Differences in the demand/withdraw pattern of communication in married and dating couples with known and unknown extradyadic involvement

Material Information

Differences in the demand/withdraw pattern of communication in married and dating couples with known and unknown extradyadic involvement
Balderrama-Durbin, Christina M
Place of Publication:
Denver, CO
University of Colorado Denver
Publication Date:
Physical Description:
xi, 42 leaves : ; 28 cm.


Subjects / Keywords:
Adultery ( lcsh )
Communication in marriage ( lcsh )
Dating (Social customs) ( lcsh )
Interpersonal conflict ( lcsh )
Adultery ( fast )
Communication in marriage ( fast )
Dating (Social customs) ( fast )
Interpersonal conflict ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Thesis (M.A.)--University of Colorado Denver, 2009. Clinical psychology
Includes bibliographical references (leaves 37-42).
General Note:
Department of Psychology
Statement of Responsibility:
by Christina M. Balderrama-Durbin.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
436223245 ( OCLC )


This item is only available as the following downloads:

Full Text


DIFFERENCES IN THE DEMANDIWITHDRA W PA TIERN OF COMMUNICATION IN MARRIED AND DATING COUPLES WITH KNOWN AND UNKNOWN EXTRADY ADIC INVOLVEMENT by Christina M. BalderramaDurbin B.A., University of Colorado at Boulder, 2006 A thesis submitted to the University of Colorado Denver in partial fulfillment of the requirements for the degree of Master of Arts Clinical Psychology 2009 r


2009 by Christina M. Balderrama-Durbin All rights reserved.


lbis thesis for the Master's of Arts degree by Christina M. Balderrama-Durbin has been approved Peter Kaplan /z.o/ z.oo9 Date


Balderrama-Durbin, Christina, M. (Master's of Arts, Clinical Psychology) Differences in the Demand/Withdraw Pattern of Communication in Married and Dating Couples with Known and Unknown Extradyadic Involvement Thesis directed by Assistant Professor Elizabeth Allen, Ph.D. ABSTRACT Infidelity or extradyadic involvement (ED I) is one of the most damaging issues facing individuals' in romantic relationships. EDI and the demand/withdraw communication pattern have been topics of valuable research in predicting relationship satisfaction, but to our knowledge these topics have not been concurrently investigated. The current study investigated the differences in couple level demand/withdraw in romantic relationships where there is no reported EDI, where there is known EDI, and where there is unknown EDI in a community sample of married and dating couples. Additionally, the subgroups ofEDI (none, known and unknown) were evaluated on their couple level relationship satisfaction. Couples with unknown EDI bad significantly higher demand/withdraw in their conflict communication when compared to couples with no EDI. Couples with known EDI had a greater level of demand/withdraw compared to couples with no EDI, but lower levels of demand/withdraw compared to those with unknown ED I, these differences were not significant. Couples with known EDI in their relationship reported significantly lower couple level relationship satisfaction compared to couples with no EDI. Couples with unknown EDI reported greater


satisfaction than the couples with known EDI, but are less satisfied than the couples with no EDI, neither of these differences were significant. Lastly, infidelity (present or not present) could predict couple level relationship satisfaction above and beyond the effect of demand/withdraw. Individual characteristics such as the desire for closeness and the need for autonomy overlap in the literature on deception, infidelity, and the demand/withdraw pattern. Heightened levels of demand/withdraw in couples with unknown EDI may reflect asymmetrical desires for closeness and/or need for autonomy which appear salient for behaviors of deception and infidelity. Future research on the individual level could provide greater insight into the pathway by which these outcome variables (relationship satisfaction and demand/withdraw) manifest. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. Eiizabeth S. Allen


DEDICATION PAGE I would like to dedicate my thesis to my loving and supportive family from whom I extract much of my motivation. Specifically, I would like to dedicate this work to my loving husband, Tony, for seeing me through this enlightening process.


ACKNOWLEDGMENT I would like to sincerely thank Dr. Peter Kaplan, and Dr. Joy Berrenberg for their support in the completion of my thesis project. I would also like to extend a very special thank you to Dr. Elizabeth Allen for making this project a reality through her support and thoughtful guidance. Additionally, thank you to Ryo Morimoto for help in the revisions of my thesis. Lastly, many thanks to all of the coding team members throughout the years who have dedicated their time and efforts to make this project possible


TABLE OF CONTENTS CHAPTER 1. INTRODUCTION .................................................................................... 1 Extradyadic Involvement (EDQ and Relationship Satisfaction ......... 2 Deception Research and EDI ............................................................. 3 Conflict Communication and Relationship Satisfaction .................... 5 Communication and EDI .................................................................... 6 Demand/withdraw Pattern in Communication ................................... 6 2. METI-IOD ............................................................................................... 14 Participants ....................................................................................... 14 Procedure .......................................................................................... 16 Measures ........................................................................................... 17 Relationship Satisfaction ......................................................... 17 Infidelity Deception ................................................................. 18 Demand/withdraw Communication Pattern ............................. 19 3. RESULTS ............................................................................................... 22 Replication Analyses ........................................................................ 22 EDI and Couple Level Relationship Satisfaction ............................. 23 EDI and Demand/withdraw .............................................................. 25 viii


4. DISCUSSION ........... ............................................................................ 29 Limitations ........................ ........................................ ............. 35 Swnmary .................................................................................. 36 REFERENCES .................................................................................................. 37 vi iii


LIST OF FIGURES Figure 1.1 EXAMPLE OF SELF INFLUENCE MODEL .............................................. 9 1.2 EXAMPLE OF RELATIONAL INFLUENCE MODEL ............................ 10 3.1 RELATIONSHIP SATISFACTION AND EDI (PARTICIPATED OR NO PARTICIPATION) ...................................................................................... 23 3.2 RELAITONSHIP SATISFACTION AND EDI (NONE, UNKNOWN, AND KNOWN) .................................................................................................... 25 3.3 COUPLE LEVEL DEMANDIWITHDRA WAND EDI (NONE, UNKNOWN, AND KNOWN) ..................................................................... 26 X


LIST OFT ABLES Table 2.1 Percentage of Race!Ethnicity for Male and Female Participants ................. 15 3.1 Predicting Relationship Satisfaction, Results from Multiple Regression .... 28 xi


