vary in how severe the sexual harassment depicted in them was. Scenario severity did not interact with most of the factors; however, there were several significant interactions between scenario severity and challenge messages. Low challenge messages were rated more validating in the low severity scenario compared to the moderate severity scenario. Message validation means that the message accepted the feelings and experience of the recipient and was not judgmental or emotionally cold. The low severity scenario depicted a professor telling sexist jokes in class, compared to the moderate severity scenario which portrayed a professor invading a student’s personal space and putting an arm around them, which made the student uncomfortable. Perhaps the low challenge messages were rated more validating for the low severity scenario because high challenge messages would be viewed as an excess of support in a less serious situation. If the recipient does not view the low severity condition as serious, they may wonder why a support provider would give them support that would try to change their cognitions surrounding the event, when they did not blame themselves in the first place. In the more serious situation, the moderate severity condition, low challenge messages are perceived as less validating because the situation is more severe, and a higher quality message is needed to attend to the emotions and cognitions that recipients are experiencing. Additionally, there was a curvilinear effect with message appropriateness ratings of low challenge messages and scenario severity. Low challenge messages were rated more appropriate in the low severity scenario compared to the moderate severity scenario, and they were also rated 41 more appropriate in the high severity scenario compared to the moderate severity scenario. Message appropriateness means that the message did not make any rude or inappropriate remarks and the message was suitable in the context of the situation, in this case a disclosure of sexual harassment. The high severity scenario described a professor repeatedly asking a student to hook up with him for a better course grade, compared to unwanted touching depicted in the moderate severity condition and sexist jokes depicted in the low severity condition. The dual- process theory of supportive communication outcomes may offer an explanation for this finding (Bodie & Burleson, 2008). This theory indicates that when motivation and/or ability to process supportive messages is low, recipients don’t distinguish between message quality as much as they do when motivation and ability to process messages are high. Perhaps in low severity conditions, recipients are rating lower-quality messages (low challenge messages) as more appropriate because they do not perceive the scenario as being that serious, so they are not as motivated to process the messages in the low severity condition. In the high severity condition, the ability to process the messages may be decreased because the situation is too severe and emotionally upsetting, so recipients may not distinguish between message quality as much. If participants were adequately processing messages, another explanation is that high challenge messages may be an excess of support in the low severity condition because it is viewed as less serious, so low challenge messages are rated more appropriate. On the other hand, high challenge messages may be perceived as too confrontational or assertive because of the situation’s severity, so low challenge messages are rated more appropriate. Initially it was thought that females and males may perceive the scenarios differently in terms of realism, severity, and how emotionally upsetting they were because females are more frequently victims of sexual harassment. Preliminary analyses indicated significant differences 42 between males and females in realism, perceived severity, and emotional upset ratings, so sex was added as a factor in the series of repeated-measures ANOVAs. Biological sex interacted with challenge messages in terms of validation, motivation, and appropriateness ratings. Males rated low challenge messages more validating, motivating, and appropriate than females did. Furthermore, females rated high challenge messages more validating and appropriate than males did. Research on sex differences in processing supportive messages showed that females have greater ability and motivation to process messages than males do, such that females can better differentiate between better and worse comforting messages (Burleson et al., 2011). This may be one explanation why females favor high challenge messages and males favor low challenge messages, as high challenge messages are thought to be higher quality than low challenge messages. Another explanation is that males may prefer the less intrusive low challenge messages, compared to high challenge messages which push for reappraisal of one’s cognitions and behaviors. Since males generally receive less support than females, they may be less comfortable talking through their feelings and issues (Antonucci & Akiyama, 1987; Burda, Vaux, & Schill, 1984). Additionally, males experience sexual harassment far less frequently than females, so this is already a hypersensitive and face-threatening issue because it is more atypical for a male to be harassed, which may be why they prefer these low challenge messages that do not suggest reappraisal (Berdahl & Moore, 2006; Terpstra & Baker, 1987; Thakur & Paul, 2017). Perhaps males prefer validation over challenge, whereas females favor the reappraisal component present in challenge messages. Another explanation may stem from the sex differences in perceived severity, realism, and emotional upset ratings. Males rated all scenarios significantly lower than females in terms of how severe, realistic, and emotionally upsetting they perceived them to be. Because of these perceptions, males may not see the need for high-quality 43 messages, and they may believe these low challenge messages suffice. Future research should further examine this interaction. The present study extends confirmation theory to several new contexts. Most confirmation theory research has been done through the lens of parent-child relationships or romantic couples (e.g., Dailey, 2006; Dailey et al., 2010a). The present study uses a less interdependent relationship – close friends – yet the results are in line with past confirmation research, thus extending the boundaries of confirmation theory and what relationships it is applicable in. The present study also adds message appropriateness and facilitation of reappraisal as dependent variables, further specifying what