2&5”. . ”mg. .21.». w: fife five}! (.3. l. - a nhfifz: .. ... . . 3.1...)- I ~ : It: It «”5 ‘0'!!!" .- 3:52.11. 35%.... 1:... a . u}. . 5......» . . it}?! . 5H... . «.55.. .9439 2 A. Haggis . a: . vy) ‘1‘ . huiiuuuub... «£33.... 33... 1.3.5.... \ . .3. gig...» i. £2313... HIE. 5:35 .2. “05...“? 2:2! ... 2.1. In. E. v i... 54.... . .. 1:193:13 . 5......132 . uglézunnfiud 1...)!!! 2.17.15 5.1 1.1.}; it? I 5:03.1- afim.w..a§§.ix..ai:n . L . 11417879 iobl’ This is to certify that the thesis entitled Can't Live with Them or Can't Live without Them?: Effects of Betrayal on Relational Outcomes presented by Merissa Ferrara has been accepted towards fulfillment of the requirements for MoA. degree incommunication Major professor _/ / Date “fr/gs/Oa/ 0-7 639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State University PLACE IN RETURN Box to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE ' DATE DUE OCT 2 5 2005 “crfi'z‘s‘zbgyi 051506 6/01 cJCIRC/DateDuepGS-sz CAN’T LIVE WITH THEM OR CAN’T LIVE WITHOUT THEM: EFFECTS OF BETRAYAL ON RELATIONAL OUTCOMES By Merissa Hart Ferrara A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Communication 2002 ABSTRACT CAN’T LIVE WITH THEM OR CAN’T LIVE WITHOUT THEM?: EFFECTS OF BETRAYAL ON RELATIONAL OUTCOMES By Merissa Hart Ferrara The focus of this investigation is on explaining and predicting relational satisfaction and stability after incidents of betrayal. Theoretical predictions from Social Exchange Theory, specifically Rusbult’s Investment Model, were tested along with the effects of various communication strategies. Participants (N=155) were asked to recall a time in which they were betrayed and fill out a questionnaire. The results indicated that both the communication strategy ratings and the variables from the investment model were significant predictors of relational stability. The data, however, were inconsistent with the specific functional relationships predicted by the investment model. Instead, relational satisfaction was the best predictor of relational stability. For those who said it could be done and done well, thank you for your unconditional support. iii ACKNOWLEDGEMENTS To my advisor, Dr. Tim Levine. Thank you for last-minute meetings, late night phone calls, and long talks. You would have gone to great lengths to help, and did. You are an excellent advisor, researcher, and friend. I have learned so much from you about the field of communication, research, Michigan State, and myself. To my committee members, Dr. Brad Greenberg and Dr. Steve Mc Cornack. Thank you for your patience, encouragement, and brilliant feedback. It was wonderful being surrounded by great minds. I know that my thesis and viewpoint of the field is better because of what you taught. To my family. Throughout this experience I realized how important it is to have such an excellent support system. You have been incredible. With you all behind me I know that there is nothing that I cannot accomplish. Thank you. To the communication staff, especially Marge and Deb. I am in debt to you all for your help, patient ear, humor, and sound advice. Thank you for helping me grow. To Tisha. Thank you for being my thesis pal. It definitely made a difference to have someone to go through this with. Without your kindness I would have spent many more nights on the couch in the mailroom. TABLE OF CONTENTS INTRODUCTION BETRAYAL LITERATURE Betrayers Types of Betrayal Betrayal Motives Discovery and Emotional Reactions Relational Outcomes Forgiveness INVESTMENT MODEL Theory components Theory implications METHOD Participants Procedures MEASURES Commitment Level Investment Level Quality of Alternatives Comparison Level Apology Strategy Responsibility Acceptance Strategy Justification Strategy Promise Change Strategy Relational Importance Strategy Satisfaction Level Stability RESULTS Communication Strategies Investment Model Supplemental Analysis DISCUSSION APPENDICES BIBLIOGRAPHY 8-12 13-18 14-15 15-18 19-20 19 19-20 21-24 21 21 21 22 22 22 22 23 23 23 24 25-31 25-26 27-29 30-3 1 32-40 41-50 51-55 LIST OF TABLES Variable Means, Standard Deviations, and Zero-Order Correlations vi 26 LIST OF FIGURES Initial path model for Investment Model Path model results for Investment Model vii 17 29 Introduction The old phrase, “Can’t live with them. Can’t live without them!” has some merit. A growing abundance of literature suggests that while the intimacy found in close relationships is necessary to maintain individual health, to minimize the effects of stress, and to prevent some psychological disturbances (Duck, 1994), intimacy and companionship also include the potential for great personal risk and cost (Lawson, 1988; Nunnally, Chilman, & Cox, 1988). The people who are our important relational partners (e. g., friends, family, lovers) are often the cause of pain, stress, disappointment, and grief (Fischer, 1982; Metts, 1989; Miller, Mongeau, & Sleight, 1986). To better understand relationships, sources of relational discomfort should be critically analyzed. One such source, betrayal, or disloyalty to an intimate relationship, is an important type of relational transgression (Metts, 1989). Investigators have conceptualized betrayal, (Feldman, Cauffman, Jensen, & Amett, 2000; Jones & Burdette, 1994; Metts, 1994), betrayal types (Hansson, Jones, & Fletcher, 1990; Jones & Burdette, 1994; Jones, 1988; Roscoe, Cavanaugh, & Kennedy, 1988), betrayal motives (Fincham & Bradbury, 1989; Holmes & Rempel, 1989; Jones, Sanchez, & Merrell, 1989), discovery management (Feldman & Cauffman, 1998; Feldman et al., 2000; Metts, 1994; Scott & Lyman, 1968), relational outcomes (Metts, 1994; Jones & Burdette, 1994), and forgiveness (McCullough, Sandage, Brown, Rachal, Worthington, & Hight, 1998). In recent research on betrayal, two noteworthy gaps remain in this literature. First, Shackelford and Buss (1996) argue that betrayal research has lacked theoretical direction. Theories add structure to research, allow one to make predictions about human behavior, aid in proposing logical hypotheses, and help explain research results. It is through theoretical frameworks that the second gap in the literature can be addressed. Currently, the relational outcomes associated with betrayal are described in the literature, but little is known about how to predict whether or not the betrayal will result in termination of the relationship. Theoretical research is necessary to aid in making logical predictions about which variables affect relational outcomes and to analyze how betrayed individuals assess acts of betrayal and then decide upon the fate of their relationships. Metts (1994) affirms the importance of the second gap when she states, “A transgression (e.g., betrayal) need not destroy an otherwise viable and satisfying relationship. The fact that it sometimes does is perhaps the clearest signal that additional research into the coping process is necessary” (p. 238). The purpose of this investigation is to explain and predict relational satisfaction and stability after incidents of betrayal through the investment model and various communication strategies. First, the investment model emphasizes the nature of interdependence and reward/cost ratios in ongoing relationships, such that rewards of the relationship are greater than the costs for the people in the relationship. Popular with social exchange researchers, it has been used to examine romantic relationships, emotion- arousing events, and breakup decisions (Drigotas & Rusbult, 1992; Rusbult, 1980, 1983; Rusbult & Martz, 1995). Second, this thesis explores the nature of forgiveness and what communication strategies are effectively used by the betrayer to prevent relational termination. Thus, the intent of this thesis is to: (1) review the literature on betrayal; (2) examine the investment model and its application to betrayal; (3) explore communication strategies and their effect on relational outcomes, (4) test theoretical predictions concerning the relational outcomes of betrayal, and (5) provide conclusions and discussion of these predictions. Betrayal Literature Betrayal is defined as being unfaithful to relational expectations (Feldman et al., 2000). Individuals develop expectations of what they value in any relationship as well as what they expect from a specific relationship. Relational expectations may be mutually or individually defined. Thus, what one perceives as betrayal may not necessarily be thought of as disloyalty by another. However, when partners act in ways that conflict with such expectations, betrayal may be perceived. Metts (1994) views betrayal as “the unintended consequence of human fallibility and individual needs conflicting with relational needs” (p. 238-239). Individuals evaluate external (i.e., friends, environment) and internal (i.e., self-esteem, need for independence) factors, and weigh loyalties to both. Overwhelmed by multiple factors, a loyalty to one factor could create a temporary disloyalty to another. When unfaithfulness affects a personal relationship, the partner could feel betrayed. Even if betrayal is only a temporary lapse in an individual’s commitment to the other and to the relationship, such instances of betrayal affect the overall success of the relationship (Jones & Burdette, 1993; Rusbult, Drigotas, & Verette, 1994). Betrayers. Two extensive studies provide evidence that memorable and intense cases of betrayal often occur among members of close relationships. Jones and Burdette (1994) used a Social Network List instrument, and participants recalled instances of betrayal and how close they were to the person(s) involved. Nearly half of the participants (45%) admitted they were disloyal to at least one member of their current social network. In addition, more than half of the participants (52%) felt betrayed by their more intimate relationships. Romantic partners or spouses were most frequently reported as sources or victims of betrayal. In another study of close relationships, Feldman and Cauffmann (1998) questioned the prevalence of sexual betrayal between adolescent dating partners and found that almost two-thirds of the adolescents surveyed had either been the perpetrator, the aggrieved, or both. Studies have also identified gender differences in the self—reports (Feldman & Cauffmann, 1998; Jones & Burdette, 1994; Lawson, 1988). Interestingly, women reported roughly equal amounts of disloyalty by and toward romantic partners, while men were four