CHAPTER I INTRODUCTION Divorce rates in the United States have steadily increased from the 1920's and have virtually stabilized resulting in one half to two thirds of current marriages ending in divorce (National Center of Health Statistics, 2004). Raley and Bumpass (2003) suggest that this could be a superficial stabilization since subgroups, such as African American women, are still experiencing declines in martial success rates. These substantial failure rates have enhanced research interest in dyadic relationship functioning. Whisman, Dixon and Johnson (1997) surveyed a national sample of therapists who indicated that communication is among the most frequent problems encountered in couple therapy. Furthermore, couples' therapists cite extramarital affairs as one of the most damaging issues facing couples in therapy. Both infidelity and couple communication patterns have been topics of valuable research in relationship functioning, but few studies have investigated these topics concurrently. Conflict communication and infidelity have each attracted vast amounts of research interest in the area of relationship functioning and justifiably so. Almost one third of divorced woman cite an affair as the main cause for the breakdown of the relationship (Burns, 1984 ). Furthermore, a longitudinal study of premarital I


couples indicated that the conflict communication style prior to marriage is a primary predictor of marital outcomes such as relationship satisfaction and divorce Stanley & Markman, 2004). Positive and negative communication patterns not only predict relationship dissolution, but serve as precursors to infidelity (Allen et al., 2008). Negative conflict communication patterns and infidelity do not occur in exclusion; instead one may exacerbate the other's detrimental effects. A review of the literature on both infidelity and conflict communication styles will help illustrate the potential relationship between these two important aspects of relationship functioning and the objectives of the current study. Extradyadic Involvement (EDI) and Relationship Satisfaction Extradyadic involvement (ED I) is often defined as the engagement in intercourse with a person outside the primary dyad, a spouse or exclusive dating partner (Allen and Baucom, 2004); certainly other types of extramarital involvements can occur. Although a large majority of Americans believe that EDI while in a committed relationship is wrong (Glass & 1992), recent studies estimate that 22-25% of men and 11-15% of women have engaged in extramarital sex (for a review see Allen et al., 2005). Gender is one of the most constant predictors of infidelity, with males historically reporting higher rates of ED I, though the gender gap is gradually diminishing (Atkins, Baucom & Johnson, 2001; 2


Buss & Shackelford, 1997). Another common predictor for infidelity is marital dissatisfaction (Buss & Shackelford, 1997; Glass & Wright, 1992). Using a national, random sample, Atkins, Baucom and Jacobson (200 1) demonstrated the powerful negative association between EDI and relationship satisfaction. Atkins et al. (2001) reported that those who described their marriage as "not too happy" were four times more likely to report extramarital sex than those who were "very happy" in their marriage. Furthermore, those who reported their marriage as "pretty happy" were two times more likely than "very happy" couples to have engaged in extramarital sex, demonstrating a graded association between the likelihood of engagement in EDI and satisfaction. Deception Research and EDI A large amount of research exists on general deception in close relationships, but few studies have focused on EDI deception specifically. Following engagement in an ED I, the betrayal partner must determine whether to disclose to or deceive his/her primary partner. Research suggests partner deception related to an EDI may have a unique association with relationship satisfaction. The majority of spouses who have committed an EDI report that their partner is unaware of their sexual engagement (Fats-Stewart et al., 2003), and these hidden affairs appear to be related to the most relationship distress overall when compared to known affairs in a small clinical sample (Atkins et al., 2005). Additional 3


research is necessary to determine if this finding is unique to a clinical sample of couples seeking marital therapy. In more general deception DePaulo, Ansfield, Kirkendol & Boden (2004) investigated the association between the frequency of serious lies, lies that could threaten the relationship (including infidelity), and closeness of the relationship. Relationship closeness was classified into three categories: casual (strangers and acquaintances), intermediate (friends other than best friends, and family other than parents, children or spouse) and close (parents, romantic partners, children and best friends). Analyses based on these three categories demonstrated that the closer the relationship the higher the frequency of serious lies told. DePaulo et al. (2004) asserted that the closer the relationship, the higher the perceived cost of disclosure and the more motivated a person is to deceive his/her partner. When evaluating deception within romantic relationships specifically, research suggests that those in lower quality relationships, with lower commitment and higher suspicion use deception techniques more frequently than those in higher quality relationships, with higher commitment and lower suspicion (Cole, 200 I; Kalbfleisch, 2001). An individual's attachment style is also associated with their frequency of lying in a romantic relationship. Individuals who fearful or uncomfortable with intimacy from their partner are said to have an avoidant attachment style, while individuals who long for intimacy and closeness from their 4


partner are said to have an anxious attachment style. Both anxiously and avoidantly attached individuals are more likely to use deception in their romantic relationships (Cole, 2001). Other research suggests that individuals who use deception as a mode of conflict avoidance are less satisfied in their relationships (Peterson, 1996). A review of the deception literature suggests that EDI deception might occur in both satisfied and dissatisfied relationships. For those in satisfactory relationships, there is more to lose if their partner discovers the betrayal, whereas other literature finds the use of deceptive techniques are more frequent in less satisfied relationships. Conflict Communication and Relationship Satisfaction Conflict is a nonnal part of many relationships, but how the conflict is handled can be an important indicator of relationship functioning. Dissatisfied couples are more likely to engage in destructive conflict communication styles including criticism, defensiveness, contempt, and withdrawal in conflict interactions. These negative conflict communication patterns prior to marriage predict lower satisfaction and divorce longitudinally (Clements, Stanley, & Markman, 2004). Additionally, decreased satisfaction can lead to negative conflict patterns (Marchand & Hock, 2000), and reduction in negative conflict patterns can lead to increased satisfaction (Markman, Floyd, Stanley & Storaasli, 1988). This suggests that a bidirectional relationship exists between conflict communication style and relationship quality. 5


Communication and EDI Communication patterns not only predict relationship dissolution, but serve as a precursor to infidelity. A recent longitudinal study of premarital couples conducted by Allen et al. (2008) reported that female infidelity is preceded by lower levels of positive communication by the female, and higher levels of negative communication by both genders. Male infidelity was predicted by their own premarital sexual dissatisfaction, low positive communication by the male and higher invalidation during communication exhibited by the female. Another study evaluating communication and occurrence of EDI found that the ease of discussing sexual matters within the relationship was negatively associated with likelihood of EDI occurrence for Hispanic males (Choi, Cantania & Dolcini, 1994). Thus, the limited existing research on EDI and communication patterns suggests that EDI is associated with problematic couple communication; however, the link between demand/withdraw communication and EDI has not been examined. Demand/withdraw Pattern in Communication The demand/withdraw pattern in conflict communication is a negative interaction pattern that has a strong negative correlation with relationship satisfaction. This communication pattern, observed and measured by Christensen and Heavey ( 1990), has been widely studied and increasingly well characterized. Frequently during conflict interactions, distressed couples display an asymmetrical conflict response where one spouse blames, nags, criticizes and pressures the other 6


