“I DIDN’T EXPECT YOU TO GOSSIP ABOUT ME!”: HOW GOSSIP VALENCE AND SOURCE EFFECT EMOTIONS AND BEHAVIORAL INTENTIONS FOR THE TARGET OF GOSSIP By Amanda Allard A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Communication – Doctor of Philosophy 2023 ABSTRACT This piece explores how positive and negative gossip communicated by either a friend or an acquaintance impacts a target of gossip’s expectancy violations, emotions (i.e., pride, shame, and hurt), state self-esteem, and desire to interact with the gossiper. Two pilot studies were conducted to ensure realistic scenarios were employed for the main analyses. Participants (N = 602) were recruited from a survey-based platform, Prolific. The survey was a 2 (gossip valence) x 2 (relationship with gossiper) x 2 (gossip scenario) between subject factorial design created through Qualtrics. Results revealed that gossip valence had a significant effect on expectancy violations such that positive gossip was perceived as a positive expectancy violation whereas negative gossip was perceived as a negative expectancy violation. Similarly, gossip valence impacted one’s desire to affiliate with the gossiper. Results indicated individuals who were positively gossiped about rated intentions to affiliate with the gossiper higher than individuals who were negatively gossiped about. Next, a series of multiple regressions were conducted to determine how expectancy violations influenced emotions and in turn how those emotions influenced one’s state self-esteem and desire to affiliate with the gossiper. Finally, three serial mediation models were analyzed to explore whether the relationship between gossip valence and intentions to affiliate with the gossiper were mediated by positive expectancy violations and pride, negative expectancy violations and hurt, and negative expectancy violations and shame. The theoretical and pragmatic implications of these findings are discussed throughout the manuscript. This dissertation is dedicated to all first-generation students. You are more capable than you could ever imagine. iii ACKNOWLEDGEMENTS I would first and foremost like to thank God for allowing me the intellectual capability to obtain a higher level of education, for providing me with the opportunities to connect with others, and for being my peace during difficult times. This gift He has given me will be used to help others connect more effectively with one another. In addition, I would like to thank and acknowledge the individuals God allowed into my life that helped support me throughout the completion of my dissertation. To my parents, Robert and Linda Allard, thank you for always believing in me, supporting me, and providing an environment where learning was always accessible to me. I know that whatever happens in life that I have a safe and secure foundation to come back to. Without your love and support I would not have the drive or ambition I have today, nor would I be the woman I am today. Thank you for everything! I would also like to thank my brother, Robert Allard II, for making me a big sister. You taught me at a young age the power empathy and selflessness can have on an individual and I will always love, respect, and treasure our relationship because of it. A special thanks is required to my advisor, Dr. Monique Turner. Your guidance on this project has sparked a love for education and research that I will cherish forever. Without your support I would not have been able to complete this project or my program at Michigan State University. You are the epitome of what every academic should model themselves after, and please know you have made a huge difference in not only my life but the lives of so many graduate students. Thank you! Similarly, I would like to thank my dissertation committee, Dr. Amanda Holmstrom, Dr. William Chopik, and Dr. Elizabeth Dorrance-Hall. The expertise you iv all provided towards this project pertaining to relationships and gossip is priceless. Working with each one of you helped me grow as a scholar, and I am eternally grateful. Next, I would like to give a special thanks to every graduate student who I had the pleasure of meeting or working with during my time at Michigan State University. You all had a crucial role in supporting me through my programs which ultimately assisted me in completing my doctorate. However, a special thanks is required to Matt Olah who is one of my best friends and closest PhD cohort members. Thank you for being my friend, allowing me to sleep on your futon in times of need, and always supporting me. Our daily phone calls for advice, jokes about academia, and support to follow my dreams has not been taken for granted. You are one of the few people I would drop everything to work on a project with, and I can’t wait to see how our friendship grows in the future. Finally, I would like to say thank you to my soon to be husband, Dylan Schuster. Thank you for believing in me, thank you for seeing me when others didn’t, and thank you for showing me a love few have the pleasure of experiencing in this lifetime. You mean so many things to me and this project would not have been completed without your love, support, and dedication to our relationship. You amaze me every day, you motivate me to be the best version of myself, and you are someone I am honored to share this space and time with. I’m forever yours and will love and respect you always. v TABLE OF CONTENTS INTRODUCTION..........................................................................................................................1 METHOD .....................................................................................................................................20 PILOT STUDY 1 ..........................................................................................................................21 PILOT STUDY 2 .........................................................................................................................24 PILOT STUDY 2 RESULTS ......................................................................................................27 PILOT STUDY 2 DISCUSSION ................................................................................................31 MAIN STUDY ..............................................................................................................................33 MAIN STUDY RESULTS ...........................................................................................................37 DISSCUSION ..............................................................................................................................50 CONCLUSION ............................................................................................................................65 REFERENCES .............................................................................................................................66 APPENDIX A: Tables and Figures ...........................................................................................74 APPENDIX B: Pilot Study Survey .............................................................................................89 APPENDIX C: Qualitative Themes from Pilot Study 1 ...........................................................95 APPENDIX D: Pilot Study 2 Survey ..........................................................................................97 APPENDIX E: Instruments for Pilot Study 2 .........................................................................102 APPENDIX F: Main Study Instruments .................................................................................107 APPENDIX G: Main Study Survey .........................................................................................117 ix INTRODUCTION Gossip, evaluative communication about an absent third-party, has been estimated to comprise up to 70 percent of individuals’ daily conversations (Beersma & Kleef, 2012; Foster, 2004; Torres, 2019). Although people often hold the belief that all gossip is distasteful or “nasty”, gossip can be positive or negative (Turner et al., 2003). Gossip is positively valenced when evaluations of an absent third-party are deemed to be about “socially approved behavior”, whereas gossip is classified as negatively valenced when the content regards socially unapproved of behavior or characteristics (Turner et al., 2003, p. 132). Previous research has attended to both the gossiper (Farley et al., 2010; Turner et al., 2003) and the gossip recipient (Martinescu et al., 2014) finding that gossip affects credibility, self-evaluation, likeability, and trustworthiness (though the nature of the effect depends on other factors such as the valence of the gossip). Turner et al., (2003) conducted an experiment to understand how the gossip valence (positive or negative) interacted with the relationship with the gossiper (friend or stranger) to influence perceptions of the gossiper. Results showed that relationship type significantly interacted with the gossip valence to influence perceptions of trust, liking, and expertise of the gossiper. For example, gossip valence did not play a role in influencing perceptions of liking when the gossiper was a friend, but it did negatively influence liking for strangers when they received negative gossip from them. The authors also found that regardless of valence of gossip or relationship type, credibility of the gossiper was reduced when they engaged in the act of gossiping. These findings show that understanding the nuances of gossiping and its effects is important because gossiping can affect our perceptions of individuals and the relationship quality we have with others (Turner et al., 2003). 1 However, there are at least three parties involved in gossip: the gossiper, the gossip recipient(s), and the target of the gossip. Still, there is scant research regarding how being gossiped about affects the gossip target. This leaves open a critical question: What happens to targets when they find out people gossiped about them? Martinescu et al., (2019) examined the emotions gossip targets felt after learning they had been gossiped about by a co-worker, and how those emotions affected their desire to interact with the gossiper. Their data showed that targets of positive gossip were more likely to experience a positive emotion (i.e., pride) relative to targets of negative gossip who were more likely to experience a negative emotion (i.e., guilt), revealing that the valence of gossip is a critical determinant of emotional outcomes. However, Martinescu et al.’s study only examined work relationships. Given that people have different expectations of friends than they may have for co-workers, it is unclear if Martinescu et al.’s findings apply to other interpersonal relationships such as with friends or acquaintances. The purpose of this dissertation is to build upon both Turner et al., (2003) and Martinescu et al’s., (2019) research by analyzing the effects gossip has on targets of gossip when they are gossiped about by a friend versus an acquaintance, and to enhance our knowledge on gossip as a communicative phenomenon. Informed by Burgoon and Hales’s (1988) expectancy violation theory (EVT) this study will attempt to uncover how one’s expectations of the gossiper affect emotional consequences for the target of gossip when they are gossiped about by either a friend or an acquaintance who gossip in a negative or positive manner and attempt to understand how those emotions could then affect their state self-esteem and relational quality with the gossiper. This study will help enhance the explanatory power of understanding how certain forms of communication, in this case gossip, can cause detrimental or positive consequences dependent 2 upon the relationship with the gossiper and valence of gossip. Pragmatically, findings from this study may be able to assist in identifying how communication can impact a crucial interpersonal relationship in one’s life, friendships. In addition, findings from this study may be beneficial for the public because friendships are critical relationships that assist in enhancing one’s overall well-being (Bliezner et al., 2019). Gossip Gossip is evaluative communication about an absent third-party (Turner et al., 2003) and is distinctive from similar concepts like rumors. Meng et al. (2022) defined rumors as “information which (is) ambiguous or unconfirmed on authenticity but spreads rapidly among people” (p. 92). The authors noted that people can gossip about rumors illustrating an inherent difference between the two concepts. Pheko (2018) addressed the similarities and differences between rumors and gossip. She defined rumors as pieces of information that are spread amongst individuals to assist them in reducing uncertainty and to assist them in making judgments regarding protentional threats. Although Meng et al., (2022) and Pheko (2018) utilized two different conceptual definitions of rumors, both pieces highlight that gossip is conceptually different than rumors because gossip requires there to be an evaluation made about another person whereas rumors do not. Functionally, gossip can assist people in (a) adapting to social interactions, (b) the maintenance and development of relationships, and (c) allowing individuals the ability to compare themselves to others by saving face (Foster, 2004). Dunbar (2004) described gossiping as “the core of human social relationships… (and) society itself” (p. 100). Although gossipers may have a variety of reasons for gossiping, it should be clear that the target of the gossip would typically not be privy to that rationale. Instead, if a gossip target learns that they were gossiped 3 about, it is probable that they would appraise (i.e., interpret) the event. One portion of that appraisal would be to examine what was said about them; that is, individuals may use the valence of the gossip and perhaps the topic to guide how they feel and their intentions of the future of the relationship. Gossip Valence Despite common misperceptions of gossip being a malicious communicative act, gossip can be complimentary in nature (Foster, 2004). Gossip that is positively valenced includes evaluations made about an absent third party that regards socially approved behavior or compliments (e.g., “I think it is great John got that big promotion”) whereas negative gossip occurs when evaluations are made about an absent third party regarding socially unapproved behavior or insults (e.g., “I don’t think John is competent enough to receive that promotion”) (Turner et al., 2003). The outcomes associated with positive and negative gossip may vary depending on whether one is analyzing the gossiper or the target of gossip. For example, some research has highlighted that perceptions of gossipers can be overwhelming negative (Peters and Kashima, 2015). Similarly, Turner and colleagues (2003) found that gossip, regardless of its valence and who it was shared with (a stranger or friend), caused the recipient of gossip to view the gossiper as less credible. Turner, Allard, and Patel (under review), however, were unable to replicate Turner et al.’s 2003 findings and instead showed that if the gossiper was a friend the valence of the gossip did not affect perceptions of likeability and trustworthiness. However, when the gossiper was a stranger, findings revealed that positive gossip enhanced perceptions of likeability and trustworthiness whereas negative gossip decreased perceptions of likeability and trustworthiness, which they explained through the lens of expectancy violations theory (Burgoon & Hale, 1988). This shows that being perceived as a gossiper has the potential to result in others 4 making negative attributions about the speaker. Less is understood, however, with how it might affect the target. Martinescu et al.’s study (2019) is one of the few studies to analyze the outcomes for targets of positive or negative gossip. They conducted three studies to understand the effects gossip has on a target’s performance in an organizational setting. The first study recruited students from a Dutch University and asked participants to read either positive or negative gossip about themselves. The purpose of this study was to analyze the mediation effects of gossip valence on behavior through one’s emotional reaction to the gossip. Results indicated that negative gossip resulted in participants experiencing negative emotions (i.e., guilt) and a desire to repair the relationship more than participants who were gossiped about positively. Moreover, participants who were gossiped about positively reported experiencing positive emotions such as pride and an enhanced desire to affiliate with the gossiper. The authors conducted a second study to determine if findings from study one could be replicated, and to determine whether the findings differed based on if gossip was substituted for feedback. Participants who worked at least 20 hours a week in the United States were recruited from Amazon MTurk and were asked to read and imagine a scenario where they received either a positive or negative feedback in the form of an email from a coworker, or they overheard their coworker gossiping about them in a positive or negative manner. Results were harmonious with the first study and there was a distinguishable difference between feedback and gossip such that participants who received feedback experienced more intense negative emotions than those who were gossiped about. Finally, the purpose of part three was to identify if an individual’s self- evaluation or concern for their reputation interacted with the gossip valence to impact the outcomes of interest. Dutch employees from various occupations were recruited for part three of 5 this study. Participants were randomly assigned to either a positive or negative gossip condition and were asked to recall a time when a coworker gossiped about them before completing dependent measures. Results were consistent with the previous studies and reiterated that negative gossip led participants to feeling stronger negative self-conscious emotions (e.g., guilt) and positive gossip led participants to feeling stronger positive self-conscious emotions (e.g, pride). In addition, targets of negative gossip experienced a greater desire to retaliate against the gossiper because they experienced emotions such as anger whereas when they were gossiped about positively, they generated feelings of liking for the gossiper. However, there are additional outcomes targets of gossip may experience based on the valence of gossip. Targets of Negative Gossip When targets of gossip discover they have been gossiped about in a negative way it can impact their actions and perceptions. For example, research has found that some individuals would rather self-disclose sensitive information about themselves, such as having HIV, than be gossiped about negatively (Stutterheim et al., 2017). In terms of negative gossip impacting perceptions, Wu et al., (2018) analyzed the effects of negative gossip on targets in an organizational setting and found that when individuals perceived there to be negative workplace gossip about them it negatively impacted their perception of the company in its entirety. Hence, negative gossip has the potential to influence one’s communicative actions and perceptions, but negative gossip can also affect an individual emotionally. As previously mentioned, Martinescu et al., (2019) found that targets of gossip perceived negative gossip about them to be associated with negative emotional outcomes (i.e., guilt). Another negative emotion targets of negative gossip may experience is hurt (De Gouveia et al., 2005). An individual experiences the emotion of hurt “when they perceive they have been 6 harmed (or) some transgression has been committed against them” (Young et al., 2005; p. 123). Vangelisti et al., (2005) explained that this emotion results in feelings of hurt because of others violating their private matter, judging their actions, or through the manipulation of reality. In fact, individuals may experience a variety of negative self-conscious emotions (e.g. anger, fear, shame) when they are negatively gossiped about because they are unable to control the information and evaluations that are being shared with others (Petronio, 2002). Thus, it can be concluded that negative gossip can result in unfavorable outcomes for an individual. Targets of Positive Gossip Similarly, positive gossip can beget positive outcomes. Support for this claim is derived from research that found when individuals are gossiped about in a positive manner it can enhance their desire to associate with the gossiper (Martinescu et al., 2019). This finding is in line with previous literature which explained that feelings of similarity can assist individuals in experiencing feelings of closeness with another individual (Weaver & Bosson, 2011) and can aid in the development or maintenance of a friendship (Vieth et al., 2022). Therefore, previous research has highlighted how positive gossip can result in positive relational and emotional outcomes (Martinescu et al., 2019). Although gossip can be a powerful communicative tool that can have varying outcomes for the target of gossip based on the valence of gossip (positive or negative), outcomes for the target of gossip may also be influenced by their relationship with the gossiper (Dunbar, 2004). Gossip and Relationships Previous research conducted by Turner et al., (2003) found that the relationship a gossip recipient has with the gossiper can impact perceptions of the gossiper. This finding indicates that the relationship one has with the gossiper is influential in swaying interpretations of the 7 communicative act of gossiping. The two relationships targets have with the gossiper that are explored in this dissertation are friendships and acquaintances. Foster (2004) identified friendship, or the “cooperative alliance between genetically unrelated conspecifics” (Krems & Williams, p. 978), as being one of the major reasons people gossip, and explained that gossip typically occurs amongst friends more often than acquaintances. Friendships are important interpersonal relationships that allow individuals to be socially connected to others as well as enhance their overall psychological well-being (Blieszner et al., 2019). Communicating within these relationships is imperative to maintaining them and ensuring their longevity (Roberts & Dunbar, 2011). Thus, it can be inferred that since gossip comprises a great deal of everyday conversations, and that friends will communicate often to maintain their relationships, gossip most likely occurs between friends (Dunbar, 2004). However, gossip can occur amongst individuals who have little to no relationship with each other to create assimilation in groups or to begin developing the interpersonal relationship (Dunbar, 2004, Foster 2004, Weaver & Bosson, 2011). This can be seen in Martinescu et al.’s (2019) piece which analyzed outcomes targets of gossip experienced when they were gossiped about by an individual with a relationship like an acquaintance, a co-worker. The key differences between a friend and an acquaintance are that in an acquaintance relationship you typically know very little about the individual and there is almost no intimacy within the relationship (Bryant & Marmo, 2012). Hall (2018) explained that individuals are easily able to make the distinction between a friend and an acquaintance. But, targets of gossip may become aware that they have been gossiped about by a friend or an acquaintance. For example, Krems and Williams (2021) found that individuals can experience feelings of jealousy when they perceive their friendship has been threatened by an 8 absent third-party, or a new friend that may steal the attention of their already established friendship. They may become aware of this threat to