Conflict Behaviors and Marital Satisfaction in Older Adulthood: A Typology by Leah Megan Williams A thesis submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Master of Science Auburn, Alabama August 2, 2014 Keywords: older adults, conflict, marital satisfaction, typology Copyright 2014 by Leah Megan Williams Approved by Amy J. Rauer, Chair, Assistant Professor of Human Development and Family Studies Thomas A. Smith, Jr., Associate Professor of Human Development and Family Studies Scott Ketring, Associate Professor of Human Development and Family Studies ii Abstract Research has generally supported the idea that older adult conflict resolution is different from that of younger adults and that more effective conflict resolution skills are positively associated with marital satisfaction at the aggregate level. However, it remains unclear how maritally satisfied older couples differ in their handling of conflict. Addressing this gap in the literature, the current study uses a person-centered typology approach to better understand the different behaviors that comprise conflict management among satisfied older adult couples. Using observed, behavioral measures of conflict resolution with self-reported measures of marital satisfaction, this study creates a typology of conflict resolution behaviors and their relationship with marital satisfaction in a sample of older couples (N = 64). Results provided support for the hypothesized associations. Four clusters emerged: Problem Solvers, Warm Couples, Even Couples, and Cool Couples. Clusters varied on problem solving skills, warmth, hostility, and denial, and these differences were linked to significant differences in spouses? marital satisfaction. Potential explanations of these findings and future directions are provided. iii Acknowledgments I would like to first thank Dr. Amy Rauer for her constant patience and guidance on this thesis along with Dr. Thomas Smith and Dr. Scott Ketring for their considerable contributions. I would also like to thank Jakob Jensen and the MARS team for their assistance. Finally, I would like to thank Garland and Kathy Williams, John Abrusci, Becca Williams, and Meg Heinicke for their endless patience, enthusiasm, and encouragement. iv Table of Contents Abstract ......................................................................................................................................... ii Acknowledgments ....................................................................................................................... iii List of Tables ............................................................................................................................... vi List of Figures???????????????????????????????...vii Introduction ................................................................................................................................. 1 Literature Review ........................................................................................................................ 4 Theoretical Background .................................................................................................. 4 Conflict ........................................................................................................................... 6 A Person Oriented Approach to Understanding Conflict ............................................... 8 The Current Study ......................................................................................................... 11 Method ...................................................................................................................................... 13 Participants .................................................................................................................... 13 Procedure ...................................................................................................................... 14 Measures ....................................................................................................................... 14 Plan of Analysis ............................................................................................................ 19 Results ....................................................................................................................................... 22 Discussion ................................................................................................................................. 28 References ................................................................................................................................. 38 Appendices ................................................................................................................................ 51 v Appendix A - Questionnaire Measures ......................................................................... 52 Appendix B ? Conflict Task ........................................................................................ 54 Appendix C ? Conflict Coding System ........................................................................ 60 Appendix D-Extended Literature????????????????????? .74 vi List of Tables Table1. Correlations and Distributional and Scale Properties of Independent and Dependent Variables............................................................................................................................47 Table2. Correlations and Distributional and Scale Properties of Cluster Variables??????????????????????????????? 48 Table3. Means of the 2 (spouse) x 4 (cluster) Repeated Measures Analysis of the Marital Clustering Variables????... ............???????????????????49 vii List of Figures Figure1. Clustering variables by cluster?????????????????????? 50 1 Introduction Conflict in marriage is inevitable (Zeidner & Kloda, 2013) but changeable (Holly, Haase & Levenson, 2013). Unfortunately, for those couples who are unable to resolve their conflict either on their own or through the help of others, marital distress often arises (Karney & Bradbury, 1995). Conflict is an ever-present and damaging issue in many marriages. In light of the potential damage conflict can do, not only to the marriage, but to the well-being of those both within the marriage and the family (Coln, Jordan, & Mercer, 2012; Fincham & Beach, 1999; Kiecolt-Glaser & Newton, 2001), it is not surprising that millions of dollars in government funds have been spent on trying to understand the nature of marital conflict, its antecedents, and its consequences (Hawkins, Amato, & Kinghorn, 2013). ). To enhance the efficacy of government efforts, it is important to not only study those who fail to resolve marital conflict, but also the couples who appear to handle conflict better (Weiss & Heyman, 1997). Although conflict can be damaging and hurtful, conflict can also be functional, depending on how it is carried out (Gottman & Krokoff, 1989). Conflict that includes wives?, but not husband?s, positive verbal behavior and expressed compliance as well as conflict discussions that do not include defensiveness, stubbornness, and interaction withdrawal tend to be functional (Gottman & Krokoff, 1989). By studying those who are able to handle their conflicts successfully, it may become clearer what conflict management skills can be taught to couples to assist them in having functional conflict over the course of their marriage. 2 One population that might be especially pertinent to study in this regard is older couples. Research has found that older couples have different conflict resolution strategies that have been found to be beneficial for their marriage. For example, older couples tend to express lower levels of anger, disgust, belligerence, and whining and utilize more affection during their conflict discussions (Carstensen, Graff, Levenson, & Gottman, 1996). Socioemotional selectivity theory suggests that older adults are influenced by time and their own mortality, which can impact the role that conflict plays in their relationship and the way that it may be handled (Carstensen et. al., 1996; Carstensen, Pasupathi, Mayr & Nesselroade, 2000). Birditt and Fingerman (2005) suggest that as a result of these developmental changes in perception of time, older adults are compelled to use loyalty strategies in handling conflict, like waiting to see if things improve, whereas younger adults tend to use active destructive, or exit, strategies. There are many factors that make up conflict resolution in older adulthood including partner collaboration, emotion regulation, and context of the problem as well as the loyalty and exit strategies mentioned above (Birditt & Fingerman, 2005; Hoppman & Blanchard-Fields, 2011). With all of these factors impacting conflict resolution, it is likely that each couple will handle conflict differently, employing different strategies that have proven successful in years past, creating a great deal of potential variability in the conflict experiences of couples (Hoppman & Blanchard-Fields). Traditionally, however, studies examining conflict in older adulthood have taken a variable-centered approach, which provides only a snapshot of the aggregate experience of conflict in older adulthood (Peter-Wight & Martin, 2011; Hoppman & Blanchard-Fields, 2011). Although this approach provides valuable information about the antecedents of conflict for this population and creates a solid foundation from which to work, it does not allow us to account for the potential differences in the ways couples handle conflict. In 3 order to identify these differences between couples, it is important to move beyond the variable- centered approach on conflict to take a more person-centered, dyadic approach. Previous studies using this type of approach have found meaningful differences between couples earlier in the lifespan in terms of understanding what makes couples happy (Kamp Dush & Taylor, 2011; Rauer & Volling, 2013), revealing the inherent diversity of couples? relationships and experiences. This study will utilize a similar approach in order to identify if there are meaningful differences between older couples in observations of problem-solving in a sample of satisfied older adults. Accordingly, the current study seeks to understand conflict in older adult couples in intact, high-functioning marriages by examining potential variability across couples and how different conflict approaches are linked with husbands? and wives? marital satisfaction. To accomplish this, we will utilize observational data of conflict discussions collected from 64 older couples. The hope is that by using a behavioral, person-centered, dyadic approach to study joint problem solving discussions in this population, we can gain understanding of the different strategies successful older couples use to help conflict become less detrimental over time within marriages. By understanding this, we may be better able to help younger couples understand how to manage conflict within their relationships in order to preserve marital satisfaction throughout the early years of the marriage and continuing into older adulthood. 