times as likely to report they had betrayed a romantic partner than their partner had betrayed them. Women reported betrayal most often from romantic relationships with same-sex friends second. Men, however, reported the most betrayal from co- workers. Types of Betrayal. Individuals are fairly consistent in what they consider to be incidents of betrayal in relationships. Roscoe et al. (1988) asked participants to describe behaviors that comprised unfaithfulness to romantic relationships. The five behavior categories identified, ranked by prevalence, included: (1) dating/spending time with another; (2) having intercourse with another; (3) keeping secrets from partner; (4) betrayal of confidence; and, (5) emotional involvement with another. Jones and Burdette (1994) expanded on this concept to include all relational types. They concluded that the following ten behaviors were reported as betrayal: extramarital affair (or cheating), lies, betrayed confidence, two-timing, jilting (rejection), lack of support, ignoring, excessive criticism, gossip, and miscellaneous. Betrayal Motives. Similar to Weiner (1986) the research on betrayal acts was detailed by Jones, Sanchez, and Merrell (1989) who placed motives for betrayal along three attributional dimensions: internal versus external locus (eg., personality vs. peer pressure); stable versus unstable causes (eg., trait vs. state); and, intentional versus unintentional motives. When reporting disloyalty to others, two-thirds of participants saw their actions as internal, unstable, and intentional, citing temporary anger as a motive. The remaining one third reported their behaviors as external, unstable, and intentional. Betrayers often claim temporary emotions (e. g., anger and jealousy) or external causes motivated them to hurt another. Yet, when evaluating the individuals who had betrayed them, they described motives in terms of being internal, stable, and intentional, offering dispositional weaknesses as the reason why betrayal occurs. Those who are hurt believe that others do so intentionally, for personal and persistent reasons. Thus, there is a “tendency for respondents to explain away their own betrayals, but at the same time hold their partners morally responsible for their betrayals” (Jones & Burdette, 1994, p. 258). Offended partners connect betrayal motives to relationship satisfaction, level of trust, knowledge of partner, knowledge of present situation, and knowledge of past situations (Fincham & Bradbury, 1989; Holmes & Rempel, 1989). Roscoe et a1. (1988) asked participants why they thought that their partners had been unfaithful. Dissatisfaction with the relationship, boredom, revenge/anger, and being insecure or unsure of the relationship were the main motives listed were ranked from most to least important factors. Discovery mid Emotional Reactions. Acts of betrayal are discovered less than half of the time (Feldman & Cauffman, 1998). When they are, it is usually due to a confession by a disloyal member (Feldman & Cauffman, 1998). Regardless of how betrayal is revealed, it should elicit predominately negative emotions from the victim (Metts, 1994; Planalp & Honeycutt, 1985). According to parallel literature on detected deception reported by McComack and Levine (1990), discovered lies can produce intense emotions, an erosion of trust, and relational dissolution. The degree of intensity and the negativity felt are positively correlated (Jones & Burdette, 1993; McComack & Levine, 1990). However, the emotional intensity can be reduced through the efforts of the betrayer to repair the relationship (Jones & Burdette, 1993; McCullough et al., 1998). Relational Outcomes. Some studies have examined the nature of the relationship after betrayal. Jones and Burdette (1994) found that 50% of men and 40% of women who had been disloyal reported that the relationship became worse or terrrrinated entirely. Roughly 40% of both genders reported initial harm but a subsequent return to the pre- betrayal state; the remainder reported that the relationship had improved because of the betrayal. However, when participants who were betrayed were asked to assess relationships outcomes, the percentages were drastically altered; 90% worsened, while 5% remained the same, and 2% improved. Similar patterns were found across several other studies (Hansson, Jones, & Fletcher, 1990; Jones, 1988; Jones, Cohn, & Miller, 1991 ). McComack and Levine (1990) investigated the determinants of relationship termination after deception is detected. The more serious the lie, the more likely the relationship ended. “More than two-thirds of the subjects who reported that their relationships had terminated since the time that the lie was discovered reported that the discovery of the lie played a direct role in their decision to end the relationship” (p. 131). Most of the participants who said that the relationship ended stated that they initiated the termination. Forgiveness. The research on hurtful behavior and forgiveness (Nelson, 1993; Rackley, 1993; Roloff & Janiszewski, 1989; Woodman, 1991) reflects the more positive relational outcomes. Forgiving is the forgoing of vengeful behavior (Heider, 1958). McCullough, Worthington, & Rachal, (1997) further explained forgiving as residing at the level of people’s motivations toward an offending relational partner. People forgive when they experience low motivation to avoid and low motivation to seek revenge (McCullough et a1 1998). The literature reveals that partners are “more willing to forgive one another for interpersonal offenses in relationships that are characterized by high satisfaction, closeness, and commitment” (McCullough et al., 1998, p. 1588). Roloff and Janiszewski (1989) observed that the only offenses more likely to be forgiven in less intimate relationships than in highly committed relationships were refusals to perform low-cost favors (e.g., picking a person up from work or cleaning the house). Similarly, McCullough et al. (1998) found a high correlation between forgiving and the relational dyadic satisfaction-commitment, concluding that forgiving is a “motivational transformation occurs more readily in satisfactory, committed relationships” (p. 1593). Rusbult et al. (1991) suggest four reasons why forgiveness is more likely to occur in close relationships. To begin with, intimate couples have a great deal invested in the relationship and gain many resources from the relationship, motivating them to keep the relationship intact. In addition, usually the closer two people get, the more their interests in self and other merge. Third, persons high in comrrritment may look at long-terrn orientation and are thus able to overlook offenses for the good of the relationship. Finally, since high-quality relationships are marked by collectivist ideals, those individuals are willing to bear some cost to preserve the unity. There are three additional reasons for the link between relational quality and forgiveness. First, persons in high-quality relationships are able to empathize with the other person since they have prior knowledge about the person’s weaknesses, thoughts, and values (Baston & Shaw, 1991). Second, Heider (1957) noted that persons in lower quality relationships may see a transgression as inappropriate or mean, while those hurt in high-quality relationships are more likely to report that the hurtful event was “for one’s own good.” Finally, offenders in high-quality relationships are more prone to seeking forgiveness because they have a vested interest in the relationship. This increases empathy in the victim, which in turn improves the chances of forgiveness (McCullough et al., 1998). Similarly, the resultant behavior is dependent upon whether or not the offender offers an apology and pursues forgiveness (McCullough, Worthington, & Rachal, 1997). Clearly, the probability of forgiveness rests to a great extent upon the nature of the offense. The perceived seriousness of the offense and its immediate implications to the relationship influence forgiveness (Girad & Mullet, 1997; McCullough et al., 1998; Ohbuchi, Kameda, & Agarie, 1989). As one might expect, the greater the perceived severity, the more difficult it is to forgive. Feldman et al. (2000) approach managing discovery from the view of the betrayed. They found that, despite overall disapproval of betrayal, the extent of disapproval was a function of three factors: whether the relational partner was romantic or a same-sex friend; the justification for the betrayal; and, the characteristics of the respondent. “There was greater acceptance of sexual betrayal than of a friend’s confidence, of male than female transgression, and by male than female respondents” (Feldman et al., 2000, p. 500). Acceptance of a romantic partner’s betrayal was positively related toward tolerance of deviation and behavioral betrayal, and negatively associated with self-restraint. However, these results were due largely to the fact that men were highly accepting of other male sexual transgression. With respect to accepting female sexual transgression, neither gender was very accepting. Acceptance of a friend’s betrayal was associated with the following personal characteristics: tolerance of deviation; lack of self-restraint; and behavioral betrayal. The current research posits that communication strategies are also a factor of forgiveness. What the betrayer says when addressing the victim after the event may affect relational outcomes. There are several methods of repair. Account methods include justifications and excuses, which allow the disloyal individual to reframe the event and then lower personal responsibility for the event (Scott & Lyman, 1968). Apologies accept responsibility and admit to the truths of the event. A commitment to change statement like, “This will never happen again.” or a reaffirmation of the value of the relationship may also affect whether or not the couple terminates the relationship. In an effort to explore the effectiveness of communication in maintaining a relationship after betrayal, a research question is proposed: ROI: What is the relationship between communication strategies (egg, justification, apology, promise to change, accepted responsibility, and relational importance.) and relational stability after betrayal occurred? Given the aforementioned observations on betrayal, the negative emotions it arouses, the related motives generated by betrayers, and the reasons for forgiveness, it