the relationship by this “new friend” through the form of gossip, and the threat of this new friend having the power to influence their established relationship through negative gossip can invoke negative feelings such as distress for an individual (Vieth, 2022). Regarding the relationship targets of gossip have with an acquaintance, Martinescu et al. (2019) found that when individuals discovered they had been gossiped about by a co-worker, a relationship which can be synonymous with an acquaintance (McAndrew et al., 2007), the targets experienced positive and negative outcomes dependent upon the valence of the gossip. These findings indicate that not only does the gossip valence matter in identifying potential outcomes of gossip but so does the relationship with the gossiper. Hence, this piece argues that the relationship between the target of gossip and gossiper, whether they are a friend or an acquaintance, should impact the psychological consequences for the target of gossip. In addition, individuals may have different expectations of being gossiped about by acquaintances versus friends. The following section explores the nuances of these expectations by employing expectancy violation theory (EVT; Burgoon, 1993; Burgoon & Hale, 1988; Burgoon & Jones, 1976). Expectancy Violation Theory (EVT) Originally, expectancy violation theory (EVT) was created to understand perceptions of how nonverbal communication violated individual’s expectations (Burgoon & Hale, 1988). Since then, EVT has been expanded and adapted to also include verbal expectancy violations (see Burgoon, 1992, 1993). A verbal expectation is defined as “patterns of anticipated verbal…behavior” (Burgoon & Hubbard, 2005, p. 151). Expectations can either be met, violated, or exceeded during an interaction (Burgoon, 1992;1996). More specifically, when an individual’s 9 expectations are met or exceeded, it is considered a positive violation, but if an individual’s expectations are not met then it is considered a negative violation (Levine et al., 2000). The way a communicative event is interpreted as either being a positive or negative violation is all dependent on how it is appraised (Planalp & Rosenberg, 2014). Lerner and Keltner (2000) posits that when people are exposed to an event (i.e., social stimuli) they come to understand it by appraising it. An individual appraises an event when they interpret nonverbal or verbal stimuli, and then use those interpretations to influence their reactions or emotions depending (Chen et al., 2023). The valence of appraisals can influence one’s overall mood such that positive appraisals beget positive outlooks whereas negative appraisals can lead to a negative state of mind (Kelly et al., 2012). Similarly, the previous section regarding outcomes of gossip illustrated how positive gossip tends to precede positive outcomes and negative gossip precedes negative outcomes. Thus, it is predicted that individuals who find themselves as the target of positive gossip will perceive the event as a positive expectancy violation whereas individuals who are the target of negative gossip will experience a negative expectancy violation. H1: a) Positively valenced gossip will have a significant and positive effect on positive expectancy violations, whereas b) negatively valenced gossip will have a significant effect on negative expectancy violations. Furthermore, Burgoon (1995) explains that individuals formulate expectations of others as the number of interactions increase. Similarly, individuals may formulate expectations of how their friends will act or communicate based on the identification of communicative patterns throughout the duration of their relationship with them. In contrast, Burgoon goes on to explain that when individuals meet strangers, they rely on societal norms to determine if communicative acts violate expectations. Hence, when an individual has little to no relationship with another 10 individual, like the relationship with an acquaintance, they will rely heavily on social norms to make sense of their actions more than with their friends. Turner et al., (under review) explained that positive gossip is a socially approved evaluation of the target that would most likely not be deemed as a violation of expectation if the gossiper was a friend. Instead, it would meet the expectations of the target of gossip because they would expect their friend to say positive things about them. However, if an acquaintance positively gossips about an individual, it will most likely exceed their expectations because they would not be expecting such a kind gesture from someone, they have little to no relationship with. In contrast, negative gossip, or a socially disapproving evaluation, made by an acquaintance would most likely violate an individual’s expectations because it is not socially acceptable to be negatively gossiped about by others (Wert & Salovey, 2004). Furthermore, individuals may not expect to be gossiped about by a friend more than an acquaintance because trust is a major component of friendships and would violate the norms of the relationship (Giardini & Wittek, 2019; Warris & Rafique, 2009). Hence, it is predicted that the relationship with the gossiper will interact with the gossip valence to influence whether expectations are positively or negatively violated. More specifically, targets of positive gossip will report having their expectations positively violated more when they are gossiped about by an acquaintance versus a friend. In contrast, a target of negative gossip will report having their expectations negatively violated more when they are gossiped about by a friend compared to an acquaintance. H2: The type of relationship between the gossiper and the target (friends versus acquaintance) will interact with the gossip valence on expectations such that a) the effect of positive gossip on positive violations will be greater when the gossiper is an 11 acquaintance and b) the effect of negative gossip on negative violations will be greater when the gossiper is a friend. This dissertation is making the case that the outcomes of being gossiped about are heavily dependent on what was said and who said. The following sections will discuss more specifically what outcomes could be expected when expectations are positively or negatively violated after an individual discovers they have been either positively or negatively gossiped about by a friend or an acquaintance. Expectancy Violations and Emotions Burgoon (1993) states that EVT’s “key concepts, principles, and predictions are applicable to… the function of emotional communication” (p. 31). This is supported by research that indicated when an individual experiences a positive expectancy violation they may experience positive emotions, yet when an individual’s expectations are not met, they may experience negative emotions (Bennett et al., 2020; Biernat et al., 1999). Similarly, research has found that targets of positive gossip can experience positive feelings such as pride (Martinescu et al., 2019), whereas targets of negative gossip can experience negative self-conscious emotions such as hurt and shame (De Gouveia et al., 2005; Zong et al., 2022). Hence, the three emotions that will be examined as outcome emotions of expectancy violations will be pride, shame, and hurt. It is critical to mention that there are two distinct types of pride, hubristic pride and authentic pride. Tracy and Robbins (2007) distinguished the difference between the two facets of pride by explaining that authentic pride is based on one’s accomplishments whereas hubristic pride occurs because of “distorted and self-aggrandized self-views” (p.265). They also provide a comparison of authentic pride to being associated with positive behaviors related to success in 12 one’s life and relationship, and hubristic pride being in congruence with interpersonal conflicts, aggression, and negative outcomes. Since this piece is focused on how gossip may or may not illicit positive or negative self-conscious emotions through expectancy violations in the target of gossip, and previous research has shown that gossip can lead to feelings of pride in a positive manner, this piece will focus on authentic pride specifically. In tandem, Hardecker & Haun, (2020) explain that pride is an emotion that is also “based on the evaluation of the action of other individuals” (p.18). In fact, when an individual is perceived in a positive manner in a social standing it is often associated with that individual experiencing feelings of pride (Chen et al., 2020). This shows that an individual who is positively gossiped about will most likely experience feelings of pride because it will be considered a positive expectancy violation. Another emotion that is analyzed throughout this manuscript is shame. Contrary to feelings of pride, shame is an emotion that occurs within an individual when they appraise an event as being a threat to their personal perception of themselves, or when they fail to meet the expectations of others (Holmstrom et al., 2021). Hence, an individual may experience feelings of shame when they discover they have been negatively gossiped about by other individuals and would ultimately be a result of a negative violation of their expectations of self. Recently, Bennett and colleagues (2020) conducted a study analyzing the relationship between expectancy violation(s) and three specific emotions (i.e., surprise, anger, and hurt). Findings revealed that there was a negative association between violation valence, or “how much violations are (considered) negative or positive,” on feelings of hurt (p.23). Hence, the more a violation is positive the less likely individuals are to experience feeling hurt. Although this piece is not utilizing surprise or anger as outcome variables, the authors make the argument that positive expectancy violation should lead to positive emotions and vice versa for negative 13 violations. Therefore, this study will expand upon this research by predicting that individuals who report experiencing a positive expectancy violation will report higher levels of pride than shame and hurt. Hence, the following hypothesis are generated: H3: Positive expectancy violation will have a a) positive impact on feelings of pride but a negative impact on feelings of b) shame c) and hurt. H4: Negative expectancy violation will have a positive impact on feelings of a) shame and b) hurt but a negative impact on feelings of c) pride. In addition, there is research that illustrates how positive emotions (i.e., pride) and negative emotions (i.e., hurt) can impact one’s state self-esteem directly (Brown & Marshall, 2001; Ruvalcaba-Romero et al., 2017). Therefore, the relationship between emotions and state self- esteem will be explored in the upcoming section. Emotions and State Self-esteem To begin, self-esteem is defined by Heatheron and Wyland (2003) as, “the evaluative aspect of the self-concept that corresponds to an overall view of the self-as worthy or unworthy” (p.220). The authors go on to explain the difference between one’s self-esteem and their self- concept. They explain that one’s self-esteem is a part of the emotional response an individual has when they evaluate themselves whereas self-concept is an individual’s overall view of themselves which is based on race, beliefs, and values. This difference in conceptualizations indicate why this piece is more concerned with self-esteem than self-concept. Another important distinction to make is the difference between trait self-esteem and state self-esteem. Trait self- esteem is referring to one’s overall self-esteem levels whereas state self-esteem is regarding one’s self-esteem during a particular instance (De Ruiter et al., 2017). Since state self-esteem 14 fluctuates based on one’s environment and context, this study will analyze state self-esteem for targets of either negative or positive gossip. Furthermore, it is essential to clarify the difference between an emotion and state self- esteem. Heatherton and Wyland (2003) clarify the difference between the two concepts by describing self-esteem as an attitude which is conceptually different than an emotion. In addition, there is research that has shown that emotions can have a direct effect on one’s self-esteem. For example, Chang and Mackenzie (1998) explain that “self-esteem is a… symptom of depression” and clarify that depression is a mood disorder that is caused by negative emotions (p. 2325). This shows how emotions, specifically a negative one, can precede or cause an individual’s state self- esteem. Experiencing negative emotions such as hurt can also negatively influence one’s state self-esteem (Salice, 2020). Salice (2020) explains that “negative emotions have the power to reduce one’s self-esteem, whereas positive emotions enhance it” (p.198). In terms of pride, Ruvalcaba-Romero et al., (2017) found that positive emotions (i.e., pride) predicted self-esteem. Similarly, Brown and Marshall (2001) found that when an individual experienced the emotion of pride it was closely associated with one’s self-esteem. The authors also found that negative emotions were also significantly associated with an individual’s self-esteem. In tandem, Vangelisti et al., (2005) discovered that perceived causes of hurt feelings were associated with one’s state self-esteem, and Budiarto and Helmi (2021) conducted a meta-analysis where they discuss that shame has been found to negatively impact one’s state self-esteem. Therefore, it is predicted that feelings of pride will result in higher ratings of state self-esteem than feelings of hurt. H5: Feelings of a) pride will have a positive impact on one’s state-self-esteem whereas feelings of b) hurt and c) shame will have a negative impact on one’s state self-esteem. 15 Furthermore, gossip (Martinescu et al., 2019), violation valence (Burgoon, 2015), and one’s emotions (McLaren & Solomon, 2008) can all influence an individual’s relational intentions. Thus, the subsequent section explores how these variables of interest work together to influence the target of gossips intention to interact with the gossiper. Relational Intentions The valence of gossip can impact whether an individual desires to affiliate with the gossiper. Martinescu et al., (2019) explored how gossip valence affected the target of gossip’s intention to interact with the gossiper and found that when individuals were positively gossiped about by a co-worker, they were more inclined to want to affiliate with them more so than if they were negatively gossiped about. Similarly, this piece predicts that individuals who are the target of positive gossip will report a greater desire to want to be affiliated with the gossiper than individuals who are the target of negative gossip. H6: Targets of positive gossip will report a greater intention to affiliate with the gossiper than targets of negative gossip. In addition, this piece expands upon Martinescu et al.’s (2019) findings by looking at how the relationship with the gossiper influences the relationship between gossip valence on intentions to interact with the gossiper. EVT (Burgoon, 1993) illustrated that individuals form expectations regarding how friends versus acquaintances are expected to behave. Moreover, research has explained that when expectations are violated it impacts one’s behavioral interactions with the violator such that positive expectation violations lead to positive interactions whereas negative violations tend to lead to negative interactions (Burgoon, 1993). Although Foster (2004) indicated that gossip is an important form of communication utilized within relationships, being 16 gossiped about negatively by a friend or an acquaintance may not be congruent with our expectations of self or with expectations regarding the relationship. The key difference between the relationship with the gossiper being either a friend or an acquaintance is that the targets of gossip will most likely still affiliate with the friend regardless of the gossip valence due to the complexity of the relationship and the history between the two parties (Tolhuizen, 1989). In addition, individuals are more forgiving of friends than acquaintances because there is a greater relational investment in friendships (Marks et al., 2018). This indicates that targets of negative gossip would still most likely interact with the gossiper if they were a friend versus an acquaintance. In terms of positive gossip, targets will most likely expect their friends to speak highly of them to others, but this probably won’t change their affiliation. In contrast, when an individual is positively gossiped about by an acquaintance, they may not have expected them to say positive things about them, and this positive expectancy violation will lead to a greater desire to interact with the gossiper (Burgoon, 1993). Hence, it is predicted that the relationship with the gossiper will interact with the gossip valence to influence intentions to affiliate with the gossiper. More specifically, targets of positive gossip will report a greater desire to affiliate with acquaintances than friends. Similarly, targets of negative gossip will report a greater desire to not be affiliated with gossipers that are an acquaintance more than friends. These predictions are depicted in the following hypotheses: H7: Relationship type will interact with gossip of valence to influence intentions to affiliate with the gossiper such that: a) An individual who is gossiped about positively by an acquaintance will have a greater intention to affiliate with them than an individual who was gossiped about by a friend. b) In contrast, an individual who is negatively gossiped 17 about by an acquaintance will have a lower intention to affiliate with them than if they were negatively gossiped about by a friend. This manuscript has argued that expectancy violations, whether they are positive or negative can impact an individual’s emotions of feeling either pride, shame, or hurt and that these emotions can impact an individual's state self-esteem. Similarly, these feelings of pride, shame, and hurt can also determine whether an individual wants to interact with an individual. In fact, Fitness (2001) explained that when expectations are negatively violated it can lead to negative self- conscious emotions. Negative emotions, like hurt or shame, can cause people to want to distance themselves from the transgressor (McLaren & Solomon, 2008). Similarly, people's thoughts and feelings about themselves can impact how they believe they are evaluated by others (Leary et al., 1998). Hence, it can be concluded that individuals who are hurt or experience shame will most likely not want to affiliate with the gossiper compared to individuals who experience pride. H8: Feelings of a) hurt and b) shame will have a negative impact on target’s desire to affiliate with the gossiper whereas c) feelings of pride will have a positive impact on a target’s desire to affiliate with the gossiper. Finally, the relationship among all the variables of interest will be tested. More specifically, it is predicted that the indirect relationship between gossip valence and intentions to interact with the gossiper will be mediated by violation valence and emotions. More specifically, it is predicted that targets of negative gossip will experience a negative expectancy violation, feelings of hurt or shame, and ultimately a desire to not affiliate with the gossiper. In contrast, targets of positive gossip will experience a positive expectancy violation which will result in heightened levels of pride, and ultimately a desire to affiliate with the gossiper. 18 H9: The indirect relationship between positive gossip and greater intentions to affiliate with the gossiper is serially mediated by positive expectancy violations and pride. H10: The indirect relationship between negative gossip and lower intentions to affiliate with the gossiper is serially mediated by negative expectancy violations and hurt. H11: The indirect relationship between positive gossip and greater intentions to affiliate with the gossiper is serially mediated by negative expectancy violations and shame. Finally, a research question is posed regarding how the relationship with the gossiper will interact with the valence of gossip on intentions to affiliate with the gossiper through the variables of interest within this piece (i.e., violation valence and emotions). More specifically, since the relationship with friends tends to be deeper and richer than with an acquaintance will it interact with gossip valence on intentions to affiliate with the gossiper through expectancy violations and emotions (pride, shame, and hurt)? Hence, the following research question is presented: RQ1: Does the relationship with the gossiper interact with the gossip valence on intentions to affiliate through expectancy violations, pride, shame, and hurt? 19 Overview METHOD Before the conduction of the main study, it was imperative that the scenarios used were created based on realistic gossip scenarios. Thus, the purpose of pilot study 1 was twofold: to ascertain realistic scenarios from participants about a time where they were either positively or negatively gossiped about by a friend and an acquaintance, and to identify perceptions regarding why they believed they were gossiped about. The themes identified in the gossip scenarios recalled were then used to create scenarios which were then analyzed in terms of realism and severity in a second pilot study. 20 Participants PILOT STUDY 1 Participants (N =100) were recruited from both a large Midwest university (n = 48) and an online participant recruiting platform called Prolific (n = 51). Collectively, participants age ranged from 18 to 74 years of age with the average age of the participants being 20 years old. Most participants identified their sex as being female (n = 59) followed by male (n = 40). Regarding race, 80% identified as White/Caucasian (n = 80) followed by Asian (12%; n = 12), Black or African American (4%; n = 4), or other/prefer not to answer (3%; n = 3). Finally, the three main areas of the United States participants identified as being their home state were Michigan (n = 33), Illinois (n = 10) and California (n = 6). Procedure All participants completed this study by partaking in a survey developed from an online survey platform, Qualtrics. After providing consent and indicating an age of 18 or older, participants were randomly assigned to recall a time they were positively or negatively gossiped about by a friend and an acquaintance for each context [friend/negative (n = 50), friend/positive (n = 48), acquaintance/negative (n = 52), acquaintance/positive (n = 48)]). Definitions for “friend” and “acquaintance” were provided for participants as reference. More specifically, the definition for friend was adapted from Fehr (1996) and stated that a friend was defined as “someone (non-family) you hang out with often and have a close relationship with”. In contrast, the definition for an acquaintance was adapted from Bryant and Marmo (2012) and was defined as “a person you know very little about yet do not seek to hang out with them or connect with them regularly.” Next, participants were asked to answer open-ended questions assessing how they found out they had been gossiped about, what exactly the person said about them, and how 21 the gossip affected them. Subsequently, they answered additional open-ended questions asking the specific emotions they felt and what actions they took, if any, after learning they had been gossiped about. Finally, participants completed the survey by answering demographic questions followed by a debrief form where mental health resources were provided to them in case the recalling activity triggered negative emotions or thoughts (see Appendix A for an overview of pilot study 1). Qualitative Analysis After data was collected, an additional coder was recruited and trained to identify themes associated with the recalled gossip scenarios (n = 198). Coders went through the qualitative data independently to identify themes, and later met to discuss findings. From the data, eight common themes, or reasons why people were gossiped about, arose. Please reference Appendix B for a full description of these themes. The coders then took these themes and analyzed the data separately. It is important to note all eight themes were present in both the positive and negative conditions. Once completed the coders reconvened and discussed discrepancies in coding (n = 45) and were able to come to verbal resolutions regarding which theme represented the data (n = 18). Next, the data was numerically coded so inter-coder reliability could be assessed. Inter- coder reliability amongst the themes identified ranged from a = 0.79 to a = 1.00 indicating an acceptable level of agreement (Krippendorff, 2004). Please reference Table 1 for the results of the inter-coder reliability for each theme. Following the second round of coding, a third-party unbiased coder was trained and acted as a tiebreaker for the scenarios that the original two- coders held discrepancies on. After the third coder reviewed the scenarios and themes, there was complete agreement amongst all the recalled scenarios. Based on the data collected from the pilot study, the researchers moved forward with analyzing the three most common themes 22 present within the data including gossip about one’s personality/characteristics (n = 58), talents (n = 31), and work ethic (n = 24) for pilot study 2. 