4 Literature Review Theoretical Framework Behavioral theories of marriage suggest that the key to understanding couples? functioning lies in the exchanges of behaviors between partners, rather than in the intrapersonal weighing of benefits and alternatives in the relationship (Karney & Bradbury, 1995). In essence, behavioral theory states that positive behaviors enhance the evaluation of the marriage, whereas negative behaviors are detrimental to the evaluation of the marriage (Karney & Bradbury; Weiss, 1984). For conflicts, behavioral theories posit that negative interactions between spouses build up over time and that these negative interactions both detract from and erode positive interactions (Karney & Bradbury). Conversely, positive interactions that build up between couples serve as protective factors for the relationship by preventing occasional negative interactions from negatively impacting overall evaluations of the marriage. Behavioral theory has its strength in suggesting that perceptions of one another alter how subsequent interactions between partners will take place (Karney & Bradbury, 1995). Specifically, Gottman (1990) suggests that happy couples who have positive interactions view positive experiences as lasting traits within their marriage and, in contrast, see negative experiences as temporary and situational. This phenomenon is described as a positive ?halo effect? (p.78). In action, happy couples who feel generally loved and respected and who experience a drop in marital satisfaction find ways to cope by doing things that they enjoy 5 together and things that bring them closer, therefore increasing positive interactions and stabilizing marital satisfaction (Gottman). This is not to suggest that happy couples do not experience negativity or negative affect. Gottman (1999) finds that couples who are happier tend to experience negative affect and negative interactions; however, during conflict, they experience at least five times as many positive interactions as negative interactions and tend to avoid four particularly severe negative interactions including: criticism, defensiveness, contempt and stonewalling. Alternatively, unhappy couples experience a negative halo effect where positive interactions are seen as situational and temporary and negative interactions are seen as lasting traits within their marriage. Additionally, unhappy couples experience more frequent instances of criticism, defensiveness, contempt, and stonewalling. A criticism of behavioral theories is that they do not account for variations in marital stability (Karney & Bradbury, 1995). For example, if negative interactions beget negative interactions, and this impacts marital satisfaction, why do some couples divorce after a few years of repeated negative interactions between them lead to destructive conflict patterns, whereas other couples, utilizing the same harmful patterns, stay together? Behavioral theory has no answer for this outcome. Although behavioral theories do not tell the entire story of how marital interactions and marital satisfaction are formed, they provide a basis for how patterns of interactions between couples may develop. Oftentimes variable-centered approaches are used in these kinds of interactions capture the aggregate experience and relationship between variables. Using a person-centered approach that looks at the relationship between variables within a person or couple, as this study does, may begin to address this gap by accounting for variations in marital satisfaction that are otherwise unaccounted for by behavioral theory by obtaining a 6 clearer picture of different configurations of couple interactions rather than solely focusing on experiences at the aggregate level. Conflict Conflict plays a vital role in relationships. Although it can be detrimental to the quality of the marriage, to spouses? mental and physical health, and the health of the family (Fincham & Beach, 1999; Kiecolt-Glaser & Newton, 2001), Gottman (1999) demonstrated that it also has an important function within the marriage depending on what type of conflict the couple is dealing with and how. For example, Gottman (1999) found that many of the couples he studied were constantly arguing about the same issues, which he termed ?perpetual problems?. For conflict, especially perpetual conflict, basic problem-solving skills were not effective because many times these issues were related to differences in personality or ?needs that were fundamental to their core definition of self? (Gottman, p.56). Rather, it was important to establish dialogue about these issues (Gottman). However, although this can make conflict appear to be a looming and destructive event, this is not always the case. Gottman and Krokoff (1989) found that some types of conflict are actually functional in marital relationships provided the interactions do not include stubbornness, defensiveness, and/or withdrawal from the interaction. Furthermore, they posited that conflict avoidant couples may, in fact, be at risk over time for marital dissolution due to not feeling as though they, as a couple, can weather conflicts together. Consistent with Gottman and Krakoff?s findings, Kerig (1996) also finds that conflict can be functional. Some couples who engage in conflict and express their feelings experience subsequent feelings of relief. Kerig makes the point that resolution of problems is the greatest predictor of couples? satisfaction with their conflict management strategies and only differences that are revealed may be solved, making conflict necessary to resolve issues. If conflict in and of itself is not uniformly 7 problematic, it is likely that what makes conflict functional are the ways that the couple engages in conflict. Birditt and Fingerman (2005) identified a wide variety of behaviors that couples utilize when engaging in conflict that appear to differentially affect their marital satisfaction. These behaviors ran the gamut from name-calling and arguing, to listening and discussing the problem, to avoiding the situation and person, to doing nothing and letting the situation blow over (Birditt & Fingerman, 2005). As previously mentioned, Birditt and Fingerman (2005) found that older adults tended to use more behaviors like doing nothing and letting the situation blow over whereas younger adults tended to argue. Consistent with these findings, in observing both couples in middle adulthood and those in later adulthood, Carstensen, Gottman, and Levenson (1995) found that, even when controlling for severity of marital problem, older adults tended to display less negative affect and more affection during conflict discussions than their younger counterparts. Specifically, they found that younger adults tended to display considerably more anger, belligerence, whining, and disgust than did older adults. The differences between the way that older adults handle conflict in comparison to younger adults are likely attributable to a number of reasons. Some of these include the idea that older adults have better emotion regulation skills and less frequent contact with their social partners, which provides them more time to calm down when issues do arise and employ more constructive strategies (Birditt & Fingerman, 2005). Another possibility is that older couples follow a behavioral sequence where, in discussing marital problems, one spouse?s neutral affect was not met with the other spouse?s subsequent negative affect, an interaction described by the authors as negative start-up, which then helps prevent negative escalation (Carstensen, Gottman, & Levenson, 1995). The authors suggest that listener positivity or neutrality is instrumental in 8 helping to regulate conflict and that maritally satisfied older couples may have attained some level of control over negative affect, as evidenced by greater instances of humor, affection, and validation as well as a lack of negative affect continuance (negative affect following negative affect) during conflict discussions that can lead to negative affect escalation (Carstensen et al., 1995). Using positivity and controlling negative escalation is associated with more than just age. In conjunction with age effects, more collaborative problem solving has been associated with marital satisfaction (Carstensen et al., 1995). The authors found that older couples with higher levels of marital satisfaction tend to utilize collaborative problem solving and show more instances of being positive or neutral toward one another as well as engaging in more humorous, affectionate and validating interactions. This differs from unhappy, dissatisfied older couples as dissatisfied couples tend to have a higher expression of negative affect. Additionally, they showed more anger, belligerence, contempt, dominance, and sadness than their happier counterparts. Thus, within a sample of satisfied older adult couples, both collaborative problem solving and behaviors such as doing nothing and letting the situation blow over appear to be viable options for conflict management. This suggests variability exists in the experiences of older adult couples and using a typology approach which seeks to identify similar groups of couples will allow for the differences between these couples to be noted. A Person-Oriented Approach to Understanding Conflict Typology approaches have been used to successfully identify a variety of different couples. Fitzpatrick (1988) first used this approach to say that there are many different types of satisfied couples. In her study, three types of satisfied middle-aged couples emerged and each handled conflict differently. The first of the three types was the traditional couples, who tended 9 to have stereotypical roles in marriage and place a higher value on developing a sense of ?we- ness? rather than individual goals. Traditional couples tended to avoid conflict on the whole, but would argue about important topics in their relationship. The second type that emerged was the independent couples who tended to have more egalitarian marriage roles and value individual privacy. Independent couples, alternatively, thrived on conflict and openly expressed disagreement. The third type of couple to emerge was the separates, who tended to value separateness and autonomy as well as have a low level of sharing and companionship. Separate couples typically avoided all marital conflict. These three categories of couples show how conflict can be handled in vastly different manners even among uniformly happy couples. Focusing on a group of couples that had more diversity in their marital functioning, Gottman (1993) used a similar typology approach with couples in their early to mid 30s participating in a conflict task. Similar to the current study, Gottman coded fifteen minute videos of spouses engaging in discussions, including discussions on conflict, using several different coding systems for specific behaviors. From this, five types of couples emerged with very different ways of approaching conflict. Of these five types, three of them were considered regulated or stable types with more positive interactions, and two were considered unregulated or unstable types with more negative interactions. Starting with the stable couples, the three types that emerged were validators, volatiles, and avoiders. These three types were comparable in their approach to conflict to those that Fitzpatrick (1988) identified years previously. Fitzpatrick?s traditional type was compared to Gottman?s validator type, whereas the independents were comparable to the volatile type, and the separates were comparable to the avoider type. Of the unstable, unregulated couples the two types that emerged were the hostile and the hostile/detached. The hostile type was characterized by a great deal of conflict accompanied by 10 defensiveness. The hostile/detached type, on the other hand, was characterized by spouses who were emotionally uninvolved with one another, but tended to experience brief occurrences of reciprocated attack with defensiveness over seemingly inconsequential matters. Gottman was able to expand on Fitzpatrick?s types of satisfied couples, while expanding to those couples who are not satisfied. Given the unique nature of conflict in older adults, the typology approach needs to be expanded to a sample of older adults that may approach conflict in different ways. Using a typology approach not based on Fitzpatrick (1988)?s model, Rauer and Volling (2013) found three different types of couple during a conflict task within a sample of happy, high-functioning couples in middle adulthood akin in functioning to the sample used in the current study. The first of the three types that emerged were the mutually engaged couples. These couples were characterized by equal expression of positive problem-solving behaviors and support. Additionally, this type used significantly more negative problem-solving behaviors than did the other two types across both spouses. The second type to emerge was the mutually supportive couples. This type was characterized by both spouses using a great deal of positive problem-solving and support along with a distinct lack of negative problem-solving behaviors. The final type to emerge was the wife compensation couples. Unlike the previous two types, this type was not characterized by mutual couple components, but rather by wives, alone, displaying high positive problem-solving behaviors and low negative problem-solving behaviors in comparison to their husbands. With the wide variety of behaviors found in conflict management, there is potential for a great deal of variability in the experiences of different couples. As evidenced by these studies, there may be meaningful differences in behaviors among satisfied older adult couples during conflict discussions that can only be revealed using a typology approach. 