for change, while the other spouse withdraws or avoids conflict (Christensen & Heavey, 1990; Eldridge, Sevier, Jones & Atkins, 2007; Ridley, Wilhelm & Surra, 2001 ). Christensen & Shenk ( 1991) reported greater use of demand/withdraw communication during problem discussions for couples who were seeking marital therapy and divorcing couples when compared to non-distressed couples. Gender asynunetry in this conflict pattern has also been relatively stable across research studies on demand/withdraw communication. Dissatisfied wives tend to demand change from their husbands while dissatisfied husbands tend to withdraw from conflict with their wives (Caughlin & Vangelisti, 2000; Christensen & 1991; Eldridge et al., 2007). The wife demand/husband withdraw communication pattern predicts both early and late divorce (Gottman & Levenson, 2000). Further, research suggests that changing which partner chooses the conflict topic, female-chosen or male-chosen, could influence the direction of demand/withdraw roles. Role reversal, where the demander instead withdraws and the withdrawer instead demands, can be influenced by various moderators such as length of relationship, distress level and novelty of topic (Eldridge et al., 2007). Three theoretical perspectives have been used to describe this gender asymmetry in conflict behavior: the gender differences perspective, the social structure perspective, and the individual differences perspective The gender differences perspective attributes the gender asymmetry during conflict communication to differences in socialization. Females are socially 7


encouraged to be expressive, affiliative and seek closeness with others. This desire for more closeness with her partner is thought to foster demanding behavior in conflict communication. Conversely, males are socialized to seek independence and to be less expressive. lbis desire for independence is believed to elicit withdrawing behavior for males in conflict communication. The social structure perspective attributes the gender asymmetry in the demand/withdraw pattern to power imbalances between men and women. This perspective has been supported in various research studies (Christensen & Heavey, 1990; Heavey, Layne & Christensen, 1993). These studies suggest that a woman's lack of power in the marriage motivates her to seek change in the relationship structure, thus engaging in demanding behavior. In contrast, husbands, who traditionally hold greater power in the relationship, work to maintain the status quo by withdrawing from conflict that could potentially threaten their position. More recent research conducted by Vogel et al. (2007) has not found evidence for the social structure explanation. In this study, when importance of conflict, socio economic status and self-reported ability in decision making were kept equal, the gender disparity of demand and withdraw behaviors still remained. Those spouses with the most demandingness in communication did not exhibit less situational power during the discussion as hypothesized by the social structure perspective. The individual differences perspective attributes the demand/withdraw gender asymmetry to differences in personal characteristics, namely the gender 8


difference in the desire for closeness versus greater independence. Research has demonstrated that women generally desire more closeness in their relationships while men desire more independence. Furthermore, desire for closeness was associated with demanding behavior; while desire for more independence was associated with increased withdraw regardless of gender (Caughlin & Vangelisti, 2000; Christensen & Heavey, 1993; Christensen & Shenk, 1991). Within the individual differences perspective there are two models proposed by Caughlin & V angelisti (2000): the self influence model and the relational influence model. In the self influence model, the individual characteristic(s) of the spouse directly influences his/her own behavior during conflict communication. One spouse's desire for closeness results in their increased tendency to demand and this demanding behavior can indirectly elicit his/her partner's tendency to withdraw. In other words, an asymmetrical desire for closeness or autonomy will directly influence the individual's level of demanding or withdrawing behavior respectively, but can also indirectly increase the level of withdrawing or demanding behavior of their partner, see Figure 1.1 from Caughlin & V angelisti (2000). + Demand/ d' Withdraw d' Desire for Closeness d' Demand/ Withdraw Figure 1.1 EXAMPLE OF SELF INFLUENCE MODEL. 9


In the relational influence model, the characteristic(s) of the spouse influences the overall interpersonal interaction (Caughlin & Vangelisti, 2000). The individual's asymmetrical desire for more closeness or autonomy in the relationship increases the likelihood of their engagement in both demanding and withdrawing behaviors. Unlike the self influence model, this model does not predict which spouse demands and which spouse withdraws. Instead, the relational influence model describes that the individual characteristic(s) of the spouse results in a dyadic phenomenon where the individual engages in both demanding and withdrawing behavior and in tum their behavior elicits demand and withdraw from their partner, see Figure 1.2 from Caughlin & V angelisti (2000). Desire for Closeness + Demand/d' Withdraw d' Desire for Closeness d' Demand/ Withdraw Figure 1.2 EXAMPLE OF RELATIONAL INFLUENCE MODEL. Particular individual characteristics of a spouse may predict both the occurrence of the demand/withdraw pattern and serve as risk factors for infidelity. Research on adult attachment and EDI (Allen & Baucom, 2004) suggests that anxiously and avoidantly attached individuals are more likely to engage in an EDI. Furthermore, avoidant individuals reported autonomous motivations for engagement in EDI, while anxious individuals reported need for intimacy as their 10


motivation for ED I. If couples who have experienced EDI have higher levels of demand/withdraw, this may suggest that EDI motivated by need for autonomy or intimacy may reflect and/or exacerbate the likelihood of demand/withdraw in conflict communication. The relational influence model suggests that a person who commits an EDI motivated by either desire for more closeness or need for autonomy will increase both partners' tendency for demanding and withdrawing behavior in conflict. In contrast, the self influence model suggests that an EDI motivated by desire for closeness would increase demanding behaviors for the individual engaged in the EDI, while eliciting withdraw in his/her partner. Furthermore, ED Is committed with autonomous motivations would increase the withdrawing behaviors for the individual engaged in the EDI, while eliciting demand behaviors from his/her partner According to both models, asymmetries in the desire for closeness or the need for autonomy will increase demand/withdraw for the couple as a system, and differences in demand/withdraw would be measureable on the couple level. For the purposes of the current study, couple level demand/withdraw was evaluated to investigate the systemic effect of the subgroups ofEDI on the demand/withdraw pattern. If a difference is found, determination of the specific pathway, either self influence or relational influence, could be explored in future studies. This research study aims to concurrently investigate the difference between the subgroups ofEDI (none, known and unknown) and the demand/withdraw II


pattern in conflict communication, and to expand the deception and disclosure literature on EDI and relationship satisfaction. The study also aims to assess if the occurrence of an EDI could predict relationship satisfaction above and beyond the demand/withdraw interaction pattern. The first two postulated hypotheses are replications of prior research in order to confirm typical associations in the current sample: HI: A higher degree of the observed demand/withdraw pattern will be associated with lower relationship satisfaction for the couple. H2: History of EDI (present or not present) will be associated with lower relationship satisfaction for the person who has engaged in the behavior or the participating partner (PP). The primary novel research objectives are to explore the relationship between the demand/withdraw interaction pattern, relationship satisfaction and the subgroups of EDI (none, known and unknown). I will also investigate ifEDI can predict relationship satisfaction above and beyond the demand/withdraw interaction pattern. Specifically, I hypothesize : H3: The presence ofEDI, known or unknown, will be positively associated with the demand/withdraw conflict communication pattern similar to prior research suggesting an association between problematic communication and EDI, as well as possible contributions of intimacy/autonomy needs to both demand/withdraw and EDI. 12