23 Participants PILOT STUDY 2 Participants were recruited from an online survey recruiting platform called Prolific (N = 199). The average age of participants was 34.12 (SD = 10.57; range = 18-70). Many participants identified as their sex being male (n = 104), followed by female (n = 91), non-binary (n = 3), and those who preferred not to answer (n = 1). In terms of race, 74.4% of participants identified as white (n = 151), followed by 9.4% identifying as Asian (n = 19), or 7.4% Black (African American) (n= 15), and the rest being of either American Indian (.5%), mixed (3.0%), or other dissent (3.4%). Finally, many of the participants reside in either California (n = 25, 12.3%), Texas (n = 22, 10.8%), or New York (n = 15, 7.4%). There was at least one participant from each US state. Procedure The purpose for the second pilot study was to ensure that the scenarios created based on pilot study 1 were perceived to be realistic and to assess the perceived valence and severity of each scenario to ensure the inductions were perceived as intended. After participants consented to engaging in the study conducted through Qualtrics, they were randomly assigned to imagine that either a friend or an acquaintance made an evaluation about them to a mutual same-sex friend/acquaintance. Next, all participants read three positive gossip and three negative gossip scenarios that were based on them being gossiped about due to personality/characteristic traits, their talent, or their work ethic. The order in which participants viewed positive and negative gossip scenarios was randomized. Similarly, the order by which each scenario within each gossip condition was presented was randomized. Participants rated each scenario based on realism, 24 severity, and perceived gossip valence before concluding the survey with demographic questions. Please reference Appendix C for a full depiction of pilot study 2. Instruments Realism. Perceived realism of each gossip scenario was assessed by utilizing a revised version of Shebib et al.’s (2020) scale. More specifically, the 3-item measure was adapted to consist of 4-items and was assessed along a 7-point Likert type scale (1=strongly disagree, 7=strongly agree). Items include statements such as “The scenario I just read [stem]… is realistic.” These items were calculated together to achieve an overall score of realism where higher scores indicated greater levels of realism. This scale was found to be internally consistent in this pilot study (a = 0.95, M = 5.43; SD = 1.00). Severity. This scale was adapted from Shebib et al.’s (2020) three item scale to a four- item scale measured along a 7-point Likert type scale (1=strongly disagree, 7=strongly agree). Items included statements such as “I would perceive this scenario as[stem]… severe.” Scores were summed together and averaged to then create an overall composite score. Like the realism scale, higher scores indicate greater levels of severity. This scale was found to be internally consistent in this pilot study (a = 0.90; M = 5.13; SD = 1.36). Perceptions of Gossiper. Participants answered three questions regarding perceptions of the gossiper and the valence of gossip. More specifically, participants were asked to assess two- items on a 7-point Likert type scale (1=strongly disagree, 7=strongly agree). These two items asked participants how much they agreed, or disagreed, with the statement that they viewed the gossip informant as “kind” and “helpful”. Manipulation Check. Participants assessed two manipulation checks throughout the survey. The first one was conducted after every scenario and participants were asked to identify 25 how they perceived the information they received from their friend/acquaintance. The choices they had to choose from were “positive”, “negative”, or “not sure”. The second manipulation check appeared after participants completed viewing all 6 scenarios. They were asked to identify the relationship they had with the gossip informant (i.e., friend, acquaintance, not sure). Demographic Questions. Participants answered questions regarding their age, biological sex, gender, ethnicity, race, and location. Please reference Appendix D for a complete depiction of all the instruments used in this pilot study. 26 PILOT STUDY 2 RESULTS Manipulation Checks. This study was a 2 (relationship: friend or acquaintance) x 3 (scenario: personality, talent, work ethic) x 2 (gossip valence (positive or negative) mixed methods design. The independent variables in this survey are all categorically measured and included the valence of gossip (positive or negative), the relationship to the gossiper (friend or acquaintance), and the scenarios (personality, talent, and work ethic). The dependent variables included perceived realism and severity both of which are measured continuously. Before conducting the main analyses two manipulation checks were conducted. The manipulation check was to ensure participants had the correct relationship (friend or acquaintance) in mind throughout the entire duration of the survey. Results from a chi-square showed that the manipulation was significant, 𝑥2 (1, N = 196) =176.53, p < .001. More specifically, 97.9% of participants that were assigned to the friend condition accurately identified the relationship with the gossiper as friend, and 96.9% of participants who were assigned to the acquaintance condition accurately identified the gossiper as an acquaintance. However, there were some participants who did fail the manipulation check (N =5). Since many of the participants passed the manipulation check the few participants who failed the manipulation check were not removed from the data as their results would not influence drastic changes for the outputs. The second manipulation check was conducted to ensure positive gossip scenarios were perceived as positive and negative gossip scenarios were perceived as negative. Frequency analyses revealed that 97.2% of participants perceived positive gossip scenarios as positive and negative gossip scenarios were perceived as negative. Since the chi-square analysis and the frequency tables indicated participants had the correct relationship in mind, and the scenarios were properly perceived in terms of gossip valence, the remaining analyses were conducted. 27 Realism. To determine if participants viewed the scenarios as realistic a 2 (relationship: friend or acquaintance) x 3 (scenario: personality, talent, work ethic) x 2 (gossip valence (positive or negative) repeated measures analysis of variance (RM-ANOVA) tests was conducted. Results revealed that there was not a significant interaction between the relationship with the gossiper (friend/acquaintance) and the scenario’s presented (Wilk’s L =0.97, F (5, 199) = 0.002, p = 0.96, partial ηp 2 = 0.000). However, results revealed there was a significant difference within the scenarios (Wilk’s L =0.49, F (5, 199) = 128.80, p < 0.001, partial ηp 2 = 0.40). To uncover these difference, two one-sample t-tests were conducted to determine within each gossip condition (positive or negative) which scenarios (personality/characteristics, talent, or work ethic) were perceived to be statistically significant above the scale midpoint of 4.0. The first one sample t-test assessed the positive gossip condition. Results revealed that gossip scenarios based on personality/characteristics (t(198) = 15.52, p < .001, Cohen's d = 1.10), talent (t(198) = 11.50, p < .001, Cohen's d = 0.82), and work ethic (t(198) = 20.16, p = p < .001, Cohen's d = 1.43) were all significantly above the scale midpoint. More specifically, gossip scenarios involving content regarding one’s work ethic (M = 5.43; SD =1.00) were perceived as the most realistic by participants followed by the scenarios regarding one’s personality (M = 5.31; SD =1.20) and then the gossip scenarios regarding one’s talents (M =5.06; SD = 1.30). For the negative gossip condition, results showed that scenarios based on gossip regarding one’s personality/characteristics (t(198) = 2.95, p = .004, Cohen's d = 0.21) was statistically significant above the scale midpoint whereas negative gossip about on one’s talents (t(198) = - 2.26, p = .03, Cohen's d = -0.16) was statistically below the scale midpoint. Furthermore, negative gossip scenarios about one’s work ethic were not statistically significant above the scale midpoint (t(198) =0.78, p = 44, Cohen's d = 0.05). Negative gossip scenarios based on one’s 28 personality/characteristics (M = 4.34; SD =1.61) was perceived as the most realistic, followed by negative gossip scenarios based on work ethic (M =4.09; SD =1.70), and then those based on one’s talent (M =3.73; SD =1.70). Please reference Table 2 for a depiction of this output. Severity. Another 2 (relationship: friend or acquaintance) x 3 (scenario: personality, talent, work ethic) x 2 (gossip valence (positive or negative) repeated measures analysis of variance (RM-ANOVA) tests was conducted to determine which conditions and which scenarios were perceived to be the most severe. Results revealed that there was not a significant interaction between the relationship with the gossiper (friend/acquaintance) and the scenario’s presented (Wilk’s L =0.97, F (5, 199) = 2.02, p = 0.07, partial ηp 2 = 0.01). However, results revealed there was a significant difference within the scenarios (Wilk’s L =0.16, F (5, 199) = 698.50, p = .000, partial ηp 2 = 0.80). To uncover these difference, two one-sample t-tests were conducted to determine within each gossip condition (positive or negative) which scenarios (personality/characteristics, talent, or work ethic) were perceived to be statistically significant above the scale midpoint of 4.0. The first one sample t-test assessed the positive gossip condition. Results revealed that gossip scenarios based on personality/characteristics (t(198) = -34.17, p < .001, Cohen's d = -2.42), talent (t(198) = -31.88, p < .00, Cohen's d = -2.26), and work ethic (t(198) = - 36.85, p = p < .001, Cohen's d = -2.61) were all significantly below the scale midpoint. More specifically, gossip scenarios regarding one’s talents (M = 2.08; SD =0.85) were perceived as the most severe by participants followed by the scenarios regarding one’s work ethic (M =2.04; SD =0.75) and then the gossip scenarios regarding one’s personality/characteristics (M =1.98; SD = 0.84). For the negative gossip condition, results revealed that gossip scenarios based on personality/characteristics (t(198) = 11.31, p < .001, Cohen's d = 0.80), talent (t(198) = 11.64, p 29 < .001, Cohen's d = 0.83), and work ethic (t(198) = 11.73, p = p < .001, Cohen's d = 0.83) were all significantly above the scale midpoint. Negative gossip scenarios based on one’s talents (M = 5.20; SD = 1.42) was perceived as the most severe, followed by negative gossip scenarios based on work ethic (M = 5.13; SD = 1.40), and then those based on one’s personality/characteristics (M = 5.07; SD = 1.33). Please reference Table 3 for a depiction of this output. 30 PILOT STUDY 2 DISCUSSION The main purpose of this pilot study was two-fold. The main goal was to determine which gossip scenarios were perceived as realistic for both the positive and negative gossip conditions, and to ensure the negative gossip scenarios were perceived to be severe and the positive gossip scenarios were not perceived as severe. Results revealed that in the positive condition, gossip regarding one’s work ethic was perceived to be the most realistic followed by scenarios based on one’s personality and then scenarios based on one’s talents. For the negative gossip condition, results revealed that gossip scenarios regarding one’s personality was perceived to be the most realistic followed by scenarios based on one’s work ethic and then those based on one’s talents. In terms of severity, the outputs revealed that in the positive condition gossip scenarios regarding one’s talents were perceived as the most severe by participants followed by the scenarios regarding one’s work ethic and then the gossip scenarios regarding one’s personality/characteristics. However, it is important to note that all three scenarios were statistically significant below the scale midpoint of 4.0. This is to be expected as positive gossip scenarios should not render severe outcomes (Kelly et al., 2012; Martinescu et al., 2019). However, for the negative gossip condition, the data indicated that all three scenarios were statistically significant above the scale midpoint of 4.0. Negative gossip scenarios based on one’s talents were perceived as the most severe, followed by negative gossip scenarios based on one’s work ethic and then those based on one’s personality/characteristics. For the main study it is critical that a scenario is chosen that is realistic for both positive and negative gossip conditions. Since all three negative gossip scenarios were perceived as severe and all three positive scenarios were perceived as not severe in nature the decision 31 regarding which scenario to use in the main study was based on perceptions of realism. In addition, both scenarios regarding one’s personality and work ethic had the highest mean scores in terms of realism for both the negative and positive gossip conditions. Therefore, these two scenarios will be used in the main study. This prompted the researchers to ask an additional research question: RQ2: What effect, if any, will the scenario have on the independent and dependent variables of interest? 32 Participants MAIN STUDY Participants (N = 602) were recruited from an online survey platform called Prolific. Of these participants, 51% identified as male (n = 307), 46 % as female (n = 277), 1.5% as non- binary (n = 9), and 1.2% selected “other” (n = 7). The average age of the participants was 38.36 (SD = 13.3) years of age with our youngest represented age being 18 and the oldest being 77. In terms of race, 77.6% identified as White/Caucasian (n = 467) followed by 7.3% identifying as Black/African American (n = 44), 6.8% Asian (n = 41), 4.7% mixed or multiple races (n = 28), and the remaining 3.6 % identified as either “other”, American Indian, or Native Hawaiian. Finally, the three most common states participants derived from were California (n = 66), Texas (n = 49), and New York (n = 39). However, it is important to note that there was at least one person from every state in the United States represented in this study. Design and Procedure The study was designed as a 2 (relationship: friend vs. acquaintance) X 2 (gossip valence: positive vs. negative) x 2 (scenario: personality vs. work ethic) between groups post-test only experiment. Once participants read and provided consent, they were instructed to imagine that they recently were informed that either a friend or an acquaintance made an evaluation about them. Participants were then directed to read either a positive or negative gossip scenario (based on pilot 2). After reading through the scenario, they were then asked to rate their levels of expectancy violation, hurt, pride, shame, state self-esteem, and intentions to affiliate with the gossiper. This was followed by a manipulation check and demographic questions before all participants were thanked for their time in completing this survey. 33 Instruments Expectancy Violation. Violation valence was assessed by using a subscale created by Afifi and Mett’s (1998). This subscale consisted of 5-items measured on a 7-point Likert-type scale. Higher values indicated a greater degree of a negative or positive expectancy violation. Participants rated statements such as, “The statements made about me[stem]….exceeded my personal expectations” and “The statements made about me [stem]…shocked me in a bad way”. The positive expectancy scale was found to be internally consistent in this study (a = 0.97; M = 3.87, SD = 2.31). Similarly, the negative expectancy scale was also found to be internally consistent in this study (a = 0.98; M = 3.74, SD = 2.50). Pride. To measure pride, 5-itmes were employed from Marshall et. al’s, (1994) 15-item State Shame and Guilt Scale (SSGS). This scale has been supported by Tracy and Robins (2007) as being an adequate scale to measure one’s state of pride. The full scale measures an individual’s state of shame, guilt, and pride by utilizing a 7-point Likert type scale (strongly disagree…Strongly Agree). However, only the 5 items assessing one’s state of pride were adapted and assessed for the current study. For example, participants in the present study assessed statements such as “The statements made about me would make me feel[stem]…good about myself” and “The statements made about me would make me feel[stem]…pleased with myself”. This scale was found to be internally consistent in this study (a = 0.98; M = 3.98, SD = 2.40). Shame. To measure feelings of shame, 5-itmes were employed from Marshall et. al’s, (1994) 15-item State Shame and Guilt Scale (SSGS). The full scale measures an individual’s state of shame, guilt, and pride by utilizing a 7-point Likert type scale (strongly disagree…Strongly Agree). However, only the 5 items assessing one’s state of pride were 34 adapted and assessed for the current study. For example, participants in the present study assessed statements such as “The statements made about me would make me feel[stem]…like sinking into the floor” and “The statements made about me would make me feel[stem]…like a bad person”. This scale was found to be internally consistent in this study (a = 0.97; M = 2.96, SD = 2.10). Hurt. Hurt feelings was measured by asking participants to rate 7 items on a 7-point Likert type scale (1 = strongly disagree…7= strongly agree). Participants rated statements such as, “The statements made about me would make me feel[stem]…unvalued” and “The statements made about me would make me feel[stem]…less than”. The first three items (e.g., hurt, disappointed, wounded) derive from a study conducted by Scott and Caughlin (2014) who created their items based on a 50-item scale created by Vangelisti et al., (2005). This scale was found to be internally consistent in this study (a = 0.98; M = 3.36, SD = 2.32). State self-esteem. An adapted version of Heatherton and Polivy’s (1991) 20-item State Self-Esteem Scale was utilized for the main study. The original scale is comprised of three subcategories including performance, social, and appearance. More specifically, this instrument was measured on a 7-point Likert type scale (1 = strongly disagree...7 = strongly agree). This study was interested more in how an individual perceives their state of self-esteem in a social manner. In addition, this study is not intending to manipulate one’s appearance or performance in the scenarios. Thus, the main study utilized the adapted version of the 7-items associated with the social subcategory of this scale. Participants were asked to assess statements such as, “The statements made about me would make me feel[stem]…self-conscious” and “The statements made about me would make me feel[stem]…worried what other people think of me”. This scale was found to be internally consistent in this study (a = 0.97; M = 4.57, SD = 2.10). 35 Behavioral Intentions. To measure the participants, desire to want to affiliate with the gossiper an adapted scale was created based on a scale utilized by Martinescu et al., (2019). To measure gossip targets desire to want to affiliate with the gossiper the authors adapted a 2-item 7-point Likert type scale ranging from strongly disagree to agree to measure affiliation intentions into a 6-item 7-point Likert type scale ranging from strongly disagree to agree. Participants rated items such as, “How likely would it be for you to do the following things [stem]… spend time with them?” and “How likely would it be for you to do the following things [stem]… continue a relationship with the gossiper”. This scale was found to be internally consistent in this study (a = 0.96; M = 3.15, SD = 1.40). Manipulation check. After participants completed answering the scales associated with the outcome’s variables of interest, they answered the following questions, “What was the relationship you had with the individuals in the scenario?” and “How would you describe the statements made about you in the scenario you previously read?”