11 Current Study In spite of the large body of work pertaining to conflict, there remains a need to better understand the behavioral processes underlying this phenomenon among certain populations. Although it is understood that conflict can be functional (Kerig 1996), and researchers have looked at conflict behaviors and how conflict may differ between younger adults and older adults (Birditt & Fingerman, 2005; Carstensen, Gottman, & Levenson, 1995), there is little research capturing the conflict experiences of maritally satisfied, older couples. Typology approaches have been utilized to look at conflict in younger and middle adulthood (Fitzpatrick, 1988; Gottman, 1993; Rauer & Volling, 2013), but not older adult couples. The current study seeks to help fill this gap and find differences among maritally satisfied, older couples in how they handle several dimensions of conflict including problem solving skills, dominance, denial, sensitivity, balance, as well as both positive and negative affect. We further examine how differences in the utilization of these dimensions may be differentially related to marital satisfaction. In order to capture these differences, the current study will use a typology approach similar to those described previously, using cluster analyses to first separate the couples into types dependent on the dimensions of conflict used and then to see potential differences in marital satisfaction amongst them. Given the high-functioning nature of our sample, it is predicted that although all couples will demonstrate high levels of positive affect (Gottman, 1999), there will be key differences amongst these couples. I hypothesize that three couple types will emerge: (1) couples characterized by greater conflict, (2) couples who are mainly sensitive to one another and validate each other?s feelings, and (3) couples who refuse to argue and avoid the conflict discussion. Consistent with the idea that this sample is full of satisfied couples, it is hypothesized 12 that marital satisfaction will not differ significantly between the types that emerge, showing that there are many ?right? ways to handle conflict later in life. 13 Methods Participants Sixty-four married heterosexual couples were recruited as part of a larger study examining how problem solving and other aspects of marriages were related to well-being, both individual and marital, in older adulthood. Participants were recruited locally through newspaper advertisements, healthcare agencies, churches, and other community organizations in the Southeast United States. To participate, couples needed to be (1) married, (2) retired or partially retired (working less than 40 hours per week), and (3) at the level of health and functioning that they were able to drive to the on-campus research center. The health and functioning requirement was put in place in order to capture those older adults who are not yet in the role of needing or being spousal caregivers. Husbands and wives were approximately 71 (SD = 7.4) and 70 years old (SD = 7.0) respectively, and primarily European American (n = 61 and n = 60). Fifty-one (80%) of couples were in their first marriage and couples had been married for 42.4 years, on average (SD=15). The couples were highly educated; with 60 (94%) of the husbands and 54 (84%) of the wives attending college. Couples? average annual income was $86,000 (SD = $64,000) with an average total wealth (i.e., property, pensions, IRAs, and income) of $1,100,000 (SD = $1,300,000). Couples were either fully retired (n = 47; 73%) or one spouse was currently working for pay (n = 17). On average, the couples reported having 2.6 children (SD = 1.3; range = 0-6). 14 Procedure Couples came to an on-campus research facility for a visit that lasted 2-3 hours. During the visit, couples were invited to participate in a variety of marriage-related communication tasks. Among these were a relationship narrative task, a baseline picture-viewing task, a problem-solving task, a compassionate love task, and a support task. The 15-minute problem- solving task is the focus of the current study. Couples each received a list of common marital problems (ex: needing to be more organized, wanting more independence in the relationship, or wanting to have sex more or less often) and were asked to independently rate what they felt the biggest problems in their relationship are. The couples were then asked to discuss these issues with one another, identify an issue together that they wanted to work on, and to come up with a workable solution to the problem. At the end of the visit, couples received questionnaires that assessed individual and marital functioning. Couples were compensated $75 for their participation once their questionnaire was received via a pre-addressed, pre-stamped envelope. Measures Marital satisfaction. Marital satisfaction was measured using the Marital Satisfaction Questionnaire for Older Persons (Haynes et al., 1992). This 24-item questionnaire assessed standard topics associated with satisfaction like conflict management (e.g., ?How satisfied are you with the way disagreements are settled?? and ?How satisfied are you with the number of disagreements between you and your spouse??) as well as topics that are more developmentally tailored to older adults (e.g., ?How satisfied are you with your spouse?s physical health??). Questions were scored on a variety of scales ranging from one to four, one to five, or one to six. On all the scales, the lower end, or one, was a dissatisfied or negative response, whereas the 15 higher end (four, five, or six) depending on the scale, was a very satisfied or positive response. Reliability for this scale was .93 for husbands and .93 for wives in the current study. Conflict. Conflict discussions were later coded by trained observers on ten topics of interest, with each spouse coded separately and by different coders. Scores ranged from 1 to 7. There were both individual codes and dyadic codes for each couple. These codes are a slightly modified part of the larger Interactional Dimensions Coding System-Problem Discussion or IDCS-PD, a system designed to assess how couples interact with one another during discussions surrounding problem areas in their relationship (Kline, Julien, Baucom, Hartman, Gilbert, Gonzalez, & Markman, 2004). The IDCS-PD has been chosen for this study due to its global design in assessing behavior over lengthy conflict discussions as well as its two-part system utilizing both content and affect codes (IDCS; Kline et al.). Codes have been altered from a nine point scale to a seven point scale to match other scales utilized in the current study. A codebook has been created for use in training. For reliability, two coders were trained on a subsample of tapes until interobserver agreement was 80% or higher. Reliability was calculated via intraclass correlation on 20% of the taped conflict interactions (13 of 64 tapes). To note, given the low occurrence of many of these behaviors, we had extremely high reliability for many of these codes. Individual codes. First, individual positive affect was observed, which referred to the degree to which each spouse responded positively towards the other. A score of 1 (?extremely uncharacteristic?) indicated no positive affect during the interaction. These individuals did not smile or laugh and did not seem to enjoy the interaction. A score of 4 (?somewhat characteristic?) reflected moderate positive affect in which the spouse expressed low-level enjoyment or positive affect towards the other although these displays were not intense or 16 prolonged. There were frequent and somewhat prolonged lapses in the individual?s positive expressions. Just as often as there were positive displays, there were also pauses. A score of 7 (?extremely characteristic?) denoted continuous individual positive affect. The spouse was thoroughly enjoying the interaction with no noticeable delays in positive affect. Interrater reliability for wives? positive affect was r =1.0. Interrater reliability for husbands? positive affect was r = 1.0. Second, individual negative affect was observed, which referred to the degree to which each spouse responded negatively toward one another, including behaviors ranging from frowning to hostile tone. A score of 1 (?extremely uncharacteristic?) indicated no negative affect during the interaction. A score of 4 (?somewhat characteristic?) represented periodic negative affect in which the spouse expressed things like frowning or angry facial expressions that were neither intense nor disruptive of the flow of the interaction. A score of 7 (?extremely characteristic?) reflected extreme displays of negative affect where the spouse was markedly angry or sad toward the other. These displays were intense and frequent and disrupted the flow of the interaction. Additionally, the spouse?s tone toward their partner was negative overall. Interrater reliability for wives? negative affect was r = 1.0. Interrater reliability for husbands? negative affect was r = 1.0. Third, coders observed the spouses? problem-solving skills which assessed the spouse?s ability to define a problem and work toward a mutually satisfactory solution for the problem. Ratings were assigned based on the spouse?s ability to try to solve the problem, not on whether or not the problem was actually solved. A score of 1 (?extremely uncharacteristic?) indicated that the spouse made no effort to solve the problem. The partner may have mentioned the problem, but immediately changed the topic. A score of 4 (?somewhat characteristic?) indicated 17 that the spouse spent some time trying to solve the problem, but an equal amount of time not solving the problem. This could include those couples who jumped from topic to topic and only somewhat discussed brief solutions. A score of 7 (?extremely characteristic?) reflected a spouse with exemplary problem-solving skills. The entirety of the interaction was devoted to attempting to solve the problem. Interrater reliability for wives was r =1.0. Interrater reliability for husbands was r = 0.97. The fourth area coders observed was denial, which referred to the active rejection of personal responsibility for the problem being discussed. Examples included making excuses for his or her role in the problem area, acknowledging the problem but refusing to take any personal responsibility, or acknowledging the problem and entirely blaming his or her partner. A score of 1 (?extremely uncharacteristic?) reflected absolutely no denial. The spouse was aware of, acknowledged, and discussed the problem. A score of 4 (?somewhat characteristic?) indicated that the spouse spent about half of the interaction showing signs of denial. A score of 7 (?extremely characteristic?) reflected a spouse who showed denial throughout the entire interaction. The spouse denied any awareness of the problem or responsibility for it, and/or was unwilling to learn more about the problem from their partner. Interrater reliability was r = 1.0 for wives and r =1.0 for husbands. Fifth, coders observed dominance which referred to a spouse?s achievement of control or influence exerted over their partner during the interaction. A score of 1 (?extremely uncharacteristic?) reflected no signs of dominance during the apportioned time. The spouse either shared the floor and took turns with their partner or was completely overrun by a more dominant partner. A score of 4 (?somewhat characteristic?) indicated that the spouse spent about half of the interaction exerting signs of dominance. A score of 7 (?extremely characteristic?) 