H4: Unknown EDI will be more highly associated with the occurrence of the demand/withdraw conflict communication pattern when compared to knownEDI. H5: Unknown EDI will be associated with greater relationship distress when compared to known EDI. H6: The occurrence of EDI will predict relationship satisfaction above and beyond the demand/withdraw interaction pattern. I may have research findings similar to those of the clinical sample in Atkins et al. (2005), see H5. Couples that have remained together and participated in this study despite the disclosure of EDI will likely have additional protective factors against greater relationship distress, such as higher commitment. Furthermore, the literature on the association between demand/withdraw and relationship satisfaction would suggest that the group with the greatest relationship distress would also possess the highest level of demand/withdraw, see H4. 13


CHAPTER2 METHOD Participants Participants in the current study consisted of 112 couples from the community, composed of 47 married and 65 dating couples. Couples are a subset of participants from a larger study on relationship functioning and sexual risk behaviors. Participants were recruited through flyers, online classified.s, and other various local newspapers in a mid-Western metropolitan area. Each participant was paid $50 as compensation for their time To participate, couples were required to be together for a least one year, read and speak English fluently, and be 18 years or older. Each member of the couple was screened individually to gauge their interest in involvement. The average age for male participants was 3 5. 7 years (SD = 12.2, range 1966), and the average age for female participants was 33.1 years (SD = 11.9, range 18-67). Average years of education were 13.8 years (SD = 2.2, range 9-20) for males and 14.5 years (SD = 2.2, range I 0-20) for females. The average length of relationship for married couples was 11. 5 years, while dating couples had been together for an average of three years. Males earned a median income between $20,000-29,999 per year while females earned a median annual income between 14


$15,000 and $19,000. Eighty five percent of the couples live together, earning median dual income of $35,000 $50,000. This is slightly lower than the reported median national household income of$50,233 reported in U.S. Census Bureau, 2007. Detailed demographic information on race/ethnicity for males and females is contained in Table 2.1. Hispanic and African American populations were oversampled in order to obtain greater representation from these subgroups. Table 1.1 Percentage ofRace/Ethnicity for Male and Female Participants. Gender Race/ethnicity Male Female White 60.2 54 Hispanic!Latino 18.6 21.2 Black/ African American 17.7 16.8 American Indian or 3.5 8 Alaska Native Native Hawaiian or .9 0 Pacific Islander Asian .9 1.8 Other 4.4 3.5 Multiracial 5.3 8 15


Procedure Couples came into the research laboratory to complete all procedures. For part one, participants were asked to engage in a total of four videotaped interaction tasks: two vulnerable feeling interactions and two conflict interactions. In the vulnerable feeling interaction, couples were instructed to talk like friends about a time they felt sad, hurt, lonely, ashamed or insecure. Each interaction lasted a duration of8 minutes. For the conflict interaction task, a revised version of the Marital Agendas Protocol (MAP; Notarius & Vanzetti, 1983) was used to determine the couples' top conflict area in their relationship. The MAP questionnaire lists various areas of potential conflict in close relationships such as household tasks, money, children, jealousy, in-laws, things we do for fun, and friends. Couples were asked to individually rate items on a scale from 0 (not at all) to 100 (severe problem). Based on these ratings, a male and a female top problem area were discussed. For the second part of the procedure, couples were placed into separate laboratory rooms and asked to complete a lengthy questionnaire containing various sensitive questions, some pertaining to sexual history and relationship satisfaction. Following completion of the questionnaires, participants were asked to seal their responses in the provided envelope and place the sealed envelope into a locked drop-box to preserve confidentiality. The entire procedure took approximately two and a half hours to complete. 16


Measures RelaJionship Satisfaction Relationship satisfaction was assessed using the Quality Marriage Index (QMI; Norton, 1983). Participants were asked to rate a total of six questions, five questions were rated on a seven point Likert scale, I (very strong disagreement) to 7 (very strong agreement), and one question asked participants to rate their degree of happiness in the relationship with all things considered, measured on a ten point Likert scale, 1 (unhappy) to 10 (perfectly happy), yielding a possible total range of scores from 645. This relationship index has been found to be highly reliable (for a review and description, see Rubin, Pahn-Green, & Sypher, 1994). The current sample demonstrated a Cronbach's alpha level of .91. Scores for relationship satisfaction were assessed on the individual level by summing each individual's ratings on the six items. The average individual male satisfaction score was 39.1 (SD = 5.1, range 18-45), the average individual female satisfaction score was 38.7 (SD = 6.2, range 15-45). For the couple's relationship satisfaction score, individually summed scores were averaged yielding a mean couple satisfaction score of38.9 (SD = 4.9, range 19.5-45) for the sample. Compared to the satisfaction scores of married couples in prior research studies, this sample demonstrates relatively high marital satisfaction scores, (M = 39.8, SD = 5.2) compared to (M= 38.1, SD = 8.1) Fincham, Harold & Gano-Phillips, 2000, 17


particularly given the average length of the marriage in the current sample ( 11.5 years). Infidelity Deception Questions pertaining to the individual's own sexual activities outside the primary relationship and knowledge of the sexual activities of his/her partner were included in the questionnaire. Comparison questions were used to determine if an EDI occuned and, if so, if the EDI was known or unknown to their partner. The reported number of sexual relationships while in the current relationship by one partner was compared with the reported number of known sexual relationships by the other partner. The occurrence of known and unknown EDI was assessed on the couple level. For some couples there were a combination of known and unknown EDis reported. For purposes of this study, couples were placed in the unknown EDI group if all sexual involvements within the relationship were not known or they were not sure of their partner's involvement If there was at least one known EDI, the couple was classified as having a known EDI regardless of the number that were known, rendering three groups: 1) neither have a reported EDI (None, N = 71), 2) at least one partner has at least one known EDI (Known, N = 24), and 3) all ED Is are unknown to the partner (Unknown, N = 17). 18