. Demographic Questions. Participants answered questions regarding their age, biological sex, gender, ethnicity, race, and class standing. For a depiction of all the instruments used in the main study please reference Appendix E, and for an entire overview of the main survey in its completion please reference Appendix F. 36 Preliminary Analyses MAIN STUDY RESULTS Since many of the instruments utilized within this study were manipulated and adapted, I conducted a series of confirmatory factor analyses (CFA) to ensure all items for each instrument loaded onto one factor. Results showed that expectancy violations, pride, hurt, shame, state self- esteem, and the behavioral intentions scales all loaded onto one factor. Please view the appendices associated with the measure you are interested in to reveal the specific CFA loading, and Table 4 for the goodness of fit indices. Next, a couple manipulation checks were performed to ensure that participants answered all dependent measures with the correct relationship with the gossiper (friend or acquaintance) and gossip valence type in mind. To ensure this, chi-square tests were conducted. The first chi-square test regarding whether participants had the correct relationship in mind throughout the survey revealed that the manipulation check was significant 𝑥2(1, N = 602) = 521.91, p < .001. However, since it was imperative that participants recalled the correct relationship they had with the gossiper throughout the survey, participants who failed this manipulation check (n = 23) were removed from the participant pool. Similarly, another chi- square test was conducted to ensure participants accurately interpreted the gossip valence within each scenario. Resulted also indicated this was a significant check 𝑥2(1, N = 602) = 572.91, p < .001. However, since it was also imperative that participants interpreted the correct gossip valence within the scenarios, the participants who failed this manipulation check (n = 13) were also removed from the participant pool. While removing these participants an additional participant was identified as not completing the manipulation check. Hence, this participant was 37 also removed from the data making the total number of removed participants equal to 37. Therefore, the remaining analyses are measured based on a sample size of 565 participants. Finally, please note, that for all regressions reported within the main analyses section the unstandardized beta coefficient is reported. I am aware that typically one is supposed to report the standardized beta coefficients when discussing regression results. However, later in the results section I report regression coefficients within mediation models. Hayes (2013) recommends reporting the unstandardized coefficient. Hence, to ensure consistency throughout the manuscript, I have decided to only report the unstandardized beta coefficients throughout the results section. Main Analyses Before the main analyses were conducted, the relationship amongst all independent and dependent variables were assessed through a correlation analysis. Please reference Table 5 for a depiction of this correlation matrix. To begin, two 2 (positive gossip, negative gossip) x 2 (friend or acquaintance) x 2 (scenario: personality/work ethic) between-subjects factorial ANCOVA were conducted to test H1(a), H1(b), H2(a) and H2(b). An ANCOVA was chosen to test these hypotheses because it will test the main and interactive relationships between multiple independent variables on the dependent variables of interest while holding constant the variables that may influence the main and interaction effects. The two variables that were held constant for the ANCOVAs were sex of the participant and age. The decision was made to hold these two variables constant as there is conflicting research that has found both participant age and sex to influence the effects of gossip, whereas others that have found that age and sex did not (Massar et al., 2012; McAndrew & Milenkovic, 2002). Sex was measured categorically whereas age was measured continuously within the main study. For H1(a), H1(b), H2(a) and H2(b) the 38 independent variables are categorical and include gossip valence (positive or negative) and relationship with the gossiper (friend or acquaintance). In contrast, the dependent variables include positive expectancy violations, negative expectancy violations, and behavioral intentions to affiliate with the gossiper. All the dependent variables were measured continuously. H1(a) and H1(b) predicted that gossip valence would have a significant effect on one’s positive and negative expectancy violations such that positively valenced gossip would have a main effect on positive expectancy violations whereas negatively valenced gossip would influence negative expectancy violations. Results revealed that there was a significant relationship between gossip valence and positive expectancy violations F(1, 565) = 3496.83 p < .001, ηp 2 = .86 as well as negative expectancy violations F(1, 565) = 4712.11 p < .001, ηp 2 = .89. More specifically, findings showed that positively valenced gossip (M = 5.96, SD = 0.86) was rated higher in terms of positive expectancy violations compared to negatively valenced gossip (M = 1.67, SD = 0.87). Regarding negatively valenced gossip, results indicated that this type of gossip was perceived as more of a negative expectancy violation (M = 6.17, SD = 0.93) than a positive expectancy violation (M = 1.44, SD = 0.69). Hence, H1(a) and H1(b) are supported. H2(a) and H2(b) predicted that gossip valence would interact with the relationship with the gossiper on expectancy violations. To clarify, H2(a) predicted that the relationship with the gossiper would interact with the gossip valence on expectations such that the effect of positive gossip on positive violations would be greater when the gossiper was an acquaintance whereas H2(b) predicted that the effect of negative gossip on negative violations would be greater when the gossiper is a friend. Results revealed that gossip valence did not have a significant effect on positive F(1, 565) = 0.11 p = 0.74, ηp 2 = 0.00 or negative F(1, 565) = 0.16 p = 0.70, ηp 2 = 0.00 expectancy violations. Similarly, H7(a) and H7(b) predicted that gossip valence and relationship 39 with the gossiper would interact with one another, but specifically on intentions to affiliate with the gossiper. Results indicated gossip valence did not significantly interact with relationship of the gossiper on intentions to affiliate F(1, 565) = 3.27 p =0.07, ηp 2 = 0.01. Therefore, H2(a) and H2(b) are not supported according to these results. Next, a series of multiple regression analyses were conducted to test H3(a), H3(b), H4(a- d), and H5(a-c). Ultimately, these hypotheses predicted that expectancy violation would impact emotions and that these emotions would then affect one’s state self-esteem. The variables of interest in these hypotheses are violation valence, hurt, pride, and state self-esteem, all of which are measured continuously. Since I am analyzing the relationships between continuous variables, I used a multiple regression analysis to determine if I reject or accept the null hypothesis for these predictions. In addition, the multiple regressions indicate which variables significantly predict the outcome variable of interest. H3(a) and H3(b) predicted that positive expectancy violations would have a positive impact on feelings of pride whereas negative expectancy violations will have a negative impact on pride. The overall model fit was assessed using the coefficient of determination (R-squared) and the analysis of variance (ANOVA) results. The results revealed a significant overall model fit, F(2, 562) = 2256.29, p < .001. The model accounted for 88.9% of the variance in the outcome variable, as indicated by the coefficient of determination, R-squared = .889. The adjusted R-squared value was .889, suggesting that approximately 88.9% of the variance in the outcome variable was explained when accounting for the number of predictors and the sample size. Positive expectancy violations was found to have a significant positive relationship with feelings of pride, B = .53, t(564) = 13.26, p < .001. Similarly, negative expectancy violations had 40 a significantly negative relationship with feelings of pride, B = -.42, t(564) = -11.29, p < .001. Therefore, H3(a) and H3(b) are supported. Similarly, H4(a-c) predicted that negative expectancy violations will have a positive impact on feelings of hurt and shame whereas positive expectancy violations will have a negative impact on feelings of hurt and shame. The first model assessed negative and positive expectancy violations on feelings of hurt. Results indicated a significant overall model fit, F(2, 562) = 2560.62, p < .001. The model accounted for 90.1% of the variance in the outcome variable, as indicated by the coefficient of determination, R-squared = .901. The adjusted R-squared value was .901, suggesting that approximately 90.1% of the variance in the outcome variable was explained when accounting for the number of predictors and the sample size. Negative expectancy violations was found to have a significant positive association with feelings of hurt, B = .86, t(564) = 25.15, p < .001. In contrast, positive expectancy violations was not significantly associated with feelings of hurt, B = -.02, t(564) = -.60, p = .55. Regarding feelings of shame as the outcome variable influenced by negative and positive expectancy violations, findings showed a significant overall model fit F(2, 562) = 727.53, p < .001. The model accounted for 72.1% of the variance in the outcome variable, as indicated by the coefficient of determination, R-squared = .721. The adjusted R-squared value was .720, suggesting that approximately 72.0% of the variance in the outcome variable was explained when accounting for the number of predictors and the sample size. Negative expectancy violations was found to have a significant positive association with feelings of shame, B = .76, t(564) = 14.56, p < .001. In contrast, positive expectancy violations was not significantly associated with feelings of shame, B = .05, t(564) =.93, p = .35. Hence H4(a) and H4(b) are supported whereas H4(c) is not supported. 41 H5(a-c) analyzed the relationship between pride, shame, and hurt on feelings of state self- esteem. Results from the multiple regression revealed a significant overall model fit F(3, 561) =1311.76, p < .001. The model accounted for 87.5% of the variance in the outcome variable, as indicated by the coefficient of determination, R-squared = .875. The adjusted R-squared value was .875, suggesting that approximately 87.5% of the variance in the outcome variable was explained when accounting for the number of predictors and the sample size. More specifically, pride had a significant positive relationship with state self-esteem B = .11, t(564) =3.41, p < .001, hurt had a significant negative relationship with state self-esteem B = -.26, t(564) = -5.25, p < .001, and shame had a significant negative relationship with state self-esteem B = -.55, t(564) = - 13.68, p < .001. Hence, one can reject the null hypothesis for H5(a-c). To test H6, H7(a), and H7(b), an additional 2 (positive gossip, negative gossip) x 2 (friend or acquaintance) x 2 (scenario: personality/work ethic) between-subjects factorial ANCOVAs were conducted. H6 predicted that there would be a main effect between gossip valence and the target of gossip’s behavioral intentions of wanting to affiliate with the gossiper such that target of positive gossip would report greater intentions to affiliate with the gossiper that targets of negative gossip. Results revealed there was a significant effect between gossip valence on behavioral intentions to affiliate with the gossiper F(1, 565) = 1460.88 p < .001, ηp 2 = .73. To be specific, individuals who were positively gossiped (M = 4.31, SD = 0.58) about reported greater intentions to affiliate with the gossiper than individuals who were negatively gossiped about (M = 1.95, SD = 0.90). Furthermore, H7(a) and H7(b) predicted that gossip valence and relationship with the gossiper would interact with one another to influence one’s intentions to affiliate with the gossiper. Results indicated gossip valence did not significantly 42 interact with relationship of the gossiper on intentions to affiliate F(1, 565) = 3.27 p =0.07, ηp 2 = 0.01. Hence, based on the findings H6 is supported whereas H7(a), and H7(b) are not supported. The final multiple regression analyses conducted pertained to H8(a-c) which predicted that shame and hurt would have a negative impact on target of gossip’s desire to affiliate with the gossiper whereas feelings of pride would have a positive relationship. Results from the multiple regression revealed a significant overall model fit F(3, 561) =544.02, p < .001. The model accounted for 74.4% of the variance in the outcome variable, as indicated by the coefficient of determination, R-squared = .744. The adjusted R-squared value was .743, suggesting that approximately 74.3% of the variance in the outcome variable was explained when accounting for the number of predictors and the sample size. More specifically, pride had a significant positive relationship with one’s desire to affiliate with the gossiper B = .27, t(564) =8.57, p < .001, hurt had a significant negative relationship with one’s desire to affiliate with the gossiper B = -.40, t(564) = -8.44, p < .001, and shame had a significant positive relationship with one’s desire to affiliate with the gossiper β = .17, t(564) = 4.51, p < .001. Therefore, H8(a) and H8(c) are supported whereas H8(b) was not. Next, H9, H10, and H11 were all tested by employing Hayes (2013) PROCESS macro Model 6 with 10,000 bootstrap to analyze the relationships among the variables in the form of a serial mediation models. Both participant age and the scenario type were held constant throughout each model. In addition, since scenario (1 = personality, 0 = work ethic) and gossip valence (1 = positive gossip, 0 = negative gossip) are categorical variables they were both dummy coded within all three models. The first model pertained to H9 which predicted that the relationship between gossip valence and intentions to affiliate with the gossiper were mediated through positive expectancy violations and then pride. The outcome variable in this model was 43 desire to affiliate with the gossiper, the first mediator was positive expectancy violation, and the second mediator was pride. The overall model fit was found to be significant, F(3, 561) = 467.40, p < .001, indicating that the model explained a significant proportion of the variance in the outcome variable. The model accounted for 71.4% of the variance in desire to affiliate with the gossiper, as indicated by the coefficient of determination, 𝑅2= .714. The indirect effects of gossip valence on intentions to affiliate with the gossiper through the serial mediators (M1 = positive expectancy violations; M2 = Pride) were examined. The total indirect effect of gossip valence on intentions to affiliate with the gossiper were found to be significant (B = 1.32, SE = .1828, 95% CI [.9631,1.6779]). Indirect effect 1 through positive expectancy violations was also significant (B = .4098, SE = .1861, 95% CI [.0594, .7905]). Indirect effect 2 through pride was significant as well (B = .4412, SE = .1212, 95% CI [.2240, .7010]). Also, indirect effect 3 through both positive expectancy violations and pride reached statistical significance (B = .4691, SE = .1058, 95% CI [.2688, .6837]). Hence, H9 is supported. Please reference Figure 1 for an illustration of this serial mediation model and Table 6 for an overview of the data associated with this model. H10 predicted that the relationship between gossip valence and intentions to affiliate with the gossiper were serially mediated by negative expectancy violations and then feelings of hurt. The overall model fit was found to be significant, F(3, 561) =467.40, p < .001, indicating that the model explained a significant proportion of the variance in the outcome variable. The model accounted for 71.4% of the variance in desire to affiliate with the gossiper, as indicated by the coefficient of determination, R-squared = .7142. The total indirect effect of gossip valence on intentions to affiliate with the gossiper was found to be significant (B = 1.32, SE = .2148, 95% CI [.9178,.17576]). Next the indirect effects analyzing the relationship between gossip valence 44 and desire to affiliate with the gossiper, three pathways were examined. The first indirect effect through negative expectancy violations, was .5883(SE = 0.3060, 95% CI [0190, .8737]), the second indirect effect, through hurt, was .0891(SE =.0708, 95% CI [-.0102, 0.1857]), and the third indirect effect, through both negative expectancy violations and hurt, was .6455(SE = 0.1860, 95% CI [.2682, 1.090]). Overall, the findings suggest that gossip valence and desire to affiliate with a gossiper are mediated by negative expectancy violations and hurt, as indicated by the significant indirect effects. The direct effects of the predictor variable on the outcome variable were also significant. These results provide support for the proposed serial mediation model and indicate support for H10. Please reference Figure 2 for an illustration of this serial mediation model and Table 7 for an overview of the data associated with this model. The final serial mediation model that was conducted analyzed the prediction made in H11 stating that the relationship between gossip valence and desire to affiliate with the gossiper would be serially mediated through negative expectancy violations and shame. The overall model fit was found to be significant, F(3, 561) = 467.40, p < .001, indicating that the model explained a significant proportion of the variance in the outcome variable. The model accounted for 71.4% of the variance in desire to affiliate with the gossiper, as indicated by the coefficient of determination, R-squared = .7142. Results illustrated that the total indirect effect of gossip valence on intentions to affiliate with the gossiper was significant (B = .8769, SE = .1597, 95% CI [.5851, 1.21]). The first indirect effect through negative expectancy violations, was .7643(SE = .1867, 95% CI [.4253,1.1602]), the second indirect effect, through shame, was -.0044(SE =.0158, 95% CI [-.0367,.0320]), and the third indirect effect, through both negative expectancy violations and shame, was .1170 (SE =.0831, 95% CI [-.0483, .2776]). Overall, the findings suggest that gossip valence and desire to affiliate with a gossiper are not serially mediated by 45 negative expectancy violations and shame, as indicated by the non-significant indirect effects. These results do not provide support for the proposed serial mediation model and indicate support for H11. Please reference Figure 3 for an illustration of this serial mediation model and Table 8 for an overview of the data associated with this model. To reference all the indirect effects for the three tested serial mediation models please review Table 9. RQ1 asked whether the relationship with the gossiper interacted with the gossip valence to impact intentions to affiliate through expectancy violations, pride, shame, and hurt. Based on the previously conducted ANOVA, results indicated there was not a significant interaction between gossip valence and relationship with the gossiper on intentions to affiliate with the gossiper F(1, 565) = 3.27 p = .07, ηp 2 = 0.01. Therefore, it can be determined that for RQ1 gossip valence would not significantly interact with relationship with the gossiper to influence the relationship between gossip valence on expectancy violation or desire to affiliate with the gossiper through the variables of pride, hurt, shame or state self-esteem. Finally, RQ2 asked how the scenario interacted with the other independent variables (gossip valence and relationship with the gossiper) on the dependent variables of interest (positive expectancy violation, negative expectancy violation, pride, hurt, shame, state self- esteem, and intentions to affiliate with the gossiper). The most parsimonious way to analyze this research question is by running a MANCOVA as it will allow the researchers to analyze the main and interaction effects of multiple independent variables on multiple dependent variables all while holding participant sex and age constant. Results from the MANCOVA revealed there was not a significant interaction between the scenario employed and gossip valence on positive expectancy violations (F(1, 565) = .795 p = .37, ηp 2 = 0.001), negative expectancy violations (F(1, 565) = 1.337 p = .25, ηp 2 = 0.002), pride (F(1, 565) = .101 p = .75, ηp 2 =0.00), hurt (F(1, 46 565) = .076 p = .78, ηp 2 = 0.00), shame (F(1, 565) = .200 p = .66 , ηp 2 =0.00), or state self- esteem (F(1, 565) = .252 p = .62, ηp 2 =0.00). However, there was a significant interaction between gossip valence and scenario on one’s desire to affiliate with the gossiper (F(1, 565) = 14.994 p < .001, ηp 2 =.03) indicating that the gossip scenario’s effect on desire to affiliate with the gossiper is dependent upon the gossip valence (positive or negative). Please see Figure 4 for an illustration of this interaction. To break down this interaction I split the file to compare the scenarios that were employed and conducted an independent samples t-test. Results from the test revealed there was a significant difference regarding gossip valence in the personality-based condition (t(280) = 30.96, p = .001, Cohen's d = 3.68) and the work ethic condition (t(281) =22.85, p < .001, Cohen's d = 2.70) on desire to affiliate with the gossiper. Findings indicated that when individuals were positively gossiped about regarding their personality, they reported greater intentions to affiliate (M = 4.40, SD = .60) with the gossiper than if they were negatively gossiped about (M = 1.80, SD = .80). Similarly, when individuals were positively gossiped about in terms of their work ethic, they reported greater intentions to affiliate with the gossiper (M =4.22, SD = .55) than individuals who were negatively gossiped about in terms of their work ethic (M = 2.10, SD = .97). Following this interpretation, the interaction between scenario type and relationship with the gossiper was explored. Analogous to previous findings, results indicated there was not a significant interaction between scenario type and relationship with the gossiper on positive expectancy violations (F(1, 565) = .079 p = .78, ηp 2 = 0.00), negative expectancy violations (F(1, 565) = .721 p = .40, ηp 2 = 0.001), pride (F(1, 565) = .334 p = .60, ηp 2 = 0.001), hurt (F(1, 565) = 1.25 p = .30 , ηp 2 = 0.002), shame (F(1, 565) = .401 p = .52, ηp 2 = 0.001), or state self- esteem(F(1, 565) = .118 p = .73, ηp 2 = 0.00). However, there was a significant interaction 47 between the relationship with the gossiper and scenario on one’s desire to affiliate with the gossiper (F(1, 565) = 5.34 p = .02, ηp 2 = 0.01) indicating that the gossip scenario’s effect on desire to affiliate with the gossiper is dependent upon the relationship the target has with the gossiper (friend versus acquaintance). Reference Figure 5 for a depiction of this interaction. Another independent samples t-test was employed to decipher this interaction. Results from the test revealed there was a significant difference regarding relationship type in the personality- based condition (t(280) =.76, p = .008, Cohen's d = .08) and the work ethic condition (t(281) = 2.82, p = .01, Cohen's d =.34) on desire to affiliate with the gossiper. Findings showed when individuals were gossiped about by a friend regarding their personality, they showed greater intentions to affiliate with the gossiper (M =3.20, SD = 1.55) than if they were gossiped about by an acquaintance (M =3.06, SD = 1.40). Similarly, when individuals were gossiped about in terms of their work ethic by a friend, they were more inclined to affiliate with them (M =3.41, SD = 1.40) than if they were an acquaintance (M =2.97, SD = 1.21). Finally, the MANCOVA was analyzed to determine if there was a three-way interaction between relationship with the gossiper, gossip valence, and scenario. Results illustrated non-significant interactions for positive expectancy violations (F(1, 565) = 0.82 p = .40, ηp 2 =.001), negative expectancy violations (F(1, 565) = .03 p = .90, ηp 2 = 0.00), pride (F(1, 565) = .064 p = .80 , ηp 2 =.00), hurt (F(1, 565) = 1.40 p = .24, ηp 2 = 0.002), shame (F(1, 565) = .269 p = .60, ηp 2 = .00), state self-esteem (F(1, 565) = .210 p = .65, ηp 2 = .00), and desire to affiliate with the gossiper (F(1, 565) = .25 p = .62, ηp 2 =.00). While analyzing these interaction effects, the researchers discovered that the scenario employed had significant main effects on the following dependent variables: positive expectancy violations (F(1, 565) = 5.70 p = .02, ηp 2 =.01), negative expectancy violations (F(1, 565) = 7.90 48 p = .005, ηp 2 =.014), and state self-esteem (F(1, 565) = 3.64 p = .05, ηp 2 =.007). An independent sample t-test was employed to analyze these main effects. Results showed that there was a significant difference in ratings regarding negative expectancy violation (t(563) = .77, p = .04, Cohen's d = 0.07) and state self-esteem (t(39) = 3.21, p < .05, Cohen's d = 0.09) depending on the scenario presented. However, there was not a significant difference in ratings regarding positive expectancy violations (t(39) = 3.21, p < .05, Cohen's d = 0.06) based on the scenario employed. More specifically, negative expectancy violations were rated higher in the work ethic condition (M =3.82, SD =2.56) compared to the personality conditions (M = 3.66, SD =2.45). In contrast, self-esteem had a greater average in the personality condition (M = 4.67, SD = 2.05) compared to the work ethic scenario condition (M = 4.48, SD = 2.16). Please reference Table 10 for an overview of the means and standard deviations for the dependent variables amongst the various scenario types, gossip valence groups, and relationship types. The following sections explores the impact of these findings in greater detail. 