is not surprising that decisions about staying or leaving a relationship can be difficult and involved. People at every level of relational closeness reported intrusive thoughts and images of the offender and their hurtful acts, particularly those in closer relationships (McCullough et al., 1998). In fact, there is a direct relationship between closeness and the negative emotions experienced (McComack & Levine, 1990). Interestingly, not only are individuals more likely to feel hurt by those closest to them, but they are also more likely to forgive them as well. Research shows that people in close relationships can overcome hurtful acts and continue to grow closer. “Pre- offense closeness appears to be associated with current closeness because relational closeness facilitates prosocial processes in the offender (apology) and the offended partner (empathy) that foster the reestablishment of closeness” (McCullough et al., 1998). Thus, a paradox is apparent. People are most likely to forgive the very people who hurt them most. Given that people should reject things that cause the most pain, an interesting question is posed, “Why do people stay if they were hurt so badly?” Perhaps betrayed individuals prefer not to think whether or not they can remain in a relationship with a person who has been disloyal, but they rather attempt to answer the question, “Can I live without them?” The need for theory application is clear. Theory can provide structure and offer predictions based upon the current literature. The ensuing section will explore the Investment model in an effort to provide reasoning as to why this paradox occurs, and will suggest several hypotheses on decision-making and relational outcomes. l2 Investment Model Social exchange theorists maintain that people are motivated to maximize rewards and to minimize costs. When determining whether to remain in a relationship, people analyze the ratio of rewards to costs. If the rewards exceed the costs, they are likely to continue the relationship. Kelley and Thibaut (1959) propose a specific exchange theory based on interdependence. People are interdependent to the extent that they control the amount of rewards and costs of another, who, in turn, control the amount of rewards and cost for the other. Influenced heavily by social exchange and interdependence theory, Rusbult (1979) created the investment model of commitment. The model suggests that in order to comprehend why individuals stay in ongoing interdependent relationships, it is important to make three distinctions. First, the investment model holds that relationship stability is most directly determined by feelings of commitment (Rusbult & Martz, 1995). The second distinction is that the investment model specifies satisfaction as one indirect determinant of relational stability, mediated through commitment. Extensive literature on the interdependence theory has revealed that the amount of satisfaction in a relationship is the main determinant of whether one remains in the relationship (Kelly & Thibaut, 1978; Rusbult, 1983; Simpson, 1987). Staying in a gratifying and predominately satisfying relationship is simple and rational. Research on married couples supports this notion. Happy couples have been shown to act in more positive and rewarding ways toward each other than do unhappy couples (Billings, 1979; Gottman, 1979). Investment is the third distinction. Investment researchers claim that a primary reason why people remain during unhappy times is because they have much invested in the relationship. Commitment is affected not just by the outcome values (satisfaction evaluated by CL) of the current relationship and alternative, but also by investment size. Commitment increases over time in part because the resources put into the relationship increase the cost of withdrawing from it” (Rusbult, 1979, p. 174). Theory components. Commitment is a mediating construct in the model. Commitment is defined as the degree to which individuals intend to maintain relationships, to feel psychologically attached, and to sustain long-term orientation. Commitment is inversely related to the probability that people will leave the relationship. Satisfaction level, investment size, and quality of alternatives are three elements that affect the commitment level of a relationship (Drigotas & Rusbult, 1992; Floyd & Wasner, 1994; Rusbult, 1980, 1983; Rusbult & Martz, 1995). _S_atisfaction level. Satisfaction pertains to the amount of positive affect associated with the relationship. Since it is simple to remain in a satisfying relationship, commitment should be positively correlated with satisfaction. The level of satisfaction is determined by the rewards perceived- by the goodness of outcomes experienced and by the generalized expectations regarding the quality of the relationship (CL). Social comparison and/or prior experiences dictate what individuals idealize as a rewarding relationship. If the outcomes of their current relationship are greater than what they expect out of a relationship then they are satisfied. Thus, the more satisfied individuals are in their relationships, the more committed they will be. In addition, this model proposes that people can sometimes feel strongly committed to unsatisfying relationships, because commitment also rests upon alternatives and investments. 14 Alternative quality. The second component of the model, comparison level for alternatives (CLalt), is defined as perceived outcomes people believe they would receive if not in this relationship. CLalt is one of three predictors of commitment in this model. CLalt has an inverse direct relationship with commitment which in turn affects relational stability. When attractive alternatives are available, lower commitment levels are predominantly reported, particularly after a negative event like betrayal. On the other hand, when individuals fear that they would have undesirable or limited options should the relationship end, they are more likely to feel strong commitment to the relationship (Rusbult, 1983; Rusbult & Martz, 1995). Investment size. The final variable refers to the amount and magnitude of resources put into the relationship. Direct investments include time, disclosure, and money. Other investments, such as mutual friends and shared possessions, evolve as extraneous resources become associated with the relationship. Investments enhance the commitment to the relationship because these investments are lost if the relationship is terminated. Commitment tends to be higher with those who have been in the relationship longer, who report higher levels of disclosure, and who have more joint possessions. The literature on forgiveness is consistent with the importance of considering investment size, because people who are in relationships longer tend to see a long-terrn future, and are more likely to forgive the offender (McCoullough, 1998). Theory implications. From a social exchange perspective, an act of betrayal should be a cost. The magnitude of that cost should depend on perceptions of the severity of the betrayal. Betrayed individuals should reassess the outcomes of the relationship in relation to comparison level (CL), comparison level of alternatives (CLalt), and 15 investment level. Betrayal severity should be negatively related to outcomes, such that the more hurt felt by the victim, the lower the outcomes. Satisfaction is determined by both 0 and CL. Thus, the comparison level should interact with betrayal severity to affect satisfaction (Figure 1). l6 :0 E2582: ecu .moEaEB—m .«o .26. 53858 .cozoémsam no 353:: ”Eco—2 32582: .8 .32: 53 325 ._ 053m .EoEzfiEoo .3 3362: :23, .3593 35:22 2.2582: . 0N: .9535 22:35:50 :30 um: 9N: «NI AU X bto>om cezoflmzam _ I Ransom 17 Research on relationship violence is consistent with the model’s structure. Following an episode of violence, only about half of dating relationships are terminated and more than half of these victims plan on marrying their partners (Flynn, 1990). As commitment increases, so does the cost associated with leaving (Rusbult, 1980). It has been found that a woman’s decision to stay in an abusive relationship was based on perceived rewards and costs, perceived relational alternatives, and investment size (Frisch & MacKenzie, 1991). Other data suggest that commitment level is strongly correlated with whether or not women returned to their abuser after being in a shelter (Rusbult & Martz, 1995). Once betrayal is uncovered, commitment to the relationship is evaluated by reconsidering satisfaction, alternatives, and investment. In accordance with the model (Figure 1), several hypotheses can be proposed. The investment model predicts the following: Hypothesis 1: betrayal severity interacts with comparison level to have a negative multiplicative affect on the level of satisfaction. Hypothesis 2 predicts that satisfaction (2a) and investment size (2c) have positive direct effects on commitment, and comparison level for alternatives (2b) is inversely related to commitment. Hypothesis 2d proposes that the effects of satisfaction, investment size, and comparison level for alternatives on relationship stability are mediated through commitment, which has a direct effect on stability. 