Demand/withdraw Communication Pallern Observational coding teams were trained according to the Sevier, Simpson and Christensen (2004) coding manual. Two teams of coders were trained and assigned weekly videotaped interactions to code the target individual according to each of the five dimensions of demand/withdraw. Coding assignments were counterbalanced and randomly assigned according to gender and chosen topic. Ratings are made on a 9 point Likert scale I (none) to 9 (a lot). The demand scale consists of two dimensions: Blame (blames, accuses, or criticizes the other partner for being the causal agent of the problem) and Pressures for Change (requests, demands, nags, or otherwise pressures the other partner to change). The withdraw scale is represented by three dimensions: Withdraw (withdraws, becomes silent, refuses to discuss the topic, or disengages from the discussion), A voidance (hesitation, changing topics, diverting attention, delaying the discussion, or minimizing the importance of the problem) and reverse coded Discussion (level of engagement or involvement in the discussion of the conflict). Prior coding teams of undergraduate coders have demonstrated a high range of inter-rater reliability for the demand/withdraw dimensions: Blaming, .86 to .94; Pressures for Change, .76 to .88; Withdraw, .60 to .87; Avoidance, .62 to .84; and Discussion, .69 to .83. On the subscale level the inter-rater reliability ranged from .87 to .93 for demand and 75 to.88 for withdraw (Sevier, Simpson & Christensen, 2004). 19


Throughout data interclass correlation coefficients (ICC) were calculated weekly for the current study in order to detennine the inter-rater reliability (Strout & Fleiss, 1979) and drift between coders. A cutoff of. 70 was used with an average ICC score of .80 (SD = .16). Interactions that did not reach this desired reliability score were re-coded at a later time. If sufficient reliability could not be reached after re-coding, it was rated for a final time and a group census was made on the scores. Individual scores were obtained by coding either the male or the female chosen topic for each couple. Scores from each of the five demand/withdraw dimensions (blame, pressures for change, withdraw, avoidance and reverse coded discussion) were averaged to obtain an individual demand/withdraw score. Couple level demand/withdraw was calculated as a composite score from the male and female individual scores. The average couple level demand/withdraw was 2.3 (SD = .4, range 1.6-3.7) for the current sample. Compared to prior research studies on demand/withdraw, this is a low average. Caughlin and Vangelisti, 2000, reported an average demand/withdraw score of3.5, this is an average of the wife demand/husband withdraw and the husband demand/wife withdraw scores obtained from their sample of married couples. The low demand/withdraw may be due to the relatively high levels of satisfaction in the current sample. Additionally, this low average could be due to the extensive averaging required to obtain a couple level demand/withdraw score. Averaging across the five scales means that an individual 20


who is high on the blame dimension, but low on withdraw would average to a moderate score for demand/withdraw overall. To receive a high individual demand/withdraw score would require a person to be high on both the demand and withdraw dimensions. Furthermore, for a couple to receive a high couple level demand/withdraw score, each individual partner would have to be high on both demanding and withdrawing behaviors, which is not typical for this interaction pattern. 21


CHAPTERJ RESULTS Replication analyses To confinn the negative association between demand/withdraw and relationship satisfaction, a correlation analysis between the couple level demand/withdraw and the average couple satisfaction was performed. There was a significant negative correlation between couple level satisfaction and demand/withdraw, r = -.19, p (one-tailed) < .05, supporting predictions outlined in HI. For the second confirmatory analysis, at-test was used to determine the relationship between the participating partner (PP), the individual engaged in the EDI, and individual relationship satisfaction. Those persons who participated in EDI and those who did not were compared on their level of individual relationship satisfaction. For males, those who participated in EDI (either known or unknown) had significantly lower relationship satisfaction (M= 37.1, SD = 6.3) when compared to males in a relationship with no EDI (M= 39.7, SD = 4.6), 1(110) = 2.38, p < .05, this was a medium effect d = .53. Females who participated in an EDI (either known or unknown) had significantly lower relationship satisfaction (M= 35.4, SD = 6.2) compared to females with no EDI in their relationship (M = 39.5, SD = 6.0), 1(110) = 2.94,p < .01, this was a medium effect d = .70. These results 22


support H2, such that history ofEDI (present or not present) was associated with lower relationship satisfaction scores for the PP. See Figure 3.1 for a graphical representation of male and female relationship satisfaction by EDI. 41 40 c: 39 0 .:: eM 38 In -Male Satisfaction 37 Cll .... : Female Satisfaction Q, ' :a 36 . / 35 .:: '7 -34 n J EDI NoEDI ---------------Figure 3.1 RELATIONSHIP SATISFACTION AND EDI (PARTICIPATED OR NO PARTICIPATION). p= .05, .. p = .01. EDI and Couple Level Relationship Satisfaction Contrary to prior research on EDI and relationship satisfaction (such as the findings replicated above), this investigation evaluated participants on the couple level. Evaluating data on the couple level provides the ability to evaluate the overall effect ofEDI on systemic relationship satisfaction. Evaluation on an individual level can be explored in future analyses. To assess the differences between the subgroups ofEDI and couple level relationship satisfaction, an ANOVA was performed where EDI status (none, 23


known or unknown) was the independent grouping variable and the aggregate scores of couple satisfaction was the dependent variable. There was a significant effect ofEDI (none, known or unknown) on couple level relationship satisfaction, F(2, 109) = 9.48,p < .01, w = .36. To determine which EDI subgroups (none, known and unknown) contributed to this significant mediwn effect and the direction of the effect, the Least Significant Difference (LSD) post hoc analysis was performed. Results are reported in mean difference (MD) scores. The post hoc analysis revealed that couples with no EDI had significantly higher relationship satisfaction compared to couples with a known EDI, MD= 4.64, p < .0 I. There was no significant difference in relationship satisfaction between couples with no EDI and those with an unknown ED I, MD = 2.26, p = .07. No significant difference between couples with unknown EDI and known EDI was found, MD= 2.37,p = .11, refer to Figure 3.2 for a graphical representation of average satisfaction by EDI status. The prediction outlined in H5, that couples with unknown EDI would be associated with more relationship distress, was not supported. 24