18 reflected a spouse who demonstrated a remarkably intense level of denial throughout the entire interaction. Interrater reliability was r = 1.0 for wives and r = 1.0 for husbands. The sixth area observed by coders was sensitivity/support which was the degree to which a spouse listened to their partner, perceived and interpreted feelings and signals accurately, and responded appropriately. A score of 1 (?extremely uncharacteristic?) reflected a spouse who had little regard or consideration for their partner. To get a score of 1, expressed desires or comments of their partner received no response, or a very delayed or a negative response, which may have created distress. A score of 4 (?somewhat characteristic?) indicated that the spouse showed moderate responsivity and support. The spouse responded to comments and needs fairly often, sometimes neutrally and sometimes sensitively. The spouse was not blatantly insensitive, but was also not particularly sensitive. A score of 7 (?extremely characteristic?) reflected a spouse who was characteristically responsive, sensitive, and supportive. The spouse was responsive and attentive to the desires and actions of the other, especially to dissatisfaction and distress and responded quickly and appropriately. Interrater reliability wives? sensitivity and support was r =1.0. Interrater reliability for husbands was r = 1.0. The final individual code observed was conflict. Conflict referred to the expressed struggle between two individuals with incompatible goals or opinions. Observers looked to the level of tension, hostility, disagreement, antagonism or negative affect an individual displayed. A score of 1 (?extremely uncharacteristic?) reflected a spouse showing no affective or content signs of conflict. A score of 4 (?somewhat characteristic?) indicated that the spouse showed signs of conflict for approximately half of the interaction. A score of 7 (?extremely characteristic?) reflected a spouse who showed remarkably intense signs of conflict throughout the entire interaction. Interrater reliability was r =1.0 for wives and r =1.0 for husbands. 19 Dyadic code. Coders observed balance/reciprocity, which conceptualized the relative contributions of each partner to the interaction along such dimensions as control, turn-taking, and equity. Unlike the other codes, this area was coded for the couple as a whole and not each individual spouse. A score of 1 (?extremely uncharacteristic?) indicated that a couple?s interaction was characterized by the dominance of one partner over the other. The couple was in disequilibrium and one partner was likely to control the interaction while rarely considering the other's perspective. One member may have been so passive that they relinquished power to the other. A score of 4 (?somewhat characteristic?) indicated that the spouses were fairly consistent in including the other partner, particularly through the solicitation of opinions and response. Although control of the interaction may have shifted periodically, one partner was responsible for the way the interaction progressed. A score of 7 (?extremely characteristic?) indicated that the couple seemed to be in complete equilibrium. There was a readiness to share responsibility for the interaction and a willingness to listen to and include the other partner and both partners contributed equally to the interaction. Interrater reliability was r =0.90. Controls. Although the interactions could last up to fifteen minutes, many couples did not have discussions that lasted the full duration, therefore, the current study will control for discussion length. Plan of Analysis The data analysis plan is guided by Hair and Black (1998)?s six stages of cluster analyses. Cluster analyses are used to group objects, including people, based on their shared characteristics. In terms of the current study, it is a way to group couples based on how they handle conflict and see what types emerge in order to better understand how conflict looks in older adulthood. The first stage involves selecting objectives (Hair & Black). For the current 20 study, this has been completed through a review of the literature and the development of a research question. The current study seeks to understand differences among maritally satisfied, high functioning, older adult couples in terms of how they handle conflict in the key dimensions of affect, problem solving skills, denial, dominance, sensitivity, conflict, and reciprocity. The second stage is the research design. The goal of this stage is to detect if there are outliers in the data and recognize if the data need to be standardized. It provides a preliminary structure to the data before cluster analyses evaluate the partitions. To satisfy this stage, a series of preliminary analyses will be conducted to identify descriptive statistics (e.g., means, standard deviations, skewness) and detect possible outliers. Stage three addresses assumptions of sample representativeness as well as issues of multicollinearity. Should preliminary correlational analyses reveal that several variables appear highly correlated, appropriate composites will be created to reduce these issues. Stage four is referred to as derivation of clusters and assessment of overall fit (Hair & Black, 1998). This stage will be a two part process using a combination of both hierarchical and nonhierarchical methods, which has been recommended for family scholars (Henry, Tolan, & Gorman-Smith, 2005). Using SAS due to the ability to retrieve the pseudo T statistic, the hierarchical technique will determine the number of clusters and profile the cluster centers or means (Hair & Black). This shows where the initial clusters should be placed. Once these initial clusters are identified, the nonhierarchical technique will fine tune the hierarchical clusters in SPSS by using the number of clusters identified in the first step and assigning specific observations to these clusters (Hair & Black). Using the number of clusters and the means identified using the hierarchical technique as seed points, the K-means algorithm then assigns each specific case to its most similar cluster based on the case?s distance to the cluster mean. 21 Following each case?s assignment, the means for each variable in each cluster are measured again and the cases are reassigned to clusters based on their similarity to the new cluster means. This process cycles through repeatedly until all cases stop changing cluster membership. Essentially, the first, hierarchical step identifies how many clusters to expect and the second, nonhierarchical step determines what characteristics each of those clusters have. Stage five is the interpretation of the clusters (Hair & Black). In this stage, the clusters will be named based on the differentiating or unique characteristics encompassed within them. Finally, stage six is the validation and profiling stage (Hair & Black). To begin to accomplish this step, the clusters will be validated by running a 2 (spouse) X n (where n is the number of clusters) repeated measures ANOVA with spouse as a repeated factor, cluster as a between-group factor, and marital satisfaction as the dependent variable to see if couples both within and across the different clusters differ in their satisfaction with their marriage 22 Results The results are presented in three parts. First, preliminary analyses were conducted, including descriptive statistics for all of the variables. Second, to examine variability among satisfied, high functioning, older adult couples in terms of how they handle conflict in the key dimensions of affect, problem solving skills, denial, dominance, sensitivity, conflict, and balance/reciprocity, a combination of hierarchical and nonhierarchical cluster analyses were conducted and the solution was confirmed using self-report information from both spouses on their marital relationship. Third, a 2 (spouse) by n (clusters) analysis of variance (ANOVA) was conducted using marital satisfaction as the dependent variable to examine how the different clusters differ in terms of marital satisfaction. Preliminary Analyses Descriptive statistics are presented in Table 1. Preliminary analyses found that, on average, the conflict discussions for each couple lasted 8.9 minutes. During these interactions husbands and wives exhibited similar behaviors. Husbands and wives tended to display, on average, moderate levels of positive affect and low levels of negative affect. All of these couples were able to identify a problem area in their marriage and had moderate to low levels of problem solving skills. Spouses displayed low levels of conflict, dominance, and denial with husbands showing slightly lower levels on all except denial. Spouses showed moderate to high levels of sensitivity and support toward one another. On average balance reciprocity was fairly high with both husbands and wives contributing to the conflict discussion. Paired-samples t-tests were 23 conducted for all clustering variables to test for gender differences. There was a significant differences in the negative affect scores for husbands and wives t(63) = 2.65, p < .01. This suggests that wives tended to show significantly more negative affect than husbands on average. Correlations were conducted for both husbands and wives for the observational coding measures. It was found that negative affect and conflict were highly correlated for husbands, r = .79, p < .01 and for wives r = .87, p < .01. Additionally negative affect was highly correlated with dominance for husbands, r = .59, p < .01 and for wives r = .42, p < .01. Conflict and dominance were highly correlated for husbands, r = .41, p <.01 and for wives r = .50, p < .01. Due to the high correlation of all three variables and to reduce issues of multicollinearity for the cluster analyses, a composite score was created for each spouse by averaging the spouse?s negative affect, conflict, and dominance scores. This composite is referred to as hostility and referred to antagonism in conflict discussions as evidenced by displays of conflict, dominance, and negative affect. It was also found that positive affect was highly correlated with sensitivity and support for husbands r = .51, p < .01 and wives