18 Method Participants. Participants were 165 students enrolled in undergraduate communication courses at a large midwestern university. Subjects participated on a voluntary basis, signing up for the study in advance. Respondents fulfilled mandatory research points or extra credit for their efforts. When signing up, participants were told that their responses would be confidential and would be analyzed only after participants sign the consent form (See Appendix A). Of the total number of participants, 93% (155) could recall and report extensively on romantic relationships involving betrayal. Ten were discarded. Eight were discarded because they did not complete the entire questionnaire and 2 because they reported on relationships that were not targeted. Those remaining ranged in age from 18 to 31 years (M = 21, S_D_ = 1.7) and 65% were female. Relationships labeled exclusively dating accounted for the highest proportion of relationships (73.7%), followed by casual dating (23.1%), married (1.3%), and other (1.9). Most of the relationships were formed well before the betrayal occurred (M = 22.8, S_D = 20.5), and had existed romantically for some time before the event as well (M = 13.5, §I_) = 12.5). On average, the participants reported a betrayal event that occurred 22.8 months earlier @= 19.9). Of the participants, 62% stated that this was their first time experiencing betrayal, 14% stated that this was not the first time they had been betrayal by this specific partner, and 21% admitted that they had betrayed this person before they were the betrayed. Procedures. Participants were asked to recall a recent situation in which a relational partner had betrayed them. Betrayal was defined to participants as when someone is disloyal to you and violated your trust. Participants completed an extensive questionnaire (See Appendix B). Following the completion of the questionnaire participants were thanked for their involvement. Then they were reassured that their answers would remain confidential. Participants who could not recall an instance of betrayal were offered an alternate task for equal credit. In order for participant responses to be included in the results, three criteria must have been met. To begin with, participants must have been able to think of an episode in which betrayal occurred. Next, the instance of betrayal must have occurred in a romantic relationship. Finally, the participants must have been the person betrayed or the victim. Each participant received a packet of questionnaires containing: Commitment Level Scale, Investment, Comparison level for alternatives, Comparison level, information importance (perceived betrayal severity), communication strategies, Satisfaction, and Stability (See Appendix B). They completed Commitment, investment, CLalt, CL, and information importance assessing the relationship prior to the betrayal, and then completed the communication strategies, Satisfaction, and Stability scales evaluating their relationship after the betrayal occurred. 20 Measures Commitment Level. Commitment level refers to the intentions to remain in the relationship after the betrayal incident. A 5 item, 7- point Likert scale (1: strongly agree, 7: strongly disagree) modified from Parks and Floyd's (1996) level of development in on- line relationships scales was used. For the current study acceptable internal consistency was obtained for the measure of commitment (g = .86). Investment Level. The measurement of investment size was based on previous research (Rusbult, 1983; Rusbult & Martz, 1995; Truman-Schram et al., 2000). In previous studies, Investment size was measured with 9-point rating scales with reliabilities ranging from g = .41-.94. Modifications for this study were made. Nine items using 7-point Likert scales (1: strongly disagree, 7: strongly agree) were used in this study. The questions evaluated objective investments (e.g., hours per week spent together, activities unique to the relationship) and subjective investments (e.g., rating of shared memories, how much would be/was lost if the relationship terminated). For the current study acceptable internal consistency was obtained for the measure of investment (g = .83). Quality of Alternatives. The quality of relationship alternatives measure was developed for the present investigation. This scale consisted of 5 Likert-type items, using a 7-point response format (1: strongly disagree, 7: strongly agree). One sample item used was, “I could have done better than this relationship.” For this study minimally acceptable internal consistency was obtained for the measure of comparison level for alternatives (g = .62). 2] Comparison Level. The measurement of comparison level was developed for the purpose of this investigation. Five general questions assessed comparison level on a 7- point Likert scale (1: strongly disagree, 7: strongly agree). One sample item for this scale was, “ Generally, I have high standards for my romantic relationships.” For this study minimally acceptable internal consistency was obtained for the measure of comparison level (g = .68). Apology Strategy. The measurement of the communication strategy apology was created for the purpose of this investigation. Apology was tested through a two-item 7 point Likert scale (1: strongly disagree, 7 = strongly agree). “My partner said they were sorry that the betrayal happened” was one item used to measure apology. For this study acceptable internal consistency was obtained for this measure (g = .93). Responsibility Acgtance Strategy. The measurement of the communication strategy response acceptance was created for the purpose of this investigation. It was tested through a two-item 7 point Likert scale (1: strongly disagree, 7 = strongly agree). “My partner took the blame for the betrayal” was one item used to measure responsibility acceptance. For this study acceptable internal consistency was obtained for this measure (g = .91). Justification Strategy. The measurement of the communication strategy justification was created for the purpose of this investigation. Justification was tested through a two-item 7 point Likert scale (1: strongly disagree, 7 = strongly agree). “My partner offered excuses for why they betrayed me” was one item used to measure justification. For this study acceptable internal consistency was obtained for this measure (g = .83). 22 Promise Change Strategy. The measurement of the communication strategy promise change was created for the purpose of this investigation. It was tested through a two-item 7 point Likert scale (1: strongly disagree, 7 = strongly agree). “My partner promised that they would make changes in their behavior” was one item used to measure promise change. For this study acceptable internal consistency was obtained for this measure (g = .90). Relationship Importance Strategy. The measurement of the communication strategy relationship importance was created for the purpose of this investigation. It was tested through a two-item 7 point Likert scale (1: strongly disagree, 7 = strongly agree). “My partner told me how important the relationship was to them” was one item used to measure relationship importance. For this study acceptable internal consistency was obtained for this measure (g = .92). Satisfaction Level. Satisfaction in the relationship was measured through combining two different scales. The first is a 5-item, 7-point Likert scale (1: strongly disagree, 7: strongly agree) created by Ohira (2001). The second part of the scale, following previous research (e.g., Drigotas & Rusbult, 1992; Rusbult, 1980), consists of three items that assessed global satisfaction with the current relationship (GL-CUR). The items were measured on a 7-point Likert scale (1: not satisfied, 7: very satisfied). For example, “I was happy with the state of our relationship,” “ Our relationship was fulfilling,” and “ I felt content in the relationship” were all items used to measure satisfaction. Drigotas and Rusbult (1992) found the items designed to measure GL-CUR were strongly related (g = .83). For the current study acceptable internal consistency was obtained for the combined measure of satisfaction (g = .95). 23 Stability. Relationship stability was measured through a dichotomous, one item- measure, “Are you still romantically involved with this person,” and an open-ended question describing the nature of the relationship after the betrayal occurred. In addition, participants completed open-ended questions describing the nature of the relationship prior to betrayal, the betrayal event, when the betrayal occurred, relationship length, and how the betrayal was discovered. Also, participants completed dichotomous, one-item measures to assess whether they have been betrayed before, if the same person before has betrayed them, and if they have betrayed this person in the past. 24 Results Communication Straflegies. In order to further explore the relationship between communication strategies and relational stability, research question (RQl) one addressed the relationship among the variables (apology, responsibility acceptance, justification, promise change, and value relationship) and relational stability. When considering the zero-order correlations (i.e., Pearson’s correlation coefficients) apology, r (153) =.17, p_=.04, promise to change, r (153) =.26, p=.001, and value the relationship, r (153) .---.22, 9:007, are all significantly correlated with relational stability (see Table 1). So, the more the betrayers apologize, promise to change, and state that they value the relationship, the better chance they have of keeping the relationship intact. Specific communication justifying the betrayal and accepting responsibility did not have significant effects on relational stability. Because the communication strategies were highly correlated,l it was possible that the zero-order correlations reported above rrright be spurious. Therefore, relational stability was regressed onto each strategy with logistic regression. Together as block the communication strategies predicted relational termination; °_’ (5) = 13.06, p_ = .03, 32 = .08. None of the unique effects for individual strategies, however, were statistically significant when controlling for the other strategies. 1 All of the communication strategies except excuses were highly correlated (i.e., ; < .50) with each other. Examination of the inter-item correlation matrix and reliabilities suggested that the strategies were separate but highly related constructs rather than a single broader construct. For example, combining the strategies resulted in a reduction in reliability. Therefore, an analytic approach of loading the strategies in as a block in logistic regression was adopted rather than just exarrrining zero-order correlations or treating each variable as an independent predictor in regression. 25 .26. .o. 2.. as EmoE:m.m m. 5.3.280 I .o>o. no. 2.. 8 2.85:»?- m. 5.8.280 ... m.. 0.. in. tau. :06 .8. L... 2. .c- .25.- ac. an: .m. b...0£m.a:o:~.om .N. - 1.9.. 5. mm. :mm. .1. :NN. .32.. 0m. .33.. mo. «N. .m.n EoE..EEoU .2 9.5.8033. - 0o. Inn. 31.0. m.. 3mm. :5. ..0.. war 8.. mod n5. E83. .0. - ~.. .0.- mo. 00. 5. «run: 1.3.- .1...- w... 0:” 308.38 .m 3.53.3.3. - in... 1.9.. 3&0. 2&0. m.. 0o.- ao. vmd 2.6 2.32.5 .w - 10m. :50. tan. 3.. *0..- S. .md 02.. omSEU 8.805 H - saw. :3. N.. .o. No. N..~ and $.33 .0 _- :3 .2. 2.- 8.- 8.. 3:. .322? .n - 1.0m. aim.- 5. mm. 8.0 .5539»:— .v - to. 8.- 2.. men begofieam .m - «0.. R. vwé . =30 .N - ma. 50.0 40.. .. o. a w \- 0 w v m m . Om .2 20028.50 EEO-EoN 0:35.532. 03055 .mcrmoS. 