message effectiveness means and how acceptance and challenge interact. Additionally, confirmation theory was tested in the present study with singular messages, rather than across relationships or interactions, as has been done in past research (e.g., Cranmer et al., 2018; Dailey, 2010; Dailey et al., 2010a). One-shot messages appear to have similar effects as multiple messages, which is useful in future confirmation theory study design. Lastly, the present study examines confirmation theory in relation to disclosure of sexual harassment, and confirmation proves to be effective in this context. High challenge messages stand out as particularly useful, since they can help the recipient reappraise the situation, aid them in believing they were not at fault, and can help them regain their power in the situation. Pragmatic Implications It can be difficult to know what to say after a friend discloses an upsetting experience, such as experiencing sexual harassment. If a victim discloses their harassment to someone, they are often seeking support from that person in a way to externally cope with the situation. When that disclosure occurs, it is imperative to know how to respond, and how to respond in a positive 44 and helpful manner to encourage a healthy recovery instead of hindering it (Ullman, 1996). The present study’s results, in congruence with past research, show that the best responses validate the recipient and help them reappraise the situation (Freyd, 2017; Ullman & Peter-Hagene, 2014). High acceptance/high challenge messages continuously proved to be the best response in terms of validation, motivation, appropriateness, and facilitation of reappraisal ratings. Pragmatically, these results can help informal support providers (e.g., friends, family members, coworkers, etc.) support someone experiencing sexual harassment. Being believed after either informal or formal disclosure is critical for adaptive coping and allows the victim to regain some of their agency in the situation (Ullman & Peter-Hagene, 2014). Experiencing positive social reactions to disclosure of sexual harassment may help the victim perceive greater control over their recovery, gain the confidence to disclose again, and/or take more formal action against their harasser (Ullman & Peter-Hagene, 2014). Even further, these results could be useful for formal support providers, such as police officers, medical professionals, teachers/professors, supervisors, and so on. Although formal support providers are in a role to provide assistance for more formal or legal repercussions, believing the victim and assuring them is also a crucial part of supporting them. Additionally, many victims avoid formally reporting for fear of embarrassment, more severe harassment, or retaliation (Fitzgerald et al., 1995b). Formal support providers should try their best to reassure victims that they would be protected against these negative outcomes, and they could do so through these validating and motivating strategies. Limitations and Future Directions A few limitations existed in the sample used for the present study. The sample was current or recent college students, so this does not capture the varied harassment experiences of a 45 larger population. However, the sexual harassment scenarios were written to be college-specific so participants could more easily relate to them and find them realistic. In future studies, a non- college sample should be used to see if the same results emerge. Additionally, most of the sample (73.5%) reported that they had never been sexually harassed. Replicating the present study with a sample of people who have all been sexually harassed may yield different results because they can draw on their past experiences, rather than imagining themselves in a hypothetical scenario. Out of the 25.3% of participants who said they had been sexually harassed, the majority of them were female (93%). This is congruent with past research that has shown women are more frequently sexually harassed than men (Berdahl & Moore, 2006; Terpstra & Baker, 1987; Thakur & Paul, 2017). This difference may explain why males found the sexual harassment scenarios less emotionally upsetting, realistic, and severe than females did. Additionally, the scenarios had a male professor harassing the participant, which may be less realistic for males to imagine, especially if they identify as heterosexual. Although there were a few sex differences apparent in the results, relatively few emerged considering there is a large disparity between females and males being sexually harassed. However, these few differences lead to new research questions about how to provide more specialized support. Some issues also emerged with the preliminary analyses of the perceived severity and emotional upset ratings. The perceived severity ratings between the high severity and moderate severity conditions were not significantly different from one another. To review, the moderate severity scenario depicted unwanted touching, with a professor putting his arm around the student, whereas the high severity scenario described sexual coercion, with the professor repeatedly pressuring a student to hook up with him in return for a better course grade. Even 46 though these scenarios were adapted from Baker et al. (1990) and were chosen because they reflected the distinct categories of sexual harassment that Fitzgerald et al. (1995a) outlined, participants seem to view unwanted touching and sexual coercion as similar in severity. Additionally, the emotional upset ratings for the low severity condition were not significantly different from the midpoint of a 7-point scale, indicating that participants did not find the low severity scenario particularly upsetting. The low severity condition described a professor telling an inappropriate and sexist joke in class, which is classified as gender harassment. One explanation for participants not becoming too upset by this scenario is that it happens frequently so participants may have accepted it as “normal.” Further, common responses to less severe or explicit sexual harassment is to do nothing, deny the situation, or diffuse it with humor (Fitzgerald et al., 1995a; Fitzgerald et al., 1995b). Future research should consider using overall more severe scenarios in order for participants to find them all emotionally upsetting and see a significant difference between the moderate and high severity conditions. Lastly, only one message per category was tested (i.e., only one low acceptance/low challenge message, only one high acceptance/low challenge message, etc.) for each scenario, which limits the generalizability of the results. For example, it is unknown if all high acceptance/high challenge messages are most effective, since only three were tested in the present study. Furthermore, one-shot messages are not how people naturally give support. Rather, support is given more through the context of a conversation, with a back-and-forth dialogue between the provider and recipient. A future direction is to study in-person supportive conversations that utilize confirmation messages and see if the same results emerge as the present study. 