r = .58, p < .01. Accordingly, a composite score was created for each spouse by averaging each spouse?s scores for positive affect and sensitivity/support. This composite, called warmth, referred to positivity and responsiveness during conflict discussions. Variability in Older Couples? Conflict Per the suggestion of Henry and colleagues (2005), the current study used a combination of hierarchical clustering followed by nonhierarchical clustering to identify potential clusters of couples. To first identify a potential number of clusters hierarchically, Ward?s (1963) minimum- variance method was used. Ward?s (1963) minimum variance moves through a series of distinct, hierarchical stages, to propose a full range of cluster possibilities through the combination of 24 existing clusters (Hair & Black). This method is designed to minimize the Euclidean squared distance between groups by ?join[ing] clusters that result in the minimum increase of within- cluster variance? (Fals-Stewart, Schafer, & Birchler, 1993, p. 311). Based upon Milligan and Cooper (1985) and Milligan and Cooper (1988)?s comparison and judgment of the most accurate criteria for hierarchically determining number of clusters, three criteria of the Ward?s method were used to identify the number of clusters in the data in SAS. The three criteria were Sarle?s (1983) cubic clustering criterion, or CCC, the pseudo-F statistic, and the pseudo-T2 statistic. Finding the correct cluster solution is best determined by comparing these three criteria and finding concurrent peaks of the CCC and pseudo-F statistic accompanied by a concurrent valley in the pseudo-T2 statistic. The peaks of the CCC and pseudo-F statistic represent a stable and meaningful level of clusters that are most representative and a separation among all clusters at the current step respectively. The pseudo-T2 valley represents the appropriate number of clusters (Fals-Stewart et al., 1993). Taken together, after careful examination, a four cluster solution was determined to be most suitable for the data. Once the number of clusters was determined, in line with Henry et al. (2005)?s suggestion, nonhierarchical clustering was utilized. Using the number of clusters and their starting points determined in SAS, the k-means algorithm was performed to divide the cases into their respective clusters based upon their distance from each cluster mean. Means were then recalculated and individual cases were reassigned based upon their distance from the new cluster means. This cycle repeated until cases no longer changed cluster membership (Hair & Black, 1998). Interpreting the four clusters. The four clusters of older-adult couples were categorized based upon husbands? and wives? observed conflict behaviors (see Figure 1). Repeated measures 25 analyses of variance (2 spouse x 4 cluster) were later used to confirm that the variables used in the cluster analysis did significantly and meaningfully differ among the four clusters. A one-way analysis of variance was conducted to compare the dyadic code of balance/reciprocity in conflict discussions across the four clusters (see Table 4). Cluster 1 (n = 15; 23%) was characterized by significantly higher scores on both spouse?s problem solving skills than spouses in all other clusters. Additionally, couples in this cluster had significantly higher scores on warmth for both spouses in comparison to spouses in the third and fourth clusters (see Table 4). Couples in this cluster had low scores on both denial and hostility for both spouses. Additionally, spouses in this cluster also were characterized by high balance in their conflict discussions. This cluster was named the ?problem solvers? cluster based upon their high problem solving skills and high warmth in comparison to other clusters. Cluster 2 (n = 22; 34%), the largest cluster, was characterized by significantly higher warmth scores by both spouses than spouses in either the third or fourth clusters. Couples in this cluster also had significantly higher problem solving skills than did spouses in the fourth cluster, although they had significantly lower problem solving skills than those in the first cluster. Couples did not differ, however, in problem solving skills from spouses in the third cluster. Like the problem solvers cluster, couples in cluster two had low scores on both denial and hostility for both spouses. Spouses also had high balance scores for their conflict discussions. Cluster 2 was named the ?supporters? cluster due to their high degree of warmth in comparison to clusters three and four. Although, this cluster does not significantly differ from the problem solvers on warmth, their warmth is their defining feature indicating that these couples, even with moderate problem solving skills, can still demonstrate a great deal of sensitivity, support, and positive affect combined with high balance. 26 Cluster 3 (n = 19; 30%) is characterized by moderate levels of problem solving skills. Like the supporters, they are significantly lower than the problem solvers, but higher than those spouses in the fourth cluster on problem solving skills. Additionally, couples in the third cluster had low levels of denial and hostility. Couples in the third cluster had significantly lower levels of warmth than their counterparts in the problem solvers and supporters, but had significantly higher levels of warmth compared to those couples in the fourth cluster. Couples in the third cluster had high balance in their conflict discussions like the problem solvers and supporters. Cluster 3 was named the ?even? cluster due to their moderate levels of both problem solving and warmth accompanied with high levels of balance and their low levels of hostility and denial. Cluster 4, the smallest of the four clusters, (n = 8; 13%), was characterized by significantly higher levels of denial from wives in comparison to the previous three clusters. Additionally, both spouses in this cluster had significantly higher levels of hostility in comparison to the other clusters. A significant interaction in couples? hostility revealed that, for this cluster, wives in this cluster displayed significantly more hostility than did their husbands, F(1,60) = 9.51, p < .01. With their higher levels of both hostility and wife?s denial, these couples also demonstrated significantly lower levels of problem solving and warmth from both spouses in comparison to all other clusters. Balance for couples in the fourth cluster was not significantly different from the other three clusters. Based upon these slightly elevated levels of both hostility and wife?s denial in conjunction with their significantly lower levels of warmth and problem solving skills in comparison to the other clusters, Cluster 4 was named the ?cool? cluster. Observed Conflict Clusters and Marital Satisfaction To determine if marital satisfaction differed based on cluster membership, a 2 (spouse) by 4 (cluster) repeated measures ANOVA was conducted with marital satisfaction as the dependent 27 variable, controlling for conflict duration. The results indicated that there was a significant main effect for cluster membership on marital satisfaction, F (3, 60) = 10.33, p < .01. Pairwise comparisons across cluster demonstrated that couples in the cool cluster reported significantly lower levels of marital satisfaction than the couples in the problem solvers, supporters, and even clusters. Additionally, pairwise comparison revealed that couples in the even cluster reported marginally less marital satisfaction than couples in the problem solvers cluster. There was not a spousal main effect or an interaction between spouse and cluster. 28 Discussion Conflict in marriage, although inevitable, does not have to be the damaging experience it is believed to be (Gottman & Krokoff, 1989; Holly et al., 2013; Zeidner & Kloda, 2013). Although conflict can lead to marital distress if left unresolved (Karney & Bradbury, 1995), conflict can also be functional if conducted in a way that does not include defensiveness, stubbornness or withdrawal (Gottman & Krokoff). By studying those who are able to handle their conflicts successfully, beneficial conflict management behaviors can be isolated and later potentially be used in interventions with couples who have difficulties with conflict. Therefore, to begin to determine the nature of beneficial conflict management behaviors and how these may differ both within and across couples, the current study sought to find meaningful differences in the conflict experiences among a sample of satisfied older adults and to understand if the experiences were related to their marital satisfaction. It was hypothesized that three types of couples would emerge, some would be defined by either having or avoiding conflict, while others would be defined by their sensitivity and validation and that among these three clusters, marital satisfaction would not differ. As predicted, small, but significant differences were found among this sample of satisfied older adults. However, four clusters emerged from the sample and although marital satisfaction was hypothesized to remain constant, variation was found across these clusters. The four clusters that emerged are as follows: (1) Problem Solvers (characterized by high problem solving skills, high warmth, and high balance); (2) Supporters (characterized by high warmth and high balance); (3) Balanced (characterized by high balance/reciprocity); and (4) 29 Cool (characterized by significantly higher hostility and denial). Of these clusters, the cool couples reported less satisfaction with their marriage than did spouses in the other clusters. It is important to note that although marital satisfaction differed significantly in this cluster, spouses in all clusters reported from moderate to very high marital satisfaction. Therefore, even within a sample of couples with low risk for marital distress, variability was found in their behaviors in the context of conflict discussion and consequently in the reports of their marital satisfaction. Meaningful Differences Between Clusters Small differences surfaced among the four clusters, which helped to tease apart the different ways that couples can do conflict ?right? while still maintaining high levels of marital satisfaction. Looking first at the most prevalent couple type, the supporters cluster, the notable warmth and balance of this type of couple most closely resembled Gottman (1993)?s validator type and Fitzpatrick (1988)?s traditional couple. The traditional and validator types are described as those who will generally avoid conflict, but will argue about important topics in their relationship (Gottman). The supporter?s moderate problem solving skills matches this idea as the supporters may not argue often, but they have the skillset necessary to do so. Their high warmth and balance/reciprocity supported the traditional/validator description of promoting ?we-ness?. ?We-ness? can be demonstrated through a variety of means including being responded to positively (Reid, Dalton, Laderoute, Doell, & Nguyen, 2006), and the high warmth behaviors in this cluster indicate that the couples are responding to one another often and appropriately are demonstrative of a sense of ?we-ness? that the couple likely has. Looking next at the second most prevalent couple type, the balanced cluster, the high balance/reciprocity and moderate demonstration of all other behaviors most closely resembled Gottman (1993)?s avoider type and Fitzpatrick (1988)?s separate type. These types of couples 30 have previously been described as avoiding marital conflict and valuing separateness and autonomy. Their problem solving skills reflected this as they