49 DISCUSSION The present study aimed to investigate the effects of gossip valence (positive or negative), relationship with the gossiper (friend or acquaintance), and scenario (personality or work ethic) on various psychological outcomes, including positive and negative expectancy violations, pride, hurt, shame, state self-esteem, and behavioral intentions to affiliate with the gossiper. Main and interaction effects were explored along with three serially mediated models. Results showed that gossip valence had a main effect on positive and negative expectancy violations such that positive gossip had a positive effect on positive expectancy violations more so than negative gossip. Similarly, findings revealed negative gossip had an impact on negative expectancy violations such that participants rated negative expectancy violations higher in negative gossip conditions more so than in positive gossip conditions. Regarding interactions between gossip valence and the relationship presented, the output indicated there was not a significant interaction between these two variables on positive expectancy violations, negative expectancy violations, or desire to affiliate with the gossiper. Due to these non-significant interactions, it was concluded that the relationship with the gossiper would not significantly moderate the proposed serial mediation models that were later tested. Next, a series of multiple regression analyses were conducted to determine how expectancy violation would impact emotions and how those emotions would then influence one’s state self-esteem and desire to affiliate with the gossiper. Positive expectancy violations were found to have a significant positive influence on feelings of pride, whereas negative expectancy violations had a significantly negative influence on feelings of pride. In addition, negative expectancy violations were found to have a significant positive association with feelings of hurt whereas positive expectancy violations were not significantly associated with feelings of hurt. 50 Regarding feelings of shame as the outcome variable influenced by negative and positive expectancy violations, findings showed that negative expectancy violations was found to have a significant positive association with feelings of shame, yet positive expectancy violations was not significantly associated with feelings of shame. Then the researchers assessed how the relationship between pride, shame, and hurt on feelings of state self-esteem. Pride was found to have a significant positive influence on state self-esteem, hurt had a significant negative relationship with state self-esteem, and shame had a significant negative relationship with state self-esteem. This was followed by the final multiple regression analyses conducted which predicted that shame and hurt would have a negative impact on target of gossip’s desire to affiliate with the gossiper whereas feelings of pride would have a positive relationship. Results from the multiple regression revealed pride had a significant positive relationship with one’s desire to affiliate with the gossiper, hurt had a significant negative relationship with one’s desire to affiliate with the gossiper, and shame had a significant positive relationship with one’s desire to affiliate with the gossiper. However, this piece was not only interested in the intrapersonal effects gossip had on targets of gossip but also the interpersonal implications gossip held. Hence, it was also predicted that there would be a main effect between gossip valence and the target of gossip’s behavioral intentions of wanting to affiliate with the gossiper such that target of positive gossip would report greater intentions to affiliate with the gossiper that targets of negative gossip. Results revealed there was a significant effect between gossip valence on behavioral intentions to affiliate with the gossiper. For instance, individuals who were positively gossiped about reported greater intentions to affiliate with the gossiper than individuals who 51 were negatively gossiped about. These findings were followed by analyzing the relationship amongst all the variables in three separate serial mediation models. More specifically, it was predicted that the relationship between gossip valence on intentions to affiliate with the gossiper would be serially mediated through positive expectancy violations on pride, negative expectancy violations on hurt, and negative expectancy violations on shame. Results illustrated a significant serial mediation between gossip valence on intentions to affiliate with the gossiper through positive expectancy violations and pride. In congruence, the second model analyzing gossip valence on intentions to affiliate with the gossiper through negative expectancy violations and hurt were also considered a significant serial mediation model. However, the third model analyzing whether the relationship between gossip valence on intentions to affiliate with the gossiper were serially mediated by negative expectancy violations and shame were not found to be a significant serial mediation model. An additional finding made by the researchers pertained to the scenarios utilized within the study. Findings revealed that the scenarios used influenced positive expectancy violations, negative expectancy violations, and state self-esteem. More specifically, results showed that there was a significant difference in ratings regarding negative expectancy violation and state self-esteem depending on the scenario presented. However, there was not a significant difference in ratings regarding positive expectancy violations based on the scenario employed. More specifically, negative expectancy violations were rated higher in the work ethic condition compared to the personality conditions. In contrast, self-esteem had a greater average in the personality condition compared to the work ethic scenario condition. Furthermore, an interaction was found between gossip valence and the scenario presented on one’s desire to affiliate with the gossiper. To be specific findings indicated that when individuals were positively gossiped about 52 regarding their personality, they reported greater intentions to affiliate with the gossiper than if they were negatively gossiped about. Similarly, when individuals were positively gossiped about in terms of their work ethic, they reported greater intentions to affiliate with the gossiper than individuals who were negatively gossiped about in terms of their work ethic. Similarly, there was another interaction found between relationship with the gossiper and the scenario employed on one’s desire to affiliate with the gossiper. Findings showed when individuals were gossiped about by a friend regarding their personality, they showed greater intentions to affiliate with the gossiper than if they were gossiped about by an acquaintance. Similarly, when individuals were gossiped about in terms of their work ethic by a friend, they were more inclined to affiliate with them than if they were an acquaintance. These findings have theoretical implications that are explored in the subsequent section. Theoretical Implications Results showed that gossip valence had a main effect on positive and negative expectancy violations such that positive gossip had a positive effect on positive expectancy violations more so than negative gossip. Similarly, findings revealed negative gossip had an impact on negative expectancy violations such that participants rated negative expectancy violations higher in terms of negative gossip conditions more so than positive gossip conditions. These finding are in line with Kelly and colleagues (2012) who explained that how events and communication are appraised, such as gossip, can influence one’s overall mood such that positive appraisals beget positive outlooks whereas negative appraisals can lead to a negative state of mind. The findings from this piece are congruent with previous research and assist in organizing the process in which gossip is perceived. However, it also provides a level of heuristic provocativeness to the 53 subject matter of gossip which if acted upon allows for the advancement of this topic matter in various interpersonal relationships. When analyzing the interaction effect between gossip valence and the relationship presented, the output indicated there was not a significant interaction between these two variables on positive expectancy violations, negative expectancy violations, or desire to affiliate with the gossiper. Due to these non-significant interactions, it was concluded that the relationship with the gossiper would not significantly moderate the proposed serial mediation models that were later tested. These findings are contrary to Turner et al.’s (2023) study which aimed to explore the impact of gossip valence (positive or negative) and the relationship with the gossiper (friend or stranger) on people's perceptions of the gossiper. The results from their study revealed that the type of relationship significantly interacted with the gossip valence, influencing perceptions of trust, liking, and expertise of the gossiper. Specifically, when the gossiper was a friend, the gossip valence did not affect perceptions of liking. However, when strangers engaged in negative gossip, it had a negative impact on liking. The study also found that regardless of the gossip valence or relationship type, the credibility of the gossiper diminished when engaged in gossiping. These findings emphasize the importance of understanding the intricacies of gossip and its effects since it can shape our perceptions of individuals and the quality of our relationships (Turner et al., 2003). A probable reasoning for the non-significant interactions may be attuned to the fact this study did not specify the number of hours of time an individual has spent with the acquaintance versus a friend. Hall (2018) explains that it takes time to create friendship and that there are additional factors at play that influence whether individuals become friends, such as proximity. Ultimately, what they found was that if an individual needs to be considered a friend versus an acquaintance they need to spend at least 43 hours together within a 54 3-week timeframe. Turner at el., (2003) had participants bring in friends to a lab but had the requirement set that they needed to have been friends for at least 1 year. Hence, the amount of time the target has known or interacted with the gossiper may need to be specified in future studies. As previously mentioned, participants are typically able to identify the difference between friends versus acquaintances (Hall, 2018). Similarly, people can easily identify individuals within their in-groups versus their out-groups (Mussweiler & Bodenhausen, 2022). In fact, Mussweiler and Bodenhausen (2002) dsicussed from their findings that evaluating the actions of an in-group member, such as a friend, compared to evaluating an out-group member, such as an acquaintance, impacts an individual’s ability to access knowledge about oneself. The authors go on to recommend that when analyzing in-group versus out-group members, such as friends versus acquaintances, it is important to have a norm group for participants to compare to, and that by doing this it will allow for participants to answer questions about oneself more adequately. Next, positive expectancy violations were found to have a significant positive influence on feelings of pride, whereas negative expectancy violations had a significantly negative influence on feelings of pride. Negative expectancy violations were found to have a significant positive association with feelings of hurt whereas positive expectancy violations were not significantly associated with feelings of hurt. Regarding feelings of shame as the outcome variable influenced by negative and positive expectancy violations, findings showed that negative expectancy violations were found to have a significant positive association with feelings of shame, yet positive expectancy violations was not significantly associated with feelings of shame. These findings add to the body of research that argues that negative appraisals precede negative 55 emotions (Bennett et al., 2020; Chen et al., 2020; Kelly et al., 2012). In terms of emotions, the analysis output illustrated that pride was found to have a significant positive influence on state self-esteem, hurt had a significant negative relationship with state self-esteem, and shame had a significant negative relationship with state self-esteem. Furthermore, results from the multiple regression revealed pride had a significant positive relationship with one’s desire to affiliate with the gossiper, hurt had a significant negative relationship with one’s desire to affiliate with the gossiper, and shame had a significant positive relationship with one’s desire to affiliate with the gossiper. This piece utilized expectancy violation theory (Burgoon and Hale’s 1988) to rationalize the proposed predictions and findings. However, another potential reason for the findings pertaining to gossip, emotions, and intentions to affiliate with the gossiper can also be a result of the fact humans strive to belong. Hence, theories such as Leary and colleagues (1995) sociometer theory which explains that “self-esteem serves as a subjective monitor of one’s relational evaluation” (Leary & Baumeister, 2000, p. 9) can be used to help make sense of the finding between shame and intentions to affiliate with the gossiper. More specifically, the relationship between shame and intentions to affiliate with the gossiper may be predicated or mediated by one’s state self- esteem. The authors explain that emotions are a part of the influence on one’s self-esteem, but overall individuals simply want to belong, and their self-esteem influences how they behave. Regarding gossip, this is in line with the evolutionary perspective that Dunbar (2004) suggested stating that humans evolved by utilizing gossip to survive, discover social norms, and gain acceptance from others. This shows that not only do these findings confirm the predictions rationalized in the above text, but it also generates more potential reasoning behind why the relationships between gossip, emotions, and self-esteem were discovered. 56 In addition, the findings indicated there was a significant effect between gossip valence on behavioral intentions to affiliate with the gossiper. More specifically, individuals who were positively gossiped about reported greater intentions to affiliate with the gossiper than individuals who were negatively gossiped about. This is in line with previous work conducted by Martinescu et al. (2019) who investigated how the influence of gossip valence impacted an individuals' inclination to establish a connection with the gossiper. Their findings revealed that the valence of gossip can shape an individual's desire to affiliate with the gossiper. Specifically, when individuals were the subject of positive gossip by a co-worker, they exhibited a greater inclination to establish an affiliation compared to situations where negative gossip was directed towards them. Therefore, the findings from the present study are theoretically rich because they expand upon Martinescu et al.’s (2019) study by analyzing the effects gossip has on targets of gossip in a non-organizational context; interpersonal relationships such as friendships. Future research is encouraged to continue analyzing various relationships in terms of gossip and how the valence impacts a target’s desire to affiliate with the gossiper. Furthermore, results illustrated a significant serial mediation between gossip valence on intentions to affiliate with the gossiper through positive expectancy violations and pride. In congruence, the second model analyzing gossip valence on intentions to affiliate with the gossiper through negative expectancy violations and hurt were also considered a significant serial mediation model. However, the third model analyzing whether the relationship between gossip valence on intentions to affiliate with the gossiper were serially mediated by negative expectancy violations and shame were not found to be a significant serial mediation model. The findings pertaining to the serial mediations hold theoretical importance because it indicates there may be a sequence in which researchers should be analyzing the effects of gossip. However, a 57 limitation in this study is that it is a cross-sectional design meaning that participants were assessed at only one time point. Hence, the exact order in which these variables influence one another are not solidified. For example, there is literature that supports the notion that state self- esteem precedes emotions (Zong et al., 2022). Similarly, there is research that has found expectancy violations to mediate the relationship between emotions and perceptions of others (Ask & Landstrom, 2010). This supports the notion that to obtain a more accurate depiction of how the variables explored in this piece may have causal relationship amongst one another it is critical for future research to conduct longitudinal studies pertaining to the effects of gossip. In addition, it is also encouraged that the order in which these variables are placed within models are explored in greater depth. Next, main and interaction effects were found regarding the scenario employed in the surveys. The first interaction that was detected was between gossip valence and the scenario presented on one’s desire to affiliate with the gossiper. To be specific, findings indicated that when individuals were positively gossiped about regarding their personality, they reported greater intentions to affiliate with the gossiper than if they were negatively gossiped about regarding their personality. Similarly, when individuals were positively gossiped about in terms of their work ethic, they reported greater intentions to affiliate with the gossiper than individuals who were negatively gossiped about in terms of their work ethic. Next, another interaction was found between relationship with the gossiper and the scenario employed on one’s desire to affiliate with the gossiper. Findings showed when individuals were gossiped about by a friend regarding their personality, they showed greater intentions to affiliate with the gossiper than if they were gossiped about by an acquaintance. Similarly, when individuals were gossiped about in terms of their work ethic by a friend, they were more inclined to affiliate with them than if 58 they were an acquaintance. Finally, the scenarios utilized in the study had significant main effects on positive expectancy violations, negative expectancy violations, and state self-esteem. Results showed that there was a significant difference in ratings regarding negative expectancy violation and state self-esteem depending on the scenario presented. More specifically, negative expectancy violations were rated higher in the work ethic condition compared to the personality conditions. Similarly, self-esteem had a greater average in the personality condition compared to the work ethic scenario condition. However, there was not a significant difference in ratings regarding positive expectancy violations based on the scenario employed. There are a few potential reasons for these main and interaction findings pertaining to the gossip scenarios. One potential explanation for these findings can be attributed to the relevancy of the gossip. Turner et al. (in press) found that gossip valence interacted with relevancy of the gossip, or whether the information was of importance to the gossiper, on expectations such that when individuals positively gossiped about irrelevant topics rather than gossiping about negative relevant topics it effected perceptions of the gossiper. Hence, participants may have read the scenarios and deemed the positive gossip scenarios about personality and work ethic as more relevant than negatively valenced gossip. However, future research is encouraged to continue exploring how relevancy of the gossip interacts with both the gossip valence and relationship with the gossiper in various settings. In addition, an individual’s level of self-efficacy, or “an individual’s generalized beliefs about his or her own competencies in dealing with demands that influence how the individuals…act”, may be another factor that influenced this interaction effect (Groth et al., 2019; p. 2). More specifically, participants may have felt like they had more control to maintain or alter perceptions regarding their personality than their work ethic within this study. This may be true especially since the type of work they do was not specified within the 59 hypothetical scenarios. Overall, this interaction effect between gossip valence and scenario highlights a limitation within the present study of using hypothetical scenarios, but it also indicated that the context in which gossip occurs matters and should be taken into consideration when exploring the effects of gossip on interpersonal relationships. Not only do these findings exemplify theoretical importance to the field of research pertaining to gossip and interpersonal relationships, but it also warrants the discussion of the pragmatic implications of these findings. Pragmatic Implications The present study provides valuable insights into the effects of gossip valence, relationship with the gossiper, and scenario on various psychological outcomes and behavioral intentions. These findings have practical implications for understanding human behavior in social settings and may offer guidance in interpersonal interactions, workplace dynamics, and social interventions. For example, researching the communicative acts that can influence emotions and state self-esteem are critical to do as they are linked with understanding an individual’s overall well-being (Dogan et al., 2013; King & Rosa, 2019). Similarly, these findings hold value as our relationships with others, especially the quality of our friendships, impact our overall happiness and health (Bliezner et al., 2019). In fact, this study highlights the influence of gossip valence and the relationship between the gossiper and the target on the desire to affiliate. Positive gossip about an individual, particularly when coming from a friend, fosters a stronger inclination to affiliate with the gossiper. These findings suggest that positive gossip from close acquaintances can strengthen social bonds and interpersonal relationships. Encouraging positive and supportive communication among friends can contribute to building a more cohesive and supportive social network. 