47 CONCLUSION The present study adds to the existing literature on confirmation theory by expanding it to new contexts – friendships, sexual harassment, one-shot messages – and examining if scenario severity plays a role in how confirming messages are evaluated. Scenario severity played a slight role with how challenge messages were rated, and sex differences emerged in regard to challenge message ratings, which creates a future avenue to explore. 48 APPENDICES 49 APPENDIX A Severity Scenarios Low Severity Scenario: Yesterday during class, Professor X began his lecture with his usual off-color, sex-oriented joke. You knew that more would follow as fellow students roared their approval. You wondered why your fellow classmates find them funny when you considered the jokes to be inappropriate and offensive. You wonder what you or your fellow classmates did to make your professor think he could tell inappropriate and uncomfortable jokes. (66 words) Moderate Severity Scenario: Yesterday during class, Professor X was trying to explain the details of a class project to you. He moved next to you and put his arm around you as he talked. You have asked him not to put his arm around you before, but he has continued to do so. You wonder if you led him on, or if it is your fault that he keeps touching you. (68 words) High Severity Scenario: Yesterday after class, Professor X asked you if you wanted to hook up with him. He has repeatedly asked you to hook up, and you have told him you are not interested but it keeps happening. Professor X has suggested that your course grade might be bumped up if you hooked up with him. You wonder if you led him on, or if it is your fault that he keeps asking you to hook up. (75 words) 50 APPENDIX B Confirmation Messages by Scenario Scenario Severity Condition Message Low Low A/Low C Low A/High C High A/Low C High A/High C Moderate Low A/Low C Low A/High C High A/Low C High A/High C High Low A/Low C Low A/High C High A/Low C High A/High C I’m not sure what to tell you, maybe you should focus on something else right now. Understand that this behavior is not your fault, and that you didn’t do anything to make this happen. That does sound really uncomfortable, I’m sorry you had to hear that. That does sound really uncomfortable, I’m sorry you had to hear that. Understand that this behavior is not your fault, and that you didn’t do anything to make this happen. I’m not sure what to tell you, maybe you should focus on something else right now. Understand that this behavior is not your fault, and that you didn’t do anything to make this happen. That must be really frustrating that you have told him not to touch you, but he keeps doing it. That must be really frustrating that you have told him not to touch you, but he keeps doing it. Understand that this behavior is not your fault, and that you didn’t do anything to make this happen. I’m not sure what to tell you, maybe you should focus on something else right now. Understand that this behavior is not your fault, and that you didn’t do anything to make this happen. I’m really sorry that happened to you, and that he keeps asking after you said no. I’m really sorry that happened to you, and that he keeps asking after you said no. Understand that this behavior is not your fault, and that you didn’t do anything to make this happen. Note: A = acceptance; C = challenge. Participants will be given all four messages in one severity condition. 51 APPENDIX C Survey Informed Consent This survey will ask you to provide your opinions about supportive messages pertaining to sexual harassment. This research study is being conducted by Allison Mazur from the Department of Communication at Michigan State University for her master’s thesis. We are looking for volunteers who are at least 18 years of age to participate in this study. There are no right or wrong answers to any of the questions on this survey. For your responses to be most helpful, it is important that you answer each question as honestly as you can. Please make sure you answer every question. It should take no more than about 25 minutes to complete the survey. We see no risks or discomfort associated with completing this survey. Your responses to this survey will be combined with the responses of many other people to generate a statistical profile of what people think about themselves and communication. Your responses will remain private and won’t be connected to your personal information. Participants who consent to take part in this survey will be awarded SONA credits through http://msucas.sona-systems.com. The duration of this online survey is approximately 25 minutes, so participants who complete this survey will receive .25 SONA credits. Please do not complete this survey if you did not register for it on SONA. Some studies have prerequisites. If you did not see this study advertised in your SONA account (e.g., if a friend forwarded you the link), you should not complete this study. In order to receive credit for participation you MUST be registered for this study. Participation in this online survey is voluntary. You may withdraw at any time without penalty. This means that no SONA credits will be deducted from your account, nor will withdrawal have any effect on your relationship with any of your instructors. We greatly value your participation in this research study. We want to remind you that your participation in this study is completely voluntary. You may choose to refuse to answer any particular question or quit participating in this study at any time. If you have any questions about this study, such as scientific issues, how to do any part of it, or to report an injury, please contact the lead researcher (her contact information is below). Principal Investigator: Dr. Amanda Holmstrom holmstr6@msu.edu 517-432-1286 556 Communication Arts & Sciences Building East Lansing, MI 48824 Lead Researcher: 52 Allison Mazur mazurall@msu.edu 301-300-0631 If you have any questions or concerns about your role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about this research study, you may contact, anonymously if you wish, Michigan State University Human Research Protection Program at 517-355-2180, FAX 517-432-4503, or e-mail irb@msu.edu, or regular mail at: 4000 Collins Rd, Suite 136, Lansing, MI 48910. If you have concerns or questions about this study, such as scientific issues, how to do any part of it, or to report an injury, please contact the lead researcher. Continuing on with the web survey indicates that you give your consent to participate. Section I: Age & Biological Sex 1. Are you 18 years or older? 