s:: 41 .2 0 c:! 40 1;/l -i 39 IZJ 0.. 38 -..c:: 1;/l s:: 37 0 -lU "i) 36 35 Q) > Q) .....l 34 Q) 1 33 0 u Figure3.2 None Unknown Known RELATIONSHIP SATISFACTION AND EDI (NONE, UNKNOWN, AND KNOWN). p =.01. EDI and Demand/withdraw To investigate if those with EDI (known or unknown) will have higher levels of demand/withdraw communication and if unknown EDI will be associated with the highest degree of the observed demand/withdraw pattern when compared to known EDI (H3 and H4 respectively), a second ANOVA was conducted where the subgroups ofEDI (none, known or unknown) were the independent variable and couple level demand withdraw was the dependent variable. There was a significant effect of EDI status (none, known or unknown) on couple level demand/withdraw communication during conflict, F(2, Ill)= 4.22,p < .05, ro = .23. A LSD post hoc analysis was performed and reported in mean difference (MD) scores to determine which EDI subgroups (none, known and unknown) contributed 25


to this significant small to mediwn effect. The analysis revealed that couples with no EDI have significantly lower levels of demand/withdraw relative to couples with unknown EDI, MD= -.33,p < .01. There was no significant difference in demand/withdraw for couples with no EDI and those with known EDI, MD= -.12, p = .25, and no difference between the couples with known EDI and unknown EDI, MD= -.21,p = .12. These findings are consistent with H4 predicting that couples with unknown EDI would display the highest levels of demand/withdraw conflict communication behavior; although their demand/withdraw was not significantly higher than couples with known EDI. Refer to Figure 3.3 for a graphical representation of couple level demand/withdraw means by EDI status. 2.7 2.6 ..c: -2.5 2.4 a E 2.3 Q) 0 2.2 Q) > Q) 2.1 Q) -a 2 ::3 0 u Figure3.3 2.38 None Unknown Known COUPLE LEVEL DEMAND/WITHDRAW AND EDI (NONE, UNKNOWN, AND KNOWN). p = .01. 26


To assess if the occurrence ofEDI could predict relationship satisfaction above and beyond the demand/withdraw interaction pattern, a hierarchical multiple regression analysis was performed. For the model, demand/withdraw was entered first, followed by EDI (present or not present). Entering demand/withdraw into the model first allows for the variance accounted for by demand/withdraw in relationship satisfaction to be assessed before the variance of EDI (present or not present). The multiple regression reveals that EDI (present or not present) significantly predicted relationship satisfaction above and beyond demand/withdraw,l(109) = -3.64,p < .001, accounting for an additionallO.So/o of the variance in relationship satisfaction. This was consistent with H6 stating that EDI (present or not present) would predict relationship satisfaction above and beyond the effect of demand/withdraw. Overall demand/withdraw and EDI (present or not present) accounted for 14o/o of the variance in couple level relationship satisfaction. See Table 3. I for results of the multiple regression analysis. 27


Table 3.1 Predicting Relationship Satisfaction, Results from Multiple Regression. B SEb p Step 1 Constant 43.74 2.46 Demand/withdraw -2.07 1.03 -.19. Step2 Constant 46.48 2.45 Demand/withdraw -1.26 1.00 -.11 EDI (present or not -3.40 .93 -.33** present) o:l_ -. ,:.!_ Note. R -.03 for Step 1 (ps.05), and dR -.14 for Step 2 (ps.001). p < .05, p< .001. 28


CHAPTER4 DISCUSSION A primary goal of the cunent study was to evaluate how the deception or the disclosure of EDI affects overall relationship satisfaction. Numerous prior research studies have outlined the strong negative association between EDI and individual relationship satisfaction (Atkins, Baucom & Jacobson, 2001; Buss & Shackelford, 1997; Glass & Wright, 1992). However, no studies have investigated the subgroups ofEDI (none, known and unknown) and the systemic effects of disclosure or deception on overall relationship satisfaction in a community sample. Contrary to prior research fmdings, known affairs were more common than hidden affairs in this sample. This is likely due to the restrictive definition of unknown for the purposes of this study. In order to be classified as unknown, all ED Is in the relationship, past and current, had to be concealed. For the current sample, there was a significant difference between the subgroups of EDI (none, known and unknown) and relationship satisfaction. Couples with at least one known EDI had significantly greater relationship distress then couples with no ED I. Couples where one partner has engaged in an unknown EDI endorsed more distress than couples with no EDI, but reported greater satisfaction than couples with known EDI; however, neither of these differences were statistically significant. 29


This outcome was not predicted given the findings of Atkins, Eldridge, Baucom & Christensen (2005). In their clinical sample, unknown EDI was associated with the greatest relationship distress. However, one can imagine that awareness of a serious transgression such as infidelity could result in the greatest relationship distress. For couples with unknown EDI, the lack of awareness of the infidelity may simply preserve some of the overall relationship satisfaction, but is not entirely protective of other relational dysfunctions, such as negative communication patterns. In this community sample, known affairs, as opposed to unknown affairs, predicted the greatest relationship distress overall. Another central goal of the current study was to investigate the difference between the subgroups ofEDI (none, known, and unknown) and demand/withdraw conflict communication. Negative communication patterns have been described as one of the predictors for infidelity (Allen et al., 2008), but there is no research evaluating the association between infidelity and the demand/withdraw pattern, one of the most widely studied and well characterized negative communication patterns Additionally, asymmetrical relational needs for intimacy versus needs for autonomy have been implicated in both research on demand/withdraw (Caughlin & Vangelisti, 2000; Christensen & Heavey, 1993; Christensen & Shenk, 1991) and in research on motivations for EDI (Allen & Baucom, 2004). In the current study, there was a significant difference between the level of couple engagement in demand/withdraw and EDI. Couples with an unknown EDI 30


had a significantly greater level of demand/withdraw when discussing their top problem area when compared to couples with no ED I. In relationships where there was a known ED I, the level of demand/withdraw was higher than the couples with no ED I, but lower than the couples with an unknown EDI; however, these differences were not significant. Couples in this sample with a known EDI may have more adaptive ways of communicating during conflict, skiJis which helped them remain together despite infidelity, yet these couples still do not have the level of adaptive conflict communication skills possessed by couples with no ED I. The relational influence model, developed by Caughlin & Vangelisti, 2000, may aid in explaining the association between unknown EDI and heightened levels of demand/withdraw. Individuals who have both engaged in an EDI and chose to hide their engagement from their partner may possess individual characteristics that also put them at increased risk for engaging in the demand/withdraw pattern in communication. The common characteristic described in the prior literature on deception, EDI and demand/withdraw is the desire for intimacy versus the need for autonomy. The individuals who chose to conceal their infidelity may also be those individuals who are motivated by the need for autonomy or intimacy to commit the EDI. For example, clinical authors have noted that a person with high needs for autonomy may go outside of their primary relationship as a way to get some "freedom" from the relationship; moreover, maintaining the secret allows them to have something all their own, separate from their partner, that can similarly satisfy 3)