0.03:3 _ 29¢ 26 Investment Model. In accordance with investment model predictions, hypothesis one posits that comparison level (CL) and betrayal severity interact to have a multiplicative inverse affect on satisfaction level. The predicted interaction was tested with a product-term regression analysis. Following Aiken and West (1991), the main effects for comparison level and betrayal severity on satisfaction were entered on a first step, and the product of the predictors, which models the interaction, was entered on a second step. The data were not consistent with hypothesis one as the product term reflecting the predicted interaction was not statically significant, :_ = -.63, t (151) -.-— -1.05, p = .30. Instead, both main effects were significant; severity °_ = -.24, t_ (152) = -3.02, p = .003 and comparison level °_ = -. 19, t (152) = -2.47, p_ = .015. It was further hypothesized that high levels of satisfaction (2a), fewer perceived relationship alternatives (2b), and more investment (2c) would predict greater levels of commitment. The predicted effects for satisfaction, alternatives, and investment on commitment were tested with regression analysis. The effect of satisfaction on commitment was not significant, :_ = -.06, t (151) = - l .16, p_ = .24. The comparison level for alternatives on commitment was not significant, _°__ = -.07, t (151) = -1.22, p = .22. Investment level, however, was highly predictive of commitment, :_ = .77, t (151) = 14.44, p< .001. Thus, hypotheses 2a and 2b were not consistent with the data, whereas hypothesis 2c was found to be consistent with the data. Hypothesis 3 proposed that the effects of satisfaction, investment size, and comparison levels for alternatives on relationship stability are mediated through commitment. The effect of commitment on relational stability was calculated with regression analysis. Commitment was only marginally (only significant if a one-tailed 27 test is applied) predictive of relationship stability :_ = -.13, t (151) = - l .67, p_ = .10. Although commitment was marginally correlated with relational stability, neither satisfaction nor CLalt were correlated with the mediating variable, commitment mediation. Therefore, hypothesis 3 was not consistent with the data. Additionally, the causal chain predicted by Rubin (1983) was tested with path analysis on PACKAGE (Hunter et al., 1982). Path coefficients were determined using the Ordinary Least Squares (OLS) method and are reported in Figure 2. The path coefficients for betrayal severity to satisfaction (3_ = -.24, (_lf = 152, p< .05) and for comparison level [°__ (152) = -.19, p_ < .05] to satisfaction were statistically significant. Path coefficients for satisfaction on commitment (°_ = -.06, _d_f = 151, p: n_s) and comparison level for alternatives on commitment (-_ = -.07, g: = 151, p< n_s) were not statistically significant. The path coefficient for investment to commitment was statistically significant and substantial (-_ = .76, £11 = 151, p<.05). The path coefficient for commitment to relational stability was not statistically significant (g = -. 13, d_f = 153, p < .05). Since there were several path coefficients that were not statistically significant, it was clear that the model did not fit. Further, the chi-square goodness-of-fit test confirmed that the model was not consistent with fitting the data 12 (9) = 32.94. 28 5:25.558 .3 020.085 5:3 3.35m 0.5.3.2 5 52582.. 0.... £02852... .0 .68. 55558 50.65....3 .o 855...... .002). 52585. 5. 5.38. .600... :3. .m. 2230.0. .5585. 5.....5m 50505500 :30 1.0 5.8....28m b.88m .9355 29 Supplemental Analyses. Supplemental analyses were conducted to test the relative predictive utility of the communication variables verses the investment variables in predicting relational stability. For this purpose, two logistic regression analyses were conducted. In each, the communication strategies were entered into the equation in a block, while the investment variables were entered in a second block. In the first logistic regression, the communication variables were added first, and in the second, the investment model variables were added first. Interestingly, the results of the logistic regression were different depending on which block was entered first. When the communication variables were run as the first block, they were significant predictors of relational stability; -_2 (5) = 13.06, p_=.02. The investment variables contributed significantly to the prediction of stability beyond the communication strategies, -_2 (5) = 42.60, p: .001 ), and the combined model was also significant, [model °_2 (l 1): 55.67, p = .001]. However, when the investment model variables [L2 (5): 52.45, p: .001] were run as the first block and the communication strategies as the second block [°_2 (5) = 3.2, p: .67], only investment model variables were significant. This means that the communication variables are significant predictors of relational stability, but not when controlling for the investment model variables. The investment model variables, however, explain variance above and beyond the communication variables. However, of all the investment model variables, only satisfaction (13 (157) = .5, p< .001 ) predicted relational stability when controlling for other variables. 30 The logistic regression also provided the discriminant function for accurate prediction relational stability based on the communication strategies and the investment mode] variables. The discriminant function correctly classified 58.6% of the still togethers and 94.4% of the breakups. The overall percentage of correctly classified cases was 87.7%. While this number appears impressive, 81.3% accuracy could be obtained by simply predicting that the relationships would breakup in every case. Thus, although the investment model variables and communication strategies together produced a statistically significant discriminant function, the increase in predictive power was meager. 31 Discussion What is betrayal? When asking people to tell a story that involves betrayal, most people can offer several, listing the one that hurt the most first (Jones & Burdette, 1994). In fact, the literature review discussed eight varieties of what people label as betrayal types, from spending time with another, to secret telling, to excessive criticism. People can pinpoint the level of predominately negative emotions the betrayal event evoked and freely offer explanations for why the betrayer did it (Jones & Burdette, 1994). Yet, when asked again, what is betrayal, all researchers can offer are examples of it and attempts at vague descriptions. Feldman et al., (2000) defined betrayal as being unfaithful to relational expectations; Metts (1994) described betrayal as a type of unintended consequence resulting from conflicting relational and individual needs. These definitions lead this researcher to believe that betrayal is a lay term conveniently concocted to offer hurt parties a title to their stories of woe, a bandage for relational battle wounds. Perhaps this is why the concept of betrayal has received little theoretical attention. The literature also discussed the paradox that people are most hurt by those closest to them, and are also most likely to forgive these people. Thus, the phrase “Can’t live with them, Can’t live without them?” was introduced to posit what factors make individuals, hurt by betrayal, more likely to stay in their relationships. This study sought to test the variables in the investment model and explore the effect of communication variables on relational stability after a betrayal event, applying them to the area of romantic betrayal to see if they could predict stay/leave behavior. 32 One hundred and sixty-five undergraduates completed a questionnaire, reporting on a relationship involving betrayal. Measures assessing levels of commitment, investment, alternatives, comparison, apology, responsibility acceptance, justification, promise change, relational importance, satisfaction, and stability were used and answers were analyzed in order to evaluate the research question about communication strategies and the hypotheses based on the structure of the investment model. The current research reveals several interesting findings that merit future exploration. The first research question asked about the relationship between communication strategies (e. g. justification, apology, promise to change, accepted responsibility, and relational importance) and relational stability after betrayal occurred. Most of the participants reported listening to some sort of communication strategy, if not a combination of several (apology, M = 4.8, _SQ = 1.2; justification, M = 4.5, S2 = 2.12; promise to change, M = 4.2, S_D = 2.31; value relationship, M = 4.7, SE = 2.2). Those that stayed together were more likely to report having heard several from their partners. In fact, when evaluating each communication strategy on its own merit, each did not uniquely predict relational stability. However, when taken in groups, betrayers who offered an apology, promised to change their ways (or never do it again), and stated that they valued the relationship were more successful at preserving the relationship. Surprisingly, neither betrayers who attempted to justify the betrayal act nor those who accept responsibility improved their chances of saving the relationship. These results imply that efforts to communicate after the betrayal can aid in repairing the relationship. Hurt parties want the responsible individual to come to them for forgiveness (McCullough et al., 1998). Similarly, the resultant behavior is dependent 33 upon whether or not the offender offers an apology and pursues forgiveness (McCullough, Worthington, & Rachal, 1997). According to this research, they do not want to hear reasons for their behavior. Perhaps when betrayers clearly state that they are sorry, that they will not repeat this and that they realize how important the relationship is, empathy is evoked in the victim. Increases in victim empathy improve the chances of forgiveness (McCullough et al., 1998). Additional research is needed to further divulge successful relationships and their forgiveness tactics, as well as, exploring negative communication strategies such as blaming or telling someone to “get over it” that were suggested as occurring through the open-ended discussions. Interestingly, when controlling for the investment variables the communication variables were not significant predictors of relational stability. This means that the communication variables do serve as direct predictors