1. Yes 2. No (sends to End of Survey) 2. What is your age? Please enter a numerical value in the text box provided (i.e., 18, 21, etc.). 3. What is your biological sex? 1. Male 2. Female Section II: Scenarios and Messages Severity Scenarios (randomly assigned one of the three severity scenarios) Directions: Please read the following scenario about sexual harassment and imagine that you are the student experiencing this scenario and Professor X is one of your male professors. After you read the scenario, you will be presented with 4 messages and questions associated with those messages. The scenario will be repeated for your convenience. a. Low severity scenario: Yesterday during class, Professor X began his lecture with his usual off-color, sex-oriented joke. You knew that more would follow as fellow students roared their approval. You wondered why your fellow classmates find them funny when you considered the jokes to be inappropriate and offensive. You wonder what you or your fellow classmates did to make your professor think he could tell inappropriate and uncomfortable jokes. b. Moderate severity scenario: Yesterday during class, Professor X was trying to explain the details of a class project to you. He moved next to you and put his arm around you as he talked. You have asked him not to put his arm around you before, but he has continued to do so. You wonder if you led him on, or if it is your fault that he keeps touching you. c. High severity scenario: Yesterday after class, Professor X asked you if you wanted to hook up with him. He has repeatedly asked you to hook up, and you have told him you 53 are not interested but it keeps happening. Professor X has suggested that your course grade might be bumped up if you hooked up with him. You wonder if you led him on, or if it is your fault that he keeps asking you to hook up. Confirmation Messages Directions: Imagine you told a close friend the next day about what happened in the scenario you just read. They respond by saying: "(insert message here)." Answer the following questions with that message in mind. a. Low severity scenario 1. Low acceptance/Low challenge: “I’m not sure what to tell you, maybe you should focus on something else right now.” 2. Low acceptance/High challenge: “Understand that this behavior is not your fault, and that you didn’t do anything to make this happen.” 3. High acceptance/Low challenge: “That does sound really uncomfortable, I’m sorry you had to hear that.” 4. High acceptance/High challenge: “That does sound really uncomfortable, I’m sorry you had to hear that. Understand that this behavior is not your fault, and that you didn’t do anything to make this happen.” b. Moderate severity scenario 1. Low acceptance/Low challenge: “I’m not sure what to tell you, maybe you should focus on something else right now.” 2. Low acceptance/High challenge: “Understand that this behavior is not your fault, and that you didn’t do anything to make this happen.” 3. High acceptance/Low challenge: “That must be really frustrating that you have told him not to touch you, but he keeps doing it.” 4. High acceptance/High challenge: “That must be really frustrating that you have told him not to touch you, but he keeps doing it. Understand that this behavior is not your fault, and that you didn’t do anything to make this happen.” c. High severity scenario 1. Low acceptance/Low challenge: “I’m not sure what to tell you, maybe you should focus on something else right now.” 2. Low acceptance/High challenge: “Understand that this behavior is not your fault, and that you didn’t do anything to make this happen.” 3. High acceptance/Low challenge: “I’m really sorry that happened to you, and that he keeps asking after you said no.” 4. High acceptance/High challenge: “I’m really sorry that happened to you, and that he keeps asking after you said no. Understand that this behavior is not your fault, and that you didn’t do anything to make this happen.” Message Validation (Dailey et al., 2010a) 7-point Likert-type scale (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly agree) 1. This message shows he/she cares about me. 2. This message is judgmental. (reverse coded) 3. This message is conveyed in a friendly voice 4. This message accepts my feelings. 5. This message is emotionally cold. (reverse coded) 54 Message Motivation (Dailey et al., 2010a) 7-point Likert-type scale (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly agree) 1. This message presents new information, ideas, or options about the scenario I read. 2. This message pushes me to advocate for myself. 3. This message helps me channel my negative emotions about the scenario I read into more positive actions. 4. This message encourages me to explore different ideas. Message Appropriateness (Canary & Spitzberg, 1987) 7-point Likert-type scale (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly agree) 1. This message includes some things that should not have been said. (reverse coded) 2. Some of the things said in this message are embarrassing to me. (reverse coded) 3. All of the things in this message are in good taste as far as I’m concerned. 4. At least one of the remarks in the message is rude. (reverse coded) Facilitation of Reappraisal (Jones & Wirtz, 2006) 7-point Likert-type scale (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly agree) 1. My friend made me think about the events I described during the conversation. 2. I feel that I ought to re-evaluate the event now after the conversation. 3. I don’t really see the event in a different light after the conversation. (reverse coded) 4. Talking with my friend about the event helped me get my mind off it. 5. I understand the situation better now that I talked about it with my friend. Section III: Manipulation Check Perceived Severity (MacGeorge et al., 2004) 7-point Likert-type scale (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly agree) Stem: The scenario I read… 1. described a significant problem. 2. described a trivial problem. (reverse coded) 3. described a major problem. 4. described an important problem. Emotional Upset (Bodie et al., 2011) 7-point Likert-type scale (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly agree) Stem: The scenario I read… 1. was distressing. 