also fall between minimally characteristic and somewhat characteristic in their problem solving skills behaviors. Fitzpatrick (1988) described separates as having low levels of sharing. Although the even cluster had high levels of balance/reciprocity, this is not necessarily indicative of high levels of sharing. Rather, the moderate/low levels across all other behaviors indicated that spouses are likely jointly sharing little information with one another, which is demonstrative of their separateness and autonomy. The couples in this cluster interacted very little with one another. They did not argue or problem-solve, nor did they show much affection or warmth toward one another either. They were two separate people equally inclined to minimally participate in a single task devoted to their marital issues. Looking next at the least prevalent cluster, the cool cluster, the significantly higher levels of hostility as well as high levels of denial from wives most closely resembled Gottman (1993)?s volatile type and Fitzpatrick (1988)?s independent type. These couples were described by Fitzpatrick to openly express disagreement. This was supported by the cool cluster?s engaging in more hostility and denial behaviors. Hostility was directly demonstrative of openly expressed disagreement as it encompasses both conflict and negative affect. Wife?s denial also showed open disagreement because within the definition of the denial code, spouses had to openly and negatively refuse to acknowledge their part of the problem. Couples in the cool cluster reported significantly lower marital satisfaction in comparison to the other clusters, but it is important to note that, although their satisfaction was significantly lower, couples in the cool cluster reported above average marital satisfaction. Thus, although engaging in behaviors like hostility is 31 detrimental (Gottman 1999), these couples are only engaging in small amounts of these behaviors allowing for their overall strong levels of marital satisfaction to be maintained. The second least prevalent cluster, the problem solvers cluster, was perhaps the most unexpected cluster. Problem solver?s remarkably high problem solving skills as well as their high warmth and high balance did not closely resemble any of the clusters identified by Gottman (1993) or Fitzpatrick (1988). This cluster may not have emerged in previous studies as those samples included primarily younger and middle-aged adults. Socioemotional selectivity theory suggests that older adults may be different in how they approach conflict as explains that older adults? ?increased attention to emotional goals results in greater complexity of emotional experience and better regulation of emotions experienced in everyday life? (Carstensen, Fung, & Charles, 2003, p. 104). For older adults, this theory can impact their perception of time and their own mortality in relation to how they handle emotional experiences (Carstensen et. al., 1996; Carstensen et al., 2000). It has been suggested that, as a result of these developmental changes in perception of time, older adults are compelled to use more passive strategies in handling conflict, like waiting to see if things improve (Birditt & Fingerman, 2005). This makes the more proactive problem solving behaviors evident in the problem solvers cluster highly unusual. According to Blanchard-Fields, Mienaltowski, and Seay (2007) older adults prefer planful problem solving strategies and emotion regulation strategies when working on instrumental problems. Instrumental problems are defined as ?situations in which one is having difficulty achieving something that is personally relevant? while interpersonal problems are defined as ?involv[ing] social/interpersonal concerns and stem[ming] from complications that arise when one is trying to reach an outcome that involves other people?dealing with a social conflict or obstacle in a relationship? (Blanchard-Fields et al., 2007, p. 62). It is possible that the 32 problems that the couples chose to talk about from the list of common marital problems during the conflict discussion were more instrumental rather than interpersonal, for example, choosing to discuss how they can become better organized as a couple rather than discussing issues about one spouse wanting more sex than the other. This interpretation is consistent with previous work has shown that topic choice can affect the behaviors couples demonstrate during these discussions (Sanford, 2003, 2006), such that if the topic is more difficult, like one spouse not feeling supported in the marriage, the behaviors may be less positive and more negative than those demonstrated when discussing an easier topic like getting more organized. Problem Solving for Older Adults: A Developmental and Relational Perspective Although there were significant and meaningful differences between the clusters, there were also a number of similarities. First, all types of couples exhibited high levels of balance/reciprocity. Also, each type of couple demonstrated relatively little denial and hostility. Even the cool couples, whose hostility and denial were significantly higher than the other types, showed ?highly uncharacteristic? denial and hostility. As to why we would find these consistent patterns across the clusters, there are several explanations. First, this pattern of behaviors could emerge as a function of couples? developmental stage. As previously discussed, older adults tend to have better emotion regulation than their younger counterparts (Carstensen et al., 1995). This elevated emotion regulation could account for both the low hostility as well as the low denial. Hostility and denial are both overly negative behaviors. Having better control over ones emotions would make it easier to keep overtly negative behaviors under control. In relation to older adults, socioemotional selectivity theory suggests that couples? understanding of time and its links to their own mortality influence the impact that conflict plays in their relationship (Carstensen et 33 al., 1996; Carstensen et al., 2000). This can mean that older adults either do not start arguments over trivial matters because they, quite literally, do not have the time for trivial things. Marrying emotion regulation with the full explanation of socioemotional selectivity theory, this can also mean that the strategies employed by older adults may be different when confronted with conflict, like waiting for things to improve or blow over, rather than engage in more active strategies like blaming and name-calling (Birditt & Fingerman, 2005). From a relational standpoint, the high overall marital satisfaction for the sample is a potential explanation for the low observed denial and hostility and the high observed balance/reciprocity. The entire sample rated their marital satisfaction between moderate and very high. Barnett and Nietzel (1979) found that, for non-distressed couples, there were significantly fewer displeasurable instrumental behaviors demonstrated than for distressed couples. It was also found that, pleasantness of interactions was higher for nondistressed couples than for distressed couples (Barnett & Nietzel). Spouses who are more satisfied with their marriage, or nondistressed, tend to show more positive behaviors, like warmth and balance and fewer negative behaviors, like denial and hostility in their marital interactions (Barnett & Nietzel; Huston & Vangelisti, 1991; Wills, Weiss, & Patterson, 1974). The behavior described for non- distressed and highly satisfied couples is consistent with the behavior that was observed across many of the clusters in this highly satisfied sample. A third explanation for the low observed denial in particular may be a function of the coding system not capturing conflict behaviors of older adults. The coding system was originally designed to capture the conflict experience of younger and middle-aged couples (Kline et al., 2004), but the experience older adults have and the behaviors they use may be very different (Birditt & Fingerman, 2005; Carstensen et al., 1995). The denial code is defined very negatively 34 with examples of denial including making excuses, refusing personal responsibility by saying things like ?that?s not my problem?, actively blaming the other spouse for the problem, and/or blowing the problem out of proportion; all of which can be evidence of a certain amount of emotional dysregulation (Widiger, 1998). Based upon older adults? tendency to have better emotion regulation strategies, the denial code, as it is written, is unlikely to be observed. It may not necessarily mean that older adults rarely deny the existence of problems; it may be that denial looks different in older adult conflict. For example, Sillars, Coletti, Parry & Rogers (1982) suggest that denial of conflict and shifting focus away from conflict subjects are similar types of ?avoidance tactics?. It may be that denial for older adults looks more like avoiding the conflict subject instead of actively, verbally, denying the problem. Strengths and Limitations Strengths. The current study has a number of strengths. First, the study relied on observational data in addition to questionnaire responses. Many studies looking at conflict rely on self-report data or questionnaire responses exclusively (Barnett & Nietzel, 1979; Huston & Vangelisti, 1991; Wills et al., 1974). Using observational data allowed for researchers to view participants in the moment during their conflict discussions which avoids issues associated with self-report data like social desirability (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). An additional strength of the study was the dyadic nature of the data. There are a number of reasons that having dyadic data is beneficial. Marriage, and more specifically, conflict, is an interpersonal interaction. Theoretically, dyads reflect the interdependence of both relationships and their interactions in that the way one spouse interacts with the other is influenced by the behavior of the other spouse (Wittenborn, Dolbin?MacNab, & Keiley, 2013). In action, having both members of a couple allows for us to view the contributions of both spouses to their 35 dynamic and understand how the actions of one spouse can impact the other during the conflict interactions (Kenny, Kashy, & Cook, 2006). Dyads also give multiple viewpoints which allow for researchers to not be forced to rely on one spouse to speak for the reality of the couple (Wittenborn et al.), which is especially important when looking at conflict interactions. Finally, the use of a person-oriented approach revealed a number of key differences between couples that might have been overlooked using more traditional, variable-centered approaches. Many studies have focused on conflict on the aggregate level, tending to compare either younger adults versus older adults or unhappy couples versus happy couples (Birditt & Fingerman, 2005; Carstensen et al., 1995; Birchler & Webb, 1977). Although variable-centered approaches provide important information about what conflict behaviors may matter for marital satisfaction, person-centered approaches allow for researchers to delve into the potentially meaningful differences regarding the use of these conflict behaviors within satisfied couples. The current study, using a person-centered approach, was able to find significant differences among a sample of maritally satisfied older adults in the way that they handle conflict. These previously uncovered differences make an important contribution to the understanding of conflict management and begin to reveal the different behaviors that can be utilized to handle conflict and remain maritally