of relational stability, but the investment model variables explain the variance above and beyond what the communication variables predict. As a package, the investment model variables are a more effective route at predicting relational termination. This does not imply, however, that the investment model functioned as specified. It was posited in the first hypothesis that the degree of betrayal severity would interact with comparison levels to have an inverse effect on the reported amount of satisfaction. Betrayal severity and comparison levels did inversely affect levels of satisfaction, but there was no multiplicative interaction effect (BS x CL). So, independently, the greater betrayal perceived by the victim, the more unhappy or unsatisfied the individuals reported to be. Previously it was determined that betrayal events aroused predominately negative emotions (Jones & Burdette, 1993), and now the 34 assumption that the severity of betrayal events causes increases in overall relational dissatisfaction is demonstrated. Likewise, the higher standards expressed in comparison levels, the more dissatisfied the individuals were with the relationship after betrayal was discovered. As expected, people have standards for what they want their relationships to be like, and betrayal is not something they value. It would be interesting to research how comparison levels adjust after betrayal occurs, if people increase or lower their expectations for future partners. The second hypothesis predicted that satisfaction level and investment size would have a direct effect on commitment, and that comparison level for alternatives would have an inverse affect on commitment. The data were partially consistent with these predictions. Levels of investment were highly related to levels of commitment. People were more strongly tied to their relationships to the extent that their actual investments (e. g., time, energy, money) were greater. However, the current research failed to find that satisfaction or comparison level for alternatives were related to reports of commitment. Thus, neither unhappiness nor viable alternative options lowered individual commitment to a relationship, but if a person has not put a lot of resources into the relationship then they will feel less committed. These results hold interesting theoretical implications. It is interesting to note that the two variables that were added to the original Thibaut and Kelley (1959) social exchange model, commitment and investment, were highly correlated with one another, while two original variables, satisfaction and alternatives had no bearing on commitment. Previous researchers who found that all of the variables affected commitment have 35 limited their research to abusive relationships, suggesting that this model may only work in this specific area of research (Rusbult, 1983; Rusbult & Martz, 1995 ; Truman-Schram et al., 2000). Such researchers may blame the results on the use of self-report measures and the desire for participants to look socially desirable after the fact, however, if this were true then the results would have reflected an inverse relationship between comparison level for alternatives and commitment. Future research should test the scope of this model. The third hypothesis predicted that effects of satisfaction, investment size, and comparison level for alternatives on relational stability are mediated through commitment, which has direct effects on stability. The model assumption that commitment is a predictor of relational stability was only marginally supported by the data. Rusbult et al., (1991) suggested that people in highly committed relationships would look at long-term orientation and overlook offenses like betrayal for the good of the relationship. This did not occur within the current sample. The majority of both highly and loosely committed individuals terminated the relationship after the betrayal occurred. Commitment as a mediator is controversial. There are mixed results in the literature regarding its success as a predictor of relational stability (Bui et al., 1996; Lund, 1985; Rusbult, 1983; Rusbult & Martz, 1995; Truman-Schram et al., 2000). In a longitudinal study of dating relationships, Bui et al. (1996) found that quality of alternatives had both mediated and direct effects on relational stability. Bui et al (1996) concluded that investment was correlated with commitment but does not affect relational stability. Truman-Schram et al. (2000) reported that investment size, perceived 36 alternatives, and positive feelings for partner significantly affected commitment, but commitment did not affect stay/leave behavior. On the other hand, some prior research has proven a strong link between commitment and stay-leave behavior (Lund, 1985; Rusbult, 1983; Rusbult & Martz, 1995). Rusbult and Martz (1995) found the mediation role of commitment on stay versus leave behavior highly significant when running causal model analysis on abusive marital relationship stability. Regardless of the successful studies reporting commitment’s direct relationship with stability, there are enough studies like the current one to imply that the investment model should be used with caution. There were additional results that merit discussion. The presence of the satisfaction-relational stability link is the most exciting conclusion. This research revealed that the most consistent predictor of relational termination was the evaluation of satisfaction. Investment model researchers (Rusbult, 1980, 1983; Rusbult & Martz, 1995) have downplayed the importance of this original social exchange variable, focusing instead on investment and commitment. Rusbult and Martz (1995) state “research on close relationships tend to place undue emphasis on issues concerning satisfaction; it is easy to remain in a relationship that is happy and satisfying (p. 569).” This research implies that, yes, it is easy to remain in a happy relationship; it is easy for that happy relationship to turn sour after betrayal occurs; and easy to leave it based on the simple explanation of dissatisfaction, without evaluating other variables like investment and commitment. The current study’s results mirrored some findings of Truman-Schram et al. (2000). They posited that this was because of testing in the context of abusive dating relationships. However, none of the participants for this study reported situations of 37 abuse and there was still a large correlation, suggesting that satisfaction may be a larger determinant of relational stability than the investment model implies. Of course this study and the Truman-Schram et al. (2000) study focused on dating relationships, suggesting that perhaps at the level of dating relationships, satisfaction is more important than alternatives, investment, and commitment. To test the accuracy of satisfaction as the main predictor of relational stability, future research should test the differences between dating and married couples. Another interesting result was the number of breakups. Since several researchers have found that at least half of their participants reported that their relationships worsened and terminated after betrayal events (Jones & Burdette, 1994; Jones, 1988; Jones, Cohn, & Miller, 1991), it was anticipated that the majority of relationships in the current study would terminate. However, the forgiveness literature offered hope that close relationships would reflect positive relational stability after betrayal. Yet, 117 participants labeled their relationships as exclusively dating or more serious (married), and of those 98 terminated the relationship. The high level of reported termination leads to interesting implications for survey conduction. McCullough et al. (1998) reported that people are most likely to forgive those closest to them, who are also those that hurt them the most. In Planalp and Honeycutt’s (1985) research on uncertainty-increasing events, only 27% of their participants reported relational termination. Yet, many studies on betrayal report low rates of forgiveness and relational stability. Perhaps the contradiction is all in the survey approach. If researchers ask people to recall a time in which they forgave someone or a 38 time they were uncertain as opposed to a time in which they were betrayed, results will always conflict. Of course this could also be a definitional issue, since they were asked to think of a time when a romantic partner was disloyal and violated their trust. N inety-two percent of the participants reported an episode involving cheating or seeing someone else. Study replication is necessary. Limitations. Several limitations of this study that qualify the current findings and hopefully will guide future research. First, having participants recall and evaluate a relationship after the betrayal event may affect report accuracy. A future recommendation would be to construct a longitudinal study, testing participants at various points before and after betrayal occurs. This would ensure that accounts of the variables are not distorted by retrospection. Second, 22.8 months (S: 19.9) between the betrayal event and survey completion was much higher than desired. Future investigators should be cautious using these results since the recall latency could have affected response accuracy. Also, not specifying recall time limits allowed participants to selectively choose which relationship they wanted to report on. Perhaps this led them to report on their relationships where relational termination occurred due to highly memorable and severe betrayals. That is, when asked to recall an incident, people recalled their most extreme betrayal. If this is the case, it is not surprising that so many broke up. McComack and Levine (1990) recommend a one month recall limit in order to make sure that the emotion is still fresh, yet enough time has passed to adequately evaluate the relational impact of the discovered betrayal. 