autonomy needs (Scarf, 1987). In contrast, persons high in intimacy needs may go outside of their relationship to fulfill these needs, yet be weary that revealing the infidelity to their partner will result in abandonment and thus chose to keep the extramarital relationship a secret (Allen, 2001). Asymmetry in intimacy/autonomy needs theoretically exacerbates the level of demand/withdraw in the relationship. Future research could investigate the possible differential motivations for individuals engaged in known versus unknown EDI to determine if unique motivational characteristics exist between these two subgroups. The overall findings for the subgroups ofEDI (none, known and unknown) on demand/withdraw and relationship satisfaction is counterintuitive. Curiously, couples with unknown EDI have the highest level of demand/withdraw in their conflict communication, yet they are not the least satisfied. Instead, the couples with known EDI in their relationship were the most dissatisfied overall. Based on the prior research on the association of demand/withdraw and relationship satisfaction, this result would not have been predicted. However, demand/withdraw may only account for a small amount of the variance in couple level relationship satisfaction when compared to EDI occurrence. The final objective of the current study was to assess if the occurrence of an EDI could predict relationship satisfaction above and beyond the demand/withdraw interaction pattern. In this sample, the presence or absence of EDI predicted couple level relationship satisfaction above and beyond the effects of demand/withdraw. 32


For the overall model, demand/withdraw and EDI (present or not present) predicted 14% of the variance in relationship satisfaction. In the variables assessed, EDI (present or not present) accounted for an additional 10.5% of the variance above and beyond demand/withdraw. It is important to note that the married couples in this community sample reported higher satisfaction levels when compared to couples in prior marital research studies. This higher level of satisfaction could, in part, be due to the type of recruitment utilized in the current study. Explicitly advertising for couples interested in participating in a study on relationships may attract couples with a higher level of relationship functioning, as opposed to couples who are dissatisfied in their relationship. Additionally, the levels of demand/withdraw were relatively low for the current sample. It is not unusual for couples who are relatively more satisfied to also have less maladaptive communication patterns (Eldrige et al., 2007). Despite these sample characteristics, significant differences were found between EDI and demand/withdraw, and EDI and relationship satisfaction. Future studies could evaluate these same variables in a distressed sample to determine if the differences become even more pronounced. These research fmdings aid in expanding the literature on infidelity and its systemic effects on relationship functioning, including satisfaction and conflict communication. Now that an association between EDI and demand/withdraw has been found, specifically for the unknown EDI group, future research could 33


investigate the association between demand/withdraw and EDI on the individual level. Evaluation on the individual level will permit specific investigation of the association between demanding or withdrawing behavior in conflict and the partner participating in the infidelity, the participating partner (PP). Additionally, one could examine the association between demanding or withdrawing behavior and the partner being betrayed, the injured partner (IP). This would provide insight regarding the individual characteristics of the persons who participant in infidelity and how these characteristics influence their own maladaptive communication patterns in the relationship, consistent with the self influence model (Caughlin & V angelisti, 2000). It would also potentially demonstrate gender differences in demanding or withdrawing behaviors for the PP or the IP. This study also begins to explore the deception of infidelity and its unique association with relationship functioning. Future research could investigate individual relationship satisfaction for the couples with known and unknown ED I. There may be a differential in satisfaction for the PP versus the IP in the unknown infidelity subgroup. If a difference does exist, this might help explain why couple level satisfaction for the unknown EDI subgroup bisects the difference between the couples with no EDI and known EDI. For the unknown EDI subgroup, the PP may be deflating the satisfaction score (e.g., engaging in infidelity due to dissatisfaction in the relationship) while the IP is inflating the score (e.g., "blissfully" unaware of the betrayal), but an average of their individual satisfaction scores places them the 34


midrange for satisfaction. Furthermore, evaluation of satisfaction on the individual level may reveal gender differences in reported satisfaction for the PP or the IP. Limitations The current study contains various limitations. The first limitation is that the data collected is cross-sectional. Cross-sectional data only permits evaluation of demand/withdraw communication following infidelity, and does not indicate whether the demand/withdraw communication pattern preceded the occurrence of the infidelity. Using longitudinal data, one could evaluate if individual characteristics in demand/withdraw communication (being the demander or withdrawer) could serve as early indicator of risk for later infidelity. Another possible limitation is the generalizability of the research findings. The current sample had slightly lower income, greater relationship satisfaction and lower levels of demand/withdraw in their conflict communication compared to prior samples of married couples. The current sample is representative of diverse ethnic/racial groups, this serves as a strength in many respects as these groups are often under studied, but the findings based on this sample may not generalize to the population at large. Replication analyses would be necessary to confirm these findings in a sample that is more representative of the general population. Self-selection factors are another limitation of this study. The findings specific to the known EDI subgroup should be interpreted with caution. This group of individuals may not be representative of all couples who experience a known 35


ED I, as they are a group that has remained together despite a serious transgression and have agreed to participant in a study on romantic relationships. These two factors alone imply that couples in the known group possess unique characteristics. Summary Negative communication patterns and infidelity have been topics that have individually attracted a vast amount of research interest, but until now, these topics had never been concurrently investigated. By examining multiple areas of relationship functioning, one can gain greater insight into the complexity of romantic relationships. Often times, there exists multiple layers of dysfunction in dissatisfied relationships. This study evaluated various areas of functioning on a systemic level to determine if there was an overall effect of infidelity and deception on couple level relationship satisfaction and demand/withdraw. Future research could delve further into the particular individual characteristics that may manifest into these maladaptive behaviors (infidelity, deception and demand/withdraw), in order to clearly illustrate the individual characteristics that contribute to multiple layers of dyadic dysfunction. 36


REFERENCES Allen, E. S. (2001 ). Anachment styles and their relaJion to patterns of extradyadic and extramarital involvement. Unpublished doctoral dissertation, University of North Carolina, Chapel Hill. Allen. E. S., Atkins, D., Baucom, D. H., Snyder, D., Gordon. K. C., & Glass, S. P. (2005). Intrapersonal, interpersonal, and contextual factors in engaging in and responding to extramarital involvement. Clinical Psychology: Science and Practice, 12, 101-130. Allen, E. S., & Baucom, D. H. (2004). Adult attachment and patterns of extradyadic involvement Family Process, 43(4), 467-488. Allen. E. S., Rhoades, G. K., Stanley, S.M., Markman. H. J., Williams, T., Melton. J., & Clements, M. L. (2008). Premarital precursors of marital infidelity. Family Process, 47(2), 243-259. Atkins, D.C., Baucom, D.H., & Jacobson. N.S. (2001). Understanding infidelity: Correlates in a national random sample. Journal of Family Psychology, 15, 735-749. Atkins, D.C., Eldridge, K.E., Baucom, D.H., & Christensen, A. (2005). Behavioral marital therapy and infidelity: Opti-mism in the face of betrayal. Journal of Consulting and Clinical Psychology, 73, 144-150. 37