satisfied. Limitations. Despite the strengths of this study, there are still several limitations that suggest that the findings need to be interpreted with caution. First, this study had a very small, homogenous sample. They were homogenous not only in race or ethnicity, the study participants were all high functioning: comparatively healthy for their age, wealthy, and well-educated. In marriage, the context in which couples are embedded and their level of functioning makes a difference in a number of arenas pertinent to the current study, including conflict resolution, 36 marital interaction, and types of problems the couples face (Karney & Bradbury, 2005). For example, couples facing chronic stressors with fewer sources of social support tend to have lower levels of marital satisfaction and have difficulty employing the coping skills that they possess (Karney & Bradbury). Conducting this same study with a broader sample would likely result in slightly different patterns with more variability in marital satisfaction as lower functioning couples have external stressors that can impede their ability to utilize beneficial relational skills, such as warmth and sensitivity. The cross-sectional nature of the data is an additional limitation of the study. Because conflict discussions were conducted and questionnaire data were collected cross-sectionally, the clusters that emerged are limited to the particular time point and may not be able to explain couple functioning over time. If the data had been collected longitudinally, couples may have changed cluster membership based upon behaviors over time. For example, those couples in the cool cluster, with their moderate marital satisfaction, may find that as they age and their health begins to decline, their behaviors toward one another may bring them closer together, thus moving them into the warm or balanced cluster. Conclusion This study makes an important contribution to the understanding of how satisfied older adults handle conflict, as it suggests that there are many effective ways to handle marital conflict well in older adulthood. Although this has important research implications in understanding that not all satisfied older couples handle things the same way, it also has practical implications. Therapy techniques regarding conflict resolution have been informed often at the aggregate level by research comparing maritally satisfied couples versus dissatisfied couples or distressed versus nondistressed couples (Billings, 1979; Ito, 1985; Shi, 2003). The breadth of this research alone 37 indicates that there is a multitude of behaviors that a couple can engage in and remain maritally satisfied, but by staying at this variable-centered, collective level, neither researchers nor clinicians can really see the nuances of these behaviors as they are experienced within the couples. Taking a person-centered approach, however, as this study has, delves into the many ways that conflict can be handled while maintaining marital satisfaction and provides evidence that a multitude of equally effective problem-solving behaviors can be effectively utilized. The results of this study can improve the way that intervention is informed and developed as it provides more evidence for the limitations of using an all-or-nothing approach to conflict resolution. This is good news to both clinicians and couples as it eliminates the pressure that a couple has to engage in all of the behaviors that maritally satisfied couples tend to in order to maintain their satisfaction. Using the current study as an example, no particular couple has to be high in problem solving skills, warmth, and balance/reciprocity while simultaneously staying low in hostility and denial to be satisfied. Couples can engage in a combination of levels of these behaviors and still maintain their satisfaction. This evidence improves maneuverability for therapists in their work with couples who have difficulty handling conflict as there are several directions they can take and many areas that can be built upon in small ways to improve marital satisfaction into older adulthood. 38 References Barnett, L. R., & Nietzel, M. T. (1979). Relationship of instrumental and affectional behaviors and self-esteem to marital satisfaction in distressed and nondistressed couples. Journal of Consulting and Clinical Psychology, 47(5), 946. Beach, S. R., Fincham, F. D., & Katz, J. (1998). Marital therapy in the treatment of depression: Toward a third generation of therapy and research. Clinical Psychology Review, 18, 635- 661. Billings, A. (1979). Conflict resolution in distressed and nondistressed married couples. Journal of Consulting and Clinical Psychology, 47, 368-376. Birchler, G. R., & Webb, L. J. (1977). Discriminating interaction behaviors in happy and unhappy marriages. Journal of Consulting and Clinical Psychology, 45, 494-495. Birditt, K. S., & Fingerman, K. L. (2005). Do we get better at picking our battles?: Age group differences in descriptions of behavioral reactions to interpersonal tensions. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 60, 121-128. Blanchard-Fields, F., Mienaltowski, A., & Seay, R. B. (2007). Age differences in everyday problem-solving effectiveness: Older adults select more effective strategies for interpersonal problems. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 62, 61-64. 39 Bradbury, T. N., & Fincham, F.D. (1991). A contextual model for advancing the study of marital interaction. In Fletcher, G. J., & Fincham, F. D. (Eds.), Cognition in close relationships (pp. 127-147). New Jersey: Psychology Press. Bradbury T.N., Rogge R., & Lawrence E. (2001). Reconsidering the role of conflict in marriage. In: Booth A., Crouter A.C., & Clements M. (Eds.), Couples in conflict (pp.59?81). New Jersey: Erlbaum. Bramlett, M. D., & Mosher, W. D. (2002). Cohabitation, marriage, divorce, and remarriage in the United States. National Center for Health Statistics, 23, 1-32. Carstensen, L. L., Fung, H. H., & Charles, S. T. (2003). Socioemotional selectivity theory and the regulation of emotion in the second half of life.Motivation and emotion, 27, 103-123. Carstensen, L. L., Gottman, J. M., & Levenson, R. W. (1995). Emotional behavior in long-term marriage. Psychology and Aging, 10, 140-149. Carstensen, L. L., Graff, J., Levenson, R. W., & Gottman, J. M. (1996). Affect in intimate relationships. In Magai, C., & McFadden, S. H. (Eds.), Handbook of emotion, adult development, and aging (pp. 227-248). California: Academic Press. Carstensen, L. L., Pasupathi, M., Mayr, U., & Nesselroade, J. R. (2000). Emotional experience in everyday life across the adult life span. Journal of Personality and Social Psychology, 79, 644-655. Coln, K. L., Jordan, S. S., & Mercer, S. H. (2013). A unified model exploring parenting practices as mediators of marital conflict and children?s adjustment. Child Psychiatry & Human Development, 44, 1-11. 40 Connell, C. (1994). Impact of spouse caregiving on health behaviors and physical and mental health status. The American Journal of Alzheimer's Care and Related Disorders & Research, 9, 26-36. Cutrona, C. E. (1996). Social support in couples: Marriage as a resource in times of stress. Thousand Oaks, CA: Sage Publications, Inc. DeMaris, A. (2000). Till discord do us part: The role of physical and verbal conflict in union disruption. Journal of Marriage and Family, 62, 683-692. Erel, O., & Burman, B. (1995). Interrelatedness of marital relations and parent-child relations: A meta-analytic review. Psychological Bulletin, 118, 108-132. Fals-Stewart, W., Schafer, J., & Birchler, G. R. (1993). An empirical typology of distressed couples that is based on the Areas of Change Questionnaire. Journal of Family Psychology, 7, 307-321. Fincham, F. D., & Beach, S. R. (1999). Conflict in marriage: implications for working with couples. Annual Review of Psychology, 50, 47-77. Fitzpatrick, M. (1988). Between husbands & wives: Communication in marriage. Thousand Oaks, CA: Sage Publications, Inc. Fried, L. P., Tangen, C. M., Walston, J., Newman, A. B., Hirsch, C., Gottdiener, J., Seeman, T., Tracy, R., Kop, W., Burke, G., & McBurnie, M. A. (2001). Frailty in older adults: Evidence for a phenotype. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 56, M146-M157. Gottman, J.M. (1990). How marriages change. In Patterson, G. R. (Ed.), Depression and aggression in family interaction (pp. 75-101). New Jersey: Routledge. 41 Gottman, J. M. (1993). A theory of marital dissolution and stability. Journal of Family Psychology, 7, 57-75. Gottman, J. M. (1993). The roles of conflict engagement, escalation, and avoidance in marital interaction: A longitudinal view of five types of couples. Journal of Consulting and Clinical Psychology, 61, 6-15. Gottman, J. M. (1999). The marriage clinic: A scientifically-based marital therapy. New York, NY: WW Norton & Company. Gottman, J. M., & Krokoff, L. J. (1989). Marital interaction and satisfaction: A longitudinal view. Journal of Consulting and Clinical Psychology, 57, 47-52. Greenberg, L. S. (1983). Toward a task analysis of conflict resolution in Gestalt therapy. Psychotherapy: Theory, Research & Practice, 20, 190-201. Hair, J.F., & Black, W. C. (1998). Cluster analysis. In Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (Eds). Multivariate analysis (pp. 147-205). New Jersey: Prentice Hall International. Hawkins, A. J., Amato, P. R., & Kinghorn, A. (2013). Are government?supported healthy marriage initiatives affecting family demographics?: A state?level analysis. Family Relations, 62, 501-513. Haynes, S. N., Floyd, F. J., Lemsky, C., Rogers, E., Winemiller, D., Heilman, N., Werle, M., Murphy, T., & Cardone, L. (1992). The Marital Satisfaction Questionnaire for Older Persons. Psychological Assessment, 4, 473?482. Henry, D. B., Tolan, P. H., & Gorman-Smith, D. (2005). Cluster analysis in family psychology research. Journal of Family Psychology, 19, 121-132. 42 Ho, S. C., Chan, A., Woo, J., Chong, P., & Sham, A. (2009). Impact of caregiving on health and quality of life: A comparative population-based study of caregivers for elderly persons and noncaregivers. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 64, 873-879. Holly, S.R., Haase, C.M, & Levinson, R.W. (2013). Age-related changes in demand-withdraw communication behaviors. Journal of Marriage and Family, 75, 822-836. Hoppmann, C. A., & Blanchard-Fields, F. (2011). Problem-solving variability in older spouses: how is it linked to problem-, person-, and couple-characteristics? Psychology and Aging, 26, 525-531. Huston, T. L., & Vangelisti, A. L. (1991). Socioemotional behavior and satisfaction in marital relationships: a longitudinal study. Journal of personality and social psychology, 61, 721. Ito, K. L. (1985). Ho'oponopono,?to make right?: Hawaiian conflict resolution and metaphor in the construction of a family therapy. Culture, medicine and psychiatry, 9, 201-217. Kamp Dush, C. M., & Taylor, M. G. (2012). Trajectories of marital conflict across the life course: predictors and interactions with marital happiness trajectories. Journal of Family Issues, 33, 341-368. Karney, B. R., & Bradbury, T.N. (1995). The longitudinal course of marital quality and stability: A review of theory, methods, and research. Psychological Bulletin, 118, 3-34. Karney, B. R., & Bradbury, T. N. (2005). Contextual influences on marriage implications for policy and intervention. Current Directions in Psychological Science, 14, 171-174. Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic data analysis. New York: Guilford Press. 