39 Third, there were some measurement limitations. Five close-ended questions were not strong measures of what they were supposed to evaluate and had to be discarded because they lowered reliability. In an effort to increase scale validity, future researchers should replace these items with more precise measures. Also, although not discussed in the current research results, a test of uncertainty before the betrayal event and one after the event like Planalp and Honeycutt (1985) would have been beneficial to better assess uncertainty increases due to betrayal severity. One final limitation that merits discussion is the variable relational stability. Since the majority of the participants (126) reported that their relationships ended after the betrayal, there was little to compare and even less to discover about which variables are better predictors of relational stability. By limiting the amount that the variables vary, there is a restriction on the amount that they can covary. This probably resulted in the path coefficients being attenuated. Future investigators should aim for an equal number of cases where the relationship terminated and where the relationship remained intact. Conclusions. When asked “Can’t live with them or can’t live without them?” about relational stability after being betrayed by a romantic partner, most participants, regardless of the amount invested, commitment level, alternatives, said “I can live without them.” Since betrayal severity negatively affected relational satisfaction, and satisfaction was the strongest link to relational stability, it is concluded that in this sample, the increasing seriousness of the betrayal worsened relational satisfaction, and this lack of relational satisfaction led to breakups. The investment model was unsuccessful, leading this researcher to encourage others to test other models on larger samples, particularly the traditional Thibaut and Kelley (1959) social exchange theory. 40 APPENDIX A Consent Form You are invited to participate in a research project about the relational consequences following a betrayal incident in romantic couple relationships. You will be asked to remember and recall about a recent betrayal incident with your past or present romantic partner. Then you will be asked to complete a set of questionnaires including some information about your background. Full participation in this study will take about 30 to 45 minutes. Since we are asking sensitive information about your private experience, your participation is voluntary. You may refuse to answer certain questions, refuse to participate in certain procedures, or withdraw from the experiment without penalty. If you are not comfortable sharing your personal experience, or if you are a minor, do not sign this consent form and let the researcher know now. If you feel anxious or stressed after your participation in this project, you will be referred to the University Counseling Center. Your responses to the survey will be kept strictly confidential and will NOT be linked to YOU in any way. Only the researchers listed in this form will be allowed access to these questionnaires. In addition, both consent forms and completed questionnaires will be kept in a locked file cabinet. The file cabinet is located in a private office of the principal researcher. Your privacy will be protected to the maximum extent allowable by law. Your instructor will offer extra credit for participation in this research project, and you will receive credit based on the policies of that class. If you choose not to participate in this project, there is an alternative task to receive equal credit. Contact Dr. Tim Levine at (517) 432-1124 (Office: 482 COM), or Merissa Ferrara at (517) 337-6568 (Office 454) if you have any further questions or concerns regarding this study. The researchers can answer any questions you may have about the study to help you decide whether or not to participate. If you have any questions about your role and rights as a subject of research, feel free to contact Dr. David Wright, the Chairperson of the University Committee of Research Involving Human Subjects at (517) 355-2180. Thank you, Merissa Ferrara By signing below you are indicating your agreement to participate. Your Signature Today’s Date Print your name PID 4| APPENDIX B Communication Questionnaire Instructions: Please read these instructions carefully, and then complete the following questionnaire to the best of your ability, providing as much detail as possible. This survey focuses on betrayal in romantic relationships. Have you ever been betrayed by a romantic partner? If so, we need you to think about a particular instance where you felt betrayed. This situation must meet three criteria: 1. The relationship in which the betrayal occurred was a romantic relationship (and not a friendship, professional, or family relationship). 2. The situation must really involve betrayal. Betrayal is when your romantic partner was disloyal to you and violated your trust. 3. You were the person who was betrayed. Now take a few moments to think about a time that you were in this type of situation. If you cannot think of a specific situation in which you were betrayed, or if you do not want to participate in this particular study, raise your hand and you will be given another research survey for equal credit. Once you think of a specific time that you were betrayed, make sure it meets the 3 criteria above and answer the following questions about your relationship with the person who betrayed you. Put initials of the person who betrayed you here: Prior to the betrayal, how would you describe the relationship between you and the person who betrayed you? 1. We were (circle one): Casual Dating Exclusively Dating Engaged Married Other 2. Were you living together (circle one)? Yes No 3. How many months or years ago did this betrayal situation occur? Month(s) Year(s) 42 4. How long had you known this person before the betrayal occurred? Month(s) Year(s) 5. How long have you been/were you in a close relationship with this person before the betrayal? Month(s) Year(s) 6. Had this person betrayed you before (Circle one)? Yes No 7. Have you betrayed this person before (Circle one)? Yes No 8. Has another romantic partner besides this one betrayed you (Circle one)? Yes No Now, think about your relationship with the person who betrayed you. It is IMPORTANT to answer these questions based on how you felt about this person and the relationship BEFORE the betrayal occurred. SD: Strongly Disagree SA: Strongly Agree 9. I was very committed to maintaining the relationship. SD 1 2 3 4 5 6 7 SA 10. This relationship was not very important to me. SD 1 2 3 4 5 6 7 SA 1 1. This relationship was a big part of who I was. SD 1 2 3 4 5 6 7 SA 12. I made a great effort to maintain a relationship with this person. SD 1 2 3 4 5 6 7 SA 13. I did not expect the relationship to last very long. SD 1 2 3 4 5 6 7 SA 14. We had a lot of mutual friends. SD 1 2 3 4 5 6 7 SA 15. We shared a lot of memories or events together. 43 SD 1 2 3 4 5 6 7 SA 16. I invested a lot of time into the relationship. SD 1 2 3 4 5 6 7 SA 17. I put a lot of money into the relationship. SD 1 2 3 4 5 6 7 SA 18. We did several activities that were unique to the relationship. SD 1 2 3 4 5 6 7 SA 19. I invested a lot of emotion in the relationship. SD 1 2 3 4 5 6 7 SA 20. We bought tangible items together. SD 1 2 3 4 5 6 7 SA 21. A lot would be lost if the relationship ended. SD 1 2 3 4 5 6 7 SA 22. I met most of my partner’s close friends. SD 1 2 3 4 5 6 7 SA 23. I met most of my partner’s immediate family members. SD 1 2 3 4 5 6 7 SA 24. I communicated (i.e. talk, email, phone) with to my partner’s close friends regularly. SD 1 2 3 4 5 6 7 SA 25. I communicated (i.e. talk, email, phone)with my partner’s family members regularly. SD 1 2 3 4 5 6 7 SA 26. My partner’s friends expressed their support of our relationship. so i 1 2 3 4 5 6 7 SA 27. My partner’s family expressed their support of our relationship. SD 1 2 3 4 5 6 7 SA 28. My partner’s friends labeled us as “a couple.” SD 1 2 3 4 5 6 7 SA 29. My partner’s family labeled us as “a couple.” SD 1 2 3 4 5 6 7 SA 30. My partner’s friends included me in their activities. 44 SD 1 2 3 4 5 6 7 SA 31. My partner’s family included me in their activities. SD 1 2 3 4 5 6 7 SA 32. We talked face-to-face __ hours per week together. Little 1 2 3 4 5 6 7 Every moment 33. We communicated (i.e. email, phone, letters) hours per week. Little 1 2 3 4 5 6 7 Every moment 34. We spent hours per week together. Little 1 2 3 4 5 6 7 Every moment 35. My partner and I liked a lot of the same things. SD 1 2 3 4 5 6 7 SA 36. We share a lot of the same attitudes. SD 1 2 3 4 5 6 7 SA 37. The two of us have very different values. SD 1 2 3 4 5 6 7 SA 38. The two of us are very similar. SD 1 2 3 4 5 6 7 SA 39. The two of us have a similar outlook on life. SD 1 2 3 4 5 6 7 SA 40. I can accurately predict how my partner will respond in most situations. SD 1 2 3 4 5 6 7 SA 41. I can usually tell what this person is feeling inside. SD 1 2 3 4 5 6 7 SA 42. I can accurately predict what this person’s attitudes are. SD 1 2 3 4 5 6 7 SA 43. I do not know this person very well. SD 1 2 3 4 5 6 7 SA 44. I often have trouble understanding why my partner acts the way he/she does. 45 SD 1 2 3 4 5 6 7 SA 45. I could have done better than this relationship. SD 1 2 3 4 5 6 7 SA 46. I had realistic options in addition to this relationship. SD 1 2 3 4 5 6 7 SA 47. I did have good alternatives to this relationship- SD 1 2 3 4 5 6 7 SA 48. I felt stuck in this relationship. SD 1 2 3 4 5 6 7 SA 49. Things would have been worse if I was not in this relationship. SD 1 2 3 4 5 6 7 SA Now, please fill out the following survey questions concerning your general beliefs and attitudes about relationships. 50. Generally, I have high standards for my romantic relationships. SD 1 2 3 4 5 6 7 SA 51. My expectations for what should get out of a romantic relationship are minimal. SD 1 2 3 4 5 6 7 SA 52. I am not that picky about my romantic relationships. SD 1 2 3 4 5 6 7 SA 53. I do not settle for anything less than an ideal relationship. SD 1 2 3 4 5 6 7 SA 54. I am very “choosy” about who I develop a romantic relationship with. SD 1 2 3 4 5 6 7 SA 46 55. Now, recall the actual act of betrayal. Next, recall as much as you can about that incident. In as much detail as possible, describe the betrayal and the situation. Please print or write your answers neatly. 56. How did you find out that you were betrayed? 