2. upset me. 3. troubled me. 4. bothered me. Realism (Shebib et al., 2019) 7-point Likert-type scale (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly agree) Stem: The scenario I read… 1. was realistic. 55 2. was believable. 3. commonly occurs. Section IV: Individual Differences Sexual Experiences Questionnaire – Short Form (Fitzgerald et al., 1995a) 5-point Likert- type scale (1 = never, 5 = always); Yes/No response for the last item Directions: Please indicate whether you have experienced any of the following behaviors from a male professor or instructor during your time at your university. 1. Crude sexual remarks. 2. Sexist comments. 3. Offensive remarks. 4. Unwanted touching. 5. Repeated requests for dates. 6. Repeated requests for sexual favors. 7. Implicit or explicit demands for sexual favors to avoid negative consequences. 8. Have you ever been sexually harassed? (Yes/No) Section V: Demographics 1. Which of the following best describes your race? 1. White or Caucasian 2. Hispanic or Latino 3. Black or African American 4. Native American or American Indian 5. Asian or Pacific Islander 6. Multiple races 7. Other 2. Please click the option that best describes your gender. 1. Extremely feminine 2. Moderately feminine 3. Slightly feminine 4. Androgynous 5. Slightly masculine 6. Moderately masculine 7. Extremely masculine 3. Were or are you a student at Michigan State University? 1. Yes 2. No 4. What is your class standing? 1. Freshman 2. Sophomore 56 3. Junior 4. Senior 5. 5th year senior or higher 6. Graduate student 7. Recent graduate, please enter your graduation year: _____ 5. Are you a communication major? 1. Yes 2. No 6. Are you an international student? 1. Yes 2. No 7. Have you taken this survey before for another class? (Note: your response will not affect whether you receive SONA/course credit.) 1. Yes 2. No Section VI: Resources Thank you for your time! We appreciate your participation in this study. If you or someone you know is experiencing relationship violence or sexual misconduct, please see the resources below available to support you nationwide or on Michigan State University’s campus and surrounding community. National Sexual Violence Resources: National Sexual Assault Hotline: https://hotline.rainn.org/online or (800) 656-HOPE National Street Harassment Hotline: https://hotline.rainn.org/ssh-en or (855) 897-5910 National Domestic Violence Hotline: https://www.thehotline.org/ or (800) 799-SAFE National Teen Dating Abuse Online Helpline: https://www.loveisrespect.org/ or (866) 331-9474 Know Your IX: https://www.knowyourix.org/ End Rape on Campus: https://endrapeoncampus.org/ Campus and Community Resources for MSU Students: Student Health Services at Olin: http://olin.msu.edu/ Sexual Assault Program: http://endrape.msu.edu Neighborhood Student Success Collaborative: https://nssc.msu.edu/ Psychological Clinic: http://psychology.psy.msu.edu/Clinic/ Couple and Family Therapy Clinic: https://hdfs.msu.edu/clinic Health4uProgram: http://health4u.msu.edu/ Employee Assistance Program (EAP): http://eap.msu.edu/ MSU Safe Place: http://safeplace.msu.edu/ Office of Cultural & Academic Transitions (OCAT): http://ocat.msu.edu/ 57 MSU WorkLife Office: https://worklife.msu.edu/ Center for Gender in Global Context: http://gencen.isp.msu.edu/ MSU LBGT Resource Center: http://lbgtrc.msu.edu/ Ask a Spartan: http://askus.msu.edu MSU Student Food Bank: https://foodbank.msu.edu Office for Inclusion and Intercultural Initiatives: http://www.inclusion.msu.edu/index.html Student Parent Resource Center: http://studentparents.msu.edu/ Planned Parenthood (Lansing): http://www.plannedparenthood.org 58 Table 1: Descriptive and Inferential Statistics for Preliminary Analyses M SD p-value 95% CI 1.24 1.41 5.05 3.89 0.55 < .001** Low Severity Scenario Emotional Upset Realism Moderate Severity Scenario Emotional Upset Realism High Severity Scenario Emotional Upset Realism Note. M = mean; SD = standard deviation; CI = confidence intervals; **p < .001. < .001** < .001** < .001** < .001** 5.19 5.33 5.57 1.03 5.00 1.57 1.40 1.22 -0.50, 0.27 0.72, 1.39 0.60, 1.41 1.30, 1.83 0.81, 1.57 1.00, 1.66 59 Table 2: Independent Samples t-Tests of Biological Sex on Emotional Upset, Realism, Perceived Severity, and Familiarity with Sexual Harassment Males M SD M Females SD t- df p- 95% CI Cohen’s r Emotional Upset 4.25 1.57 4.97 1.51 value -2.96 168 value .004* Realism 5.03 1.33 5.50 1.04 -2.53 168 .01* Perceived Severity Familiarity with SH 5.01 1.12 5.60 1.04 -2.96 168 .004* 1.24 0.58 1.25 0.45 -0.03 168 .98 d 0.46 0.22 0.39 0.19 0.46 0.22 0.005 0.002 -1.20, -0.24 -0.83, -0.10 -0.84, -0.17 -0.16, 0.15 Note: Sex was dichotomously coded as male or female; M = mean; SD = standard deviation; t = t-test statistic; df = degrees of freedom; CI = confidence intervals; Cohen’s d and r = effect sizes; *p < .05. 60 Table 3: Paired Samples t-Tests of Acceptance and Challenge on Validation Ratings M t- (SD) value df p-value Low A, Low C vs. High A, Low C Low A, Low C vs. Low A, High C Low A, Low C vs. High A, High C Low A, High C vs. High A, High C High A, Low C vs. Low A, High C High A, Low C vs. High A, High C -2.42 (1.69) -2.65 (1.91) -3.12 (1.87) -0.47 (1.10) -0.22 (1.30) -0.70 (1.18) -18.68 169 <.001** -18.03 169 <.001** -21.71 169 <.001** -5.62 169 <.001** -2.22 169 .028* -7.69 169 <.001** 95% CI -2.68, -2.17 -2.94, -2.36 -3.40, -2.84 -0.64, -0.31 -0.42, -0.02 -0.87, -0.52 Cohen’s d r 2.87 0.82 2.77 0.81 3.34 0.86 0.86 0.40 0.34 0.17 1.18 0.51 Note: A = acceptance, C = challenge, M = mean; SD = standard deviation; t = t-test statistic; df = degrees of freedom; CI = confidence intervals; Cohen’s d and r = effect sizes; *p < .05, **p < .001. 61 Table 4: Paired Samples t-Tests of Acceptance and Challenge on Motivation Ratings M t- (SD) value df p-value Low A, Low C vs. High A, Low C Low A, Low C vs. Low A, High C Low A, Low C vs. High A, High C Low A, High C vs. High A, High C High A, Low C vs. Low A, High C High A, Low C vs. High A, High C -0.78 (1.31) -1.93 (2.03) -2.22 (1.99) -0.29 (1.13) -1.15 (1.62) -1.44 (1.62) -7.76 169 <.001** -12.41 169 <.001** -14.57 169 <.001** -3.31 169 .001* -9.30 169 <.001** -11.61 169 <.001** 95% CI -0.98, -0.58 -2.24, -1.62 -2.52, -1.92 -0.46, -0.12 -1.40, -0.91 -1.69, -1.20 Cohen’s d r 1.19 0.51 1.91 0.69 2.24 0.75 0.51 0.25 1.43 0.58 1.79 0.67 Note: A = acceptance, C = challenge, M = mean; SD = standard deviation; t = t-test statistic; df = degrees of freedom; CI = confidence intervals; Cohen’s d and r = effect sizes; *p < .05, **p < .001. 