60 However, negative gossip does and will occur between individuals (Dunbar, 2004; Foster, 2004; Turner et al., 2003). Negative gossip, as revealed in the study, can have adverse effects on feelings of hurt and shame. These findings underscore the importance of addressing negative gossip in social and organizational settings promptly. Employers and social group leaders should create an environment that discourages gossip that can lead to harmful emotional outcomes. Implementing clear communication guidelines and promoting a culture of respect and empathy may help reduce the spread of negative gossip. Hence, if one ever finds themselves in the role of a gossiper that it is better to communicate positive gossip rather than negative as it is less likely to emotionally harm the target of gossip or the relationship, they have with them. This is supported by the findings regarding the serial mediation models which emphasize the importance of positive expectancy violations and pride in driving intentions to affiliate with the gossiper. Recognizing these underlying mechanisms can be valuable in designing interventions that promote positive interactions and strengthen social bonds. Targeted interventions could be developed to foster positive gossip and enhance feelings of pride, which, in turn, can facilitate social cohesion and positive relationship-building. In addition, the findings from this piece indicate that the scenario, or what the gossip pertains to, can have main and interactive effects with other variables to influence perceptions of the gossip. Up until this study, research regarding gossip in organizational studies has dominated the field of social science research (Martinescu et al., 2019). In contrast, this study did not specify the context in which the gossip was being conducted at. However, context matters when an individual is gossiping. This is supported by the findings illustrating the different effects the two scenarios had on eliciting varying emotional responses, influencing the nature of expectancy violations and self-esteem. Understanding these differences can help individuals tailor their 61 communication and feedback to promote positive emotional outcomes and avoid potentially harmful consequences. Therefore, it is recommended that an individual is conscientious of the setting in which they share gossip. In conclusion, the findings of this study contribute to a deeper understanding of the impact of gossip on psychological outcomes and behavioral intentions. The pragmatic implications derived from these findings can inform strategies for managing gossip effectively, promoting positive interactions, and fostering healthier social dynamics in both workplace and social settings. By recognizing the influence of gossip valence, relationship dynamics, and situational context, individuals and organizations can create a more positive and supportive environment for everyone involved. However, it is essential to acknowledge the limitations of the study and consider the need for further research to fully comprehend the complexities of gossip and its implications. Limitations and Future Directions To begin, this study has a limitation in that the data was collected cross-sectionally, meaning it was gathered at a single point in time. Therefore, it is not possible to draw conclusions about outcomes that occur over time based on the data collected in this study. Since mediation implies changes that unfold over time (Cole & Maxwell, 2003), it would be valuable for future research to collect data longitudinally. Moreover, this study solely focused on data from one perspective within the reported close relationship. In addition, another limitation present in this piece pertains to the context in which the study was deployed. More specifically, this study utilized hypothetical scenarios and was administered via a participant recruiting platform. Previous studies conducted by Turner et al. (2003) and Turner et al., (under review) analyzed the effects of gossip in a lab setting. This allowed for more generalizable findings to be 62 collected and future research is encouraged to follow suit as it may lend insight into how context influences perceptions of gossip for targets. Likewise, as previously discussed throughout the manuscript, there are three individuals who are a part of a gossip triad, the gossiper, the gossip recipient, and the target of gossip. However, previous studies tend to focus on only one of the roles within the triad (see; Turner et al., 2003; Martinescu et al., 2019). Conducting lab studies would allow for researchers to analyze the intrapersonal and interpersonal effects of gossip on all three roles within the gossip triad. In addition, lab studies would potentially allow for additional interpersonal relationships to be analyzed. More specifically, one could analyze the effects gossip has on targets of gossip when the gossiper is a friend, casual friend, acquaintance, or stranger. However, as discussed in the theoretical implications section it is encouraged researchers of lab studies should control for the length of time an individual has known the gossiper (Hall, 2018). Another limitation and opportunity for additional research to be conducted lies within the fact that the sample for the present studies were all recruited from the United States. Hence, perceptions of gossip may be different in various parts of the world. More specifically, gossip is a communicative act that can assist an individual in learning cultural norms (Baumeister et al., 2004; Dunbar, 2004), and theories such as speech codes theories (Philipsen, 1992), which explains how different cultures assign meaning to language, have shown the differences between collectivistic cultures and individualistic cultures in terms of communication. For example, gossip in a collectivistic culture may be perceived differently as negative or positive evaluations would not just affect the target of gossip but potentially the reputation of the family the target of gossip is related to. Therefore, it is encouraged that researchers explore how gossip impacts intrapersonal variables and interpersonal relationships in various countries. 63 Moreover, age was found to have a significant impact on participants’ ratings of positive expectancy violations, state self-esteem, and their desire to affiliate with the gossiper. There are a few studies that have explored the impact age has on perceptions of gossip. For example, Massar and colleagues (2012) found that younger women demonstrated a greater likelihood to gossip than older women. More specifically, they found that mate value mediated the relationship between a participant’s age and their tendency to gossip. They explain this may be a result of younger women being in their reproductive years and are in competition with other women to find a mate. Hence, they may gossip more about other women or men for a variety of social reasons. However, what is unknow is how the age of the target of gossip influences perceptions of the gossip. Findings from this piece indicate this may be a variable that is held constant or explored in greater depth in future studies. Finally, a research question was posed regarding whether scenario interacted with the independent variables on the dependent variables of interest. Results indicated that the scenario did significantly interact with the independent variables on desire to affiliate with the gossiper. The theoretical and pragmatic implications of this finding were explored along with the unexpected main effects the scenario participants read had on state self-esteem and negative expectancy violations. Therefore, it is encouraged that future research explores how the wording of gossip messages, the context, and relevancy of the gossip interact to influence perceptions targets of gossip have about gossipers. 64 CONCLUSION The present study aimed to investigate the effects of gossip valence (positive or negative), relationship with the gossiper (friend or acquaintance), and scenario (personality or work ethic) on various psychological and relational outcomes in interpersonal relationships. The study explored positive and negative expectancy violations, pride, hurt, shame, state self-esteem, and behavioral intentions to affiliate with the gossiper. Results showed that positive gossip had a positive effect on positive expectancy violations, while negative gossip had a greater impact on negative expectancy violations. However, the relationship with the gossiper did not interact with the gossip valence to influence expectations or intentions to affiliate with the gossiper. Furthermore, whether an individual’s expectations were positively or negatively violated had a relationship with the emotions targets of gossip felt, and those emotions then have ramifications on a target of gossip’s desire to affiliate with the gossiper. The study had limitations, pertaining to the cross-sectional data and hypothetical scenarios within the surveys. Future research should explore longitudinal data and collect perspectives from all parties involved in the gossip triad. Additionally, studying gossip in different cultural contexts and exploring different relationships between the gossiper and target would provide further insights. However, this piece held both theoretical and pragmatic importance. More specifically, it advanced the field of interpersonal communications knowledge on a communicative phenomenon that occurs regularly, and it highlights the importance of speaking positively about others when they are not in one’s presence. 65 REFERENCES Afifi, W. A., & Metts, S. (1998). Characteristics and consequences of expectation violations in close relationships. Journal of Social and Personal Relationships, 15(3), 365-392. https://doi.org/10.1177/0265407598153004 Ask, K., & Landström, S. (2010). Why emotions matter: Expectancy violation and affective response mediate the emotional victim effect. Law and human behavior, 34(5), 392. https://doi.org/10.1007/s10979-009-9208-6 Baumeister, R. F., Zhang, L., & Vohs, K. D. (2004). Gossip as cultural learning. Review of general psychology, 8(2), 111-121. https://doi.org/10.1037/1089-2680.8.2.111 Beersma, B., & Van Kleef, G. A. (2012). Why people gossip: An empirical analysis of social motives, antecedents, and consequences. Journal of Applied Social Psychology, 42(11), 2640-2670. https://doi.org/10.1111/j.1559-1816.2012.00956.x Bennett, L. K., Scruggs, X., & Woods, J. M. (2020) Surprise, hurt, and anger as emotional responses to expectancy violations following feedback messages, Communication Research Reports, 37:1-2, 22-33. https://doi.org/10.1080/08824096.2020.1737000 Biernat, M., Vescio, T. K., & Billings, L. S. (1999). Black sheep and expectancy violation: Integrating two models of social judgment. European Journal of Social Psychology, 29(4), 523-542. https://doi.org/10.1002(SICI)1099- 0992(199906)29:4<523::AID-EJSP944>3.0.CO;2-J Blieszner, R., Ogletree, A. M., & Adams, R. G. (2019). Friendship in later life: A research agenda. Innovation in Aging, 3(1), https://doi.org/10.1093/geroni/igz005 Brown, J. D., & Marshall, M. A. (2001). Self-esteem and emotion: Some thoughts about feelings. Personality and Social Psychology Bulletin, 27(5), 575-584. https://doi.org/10.1177/0146167201275006 Bryant, E. M., & Marmo, J. (2012). The rules of Facebook friendship: A two-stage examination of interaction rules in close, casual, and acquaintance friendships. Journal of social and personal relationships, 29(8), 1013-1035. https://doi.org/10.1177/0265407512443616 Budiarto, Y., & Helmi, A. F. (2021). Shame and self-esteem: A meta-analysis. Europe's Journal of Psychology, 17(2), 131. https://doi.org/10.5964/ejop.2115 Burgoon, J. K. (1992). Applying a comparative approach to nonverbal expectancy violations theory. In J. Blumler, K. E. Rosengren, & J. M. McLeod (Eds.), Comparatively speaking (pp. 53–69). Beverly Hills, CA: Sage. 66 Burgoon, J. K. (1993). Interpersonal expectations, expectancy violations, and emotional communication. Journal of language and social psychology, 12(1-2), 30-48. https://doi.org/10.1177/0261927x93121003 Burgoon, J. K. (2015). Expectancy violations theory. In C. R. Berger, M. E. Roloff, S. R. Wilson, J. P. Dillard, J. Caughlin, & D. Solomon (Eds.), The International Encyclopedia of Interpersonal Communication, (Vol. 26, pp.1-9). John Wiley & Sons, Inc. Burgoon, J. K., & Hale, J. L. (1988). Nonverbal expectancy violations: Model elaboration and application to immediacy behaviors. Communications Monographs, 55(1), 58-79. https://doi.org/10.1080/03637758809376158 Burgoon, J. K., & Hubbard, A. E. (2005). Cross-cultural and intercultural applications of expectancy violations theory and interaction adaptation theory. Theorizing about intercultural communication, 149-171. Burgoon, J. K., & Jones, S. B. (1976). Toward a theory of personal space expectations and their violations. Human communication research, 2(2), 131-146. https://doi.org/10.1111/j.1468-2958.1976.tb00706.x Chang, A. M., & Mackenzie, A. E. (1998). State self-esteem following stroke. Stroke, 29(11), 2325-2328. https://doi.org/10.1161/01.STR.29.11.2325 Chen, Y., Li, L., Ybarra, O., & Zhao, Y. (2020). Symbolic threat affects negative self-conscious emotions. Journal of Pacific Rim Psychology, 14, e12. https://doi.10.1017/prp.2020.3 Chen, J., Song, H., Zeng, W., & Ren, L. (2023). Female Visitors’ Behavioral Intention to the Shanghai Comics Exhibition: An Application of Cognitive Appraisal Theory. Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 36(1), 46-63. De Gouveia, C. M., Van Vuuren, L. J., & Crafford, A. (2005). Towards a typology of gossip in the workplace. SA Journal of Human Resource Management, 3(2), 56-68. https://hdl.handle.net/10520/EJC95806 De Ruiter, N. M., Van Geert , P. L., & Kunnen, E. S. (2017). Explaining the “how” of self esteem development: The self-organizing self-esteem model. Review of General Psychology, 21(1), 49-68. https://doi.org10.1037/gpr0000099 Dogan, T., Totan, T., & Sapmaz, F. (2013). The role of self-esteem, psychological well-being, emotional self-efficacy, and affect balance on happiness: A path model. European Scientific Journal, 9(20). 31-42. Dunbar, R. I. (2004). Gossip in evolutionary perspective. Review of general psychology, 8(2), 100-110. https://doi.org/10.1037/1089-2680.8.2.100 67 Dunbar, R. I. (2018). The anatomy of friendship. Trends in cognitive sciences, 22(1), 32-51. https://doi.org/10.1016/j.tics.2017.10.004. Farley, S. D., Timme, D. R., & Hart, J. W. (2010). On coffee talk and break-room chatter: Perceptions of women who gossip in the workplace. The Journal of social psychology, 150(4), 361-368. https://doi.org/10.1080/00224540903365430 Fehr, B. A. (1996). Friendship Processes. Thousand Oaks, CA: Sage. Fitness, J. (2001). Betrayal, rejection, revenge, and forgiveness: An interpersonal script approach. In Interpersonal Rejection, (Ed.) MR Leary, pp.73-103. New York: Oxford University Press. Foster, E. K. (2004). Research on gossip: Taxonomy, methods, and future directions. Review of general psychology, 8(2), 78-99. https://doi.org/10.1037/1089-2680.8.2.78 Giardini, F., & Wittek, R. P. (2019). Silence is golden. Six reasons inhibiting the spread of third party gossip. Frontiers in psychology, 10, 1120. https://doi.org/10.3389/fpsyg.2019.01120 Groth, N., Schnyder, N., Kaess, M., Markovic, A., Rietschel, L., Moser, S., ... & Schmidt, S. J. (2019). Coping as a mediator between locus of control, competence beliefs, and mental health: A systematic review and structural equation modelling meta-analysis. Behaviour research and therapy, 121, 103442. https://doi.org/10.1016/j.brat.2019.103442 Hall, J. A. (2019). How many hours does it take to make a friend?. Journal of social and personal relationships, 36(4), 1278-1296. https://doi.org/10.1177/0265407518761225 Hardecker, D. J., & Haun, D. B. (2020). Approaching the development of hurt feelings in childhood. New ideas in psychology, 59, 100796. https://doi.org/10.1016/j.newideapsych.2020.100796 Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guildford press. Heatherton, T. F., & Wyland, C. L. (2003). Assessing self-esteem. In S. J. Lopez & C. R. Snyder (Eds.), Positive psychological assessment: A handbook of models and measures (pp. 219- 233). American Psychological Association. https://doi.org/10.1037/10612-014 Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social psychology, 60(6), 895. https://doi.org/10.1037/0022-3514.60.6.895 68 Holmstrom, A. J., Shebib, S. J., Mazur, A. P., Mason, A. J., Zhang, L., Allard, A., & Boumis, J. K. (2021). Self-conscious emotions and esteem support: The effectiveness of esteem support in alleviating state shame and guilt. Human Communication Research, 47(2), 105-131. https://doi.org/10.1093/hcr/hqaa015 Kelly, R. E., Mansell, W., Sadhnani, V., & Wood, A. M. (2012). Positive and negative appraisals of the consequences of activated states uniquely relate to symptoms of hypomania and depression. Cognition & emotion, 26(5), 899-906. https://dx.doi.org/10.1080/02699931.2011.613918 King, R. B., & dela Rosa, E. D. (2019). Are your emotions under your control or not? Implicit theories of emotion predict well-being via cognitive reappraisal. Personality and individual differences, 138, 177-182. https://doi.org/10.1016/j.paid.2018.09.040 Krems, J. A., Williams, K. E., Aktipis, A., & Kenrick, D. T. (2021). Friendship jealousy: One tool for maintaining friendships in the face of third-party threats?. Journal of Personality and Social Psychology, 120(4), 977. https://doi.org/10.1037/pspi0000311 Krippendorff, K. (2004). Reliability in content analysis: Some common misconceptions and recommendations. Human communication research, 30(3), 411-433. https://doi.org/10.1111/j.1468-2958.2004.tb00738.x Leary, M. R., & Baumeister, R. F. (2000). The nature and function of self-esteem: Sociometer theory. In Advances in experimental social psychology (Vol. 32, pp. 1-62). Academic Press. https://doi.org/10. 1016/S0065-2601(00)80003-9 Leary, M. R., Haupt, A. L., Strausser, K. S., & Chokel, J. T. (1998). Calibrating the sociometer: The relationship between interpersonal appraisals and the state self-esteem. Journal of personality and social psychology, 74(5), 1290. https://doi.org/10.1037/0022- 3514.74.5.1290 Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995). Self-esteem as an interpersonal monitor: The sociometer hypothesis. Journal of personality and social psychology, 68(3), 518. LeBlanc, N. J., Toner, E. R., O'Day, E. B., Moore, C. W., Marques, L., Robinaugh, D. J., & McNally, R. J. (2020). Shame, guilt, and pride after loss: Exploring the relationship between moral emotions and psychopathology in bereaved adults. Journal of affective disorders, 263, 405-412. https://doi.org/10.1016/j.jad.2019.11.164 Lerner, J. S., & Keltner, D. (2000). Beyond valence: Toward a model of emotion-specific influences on judgement and choice. Cognition & Emotion, 14(4), 473–493. https://doi- org.proxy2.cl.msu.edu/10.1080/026999300402763 69 Levine, T. R., Anders, L. N., Banas, J., Baum, K. L., Endo, K., Hu, A. D., & Wong, N. C. (2000). Norms, expectations, and deception: A norm violation model of veracity judgments. Communications Monographs, 67(2), 123-137. https://doi.org/10.1080/03637750009376500 Marks, M. J., Young, T. M., & Zaikman, Y. (2018). The sexual double standard in the real world. Social Psychology,50 (2), 1-12. https://doi.org./10.1027/1864-9335/a000362 Marschall, D., Sanftner, J., & Tangney, J. P. (1994). The state shame and guilt scale. Fairfax, VA: George Mason University. Martinescu, E., Janssen, O., & Nijstad, B. A. (2014). Tell me the gossip: The self-evaluative function of receiving gossip about others. Personality and Social Psychology Bulletin, 40 (12), 1668-1680. https://doi.10.1177/0146167214554916 Martinescu, E., Janssen, O., & Nijstad, B. A. (2019). Self-evaluative and other-directed emotional and behavioral responses to gossip about the self. Frontiers in Psychology, 9, 2603. https://doi.org/10.3389/fpsyg.2018.02603 Massar, K., Buunk, A. P., & Rempt, S. (2012). Age differences in women’s tendency to gossip are mediated by their mate value. Personality and Individual Differences, 52(1), 106-109. https://doi.org/10.1016/j.paid.2011.09.013 McAndrew, F. T., Bell, E. K., & Garcia, C. M. (2007). Who do we tell and whom do we tell on? Gossip as a strategy for status enhancement 1. Journal of Applied Social Psychology, 37(7), 1562-1577. https://doi.org/10.1111/j.1559-1816.2007.00227 McAndrew, F.T. and Milenkovic, M.A. (2002), Of Tabloids and Family Secrets: The Evolutionary Psychology of Gossip. Journal of Applied Social Psychology, 32: 1064- 1082. https://doi-org.proxy2.cl.msu.edu/10.1111/j.1559-1816.2002.tb00256.x McLaren, R. M., & Solomon, D. H. (2008). Appraisals and distancing responses to hurtful messages. Communication Research, 35(3), 339-357. https://doi.org/10.1177/0093650208315961 Meng, L.(M)., Li, T., Huang, X. and Li, S.(K). (2022), "Lift the veil of rumors: the impact of the characteristics of information sources on the effectiveness of rumors spreading", Internet Research, 32 (1), 90-119. https://doi.org/10.1108/INTR-11-2020-0620 Mussweiler, T., & Bodenhausen, G. V. (2002). I know you are, but what am I? Self-evaluative consequences of judging in-group and out-group members. Journal of Personality and Social Psychology, 82(1), 19. https://doi.org/10.1037//0022-3514.82.1.19 70 Peters, K., Kashima, Y. (2015). Bad habit or social good? How perceptions of gossiper morality are related to gossip content. European Journal of Social Psychology. 