57. Did you and your romantic partner talk about the betrayal. Describe in detail what was said, and who said what. 47 Please answer the following questions about your partner’s communication AFTER the betrayal. SD: Strongly Disagree, SA = Strongly Agree 58. My partner said they were very sorry that the betrayal happened. SD 1 2 3 4 5 6 7 SA 59. My partner apologized for the betrayal. SD 1 2 3 4 5 6 7 SA 60. My partner accepted responsibility for their actions. SD 1 2 3 4 5 6 7 SA 61. My partner took the blame for the betrayal. SD 1 2 3 4 5 6 7 SA 62. My partner offered excuses for why they had betrayed me. SD 1 2 3 4 5 6 7 SA 63. My tried to justify what had happened. SD 1 2 3 4 5 6 7 SA 64. My partner told me it would never happen again. SD 1 2 3 4 5 6 7 SA 65. My partner promised me that they would make changes in their behavior. SD 1 2 3 4 5 6 7 SA 66. My partner told me how important the relationship was to them. SD 1 2 3 4 5 6 7 SA 67. My partner emphasized all the good things about the relationship. SD 1 2 3 4 5 6 7 SA 68. My partner told me to get over it. SD 1 2 3 4 5 6 7 SA 69. My partner tried to place the blame on me. SD 1 2 3 4 5 6 7 SA 48 Not all types of betrayal are equal. Some types might be relatively minor, while others are very serious. Please use the following scales to rate your perceptions of the seriousness of this betrayal. 70. Significant 1 2 3 4 5 6 7 Insignificant 71 . Unimportantl 2 3 4 5 6 7 Important 72. Minor 1 2 3 4 5 6 7 Major 73. Relevant l 2 3 4 5 6 7 Irrelevant 74. What is the current status of your relationship? 75. Are you still romantically involved with this person (circle one) Yes No Now, please tell us about your relationship AFTER the betrayal occurred. 76. I was happy with the state of our relationship. SD 1 2 3 4 5 6 7 SA 77. I was very satisfied with the relationship. SD 1 2 3 4 5 6 7 SA 78. I was dissatisfied with the relationship. SD 1 2 3 4 5 6 7 SA 79. I was pleased with the relationship. SD 1 2 3 4 5 6 7 SA 80. Our relationship was fulfilling. SD 1 2 3 4 5 6 7 SA 81. Our relationship was disappointing. SD 1 2 3 4 5 6 7 SA 82. I felt comfortable with the relationship. 49 SD 1 2 3 4 5 6 7 SA 83. I felt content in the relationship. SD 1 2 3 4 5 6 7 SA Now, please answer the following demographic questions: 84. I am (Circle one): Male Female 85. The relationship I mentioned was a (circle one) same-sex OR opposite-sex 86. My age is (In years): ***YOU ARE DONE. TURN IN YOUR SURVEY TO THE RESEARCHER. THANKYOU FOR YOUR PARTICIPATION!*** 50 BIBLIOGRAPHY Baston, C. D., & Shaw, L. L. (1991). Evidence for altruism: Toward a pluralism of prosocial motives. Psychology Inquiry, 2, 107-122. Berger, C. R., & Kellerrnan, K. (1083). To ask or not to ask: Is that a question? In R. N. Bostrom (Ed.), Communication Yearbook 7 (pp. 342-368). Newbury Park, CA: Sage. Drigotas, S. M., & Rusbult, C. E. (1992). Should I stay or should I go? A dependence model of breakups. Jourmtl of Persona_lgrd Sflal Psychology. 62, 62-87. Duck, S. (1994). Stratagems, spoils, and a serpent’s tooth: On the delights and dilemmas of personal relationships. In W. R. Cupach, & B. H. Spitzberg (Eds) The Dark Side of Interpersonal Communication (pp. 3-24). Hillsdale, NJ: Lawrence Erlbaum Associates. Feldman, S. S., & Cauffman, E. (1998). Sexual betrayal among late adolescents: perspectives of the perpetrator and the aggrieved. Journal of Youth and Adolescence. 28, 235-258. Feldman, S. S., Cauffman, E., Jensen, L. A., & Amett, J. J. (2000). The (un)acceptability of betrayal: a study of college students’ evaluations of sexual betrayal by a romantic partner and betrayal of a friend’s confidence. Journal of Youth Q91 Adolescence. 29, 499-523. Fincham, F. D., & Bradbury, T. (1989). Attribution of responsibility in close relationships: Egocentric bias or partner-centric bias? Journal of Marriage and the Family, 51, 27-35. Fischer, C. S. (1982). To dwell among friends. Chicago: University of Chicago Press. Floyd, F. J ., & Wasner, G. H. (1994). Social exchange, equite, and commitment: Structural equation modeling of dating relationships. Journal of Family Psychology, 8- 55-73. Flynn, C. P. (1990). Sex roles and women’s responses to courtship violence. J oumal of Family Violenceé, 83-94. 51 Frisch, F. J. & MacKenzie, C. J. (1991). A comparison of formerly and chronically battered women’s cognitive and situational dimensions. Psychotherapy, 28, 339-344. Girard, M., & Mullet, E. (1997). Propensity to forgive in adolescents, young adults, older adults, and elderly people. Journal of Adult Development, 4, 209-220. Gottman, J. M. (1979). Marital Interaction. New York: Academic Press. Hansson, R. 0., Jones, W. H., & Fletcher W. L. (1990). Troubled relationships in later life: implications for support. Joumflf Social and Personal Relationships, 7, 451-463. Heider, (1958). The Psychology of Interpersonal Relagonships. New York: Wiley. Holmes, J. G., & Rempel, J. K. (1989). Trust in close relationships. In C. Hendrick (Ed.), In Review of personality and social psychology, (Vol. 10, pp. 187-220). Newbury Park, CA: Sage. Jones, W. H. (1988). Psychological and interpersonal issues in betrayal and treachery. Paper presented at the Fourth International Conference on Personal Relationships, Vancouver, British Columbia, Canada. Jones, W. H., Cohn, M. G., & Miller, C. E. (1991). Betrayal among children and adults. In Children’s interpersonal trust: Sensitivity to lying, deception, and promise violations, ed. K. J. Rotenberg, 118-134. New York: Springer-Verlag. Jones, W. H., Sanchez, D., & Merrell, J. (1989). Explanations of betrayal accounts. Paper presented at the meeting of the Southwestern Psychological Association, Houston, TX. Jones, W. H., & Burdette, M. P. (1994). Betrayal in close relationships. In A. L. Weber & J. Harvey (Eds), Perspectives on close relationships (pp. 1-14). New York: Allyn & Bacon. Kelley, H. H., & Thibaut, J. W. (1978). Interpersonal relations: A theory of interdependence. New York: Wiley-Interscience. Lawson, A. (1988). Adultegy. New York: Basic Books. Lewis, R. (1973). Social reactions and the formation of dyads: An interactionist approach to mate selection. Sociometry. 36. 409-418. 52 Metts, S. (1989). An exploratory investigation of deception in close relationships. Journal of Social and Persopal Relationships, 6, 159-179. Metts, S. (1994). Relational Transgressions. In W. Cupach & B. Spitzberg (Eds), The Dark Side Of Communication. Hillsdale: Lawrence Erlbaum Associates. McComack, S. A., & Levine, T. R. (1990). When lies are uncovered: Emotional and relational outcomes of discovered deception. Communicflion Monographs, 57, 119- 138. McCullough, M. E., Sandage, S. J., Brown, S. W., Rachal, K. C., Worthington, E. L., & Hight, T. L. (1998). Interpersonal forgiving and close relationships: II. Theoretical elaboration and measurement. Joufll of Persopality and Social Psychology, 75, 1586- 1603. McCullough, M. E., Worthington, E. L., & Rachal, K. C. (1997). Interpersonal forgiving in close relationships. Journal of Personality and Social Pchhology- 73, 321 - 336. Miller, G. R., Mongeau, P. A., & Sleight, C. (1986). Fudging with friends and lying to lovers: Deceptive communication in personal relationships. Journal of Social and PersoMl Relationships, 3, 495-512. Nelson, M. K. (1993). A new theory of forgiveness. Unpublished doctoral dissertation, Purdue University, West Lafayette, IN. Nunnally, E. W., Chilman, C. 8., & Cox, F. M. (1988). Troubled relationships. Newbury Park, CA: Sage. Ohbuchi, K., Kameda, M., & Agarie, N., (1989). Apology as aggression control: Its role in mediating appraisal of and response to harm. Journal of Personality and Social Psycholggy. 56. 219-227. Ohira, M. (2001). Eboo Topics Among Sage-Sex Friendship. Unpublished master’s thesis, Michigan State University, Lansing. Parks, M. R., & Floyd, K.(1996). Making friends in cyberspace. Journal of Communication, 46, 80-95. Planalp, S., & Honeycutt, J. (1985). Events that increase uncertainty in personal relationships. Human Communication Research. 11, 593-604. Rackley, J. V. (1993). The relationships of marital satisfaction, forgiveness, and religiosity. Unpublished doctoral dissertation, Virginia Polytechnic Institute and State University, Blacksburg. 53 Roloff, M. E., & Janiszewski, C. A. (1989). Overcoming obstacles to interpersonal compliance: A principle of message construction. Human Communication Research, 16, 33-61. Roscoe, B., Cavanaugh, E., & Kennedy, D. R. (1988). Dating infidelity: Behaviors, reasons, and consequences. Adolescence, 23, 35-43. Rubin, L. B. (1985). Just friends: The role of friendship in our lives. New York: Harper and Row. Rusbult, C. E. (1979). The effects of relationship outcome value, alternative outcome value, and investment size upon satisfaction and commitment in friendships. Unpublished manuscript, University of Kentucky. Rusbult, C. E. (1980). Commitment and satisfaction in romantic associations: A test of the investment model. Journal of Experimental Social Psychology. 16, 172-186. Rusbult, C. E. (1983). A longitudinal test of the investment model: The development (and deterioration) of satisfaction and commitment in heterosexual involvements. Journal of Persgrlity and Social Psychology. 45, 101-117. Rusbult, C. E., Drigotas, S. M., & Verette, J. (1994). The investment model: An interdependence analysis of commitment processes and relationship maintenance phenomena. In D. J. Canary & L. Stafford (Eds), Communication and Relational Maintenance (pp. 115-40). New York: Academic Press. Rusbult, C. E., & Martz, J. M. (1995). Remaining in an abusive relationship: An investment model of analysis of nonvoluntary dependence. Personalityand Socia_l Psychology Bulletin, 21, 558-571. Rusbult, C. E., Verette, J ., Whitney, G. A., Slovik, L. F., & Lipkus, I. (1991). Accommodation processes in close relationships: Theory and preliminary empirical evidence. Journal of Personality and Social Psychology, 60, 53-78. Scott, M. B., & Lyman, S. M. (1968). Accounts. American Psychological Review 33, 46-62. Shackelford, T. K., & Buss, D. M. (1996). Betrayal in mateships, friendships, and coalitions. Personalityand Social Psychology Bulletin, 22, 1151-1164. Simpson, J. A. (1987). The dissolution of romantic relationships: Factors involved in relationship stability and emotional distress. Journal of Personality and Social Psychology, 53, 683-692. 54 Truman-Schram, D. M., Cann, A., Calhoun, A., & Vanwallendael, L., (2000). Leaving an abusive dating relationship: an investment model of women who stay versus women who leave. Journal of Social and Clinical Psychology, 19, 161-183. Weiner, B. (1986). Anattribution theory of motivation and emotion. New York: Springer-Verlag. Woodman, T. (1991). The role of forgiveness in marital adiustment. ' Unpublished doctoral dissertation, Fuller Graduate School of Psychology, Pasadena, CA. 55 will l'lllllljllljljlll J3 H133 will)