62 Table 5: Paired Samples t-Tests of Acceptance and Challenge on Appropriateness Ratings M t- (SD) value df p-value Low A, Low C vs. High A, Low C Low A, Low C vs. Low A, High C Low A, Low C vs. High A, High C Low A, High C vs. High A, High C High A, Low C vs. Low A, High C High A, Low C vs. High A, High C -2.06 (1.81) -2.34 (1.97) -2.50 (2.04) -0.15 (0.89) -0.28 (1.19) -0.44 (1.10) -14.82 169 <.001** -15.51 169 <.001** -15.93 169 <.001** -2.27 169 .03* -3.11 169 .002* -5.21 169 <.001** 95% CI -2.33, -1.78 -2.64, -2.04 -2.81, -2.19 -0.29, -0.02 -0.46, -0.10 -0.60, -0.27 Cohen’s d r 2.28 0.75 2.39 0.77 2.45 0.78 0.35 0.17 0.48 0.23 0.80 0.37 Note: A = acceptance, C = challenge, M = mean; SD = standard deviation; t = t-test statistic; df = degrees of freedom; CI = confidence intervals; Cohen’s d and r = effect sizes; *p < .05, **p < .001. 63 Table 6: Paired Samples t-Tests of Acceptance and Challenge on Facilitation of Reappraisal Ratings M t- (SD) value df p-value Low A, Low C vs. High A, Low C Low A, Low C vs. Low A, High C Low A, Low C vs. High A, High C Low A, High C vs. High A, High C High A, Low C vs. Low A, High C High A, Low C vs. High A, High C -0.95 (1.35) -1.63 (1.74) -1.93 (1.67) -0.29 (1.02) -0.68 (1.32) -0.98 (1.20) -9.18 169 <.001** -12.21 169 <.001** -14.80 169 <.001** -3.76 169 <.001** -6.77 169 <.001** -10.69 169 <.001** 95% CI -1.15, -0.74 -1.90, -1.37 -2.19, -1.67 -0.45, -0.14 -0.88, -0.48 -1.16, -0.80 Cohen’s d r 1.41 0.58 1.88 0.68 2.28 0.75 0.58 0.28 1.04 0.46 1.64 0.64 Note: A = acceptance, C = challenge, M = mean; SD = standard deviation; t = t-test statistic; df = degrees of freedom; CI = confidence intervals; Cohen’s d and r = effect sizes; **p < .001. 64 Figure 1: Interaction of acceptance and challenge on message validation ratings. 65 Figure 2: Interaction of acceptance and challenge on message motivation ratings. 66 Figure 3: Interaction of acceptance and challenge on message appropriateness ratings. 67 Figure 4: Interaction of acceptance and challenge on facilitation of reappraisal ratings. 68 REFERENCES 69 REFERENCES Antonucci, T. C., & Akiyama, H. (1987). An examination of sex differences in social support among older men and women. Sex Roles, 17, 737-749. doi:10.1007/BF00287685 Arroyo, A., Segrin, C., Harwood, J., & Bonito, J. A. (2017). Co-rumination of fat talk and weight control practices: An application of confirmation theory. Health Communication, 32, 438-450. doi:10.1080/10410236.2016.1140263 Avina, C., & O'Donohue, W. (2002). Sexual harassment and PTSD: Is sexual harassment diagnosable trauma?. Journal of Traumatic Stress: Official Publication of The International Society for Traumatic Stress Studies, 15, 69-75. doi:10.1023/a:1014387429057 Baker, D. D., Terpstra, D. E., & Larntz, K. (1990). The influence of individual characteristics and severity of harassing behavior on reactions to sexual harassment. Sex Roles, 22, 305- 325. doi:10.1007/BF00288336 Banyard, V. L., Demers, J. M., Cohn, E. S., Edwards, K. M., Moynihan, M. M., Walsh, W. A., & Ward, S. K. (2017). Academic correlates of unwanted sexual contact, intercourse, stalking, and intimate partner violence: An understudied but important consequence for college students. Journal of Interpersonal Violence, 1-18. doi:10.1177/0886260517715022 Barbee, A. P. (1990). Interactive coping: The cheering-up process in close relationships. In S. Duck & R. C. Silver (Eds.), Personal relationships and social support (pp. 46–55). London, UK: Sage. Benson, D., & Thomson, G. (1982). Sexual harassment on a university campus: The confluence of authority relations, sexual interest, and gender stratification. Social Problems, 29, 236- 251. doi:10.2307/800157 Berdahl, J. L., & Moore, C. (2006). Workplace harassment: Double jeopardy for minority women. Journal of Applied Psychology, 91, 426-436. doi:10.1037/0021-9010.91.2.426 Bergman, M. E., Langhout, R. D., Palmieri, P. A., Cortina, L. M., & Fitzgerald, L. F. (2002). The (un) reasonableness of reporting: Antecedents and consequences of reporting sexual harassment. Journal of Applied Psychology, 87, 230-242. doi:10.1037//0021- 9010.87.2.230 Biaggio, M. K., Brownell, A., & Watts, D. L. (1991). Reporting and seeking support by victims of sexual offenses. Journal of Offender Rehabilitation, 17, 33-42. doi:10.1300/j076v17n01_03 70 Binder, R., Garcia, P., Johnson, B., & Fuentes-Afflick, E. (2018). Sexual harassment in medical schools: The challenge of covert retaliation as a barrier to reporting. Academic Medicine, 93, 1770-1773. doi:10.1097/acm.0000000000002302 Bingham, S. G., & Battey, K. M. (2005). Communication of social support to sexual harassment victims: Professors' responses to a student's narrative of unwanted sexual attention. Communication Studies, 56, 131-155. doi:10.1080/00089570500078767 Bodie, G. D., & Burleson, B. R. (2008). Explaining variations in the effects of supportive messages a dual-process framework. Annals of the International Communication Association, 32, 355-398. doi:10.1080/23808985.2008.11679082 Bodie, G. D., Burleson, B. R., Holmstrom, A. J., McCullough, J. D., Rack, J. J., Hanasono, L. K., & Rosier, J. G. (2011). Effects of cognitive complexity and emotional upset on processing supportive messages: Two tests of a dual-process theory of supportive communication outcomes. Human Communication Research, 37, 350-376. doi:10.1111/j.1468-2958.2011.01405.x Buber, M. (1965). The knowledge of man. New York, NY: Harper & Row. Buchanan, N. T., & Ormerod, A. J. (2002). Racialized sexual harassment in the lives of African American women. Women & Therapy, 25, 107–124. doi:10.1300/j015v25n03_08 Burda, P. C., Vaux, A., & Schill, T. (1984). Social support resources: Variation across sex and sex role. Personality and Social Psychology Bulletin, 10, 119-126. doi:10.1177/0146167284101014 Burleson, B. R., Hanasono, L. K., Bodie, G. D., Holmstrom, A. J., Rack, J. J., Rosier, J. G., & McCullough, J. D. (2011). Are gender differences in responses to supportive communication a matter of ability, motivation, or both? Reading patterns of situational responses through the lens of a dual-process theory. Communication Quarterly, 59, 37- 60. doi:10.1080/01463373.2011.541324 Burleson, B. R., Holmstrom, A. J., & Gilstrap, C. M. (2005). "Guys can't say that to guys": Four experiments assessing the normative motivation account for deficiencies in the emotional support provided by men. Communication Monographs, 72, 468-501. doi:10.1080/03637750500322636 Canary, D. J., & Spitzberg, B. H. (1987). Appropriateness and effectiveness perceptions of conflict strategies. Human Communication Research, 14, 93-120. doi:10.1111/j.1468- 2958.1987.tb00123.x Check, J. V., & Malamuth, N. M. (1983). Sex role stereotyping and reactions to depictions of stranger versus acquaintance rape. Journal of Personality and Social Psychology, 45, 344–356. doi:10.1037/0022-3514.45.2.344 71 