45, 784- 798.https://doi.org/10.1002/ejsp.2123 Pheko, M. M. (2018). Rumors and gossip as tools of social undermining and social dominance in workplace bullying and mobbing practices: A closer look at perceived perpetrator motives. Journal of Human Behavior in the Social Environment, 28(4), 449-465. https://doi.org/10.1080/10911359.2017.1421111 Philipsen, G. (1992). Speaking culturally: Explorations in social communication. Albany, NY: State University of New York Press. Planalp, S., & Rosenberg, J. (2014). Emotions in interpersonal communication. In C. R. Berger (Ed.), Handbook of communication sciences. (pp. 273-296). Berlin: Mouton de Gruyter. Pyszczynski, T., Greenberg, J., Solomon, S., Arndt, J., & Schimel, J. (2004). Why do people need self-esteem? A theoretical and empirical review. Psychological bulletin, 130(3), 435. https://doi.org/10.1037/0033-2909.130.3.435 Roberts, S. G. B., & Dunbar, R. I. M. (2011). Communication in social networks: Effects of kinship, network size, and emotional closeness. Personal Relationships, 18(3), 439– 452. https://doi-org.proxy1.cl.msu.edu/10.1111/j.1475-6811.2010.01310.x Ruvalcaba-Romero, N. A., Fernández-Berrocal, P., Salazar-Estrada, J. G., & Gallegos-Guajardo, J. (2017). Positive emotions, self-esteem, interpersonal relationships and social support as mediators between emotional intelligence and life satisfaction. Journal of Behavior, Health & Social Issues, 9(1), 1-6. https://doi.org/10.1016/j.jbhsi.2017.08.001 Salice, A. (2020). Self-esteem, social esteem, and pride. Emotion Review, 12(3), 193-205. https://doi.org/10.1177/1754073920930788 Scott, A. M., & Caughlin, J. P. (2014). Enacted goal attention in family conversations about end-of-life health decisions. Communication Monographs, 81(3), 261–284.https/doi: org.proxy2.cl.msu.edu/10.1080/03637751.2014.925568 Shebib, S. J., Holmstrom, A. J., Mason, A. J., Mazur, A. P., Zhang, L., & Allard, A. (2020). Sex and gender differences in esteem support: Examining main and interaction effects. Communication Studies, 71, 167–186. https://doi:10.1080/10510974.2019.1692886 Schoemann, A. M., Boulton, A. J., & Short, S. D. (2017). Determining power and sample size for simple and complex mediation models. Social Psychological and Personality Science, 8(4), 379-386. https://doi.org/10.1177/1948550617715068 71 Stutterheim, S. E., Brands, R., Baas, I., Lechner, L., Kok, G., & Bos, A. E. (2017). HIV status disclosure in the workplace: positive and stigmatizing experiences of health care workers living with HIV. Journal of the Association of Nurses in AIDS Care, 28(6), 923-937. https://doi.org/10.1016/j.jana.2017.06.014 Tolhuizen, J. H. (1989). Communication strategies for intensifying dating relationships: Identification, use and structure. Journal of Social and Personal Relationships, 6(4), 413- 434. https://doi.org/10.1177/0265407589064002 Torres, S. (2019). Aging alone, gossiping together: Older adults’ talk as social glue. The Journals of Gerontology: Series B, 74(8), 1474-1482. https://doi.org/10.1093/geronb/gby154 Tracy, J. L., & Robins, R. W. (2007). The nature of pride. In J. L. Tracy, R. W. Robins, & J. P. Tangney (Eds.), The self-conscious emotions: Theory and research, 263-282. New York, NY: Guilford Press. Turner, M. M., Allard, A., & Patel, S. (under review). Perceptions of Gossipers: The impact of gossip valence and relationship type on liking and trustworthiness. Department of Communication, Michigan State University. Turner, M. M., Mazur, M. A., Wendel, N., & Winslow, R. (2003). Relational ruin or social glue? The joint effect of relationship type and gossip valence on liking, trust, and expertise. Communication Monographs, 70(2), 129-141. https://doi.org/10.1080/0363775032000133782 Vangelisti, A. L., Young, S. L., Carpenter-Theune, K. E., & Alexander, A. L. (2005). Why does it hurt? The perceived causes of hurt feelings. Communication Research, 32(4), 443-477. https://doi.org/10.1177/0093650205277319 Vieth, G., Rothman, A. J., & Simpson, J. A. (2022). Friendship loss and dissolution in adulthood: A conceptual model. Current opinion in psychology, 43, 171- 175. https://doi.org/10.1016/j.copsyc.2021.07.007 Warris, A., & Rafique, R. (2009). Trust in friendship: A comparative analysis of male and female university students. Bulletin of Education and Research, 31(2), 75-84. Weaver, J. R., & Bosson, J. K. (2011). I feel like I know you: Sharing negative attitudes of others promotes feelings of familiarity. Personality and Social Psychology Bulletin, 37(4), 481- 491. https://doi.org/10.1177/0146167211398364 Wert, S. R., & Salovey, P. (2004). A social comparison account of gossip. Review of general psychology, 8(2), 122-137. https://doi.org/10.1037/1089-2680.8.2.122 72 Wu, J., Balliet, D., & Van Lange, P. A. (2016). Reputation, gossip, and human cooperation. Social and Personality Psychology Compass, 10(6), 350-364. https://doi.org/10.1111/spc3.12255 Wu, L. Z., Birtch, T. A., Chiang, F. F., & Zhang, H. (2018). Perceptions of negative workplace gossip: A self-consistency theory framework. Journal of Management, 44(5), 1873- 1898.https://doi.org/10.1177/0149206316632057 Young, S. L., Kubicka, T. L., Tucker, C. E., Chavez-Appel, D., & Rex, J. S. (2005). Communicative responses to hurtful messages in families. The Journal of Family Communication, 5(2), 123-140. https://doi.org/10.1207/s15327698jfc0502_4 Zong, B., Martinescu, E., Beersma, B., Xu, S., & Zhang, L. (2023). How multi-source gossip affects targets’ emotions and strategic behavioral responses. Journal of Business Ethics, 1-18. https://doi.org/10.1007/s10551-022-05322-8 73 Table 1 Inner Coder Reliability for Pilot Study 1. APPENDIX A: Tables and Figures Gossip Themes Percent Agreement Cohen’s Kappa Krippendorff’s Alpha N Agreement N Disagreement N Cases N Decisions Personality Talent Relationship Helpfulness 98% 95.4% 94.4% 98% Work Ethic 97.5% Status Beauty 99% 100% Future Endeavors 98.5% Non-valid response 99% 0.95 0.83 0.74 0.86 0.88 0.85 1 0.79 0.93 0.95 0.83 0.74 0.86 0.88 0.85 1 0.79 0.93 74 193 188 186 193 192 195 197 194 195 4 9 11 4 5 2 0 3 2 197 197 197 197 197 197 197 197 197 394 394 394 394 394 394 394 394 394 Table 2 Results from a One-way Test Assessing Realism in Pilot 2. Scenario Type M SD t(198) p Cohen’s d Positive Gossip Personality Talent Work Ethic Negative Gossip Personality Talent Work Ethic 5.31 5.06 5.43 4.34 3.73 4.09 1.20 1.30 1.00 1.61 1.70 1.70 21.44 16.93 27.19 .001 .001 .001 1.10 0.81 1.43 2.95 .004 0.21 -2.26 0.78 .03 .44 -0.16 0.05 75 Table 3 Results from a One-way Test Assessing Severity in Pilot 2. Scenario Type M SD t(198) p Cohen’s d Positive Gossip Personality Talent Work Ethic Negative Gossip Personality Talent Work Ethic 1.98 2.08 2.04 5.07 5.20 5.13 0.84 -34.17 0.85 -31.88 .001 .001 -2.42 -2.26 0.75 -36.85 .001 -2.61 1.33 1.42 1.40 11.31 11.64 11.73 .001 .001 .001 0.80 0.83 0.83 76 Table 4 Goodness of Fit Indices for Main Study Instruments. Variable 𝑥2 PEV NEV Hurt Shame Pride Self-esteem Affiliation 13.19** 40.82** 620.83** 118.86** 104.44** 158.76** 438.26** df 2 2 20 5 5 14 9 CFI RMSEA (90%) LCI UCI 0.99 0.99 0.94 0.98 0.99 0.98 0.92 0.10 0.19 0.23 0.20 0.19 0.14 0.29 0.05 0.14 0.22 0.17 0.16 0.12 0.27 0.15 0.24 0.25 0.23 0.22 0.16 0.31 Note. ** p< .001. CFI = comparative fit index; RMSEA = root mean square average of approximation. 77 Table 5 Zero-Order Correlation for All Key Variables. 2 3 4 5 6 7 8 9 1 -- 0.19 -- 1. Gossip Valence 2. Relationship Type 3. Scenario -0.001 -0.002 -- 4. PEV 5. NEV 6. Hurt 7. Pride -0.93** -0.02 -0.03 -- 0.95** 0.03 0.03 -0.93** -- 0.91** 0.01 0.02 -0.89** 0.95** -- -0.93** -0.02 -0.02 0.93** -0.92** -0.92** -- 8. Shame 0.80** -0.002 0.02 -0.79** 0.85** 0.93** -0.84** -- 9. Self-esteem -0.81** -0.01 -0.04 0.79** -0.86** -0.91** 0.85** -0.92** -- 10. Affiliation -0.84** -0.10* 0.02 0.83** -0.85** -0.84** 0.84** -0.74** 0.73** Note. PEV = positive expectancy violations; NEV = negative expectancy violations; *Correlation is significant at the 0.05 level; ** Correlation is significant at the 0.01 level. 78 Table 6 Model Coefficients for the Serial Mediation between Gossip Valence on Intentions to Affiliate Through Positive Expectancy Violations and Pride. Expectancy Positive Violation 𝑅2 =0.86, MSE = 0.73 Constant Gossip Valence Age Scenario Pride 𝑅2 = 0.89, MSE = 0.61 Constant Gossip Valence PEV Age Scenario Intentions to Affiliate 𝑅2 = 0.74, MSE = 0.51 Constant Gossip Valence PEV Pride Age Scenario 𝑖𝑦 𝑏1 𝑏2 𝑏3 𝑖𝑦 𝑏1 𝑏2 𝑏3 𝑏4 𝑖𝑦 𝑏1 𝑏2 𝑏3 𝑏4 𝑏5 Coeff. SE t p 1.91 4.29 -0.01 0.18 0.76 2.13 0.53 0.002 0.02 1.61 1.04 0.09 0.21 -0.003 -0.09 0.12 0.07 0.003 0.07 0.13 0.18 0.04 0.002 0.07 0.12 0.18 0.04 0.04 0.002 0.06 15.99 59.55 -3.23 2.45 5.78 11.98 13.71 0.80 0.27 13.02 5.73 2.35 5.37 -1.56 -1.54 0.00 0.00 0.001 0.01 0.00 0.00 0.00 0.42 0.79 0.00 0.00 0.02 0.00 0.12 0.12 Note. PEV = Positive Expectancy Violations; SE = standard error; t = t-value; p = p-value; Gossip was dummy coded as 1 = Positive gossip and 0 = negative gossip; Scenario was dummy coded as 1 = Personality scenario and 0 = Work Ethic scenario. 79 Table 7 Model Coefficients for the Serial Mediation between Gossip Valence on Intentions to Affiliate Through Negative Expectancy Violations and Hurt. Negative Violation 𝑅2 = 0.90, MSE = 0.66 Expectancy Constant Gossip Valence Age Scenario Hurt 𝑅2 = 0.90, MSE = 0.53 Constant Gossip Valence NEV Age Scenario Intentions to Affiliate 𝑅2 = 0.75, MSE = 0.50 Constant Gossip Valence NEV Hurt Age Scenario 𝑖𝑦 𝑏1 𝑏2 𝑏3 𝑖𝑦 𝑏1 𝑏2 𝑏3 𝑏4 𝑖𝑦 𝑏1 𝑏2 𝑏3 𝑏4 𝑏5 Coeff. SE t p 6.16 -4.73 0.003 -0.19 0.79 -0.51 0.79 -0.003 0.05 3.91 1.04 -0.12 -0.17 -0.005 -0.09 0.11 0.07 0.003 0.07 0.25 0.19 0.04 0.002 0.06 0.25 0.19 0.05 0.04 0.002 0.06 54.21 -69.10 1.03 -2.80 3.12 -2.72 20.81 -1.39 0.78 15.62 5.60 -2.53 -4.21 -2.13 -1.57 0.00 0.00 0.30 0.005 0.002 0.007 0.00 0.17 0.43 0.00 0.00 0.01 0.00 0.03 0.12 Note. NEV = Negative Expectancy Violations; SE = standard error; t = t-value; p = p-value. Gossip was dummy coded as 1 = Positive gossip and 0 = negative gossip; Scenario was dummy coded as 1 = Personality scenario and 0 = Work Ethic scenario. 80 Table 8 Model Coefficients for the Serial Mediation between Gossip Valence on Intentions to Affiliate Through Negative Expectancy Violations and Shame. Positive Violation 𝑅2 = 0.90, MSE = 0.66 Expectancy Constant Gossip Valence Age Scenario Shame 𝑅2 =0.85, MSE = 1.23 Constant Gossip Valence NEV Age Scenario Intentions to Affiliate 𝑅2 = 0.74, MSE = 0.52 Constant Gossip Valence NEV Shame Age Scenario 𝑖𝑦 𝑏1 𝑏2 𝑏3 𝑖𝑦 𝑏1 𝑏2 𝑏3 𝑏4 𝑖𝑦 𝑏1 𝑏2 𝑏3 𝑏4 𝑏5 Coeff. SE t p 6.16 -4.73 0.003 -0.019 0.43 0.13 0.74 -0.008 0.05 3.79 1.13 -0.23 -0.05 -0.005 -0.10 0.11 0.07 0.003 0.07 0.39 0.29 0.06 0.004 0.09 0.25 0.19 0.04 0.03 0.002 0.06 54.21 -69.10 1.03 -2.80 0.00 0.00 0.30 0.005 1.11 0.45 12.83 -2.33 0.55 15.07 6.08 -5.33 -1.71 -2.02 -1.65 0.27 0.65 0.00 0.02 0.58 0.00 0.00 0.00 0.09 0.04 0.09 Note. NEV = Negative Expectancy Violations; SE = standard error; t = t-value; p = p-value. Gossip was dummy coded as 1 = Positive gossip and 0 = negative gossip; Scenario was dummy coded as 1 = Personality scenario and 0 = Work Ethic scenario 81 Table 9 Indirect Effects for Serial Mediation Models Conducted in Main Study. Model 1 Model 2 Model 3 Variables Effect SE t Bootstrap 95% CI LLCI ULCI Total Direct Effect of X on Y Direct Effect of X on Y Total Indirect Effect Gossip > PEV > Affiliation Gossip > Pride > Affiliation Gossip > PEV > Pride > Affiliation Total Direct Effect of X on Y Direct Effect of X on Y Total Indirect Effect Gossip > NEV > Affiliation Gossip > Hurt > Affiliation Gossip > NEV > Hurt > Affiliation Total Direct Effect of X on Y Direct Effect of X on Y Total Indirect Effect Gossip > NEV > Affiliation Gossip > Shame > Affiliation Gossip > NEV > Shame > Affiliation 2.36 1.04 1.32 0.41 0.44 0.47 2.36 1.04 1.32 0.59 0.09 0.65 2.36 1.13 1.23 1.07 -0.006 0.16 0.06 0.18 0.18 0.19 0.12 0.11 0.06 0.19 0.21 0.31 0.07 0.19 0.06 0.19 0.22 0.26 0.02 0.12 37.39 5.73 37.39 5.60 37.39 6.08 2.24 0.68 0.96 0.06 0.22 0.27 2.24 0.68 0.92 0.02 -0.01 0.27 2.24 0.77 0.59 0.59 -0.05 -0.07 2.49 1.39 1.68 0.79 0.70 0.68 2.47 1.40 1.76 1.22 0.26 1.01 2.49 1.50 1.21 1.62 0.04 0.39 Note. PEV = Positive expectancy violation; NEV = negative expectancy violation; SE = standard error, t = t -value. 82 Table 10 Means and Standard Deviations for Dependent Variables within Scenario, Valence Group, and Relationship Type. Personality Scenario Work Ethic Scenario Positive Gossip Negative Gossip Positive Gossip Negative Gossip Friend Acquaintance Friend Acquaintance Friend Acquaintance Friend Acquaintance Dependent Variable PEV NEV Pride Hurt M SD M SD M SD M SD Shame M SD Self-Esteem M SD Affiliation M SD 6.05 0.86 1.37 0.61 6.14 0.98 1.23 0.40 1.26 0.46 6.31 0.83 4.48 0.64 6.07 0.61 1.39 0.54 6.19 0.73 1.31 0.59 1.36 0.68 6.32 0.90 4.30 0.520 1.69 0.87 6.07 1.05 1.76 0.96 5.65 1.07 4.68 1.58 2.89 1.36 1.81 0.92 1.76 0.74 6.00 0.77 1.83 0.88 5.30 1.28 4.57 1.71 3.01 1.48 1.80 0.67 5.82 0.93 1.44 0.73 6.12 0.80 1.33 0.66 1.33 0.64 6.13 1.05 4.48 0.52 5.88 0.98 1.55 0.85 6.03 0.93 1.37 0.62 1.38 0.61 6.19 0.71 3.95 0.45 1.67 0.98 6.26 1.04 1.65 1.07 5.59 1.43 4.95 1.86 2.72 1.79 2.21 1.04 1.57 0.89 6.34 0.81 1.67 0.86 5.61 1.20 4.59 1.64 2.69 1.42 1.98 0.89 Note. PEV = positive expectancy violations; NEV = negative expectancy violations. 83 Figure 1 Illustration of Serial Mediation Model 1. Positive Expectancy Violations 𝑑21= 0.53** Pride 𝑎1= 4.29** 𝑏2= 0.21** Gossip Valence 𝑎2= 2.13** 𝑏1= 0.95* c′= 1.04** Intentions to Affiliate Note. Gossip valence was dummy coded to reflect the following: 1 = Positive Gossip, 0 = Negative Gossip. *p <.01; ** p < .001. 84 Figure 2 Illustration of Serial Mediation Model 2. Negative Expectancy Violations 𝑑21= 0.79** Hurt 𝑎1= - 4.73** 𝑏2= - 0.17** Gossip Valence 𝑎2= - 0.51* 𝑏1= - 0.12* c′= 1.04** Intentions to Affiliate Note. Gossip valence was dummy coded to reflect the following: 1 = Positive Gossip, 0 = Negative Gossip. *p <.01; ** p < .001. 85 Figure 3 Illustration of Serial Mediation Model 3. 𝑑21= 0.74** Negative Expectancy Violations Shame 𝑎1= - 4.73** 𝑎2= 0.13 𝑏1= -0.23** Gossip Valence c′= 1.14** 𝑏2= - 0.05 Intentions to Affiliate Note. Gossip valence was dummy coded to reflect the following: 1 = Positive Gossip, 0 = Negative Gossip. *p <.01; ** p < .001. 86 Figure 4 Interaction between Gossip Valence and Scenario Type on Intentions to Affiliate. Note. Covariates appearing in the model are evaluated at the following values Age = 38.32, Sex = 1.52. 87 Figure 5 Interaction between the Relationship with the Gossiper and Scenario Type on Intentions to Affiliate. Note. Covariates appearing in the model are evaluated at the following values Age = 38.32, Sex = 1.52. 88 APPENDIX B: Pilot Study Survey Informed Consent for MSU Students This survey will ask you to recall a time when you were evaluated about by another person. This research study is being conducted by Amanda Allard from the Department of Communication at Michigan State University for her thesis study for her Master of Arts degree. We are looking for volunteers who are at least 18 years of age and have attended at least one college level course 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. Hence, 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 on SONA 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, Amanda Allard (her contact information is below). Principal Investigator: Dr. Monique Turner mmturner@msu.edu Communication Arts & Sciences Building East Lansing, MI 48824 89 Lead Researcher: Amanda Allard allardam@msu.edu 713-478-7329 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.   [At this time participants were randomly assigned to one of two conditions: a positive gossip condition or a negative gossip condition. They were asked to recall a time they were positively/negatively gossiped about by both a friend and an acquaintance. Below is the prompt they were shown along with the questions they were asked.] “For the next few questions, we would like you to recall a time when a friend/acquaintance spoke about you, to another person, in a positive/negative way--and you were not there. Friend definition: a friend is someone (non-family) you hang out with often and have a close relationship with. Please do not think of a family member. Acquaintance definition: an acquaintance is considered to be a person you know very little about yet do not seek to hang out with them or connect with them regularly. Question 1: Please describe as much detail regarding this event/conversation as you can. In your description, tell us: 1. How you found out that an friend/acquaintance had been talking about you. 2. What exactly did this acquaintance say about you (saying something positive/negative). 3. How what they said about you affected you. Question 2: What emotions would you say you felt when you found out a(n) friend/acquaintance spoke about you in this way to another person? Question 3: Please describe what you did after discovering an acquaintance was talking positively about you. 90 ➢ Participants were provided a section in Qualtrics where they were able to write their response. A reminder of the instructions were placed above the text box. Demographic Questions Age 1. What is your age? Please type your age as a numerical value in the box provided (e.g., 18, 26, 45). ____________ Biological Sex 1. What is your biological sex? 1. Male 2. Female Gender 1. 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 Ethnicity 1. Which of the following choices best describes your ethnicity? 1. Hispanic or Latino 2. Not Hispanic or Latino Race Which of the following choices best describes your ethnicity? 1. White 2. Black or African American 3. American Indian or Alaska Native 4. Asian 5. Native Hawaiian or Pacific Islander 6. Multiple or mixed races 7. Other/prefer not to answer Location Are you from the US? 1. Yes 91 2. No [ If yes, participants were directed to a question where they chose what state they are from. If no, participants were directed to a page where they indicated what country they were from] Class 1. What is your class standing? 1. Freshman 2. Sophomore 3. Junior 4. 4th year Senior 5. 5th year Senior 6. Beyond 5th year Senior 7. Graduate Student ComMaj 1. Are you a communication major? 1. Yes 2. No Inter 1. Are you an international student? 1. Yes 2. No SONA 1. Are you taking this survey through SONA? 1. Yes (at end will be redirected to SONA). 2. No (at end will be redirected to another survey, not linked to their responses in this survey, to fill out information to get class credit). Thank You Page Thank you for participating in the present study! Your participation was greatly appreciated and will be recorded and used to enhance empirical scholarly research in family communication. Class Credit Page Thank you for participating in the present study! Your participation was greatly appreciated and will be recorded and used to enhance empirical scholarly research in romantic relationships. Please fill out all the information to ensure you get class credit. This information is in no way linked to your responses to the survey. Your First Name: ____________ Your Last Name: ____________ Course Letters & Number for Credit (i.e., Com 425): _________ Course Section Number for Credit (i.e., 003): __________ Professor’s Name of the Course: __________ 92 Your MSU email address: ____________ Debrief Statement Dear Participant, Thank you for taking the time to participate in this experiment. The purpose of this study was to understand how some different ways an instructor responds to a student’s use of swearing are viewed by students. Your participation in this study is greatly appreciated because the responses collected from this survey can assist in the understanding of how the communication between an instructor and a student can positively or negatively impact the student/instructor relationship and a student’s learning. Although the scenarios you were presented were hypothetical, we as researchers understand that reading them may have caused some distress. If this study has made you feel uncomfortable or distressed, please follow the following link to contact MSU’s student counseling services (https://caps.msu.edu). Also, you are highly encouraged to contact the Principal Investigator, Lead Researcher, or the Institutional Review Board (IRB) if you have any additional questions or concerns regarding the survey you just took. You can find names, emails, and phone numbers located below to contact these three individuals. Again, your participation in this study is greatly appreciated. To conclude, we ask you not to discuss this survey with anyone else who is currently participating in this study or has intentions to participate in this study. We would greatly appreciate this gesture as it will allow the researchers to truly examine the effects of inappropriate communication in the classroom. Thank you! Principal Investigator: Dr. Monique Turner mmturner@msu.edu Communication Arts & Sciences Building East Lansing, MI 48824 Lead Researcher: Amanda Allard allardam@msu.edu 713-478-7329 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 93 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. 94 APPENDIX C: Qualitative Themes from Pilot Study 1 Personality/Character trait. This theme was present when the reason an individual was gossiped about revolves around a character or personality trait of theirs (i.e., the person is so kind, they stood up for me, or they have great values/morals). Example: “I heard from another friend that my friend was talking about me. My friend said that he was talking about my character and I was a nice person and had a caring attitude. This interaction affected me in a positive way. I felt very good hearing that and it brighten up my day.” Talent. This theme was present when an individual was gossiped about regarding their ability to do something well. Some examples of this may include, they play the violin well, they are a great leader, or she is the smartest in the class. Example: “A friend said I was getting to be a good painter. They said it wasn't fair that some people do well at art while others don't. My other friend told me what she had talked about with the first friend. I was happy that I was recognized for being good at something important to me, but bothered that the friend who made this comment would give up at making art.” Relationship Quality. This theme was present when an individual was gossiped about because of they were perceived as being a good or bad friend, roommate, or even boyfriend/girlfriend. Example: “I found out my friend was talking about me to my other friends and was stating how good of a friend i am. This affected me greatly, it made me very happy to hear.” Helpfulness/Supportiveness. This theme was present when the reason a person was gossiped about revolved around how well or how poorly they helped others. Example: “One of my mom's friends said that I seem like a good son with how I help around the house. I heard this from my mom and it made me proud personally and proud for my mom.” Work ethic. This theme was present when an individual was positively or negatively gossiped about regarding their ability to work well with others in a corporate or non-corporate environment. Example: “One of my coworkers told my boss that I was a really hard worker. My boss then told me what she had heard. It made me excited and happy to hear that, as I was a new employee at the time.” Status change. This theme occurs when an individual was positively or negatively gossiped bout regarding a status change of one’s title, wealth, or achievements. Example: “I was out with a big group of friends, some of them I did not know. I overheard one of the people I didn't know say something while gesturing at my direction about being poor because anytime they see me, I'm typically wearing very similar clothes. I felt embarrassed and avoided them the rest of the time.” Beauty. This theme occurred when an individual was positively or negatively gossiped about regarding their appearance. Example: “I was out with my friends in a bar, a social setting, and a friend of a friend said that I dressed like I hadn’t bought any clothes since 2017.” 95 Future endeavors. This theme was present when an individual was positively or negatively gossiped about regarding upcoming plans they had with someone else. Example: “I found out my boyfriend was going to take me out on a date because his mom told me. He said that he was excited, and it made me happy.” 96 Informed Consent APPENDIX D: Pilot Study 2 Survey This survey will ask you to rate hypothetical scenarios regarding evaluations made about you. This research study is being conducted by Amanda Allard from the Department of Communication at Michigan State University for her dissertation. We are looking for volunteers who are at least 18 years of age. 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 10 minutes to complete the survey. We see no risks or discomfort associated with completing this survey. Furthermore, although there is a chance you may become distressed recalling personal events or reading hypothetical scenarios, there is no physical, psychological, or financial risk associated with completing this study. 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 communicate. Your responses will remain private and won’t be connected to your personal information. The duration of this online survey is approximately 8-10 minutes. Hence, participants who complete this survey will receive $2.00. You may withdraw at any time without penalty. 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, Amanda Allard (her contact information is below). Principal Investigator: Dr. Monique Turner mmturner@msu.edu Communication Arts & Sciences Building East Lansing, MI 48824 Lead Researcher: Amanda Allard allardam@msu.edu 713-478-7329 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 97 this study, such as scientific issues, how to do any part of it, or to report an injury, please contact the lead researcher. Continuing with the web survey indicates that you give your consent to participate.   [Participants were randomly assigned to one of two conditions (Friendship/Acquaintance). All participants were then asked to assess both positive and negative gossip scenarios for all three scenario types: work ethic, personality, and talent in various orders.] For the remainder of the survey, you will be asked to read and assess scenarios that occurred between you and a same-sex friend/Acquaintance. Friendship definition: Although friends may be family, we would like for you to think of a friend that is not blood-related. More specifically, A friend is considered someone who you often seek out their company for intimacy or assistance, and is considered as a voluntary, non- biological personal relationship. By clicking next, you are acknowledging how a friend is being defined in this survey. Acquaintance definition: An acquaintance is considered someone who you are aware of yet have little to no intimacy with or knowledge about,and is someone you typically do not seek out their company. By clicking next, you are acknowledging how an acquaintance is being defined in this survey. Personality Positive Gossip • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how much they enjoy hanging out with you. In fact, they said, “I think they are the funniest and most positive person I have ever met”. I normally wouldn’t tell you this information, but I just thought you would want to know. (60 words) Talent • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how smart and creative they think you are. In fact, they said, “I think they are the smartest and most creative person I have ever met”. I normally wouldn’t tell you this information, but I just thought you would want to know. (60 words) Work • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how much they admire how hard you work. In fact, they said, “I admire how much time and effort they put into all of their work.” I normally wouldn’t tell you this information, but I just thought you would want to know. (60 words) Personality Negative Gossip • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how much they dislike hanging out with you. In fact, they said, “I think they are one of the most annoying people I have ever met”. I normally wouldn’t tell you this information, but I just thought you would want to know. (60 words) 98 Talent • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how unintelligent and uncreative they think you are. In fact, they said, “I think they are the dumbest and most uncreative person I have ever met”. I normally wouldn’t tell you this information, but I just thought you would want to know (60 words) Work • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how much they dislike your work ethic. In fact, they said, “I think they are one of the laziest and uncooperative individuals they have ever met.” I normally wouldn’t tell you this information, but I just thought you would want to know. (60 words) [All participants were asked to rate these scenarios in terms of realism and severity as shown below] Realism Measure Directions: Please rate the extent to which you disagree or agree with the following statements in regard to the scenario you read. Stem: The scenario I just read is… 1. realistic. 2. Believable. 3. Plausible. 4. something that could happen. Severity Measure Directions: Please rate the extent to which you disagree or agree with the following statements in regard to the scenario you read. Stem: I would perceive this scenario as….. 1. Serious. 2. Severe. 3. Upsetting. 4. Uncomfortable. Perception of Gossiper Directions: Now, imagining this conversation just happened to you, please rate the extent to which you disagree or agree with the following statements in regard to the scenario you just read. Stem: After my friend/acquaintance told me this I view them as… 1. Kind. 99 2. Helpful. Demographic Questions Age 2. What is your age? Please type your age as a numerical value in the box provided (e.g., 18, 26, 45). ____________ Biological Sex 2. What is your biological sex? 2. Male 3. Female Gender 2. Please click the option that best describes your gender. 2. Extremely feminine 3. Moderately feminine 4. Slightly feminine 5. Androgynous 6. Slightly masculine 7. Moderately masculine 8. Extremely masculine Ethnicity 2. Which of the following choices best describes your ethnicity? 2. Hispanic or Latino 3. Not Hispanic or Latino Race Which of the following choices best describes your ethnicity? 2. White 3. Black or African American 4. American Indian or Alaska Native 5. Asian 6. Native Hawaiian or Pacific Islander 7. Multiple or mixed races 8. Other/prefer not to answer Location Are you from the US? 2. Yes 3. No 100 [ If yes, participants were directed to a question where they chose what state they are from. If no, participants were directed to a page where they indicated what country they were from] Class 2. What is your class standing? 2. Freshman 3. Sophomore 4. Junior 5. 4th year Senior 6. 5th year Senior 7. Beyond 5th year Senior 8. Graduate Student 101 APPENDIX E: Instruments for Pilot Study 2 Realism Measurement Realism of Scenario for Pilot Study • Adapted from Shebib et al.’s (2020) scale • 7-point Likert-type scale 1 = strongly disagree to 7 = strongly agree Directions: Please rate the extent to which you disagree or agree with the following statements in regard to the scenario you read. Stem: The scenario I just read is… 2. realistic. 5. believable. 3. something that could happen. 102 Severity • Adapted from Shebib et al.’s (2020) scale • 7-point Likert-type scale 1 = strongly disagree to 7 = strongly agree Severity Measurement Directions: Please rate the extent to which you disagree or agree with the following statements in regard to the scenario you read. Stem: I would perceive this scenario as….. 2. Serious. 3. Severe. 5. Upsetting. 4. Awkward. 6. Uncomfortable. 6. Intense. 103 Perceptions of the Gossiper: Pilot Study 2 Perception of Gossiper Created by: Allard and Turner • 7-point Likert-type scale 1 = strongly disagree to 7 = strongly agree Directions: Now, imagining this conversation just happened to you, please rate the extent to which you disagree or agree with the following statements in regard to the scenario you just read. Stem: After my friend/acquaintance told me this I view them as… 1. Kind. 2. Helpful. 104 Demographic Questions Demographic Questions Age 1. What is your age? Please type your age as a numerical value in the box provided (e.g., 18, 26, 45). ____________ Biological Sex 1. What is your biological sex? 1. Male 2. Female Gender 1. 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 Ethnicity 1. Which of the following choices best describes your ethnicity? 1. Hispanic or Latino 2. Not Hispanic or Latino Race Which of the following choices best describes your ethnicity? 1. White 2. Black or African American 3. American Indian or Alaska Native 4. Asian 5. Native Hawaiian or Pacific Islander 6. Multiple or mixed races 7. Other/prefer not to answer Location Are you from the US? 1. Yes 2. No [ If yes, participants were directed to a question where they chose what state they are from. If no, participants were directed to a page where they indicated what country they were from] 105 Class 1. What is your class standing? 1. Freshman 2. Sophomore 3. Junior 4. 4th year Senior 5. 5th year Senior 6. Beyond 5th year Senior 7. Graduate Student ComMaj 1. Are you a communication major? 1. Yes 2. No Inter 1. Are you an international student? 1. Yes 2. No SONA 1. Are you taking this survey through SONA? 1. Yes (at end will be redirected to SONA). 2. No (at end will be redirected to another survey, not linked to their responses in this survey, to fill out information to get class credit). Thank You Page Thank you for participating in the present study! Your participation was greatly appreciated and will be recorded and used to enhance empirical scholarly research in family communication. Class Credit Page Thank you for participating in the present study! Your participation was greatly appreciated and will be recorded and used to enhance empirical scholarly research in romantic relationships. Please fill out all the information to ensure you get class credit. This information is in no way linked to your responses to the survey. Your First Name: ____________ Your Last Name: ____________ Course Letters & Number for Credit (i.e., Com 425): _________ Course Section Number for Credit (i.e., 003): __________ Professor’s Name of the Course: __________ Your MSU email address: ____________ 106 APPENDIX F: Main Study Instruments Instrument to Measure Expectancy Violation Scale for Expectancy Violations • Adapted from Afifi and Metts (1998) and Turner et al., (2003) • 7-point Likert-type scale 1 = strongly disagree to 7 = strongly agree • Items with an * indicate items assessing negative expectancy violations. Directions (Created by Allard): “Below are some statements that may or may not illustrate how you are feeling after hearing the evaluation made about you. Please rate each statement according to the 7-point scale illustrated below.” Stem: This statement made about me… Items 4. Are nicer than I would expect. 2. Are better than I would anticipate. 3. Surprised me in a good way. 1. Exceeded my personal expectations. 7.Shocked me in a bad way. * 8.Are below my hopes of what others think of me. * 5.Are worse than I would expect. * 6.Violated my personal expectations. * Factor Loading 1 2 0.98 0.98 0.97 0.89 0.99 0.98 0.98 0.97 107 Original State Shame and Guilt Scale Instrument to Measure Pride • Marshall, Safner, and Tangney (1994) • 5-point Likert-type scale 1 = Not feeling this way at all to 5 = Feeling this way strongly • Subcategories: Pride (1,4,7,10,13), Shame (2,5,8,11,14), and guilt (3,6,9,12,15) Directions: “The following are some statements which may or may not describe how you are feeling right now. Please rate statement using the 5-point scale below. Please rate each statement based on how you are feeling right at this moment.” 1. I feel good about myself. 2. I want to sink into the floor and disappear. 3. I feel remorse, regret. 4. I feel worthwhile, valuable. 5. I feel small. 6. I feel tension about something I have done. 7. I feel capable, useful. 8. I feel like I am a bad person. 9. I cannot stop thinking about something I have done. 10. I feel proud. 11. I feel humiliated, disgraced. 12. I feel like apologizing, confessing. 13. I feel pleased about something I have done. 14. I feel worthless, powerless. 15. I feel bad about something I have done. Adapted Pride Scale for Main Study • Adapted from Marshall, Safner, and Tangney (1994) • 5-point Likert-type scale 1 = Not feeling this way at all to 7 = Feeling this way strongly *A reminder of the evaluation either their friend or acquaintance made will be placed above the directions. Directions: Please rate the extent to which you disagree or agree with the following statements regarding the scenario you just read. Stem: The statements made about me would make me feel … 108 Items 2. worthwhile, valuable. 1. good about myself. 5. pleased with myself. 3. capable, useful. 4. proud. Factor Loading 1 0.99 0.99 0.98 0.98 0.98 109 Original State Shame and Guilt Scale Instrument to Measure Shame • Marshall, Safner, and Tangney (1994) • 5-point Likert-type scale 1 = Not feeling this way at all to 5 = Feeling this way strongly • Subcategories: Pride (1,4,7,10,13), Shame (2,5,8,11,14), and guilt (3,6,9,12,15) Directions: “The following are some statements which may or may not describe how you are feeling right now. Please rate statement using the 5-point scale below. Please rate each statement based on how you are feeling right at this moment.” 1. I feel good about myself. 2. I want to sink into the floor and disappear. 3. I feel remorse, regret. 4. I feel worthwhile, valuable. 5. I feel small. 6. I feel tension about something I have done. 7. I feel capable, useful. 8. I feel like I am a bad person. 9. I cannot stop thinking about something I have done. 10. I feel proud. 11. I feel humiliated, disgraced. 12. I feel like apologizing, confessing. 13. I feel pleased about something I have done. 14. I feel worthless, powerless. 15. I feel bad about something I have done. Adapted Shame Scale for Main Study • Adapted from Marshall, Safner, and Tangney (1994) • 5-point Likert-type scale 1 = Not feeling this way at all to 7 = Feeling this way strongly *A reminder of the evaluation either their friend or acquaintance made will be placed above the directions. Directions: Please rate the extent to which you disagree or agree with the following statements regarding the scenario you just read. Stem: The statements made about me would make me feel … 110 Items 2. small. 5. worthless, powerless. 4. humiliated, disgraced. 1. like sinking into the floor. 3. like a bad person. Factor Loading 1 0.97 0.97 0.96 0.95 0.92 111 Instrument to Measure Hurt Scale Used to Measure Hurt • Adapted from Scott and Caughlin (2014) • 7-point Likert-type scale 1 = Not at all to 7 = a lot • Original Items (1,2,3) / Created items by Allard (4,5,6,7,8) Directions (Created by Allard): “Below are some statements that may or may not illustrate how you are feeling after hearing the evaluation made about you. Please rate each statement according to the 7-point scale illustrated below.” Stem: The evaluation made about me makes me feel… Items 1. hurt. 7. sad. 3. wounded. 2. disappointed. 5. unvalued. 6. less than. 4.betrayed. 8. unimportant. Factor Loading 1 0.98 0.97 0.97 0.96 0.96 0.96 0.93 0.93 112 Instrument to Measure State Self-Esteem Original State Self-Esteem Scale • Heatherton and Polivy (1991) • 5-point Likert-type scale 1 = Not at all 5 = extremely • Subcategories: performance (1,4,5,9,14,18,19), social (2,8,10,13,15,17,20), and appearance (3,6,7,11,12,16) Directions: “This is a questionnaire designed to measure what you are thinking at this moment. There is, of course, no right answer for any statement. The best answer is what you feel is true of yourself at this moment. Be sure to answer all of the items, even if you are not certain of the best answer. Again, answer these questions as they are true for you RIGHT NOW.” 1. I feel confident about my abilities. 2. I am worried about whether I am regarded as a success or failure. (R) 3. I feel satisfied with the way my body look right now. 4. I feel frustrated or rattled about my performance. ( R) 5. I feel that I am having trouble understanding things that I read. ( R) 6. I feel that others respect and admire me. 7. I am dissatisfied with my weight. ( R) 8. I feel self-conscious. ( R) 9. I feel as smart as others. 10. I feel displeased with myself. ( R) 11. I feel good about myself. 12. I am pleased with my appearance right now. 13. I am worried about what other people think of me. ( R) 14. I feel confident that I understand things. 15. I feel inferior to others at this moment. ( R) 16. I feel unattractive. ( R) 17. I feel concerned about the impression I am making. ( R) 18. I feel that I have less scholastic ability right now than others. ( R) 19. I feel like I am not doing well. ( R) 20. I am worried about looking foolish. ( R) Adapted State Self-Esteem Scale for Main Study • Heatherton and Polivy (1991) • 5-point Likert-type scale 1 = Not at all 5 = extremely Directions: *A reminder of the evaluation either their friend or acquaintance made will be placed above the directions. Directions: Please rate the extent to which you disagree or agree with the following statements in regard to the scenario you just read. 113 Stem: The statements made about me would make me... Items 7. worried about looking foolish. (R) 4. worried about what other people think of me. (R) 6. concerned about the impression I made. (R) 3. displeased with myself. (R) 5. inferior to others at this moment. (R) 1. worried about whether I was regarded as a success or failure. (R) 2. self-conscious. (R) Factor Loading 1 0.97 0.96 0.94 0.93 0.93 0.92 0.89 114 Original Affiliation Scale Instrument to Measure Behavioral Intentions to Affiliate • Created by Martinescu et al., (2019) • 7-point Likert Type Scale (1= strongly disagree to 7= strongly agree) • Bolded items derive from original scale • (*) Indicates reverse coding Directions: After hearing the evaluation made by your friend/the acquaintance, please rate how much you agree or disagree with the following statements. Stem: How likely would it be for you to do the following things with this friend/acquaintance: Items 3. spend time with them. 1. hang out with them 2. continue a relationship with them. 4. take a break from the relationship. (R) 6. avoid being around the. (R) 5. interact with them. Factor Loading 1 0.98 0.97 0.97 0.88 0.86 0.78 115 Manipulation Check Participants will answer the following question to complete the manipulation check: Who made the evaluation regarding you in the scenario you used for this survey? Choice 1: Friend Choice 2: An acquaintance Choice 3: Neither friend nor an acquaintance Choice 4: I don’t remember 116 APPENDIX G: Main Study Survey Informed Consent This survey will ask you to provide your opinions about a hypothetical evaluation made about you. This research study is being conducted by Amanda Allard from the Department of Communication at Michigan State University for her dissertation in completion of her doctorate degree. We are looking for volunteers who are at least 18 years of age and have attended at least one college level course 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 6 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 $1.20 credits through Prolific. The duration of this online survey is approximately 5-6 minutes. Participation in this online survey is voluntary. You may withdraw at any time without penalty.  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 117 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, Amanda Allard (her contact information is below).  Principal Investigator: Dr. Monique Turner mmturner@msu.edu Communication Arts & Sciences Building East Lansing, MI 48824 Lead Researcher: Amanda Allard allardam@msu.edu 713-478-7329 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 with the web survey indicates that you give your consent to participate. [Participants were randomly assigned to either the friend or an acquaintance condition. In these conditions they either read one of the following combos: positive gossip about work ethic, positive gossip about personality, negative gossip about work ethic, or negative gossip about personality.] Relationship Definitions: For the remainder of the survey you will be asked to read and assess scenarios that occurred between you and a same-sex friend/acquaintance. Friend: Although friends may be family, we would like for you to think of a friend that is not blood-related. A friend is considered someone who you often seek out their company for intimacy or assistance, and is considered as a voluntary, non-biological personal relationship. 118 Acquaintance: An acquaintance is considered someone who you are aware of yet have little to no intimacy with or knowledge about, and is someone you typically do not seek out their company. By clicking next, you are acknowledging how a(n) friend/acquaintance is being defined in this survey. Scenarios used in Main Survey [Please imagine a same-sex friend/acquaintance says the following to you] Personality Positive Gossip • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how much they enjoy hanging out with you. In fact, they said, “I think they are the funniest and most positive person I have ever met”. I normally wouldn’t tell you this information, but I just thought you would want to know. (60 words) Work • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how much they admire how hard you work. In fact, they said, “I admire how much time and effort they put into all of their work.” I normally wouldn’t tell you this information, but I just thought you would want to know. (60 words) Personality Negative Gossip • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how much they dislike hanging out with you. In fact, they said, “I think they are one of the most annoying people I have ever met”. I normally wouldn’t tell you this information, but I just thought you would want to know. (60 words) Work • Hey, the other day I was talking with a mutual (acquaintance/friend) of ours and they mentioned to me how much they dislike your work ethic. In fact, they said, “I think they are one of the laziest and uncooperative individuals they have ever met.” I normally wouldn’t tell you this information, but I just thought you would want to know. (60 words) [All participants were asked to rate these scenarios in terms of the instruments listed below.] Scale for Expectancy Violations *Reminder of scenario* Directions: Now, imagining this conversation just happened to you, please rate the extent to which you disagree or agree with the following statements in regard to the scenario you just read. 119 Stem: The statements made about me.... 1. Exceeded my personal expectations. 2. Are better than I would anticipate. 3. Surprised me in a good way. 4. Are nicer than I would expect. 5. Are worse than I would expect. * 6. Violated my personal expectations. * 7. Shocked me in a bad way. * 8. Are below my hopes of what others think of me. * Hurt Measurement *Reminder of scenario* Directions: Now, imagining this conversation just happened to you, please rate the extent to which you disagree or agree with the following statements in regard to the scenario you just read. Stem: The statements made about me.... 1. Hurt 2. Disappointed 3. Wounded 4. Betrayed 5. Unvalued 6. Less than 7. Sad 8. Unimportant Pride Measurement Directions: *Reminder of scenario* Directions: Now, imagining this conversation just happened to you, please rate the extent to which you disagree or agree with the following statements in regard to the scenario you just read. Stem: The statements made about me.... 1. Feel good about myself. 2. Feel worthwhile, valuable. 3. Feel capable, useful. 120 4. Feel proud. 5. Feel pleased with myself. Shame Measurement Directions: *Reminder of scenario* Directions: Now, imagining this conversation just happened to you, please rate the extent to which you disagree or agree with the following statements in regard to the scenario you just read. Stem: The statements made about me.... 1. Like sinking into the floor. 2. Small. 3. Like a bad person. 4. Humiliated, disgraced. 5. Worthless, powerless. State Self-Esteem Measurement Directions: *Reminder of scenario* Directions: Now, imagining this conversation just happened to you, please rate the extent to which you disagree or agree with the following statements in regard to the scenario you just read. Stem: The statements made about me.... 1. Be worried about whether I was regarded as a success or failure. (R) 2. Feel self-conscious. ( R) 3. Feel displeased with myself. ( R) 4. Be worried about what other people think of me. ( R) 5. Feel inferior to others at this moment. ( R) 6. Feel concerned about the impression I made. ( R) 7. Be worried about looking foolish. ( R) Affiliation Intention Scale: *Reminder of scenario* Directions: Now, imagining this conversation just happened to you, please rate the extent to which you disagree or agree with the following statements in regard to the scenario you just read. 121 Stem: The statements made about me.... 1. Hang out with them 2. Continue a relationship 3. Spend time with them 4. Take a break from the relationship * 5. Interact with them 6. Avoid being around them* Manipulation Checks 1. How would you describe the statements made about you in the scenario you previously read? Choice 1: Positive Choice 2: Negative Choice 3: Not Sure 2. Who made the evaluation regarding you in the scenario you used for this survey? Choice 1: Friend Choice 2: an acquaintance Choice 3: Neither friend nor an acquaintance Choice 4: I don’t remember Demographic Questions Age What is your age? Please type your age as a numerical value in the box provided (e.g., 18, 26, 45). ____________ Biological Sex What is your biological sex? 1. Male 2. Female Gender 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 122 Ethnicity Which of the following choices best describes your ethnicity? 1. Hispanic or Latino 2. Not Hispanic or Latino Race Which of the following choices best describes your ethnicity? 1. White 2. Black or African American 3. American Indian or Alaska Native 4. Asian 5. Native Hawaiian or Pacific Islander 6. Multiple or mixed races 7. Other/prefer not to answer Location Are you from the US? 1. Yes 2. No [ If yes, participants were directed to a question where they chose what state they are from. If no, participants were directed to a page where they indicated what country they were from] Thank You Page Thank you for participating in the present study! Your participation was greatly appreciated and will be recorded and used to enhance empirical scholarly research in interpersonal communication. 123