Bums, A. (1984). Perceived causes of marriage breakdown and conditions of life. Journal of Ma"iage and the Family, 46(3), 551-562. Buss, D. M., & Shackelford, T. K. ( 1997). Susceptibility to infidelity in the first year of marriage. Journal of Research in Personality, 31, 193-221. Caughlin, J.P., & Vangelisti, A. L. (2000). An individual difference explanation of why married couples engage in the demand/withdraw pattern of conflict. Journal ofSocial and Personal Relationships, /7(4-5), 523-551. Choi, K. H., Catania, J. A., & Dolcini, M. (1994). Extramarital sex and HIV risk behavior among US adults: Results from the National AIDS Behavioral Survey. American Journal of Public Health, 84(12), 2003-2007. Christensen, A., & Heavey, C. L. (1993). Gender differences in martial conflict: The demand/withdraw interaction pattern. In S. Oskamp & M. Costanzo (Eds. ), Gender issues in contemporary society (pp. 113-141 ). Newbury Park, CA: Sage. Christensen, A., & Heavey, C. L. (1990). Gender and social structure in the demand/withdraw pattern of marital conflict. Journal of Personality and Social Psychology, 59(1), 73-81. Christensen, A., & Shenk, J. L. (1991). Communication, conflict, and psychological distance in nondistressed, clinic, and divorcing couples. Journal of Consulting and Clinical Psychology, 59(3), 459-463. 38


Clements, M. L., Stanley S.M., & Markman H. J. (2004). Before they said "I Do'': Discriminating among marital outcomes over 13 years. Journal of Marriage and Family, 66, 613-(,26. Cole, T. (2001), Lying to the one you love: The use of deception in romantic relationships. Journal ofSocial and Personal Relationships, /8(1), 107-129. DePaulo, B. M., Ansfield, M. E., Kirkendol, S. E., & Boden, J. M (2004). Serious lies. Basic and Applied Social Psychology, 26(2&3), 147-167. Eldridge, K. A., Sevier, M., Jones, J., & Atkins, D. C. (2007). Demand-withdraw communication in severely moderately distressed, and non distressed couples : Rigidity and polarity during relationship and personal problem discussions. Journal of Family Psychology, 2/(2), 218-226. Fats-Stewart, W., Birchler, G. R., Hoebbel, C., Kashdan, T. B., Golden, J., & Parks, K. (2003). An examination of indirect risk of exposure to HIV among wives of substance abusing men. Drug and Alcohol Dependence, 70, 65-76. Fincham, F. D., Harold, G. T., & Gano-Phillips, S. (2000). The longitudinal association between attributions and marital satisfaction: Direction of effects and role of efficacy expectations. Journal of Family Psychology, 14 (2), 267-285. Glass, S. P., & Wright, T. L. (1992). Justifications for extramarital relationships: The association between attitudes, behaviors and gender. The Journal of Sex Research, 29(3), 361-387. 39


Gottman, J. M., & Levenson, R. W. (2000). The timing of divorce: Predicting when a couple will divorce over a 14-year period. Journal of Marriage and the Family, 62(3), 737-745. Heavey, C. L., Layne, C. & Christensen, A. (1993). Gender and conflict structure in marital interaction: A replication and extension. Journal of Consulting and Clinical Psychology, 61(1), 16-27. P. J. (2001). Deceptive message intent and relational quality. Journal of Language and Social Psychology, 20, 214-230. Marchand, J. F., & Hock, E. (2000). Avoidance and attacking conflict-resolution strategies among married couples: Relations to depressive symptoms and marital satisfaction. Family Relations, 49,201-206. Markman, H.J., Floyd, F. J., Stanley, S.M., & Storaasli, R. D. (1988). Prevention of marital distress: A longitudinal investigation. Journal of Consulting and Clinical Psychology, 56(2), 210-217. National Center for Health Statistics, U.S. Dept. of Health and Human Services. (2004). Marriage and Divorce Rates, 1920-2002. World Almanac & Book of Facts. Norton, R. (1983). Measuring marital quality: A critical look at the dependent variable. Journal of Marriage and the Family, 45, 141-151. 40


Notarius, C. I., & Vanzetti, N. (1983). Marital Agendas Protocol. In E. Filsinger (Ed.), Marriage and family assessmenJ: A sourcebook for family therapy (pp. 209-227). Beverly Hills, CA: Sage. Peterson, C. ( 1996). Deception in intimate relationships. lnlernational Journal of Psychology, 3/(6), 279-288. Ridley, C. A., Wilhelm, M.S., & Surra, C. A. (2001). Married couples' conflict responses and marital quality. Journal of Social and Personal Relationships, 18(4), 517-534. Raley, R. K., & Bumpuss, L. (2003). The topography of the divorce plateau: Levels and trends in union stability in the United States after 1980. Demographic Research, 8(8), 245-260. Rubin, R. 8., Palmgreen, P., & Sypher, H. E. (1994). Communication research measures. New York: Guilford Press. Scarf, M. (1987). Intimate partners: Patterns in love and ma"iage. New York: Random House. Sevier, M., Simpson, L. E., & Christensen, A. (2004). Observational Coding of Demand-Withdraw Interactions in Couples. In Kerig, P. K. & Baucom, D. H. (Eds. ), Couple Observational Coding Systems (pp. 159-171 ). Routledge. Strout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin, 86(2), 420-428. 41


Vogel, D. L., Murphy, M. J., Werner-Wilson, R. J., Cutrona, C. E., & Seeman, J. (2007). Sex differences in the use of demand and withdraw behavior in marriage: Examining the social structure hypothesis. Journal of Counseling Psychology, 54(2), 165-177. Weger, H. (2005). Disconfirming communication and self-verification in marriage: Associations among the demand/withdraw interaction pattern, feeling understood and marital satisfaction. Journal of Social and Personal Relationships, 22( 1 }, 19-31. Whisman, M. A., Dixon, A. E., & Johnson, B. (1997). Therapists' perspective of couple problems and treatment issues in couple therapy. Journal of Family Psychology, 11(3), 361-366. 42