Cortina, L. M., Swan, S., Fitzgerald, L. F., & Waldo, C. (1998). Sexual harassment and assault: Chilling the climate for women in academia. Psychology of Women Quarterly, 22, 419- 441. doi:10.1111/j.1471-6402.1998.tb00166.x Cortina, L. M., & Wasti, S. A. (2005). Profiles in coping: Responses to sexual harassment across persons, organizations, and cultures. Journal of Applied Psychology, 90, 182-192. doi:10.1037/0021-9010.90.1.182 Cranmer, G. A., Brann, M., & Weber, K. D. (2018). Challenge me!: Using confirmation theory to understand coach confirmation as an effective coaching behavior. Communication & Sport, 6, 239-259. doi:10.1177/2167479516684755 Dailey, R. M. (2006). Confirmation in parent-adolescent relationships and adolescent openness: Toward extending confirmation theory. Communication Monographs, 73, 434-458. doi:10.1080/03637750601055432 Dailey, R. M. (2008). Assessing the contribution of nonverbal behaviors in displays of confirmation during parent-adolescent interactions: An actor-partner interdependence model. Journal of Family Communication, 8, 62-91. doi:10.1080/15267430701573599 Dailey, R. M. (2010). Testing components of confirmation: How acceptance and challenge from mothers, fathers, and siblings are related to adolescent self-concept. Communication Monographs, 77, 592-617. doi:10.1080/03637751.2010.499366 Dailey, R. M., Crook, B., Glowacki, E., Prenger, E., & Winslow, A. A. (2016). Meeting weight management goals: The role of partner confirmation. Health Communication, 31, 1482– 1494. doi:10.1080/10410236.2015.1089398 Dailey, R. M., Richards, A. A., & Romo, L. K. (2010a). Communication with significant others about weight management: The role of confirmation in weight management attitudes and behaviors. Communication Research, 37, 644–673. doi:10.1177/ 0093650210362688 Dailey, R. M., Romo, L. K., & McCracken, A. A. (2010b). Messages about weight management: An examination of how acceptance and challenge are related to message effectiveness. Western Journal of Communication, 74, 457–483. doi:10.1080/10570314.2010.512279 Dailey, R. M., Romo, L. K., & Thompson, C. M. (2011). Confirmation in couples’ communication about weight management: An analysis of how both partners contribute to individuals’ health behaviors and conversational outcomes. Human Communication Research, 37, 553– 582. doi:10.1111/hcre.2011.37.issue-4 Dorris, J. (2017, November 8). After Kevin Spacey, gay men need a whisper network, too. The New Yorker. Retrieved from https://www.newyorker.com/culture/culture-desk/after- kevin-spacey-gay-men-need-a-whisper-network-too Ellis, K. (2002). Perceived parental confirmation: Development and validation of an 72 instrument. Southern Communication Journal, 67, 319-334. doi:10.1080/10417940209373242 Fair Employment Practices Guidelines. Waterford, CT: Bureau of Business Practice, 1978-1982. Farrow, R. (2017, October 23). From aggressive overtures to sexual assault: Harvey Weinstein's accusers tell their stories. The New Yorker. Retrieved from https://www.newyorker.com/news/news-desk/from-aggressive-overtures-to-sexual- assault-harvey-weinsteins-accusers-tell-their stories Feng, B., & MacGeorge, E. L. (2010). The influences of message and source factors on advice outcomes. Communication Research, 37, 553-575. doi:10.1177/0093650210368258 Fitzgerald, L. F. (1990, March). Assessing strategies for coping with harassment: A theoretical/empirical approach. Paper presented at the Midwinter Conference of the Association of Women in Psychology, Tempe, AZ. Fitzgerald, L. F., Gelfand, M. J., & Drasgow, F. (1995a). Measuring sexual harassment: Theoretical and psychometric advances. Basic and Applied Social Psychology, 17, 425- 445. doi:10.1207/s15324834basp1704_2 Fitzgerald, L. F., Shullman, S. L., Bailey, N., Richards, M., Swecker, J., Gold, Y., Ormerod, A. J., & Weitzman, L. (1988). The incidence and dimensions of sexual harassment in academia and the workplace. Journal of Vocational Behavior, 32, 152-175. doi:10.1016/0001-8791(88)90012-7 Fitzgerald, L. F., Swan, S., & Fischer, K. (1995b). Why didn't she just report him? The psychological and legal implications of women's responses to sexual harassment. Journal of Social Issues, 51, 117-138. doi:10.1111/j.1540-4560.1995.tb01312.x Freyd, J. J. (2017, November 22). How to listen when someone you know discloses sexual harassment or assault [Blog post]. Retrieved from Psychology Benefits Society website: https://psychologybenefits.org/2017/11/22/how-to-listen-when-someone-you-know- discloses-sexual-harassment-or-assault/ Friborg, M. K., Hansen, J. V., Aldrich, P. T., Folker, A. P., Kjær, S., Nielsen, M. B. D., Rugulies, R., & Madsen, I. E. H. (2017). Workplace sexual harassment and depressive symptoms: A cross-sectional multilevel analysis comparing harassment from clients or customers to harassment from other employees amongst 7603 Danish employees from 1041 organizations. BMC Public Health, 17, 675-12. doi:10.1186/s12889-017-4669-x Gruber, J. E., & Bjorn, L. (1986). Women's responses to sexual harassment: An analysis of sociocultural, organizational, and personal resource models. Social Science Quarterly, 67, 814-826. 73 Gutek, B. A., & Koss, M. P. (1993). Changed women and changed organizations: Consequences of and coping with sexual harassment. Journal of Vocational Behavior, 42, 28-48. doi:10.1006/jvbe.1993.1003 Gutek, B. A., Morasch, B., & Cohen, A. G. (1983). Interpreting social-sexual behavior in a work setting. Journal of Vocational Behavior, 22, 30-48. doi:10.1016/0001-8791(83)90004-0 Gutek, B. A., Nakamura, C. Y., Gahart, M., Handschumacher, I., & Russell, D. (1980). Sexuality and the workplace. Basic and Applied Social Psychology, 1, 255-265. doi:10.1207/s15324834basp0103_5 Holland, K. J., & Cortina, L. M. (2017). "it happens to girls all the time": Examining sexual assault survivors' reasons for not using campus supports. American Journal of Community Psychology, 59, 50-64. doi:10.1002/ajcp.12126 Holmstrom, A. J., Bodie, G. B., Burleson, B. R., McCullough, J. D., Rack, J. J., Hanasono, L. K., & Rosier, J. G. (2015). Testing a dual-process theory of supportive communication outcomes: How multiple factors influence outcomes in support situations. Communication Research, 42, 526-546. doi:10.1177/0093650213476293 Houle, J. N., Staff, J., Mortimer, J. T., Uggen, C., & Blackstone, A. (2011). The impact of sexual harassment on depressive symptoms during the early occupational career. Society and Mental Health, 1, 89-105. doi:10.1177/2156869311416827 Jensen, I. W., & Gutek, B. A. (1982). Attributions and assignment of responsibility in sexual harassment. Journal of Social Issues, 38, 121-136. doi:10.1111/j.1540- 4560.1982.tb01914.x Laing, R. D. (1961). The self and others. London, UK: Tavistock Publications. Loy, P. H., & Stewart, L. P. (1984). The extent and effects of the sexual harassment of working women. Sociological Focus, 17, 31-43. doi:10.1080/00380237.1984.10570460 Lucero, M. A., Allen, R. E., & Middleton, K. L. (2006). Sexual harassers: Behaviors, motives, and change over time. Sex Roles, 55, 331-343. doi:10.1007/s11199-006-9087-y MacGeorge, E. L., Graves, A. R., Feng, B., Gillihan, S. J., & Burleson, B. R. (2004). The myth of gender cultures: Similarities outweigh differences in men's and women's provision of and responses to supportive communication. Sex Roles, 50, 143-175. doi:10.1023/B:SERS.0000015549.88984.8d Magley, V. J. (2002). Coping with sexual harassment: Reconceptualizing women's resistance. Journal of Personality and Social Psychology, 83, 930-946. doi:10.1037/0022-3514.83.4.930 McDermut, J. F., Haaga, D. A. F., & Kirk, L. (2000). An evaluation of stress symptoms 74 associated with academic sexual harassment. Journal of Traumatic Stress, 13, 397-411. doi:10.1023/A:1007725022534 McDonald, P. (2012). Workplace sexual harassment 30 years on: A review of the literature. International Journal of Management Reviews, 14, 1–17. doi:10.1111/j.1468- 2370.2011.00300.x Mengo, C., & Black, B. M. (2016). Violence victimization on a college campus: Impact on GPA and school dropout. Journal of College Student Retention: Research, Theory & Practice, 18, 234-248. doi:10.1177/1521025115584750 Morganson, V. J., & Major, D. A. (2014). Exploring retaliation as a coping strategy in response to customer sexual harassment. Sex Roles, 71, 83-94. doi:10.1007/s11199-014-0373-9 O'Leary-Kelly, A. M., Bowes-Sperry, L., Bates, C. A., & Lean, E. R. (2009). Sexual harassment at work: A decade (plus) of progress. Journal of Management, 35, 503-536. doi:10.1177/0149206308330555 Pinchevsky, G. M., Magnuson, A. B., Augustyn, M. B., & Rennison, C. M. (2019). Sexual victimization and sexual harassment among college students: A comparative analysis. Journal of Family Violence, doi:10.1007/s10896-019-00082-y Reilly, M. E., Lott, B., & Gallogly, S. M. (1986). Sexual harassment of university students. Sex Roles, 15, 333–358. doi:10.1007/bf00287976 Rodino-Colocino, M. (2018). Me too, #MeToo: countering cruelty with empathy. Communication and Critical/Cultural Studies, 15, 96-100. doi:10.1080/14791420.2018.1435083 Rodriguez, N., Bingham Mira, C., Myers, H. F., Morris, J. K., & Cardoza, D. (2003). Family or friends: Who plays a greater supportive role for Latino college students? Cultural Diversity and Ethnic Minority Psychology, 9, 236-250. doi:10.1037/1099-9809.9.3.236 Rospenda, K. M., Richman, J. A., & Shannon, C. A. (2009). Prevalence and mental health correlates of harassment and discrimination in the workplace: Results from a national study. Journal of Interpersonal Violence, 24, 819-843. doi:10.1177/0886260508317182 Scarduzio, J. A., Sheff, S. E., & Smith, M. (2018). Coping and sexual harassment: How victims cope across multiple settings. Archives of Sexual Behavior, 47, 327-340. doi:10.1007/s10508-017-1065-7 Schrodt, P., Ledbetter, A. M., & Ohrt, J. K. (2007). Parental confirmation and affection as mediators of family communication patterns and children’s mental well-being. Journal of Family Communication, 7, 23-46. doi:10.1080/15267430709336667 Setoodeh, R., & Wagmeister, E. (2017, November 29). Matt Lauer accused of sexual harassment 75 by multiple women. Variety. Retrieved from http://variety.com/2017/biz/news/matt- lauer-accused-sexual-harassment-multiple-women-1202625959/ Shebib, S. J., Holmstrom, A. J., Summers, M. E., Clare, D. D., Reynolds, R. M., Poland, T. L., Royer, H. R., Mazur, A. P., & Moore, S. (2019). Two experiments testing order, interaction, and absolute effects of esteem support messages directed toward job seekers. Communication Research. Advanced online publication. doi:10.1177/0093650219831813 Street, A. E., Gradus, J. L., Stafford, J., & Kelly, K. (2007). Gender differences in experiences of sexual harassment: Data from a male-dominated environment. Journal of Consulting and Clinical Psychology, 75, 464-474. doi:10.1037/0022-006X.75.3.464 Symonds, M. (2010). The “Second injury” to victims of violent acts. The American Journal of Psychoanalysis, 70, 34–41. doi:10.1057/ajp.2009.38 Terpstra, D. E., & Baker, D. D. (1987). A hierarchy of sexual harassment. The Journal of Psychology, 121, 599-605. doi:10.1080/00223980.1987.9712689 Thakur, M. B., & Paul, P. (2017). Sexual harassment in academic institutions: A conceptual review. Journal of Psychosocial Research, 12, 33-40. Till, F. (1980). Sexual harassment: A report on the sexual harassment of students. Washington, DC: National Advisory Council on Women’s Educational Programs. Ullman, S. E. (1996). Social reactions, coping strategies, and self-blame attributions in adjustment to sexual assault. Psychology of Women Quarterly, 20, 505-526. doi:10.1111/j.1471-6402.1996.tb00319.x Ullman, S. E. (2000). Psychometric characteristics of the Social Reactions Questionnaire: A measure of reactions to sexual assault victims. Psychology of Women Quarterly, 24, 257– 271. doi:10.1111/j.1471-6402.2000.tb00208.x Ullman, S. E., & Peter-Hagene, L. (2014). Social reactions to sexual assault disclosure, coping, perceived control, and PTSD symptoms in sexual assault victims. Journal of Community Psychology, 42, 495-508. doi:10.1002/jcop.21624. U.S. Department of Justice. (2019, January 3). Sexual assault. Retrieved from https://www.justice.gov/ovw/sexual-assault U.S. Merit Systems Protection Board. (1981). Sexual harassment of federal workers: Is it a problem? Washington, DC: U.S. Government Printing Office. U.S. Merit Systems Protection Board. (1988). Sexual harassment of federal workers: An update. Washington, DC: U.S. Government Printing Office. 76 U.S. Merit Systems Protection Board. (1995). Sexual harassment in the federal workplace: Trends, progress, continuing challenges: A report to the President and the Congress of the United States. Washington, DC: US Government Printing Office. Watzlawick, P., Beavin, J. H., & Jackson, D. D. (1967). Pragmatics of human communication. New York, NY: Norton. Wilson, S. R., Dorrance Hall, E., Gettings, P. E., & Pastor, R. G. (2015). A multiple goals analysis of families attempting to encourage U.S. service members to seek behavioral health care: Linking the GPA model and confirmation theory. Communication Research. doi:10.1177/0093650215617507 Zaleski, E. H., Levey-Thors, C., & Schiaffino, K. M. (1998). Coping mechanisms, stress, social support, and health problems in college students. Applied Developmental Science, 2, 127- 137. doi:10.1207/s1532480xads0203_2 77