43 Kerig, P. K. (1996). Assessing the links between interparental conflict and child adjustment: The conflicts and problem-solving scales. Journal of Family Psychology, 10, 454-473. Kiecolt-Glaser, J. K., Glaser, R., Cacioppo, J. T., MacCallum, R. C., Snydersmith, M., Kim, C., & Malarkey, W. B. (1997). Marital conflict in older adults: Endocrinological and immunological correlates. Psychosomatic Medicine, 59, 339-349. Kiecolt-Glaser, J. K., & Newton, T. L. (2001). Marriage and health: His and hers. Psychological Bulletin, 127, 472-503. Kline, G. H., Julien, D., Baucom, B., Hartman, S., Gilbert, K., Gonzales, T., & Markman, H. J. (2004). The interactional dimensions coding system: a global system for couple interactions. In Kerig, P., & Baucom, D. H. (Eds.), Couple observational coding systems (pp. 113-126). New Jersey: Taylor & Francis. Levinson, D. J. (1986). A conception of adult development. American Psychologist, 41, 3-13. Milligan, G. W., & Cooper, M. C. (1985). An examination of procedures for determining the number of clusters in a data set. Psychometrika, 50), 159-179. Milligan, G. W., & Cooper, M. C. (1988). A study of standardization of variables in cluster analysis. Journal of classification, 5, 181-204. Murphy, C.M. & O'Farrell, T.J. (1994). Factors associated with marital aggression in male alcoholics. Journal of Family Psychology, 8, 321-335. O'Leary, K. D., Malone, J., & Tyree, A. (1994). Physical aggression in early marriage: Prerelationship and relationship effects. Journal of Consulting and Clinical Psychology, 62, 594-602. Peter-Wight, M., & Martin, M. (2011). When 2 is better than 1+ 1: Older spouse?s individual and dyadic problem solving. European Psychologist, 16, 288-294. 44 Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of applied psychology, 88, 879. Rauer, A., & Volling, B. (2013). More than one way to be happy: A typology of marital happiness. Family Process, 52, 519-534. Reid, D. W., Dalton, E. J., Laderoute, K., Doell, F. K., & Nguyen, T. (2006). Therapeutically induced changes in couple identity: The role of we-ness and interpersonal processing in relationship satisfaction. Genetic, social, and general psychology monographs, 132, 241- 284. Sanford, K. (2003). Problem?solving conversations in marriage: Does it matter what topics couples discuss?. Personal Relationships, 10, 97-112. Sanford, K. (2006). Communication during marital conflict: When couples alter their appraisal, they change their behavior. Journal of Family Psychology, 20, 256-265. Sarle, (1983). N.S. Sarle Cubic clustering criterion (SAS technical report A-108). Cary: SAS Institute, Inc. Shi, L. (2003). The association between adult attachment styles and conflict resolution in romantic relationships. American Journal of Family Therapy, 31, 143-157. Sillars, A. L., Coletti, S. F., Parry, D., & Rogers, M. A. (1982). Coding verbal conflict tactics: Nonverbal and perceptual correlates of the ?avoidance?distributive?integrative? distinction. Human Communication Research, 9, 83-95. South, S. C., & Krueger, R. F. (2013). Marital satisfaction and physical health: Evidence for an orchid effect. Psychological Science, 24, 373-378. 45 Van den Broucke, S., & Vandereycken, W. (1989). The marital relationship of patients with an eating disorder: A questionnaire study. International Journal of Eating Disorders, 8, 541- 556. Weiss, R. L. ( 1984). Cognitive and behavioral measures of marital interaction. In K. Hahlweg & N. S. Jacobson ( Eds.), Marital interaction: Analysis and modification (pp. 232? 252). New York: Guilford. Weiss, R. L., & Heyman, R. E. (1997). A clinical-research overview of couples interactions. Hoboken, NJ: John Wiley & Sons Inc. Widiger, T. A. (1998). Four out of five ain't bad. Archives of General Psychiatry,55, 865-866. Wills, T. A., Weiss, R. L., & Patterson, G. R. (1974). A behavioral analysis of the determinants of marital satisfaction. Journal of consulting and clinical psychology, 42, 802. Wittenborn, A. K., Dolbin?MacNab, M. L., & Keiley, M. K. (2013). Dyadic research in marriage and family therapy: Methodological considerations. Journal of Marital and Family Therapy, 39, 5-16. Zeidner, M., & Kloda, I. (2013). Emotional intelligence (EI), conflict resolution patterns, and relationship satisfaction: Actor and partner effects revisited. Personality and Individual Differences, 54, 278-283. 46 Tables and Figures 47 Table 1. Correlations and Distributional and Scale Properties of Independent and Dependent Variables (N = 64 Couples.) Note. Correlations above the diagonal are for the husband and correlations below the diagonal are for the wives. Correlations across spouses are underscored and in the diagonal. Variables 1 2 3 4 5 6 7 8 9 1. Positive Affect .83** -.30* .42** -.13 -.03 .51** -.39** .04 .44** 2. Negative Affect -.42** .45** -.04 .41** .59** -.36** .79** -.02 -.04 3. Problem Solving Skills .42** -.21 .94** -.15 .01 .31* -.12 .13 .31* 4. Denial -.13 .40** -.24 -.03 .17 -.45** .43** .09 -.07 5. Dominance -.27* .42** -.15 .26* -.06 -.17 .41** -.04 .04 6. Sensitivity/Support .58** -.62** .28* -.18 -.44** .77** -.49** .30* .28* 7. Conflict -.45** .87** -.13 .29* .50** -.68** .58** -.03 -.09 8. Balance/Reciprocity .20 -.31* .15 -.01 -.31* .44** -.25* -- .14 9. Marital Satisfaction .37** -.50** .23 -.44** -.41** .47** -.47** .14 .57** M (SD) Husband Measures 4.19 (1.30) 1.16 (0.57) 3.94 (0.99) 1.16 (0.37) 1.03 (0.18) 4.75 (0.82) 1.27 (0.54) 5.16 (0.67) 116.36 (18.07) M (SD) Wife Measures 4.22 (1.40) 1.39 (0.75) 3.98 (0.95) 1.13 (0.42) 1.11 (0.32) 4.66 (0.88) 1.31 (0.59) 5.16 (0.67) 117.55 (14.57) *p< .05, **p < .01 48 Table 2. Correlations and Distributional and Scale Properties of Cluster Variables (N = 64 Couples.) Note. Correlations above the diagonal are for the husband and correlations below the diagonal are for the wives. Correlations across spouses are underscored and in the diagonal. *p< .05, **p < .01 Variables 1 2 3 4 5 6 1. Problem Solving Skills .94 -.15 -.07 .43** .13 .31* 2. Denial -.24 -.03 .43** -.29* .09 -.07 3. Hostility -.20 .38** .46** -.43** -.32** -.06 4. Warmth .41** -.17 -.60** .84** .33** .43** 5. Balance .15 -.01 -.03 .16 -- .14 6. Marital Satisfaction .23 -.44** -.53** .46** .14 .57** M (SD) Husband Measures 3.94 (0.99) 1.16 (0.37) 1.15 (0.38) 4.47 (0.93) 5.16 (0.67) 116.36 (18.07) M (SD) Wife Measures 3.98 (0.95) 1.13 (0.42) 1.27 (0.49) 4.44 (1.02) 5.16 (0.67) 117.55 (14.57) 49 Table 3. Means of the 2 (spouse) x 4 (cluster) Repeated Measures Analysis of the Marital Clustering Variables Cluster Membership Spouse Outcome Problem Solvers N=15 Supporters N=22 Even Couples N=19 Cool Couples N=8 F(3,60) Wife Husband F(1,60) Problem Solving Skills 5.27a 3.61b 3.84b 2.75c 46.17*** 3.89 3.85 .58 Denial 1.03a 1.11a 1.16b 1.38c 3.09* 1.18 1.16 .02 Hostility 1.11a 1.07a 1.22a 1.77b 10.83** 1.38 1.20 9.51** Warmth 4.98a 5.11a 4.00b 2.72c 65.14** 4.17 4.24 1.03 Balance 5.27 5.23 5.21 4.63 2.02 Marital Satisfaction 123.02a 120.88a,b 116.11b 96.77c 10.33** 114.52 113.87 1.68 ?p? .10 *p ? .05 **p ? .01 Note: Means with different subscripts are significantly different from one another (p ? .05). 50 Figure 1. Clustering variables by cluster. 1 2 3 4 5 6 7 Problem Solvers n=15 Supporters n=22 Even Couples n=19 Cool Couples n=8 Wife Problem Solving Skills Husband Problem Solving Skills Wife Denial Husband Denial Wife Hostility Husband Hostility Wife Warmth Husband Warmth Balance/Reciprocity Cluster Membership 51 Appendices 52 Appendix A-Questionnaire Measures Marital Satisfaction Questionnaire for Older Persons (Haynes et al., 1992). 1 2 3 4 5 6 Very dissatisfied Dissatisfied Somewhat dissatisfied Somewhat satisfied Satisfied Very satisfied 1. The amount of time my spouse and I 1 2 3 4 5 6 spend in shared recreational activities. 2. The degree to which my spouse and I 1 2 3 4 5 6 share common interests. 3. The day-to-day support and encouragement 1 2 3 4 5 6 provided by my spouse. 4. My spouse?s physical health. 1 2 3 4 5 6 5. The degree to which my spouse motivates me. 1 2 3 4 5 6 6. My spouse?s overall personality. 1 2 3 4 5 6 7. The amount of consideration shown by my spouse. 1 2 3 4 5 6 8. The manner in which affection is expressed 1 2 3 4 5 6 between my spouse and me. 9. How my spouse reacts when I share feelings. 1 2 3 4 5 6 10. The way disagreements are settled. 1 2 3 4 5 6 11. The number of disagreements between my spouse 1 2 3 4 5 6 and me. 12. My spouse?s philosophy of life. 1 2 3 4 5 6 13. My spouse?s values. 1 2 3 4 5 6 14. My spouse?s emotional health. 1 2 3 4 5 6 15. The frequency of sexual or other physically 1 2 3 4 5 6 intimate relationships with my spouse. 16. The quality of sexual or other physically intimate 1 2 3 4 5 6 relations with my spouse. 17. The frequency with which my spouse and I 1 2 3 4 5 6 have pleasant conversations. 53 18. My overall compatibility with my spouse. 1 2 3 4 5 6 19. How decisions are made in my marriage. 1 2 3 4 5 6 20. How well my spouse listens to me. 1 2 3 4 5 6 21. Of all the attention you receive from your spouse, 1 2 3 4 what percent is pleasant or positive? 0-25% 26-50% 51-75% 76-100% 22.Overall, how satisfied are you with your marriage 1 2 3 4 5 6 Very Much less Less Satisfied More Very dissatisfied satisfied satisfied satisfied satisfied 23. In the past year, how often have you had significant 1 2 3 4 problems in your marriage? Very often Often Seldom Never 24. Compared to five years ago, how satisfied are you 1 2 3 4 5 with your marriage? much less less equally more much more right now 54 Appendix B-Conflict Task Areas of Disagreement ? Wife Read through the list below and indicate whether any of these issues are an area in your marriage that you might like changed. This list was compiled from responses from other marital couples when asked to make a list of potential areas of disagreement in a marriage. Read each issue and indicate whether it is a problem or not a problem for your marriage and if it is a problem, whether it is a small or serious problem. Circle a 0 if it is not an issue at the present time. Circle 1 if it is only a small problem and 5 if it is a major problem. We?d like you to consider these issues as they pertain to the past few months. 1) I would like us to talk to each other more. 0 1 2 3 4 5 2) I would like us to have more independence in this marriage. 0 1 2 3 4 5 3) I would like it if we were more organized. 0 1 2 3 4 5 4) I would like it if my spouse spent more time with me. 0 1 2 3 4 5 5) Our problems center on doing household chores. 0 1 2 3 4 5 6) I would like my spouse?s relationships with our families to improve. 0 1 2 3 4 5 Not an issue Small problem Minor problem Moderate problem Somewhat major problem Major problem 0 1 2 3 4 5 55 7) I would like us to go to church, mosque, or synagogue together. 0 1 2 3 4 5 8) I would like us to have more fun together. 0 1 2 3 4 5 9) I would like to have fewer problems with my jealousy. 0 1 2 3 4 5 10) I would like to have fewer problems with my partner?s jealousy. 0 1 2 3 4 5 11) I would like us to have more friends in common. 0 1 2 3 4 5 12) I would like to be consulted on important decisions. 0 1 2 3 4 5 13) I would like my partner to show more physical affection toward me. 0 1 2 3 4 5 14) I want us to go out on more ?dates? together. 0 1 2 3 4 5 15) I would like my partner to watch less television and talk to me instead. 0 1 2 3 4 5 16) I want us to make love more often. 0 1 2 3 4 5 17) I want more help with the finances. 0 1 2 3 4 5 18) I would want to receive more appreciation for what I do. 0 1 2 3 4 5 19) I would like for us to have fewer problems with in-laws. 0 1 2 3 4 5 56 20) I would like us to agree more about saving money. 0 1 2 3 4 5 21) I don?t feel that my partner listens to me when I am upset. 0 1 2 3 4 5 22) I don?t feel supported in this marriage. 0 1 2 3 4 5 23) I would like for us to take more trips together. 0 1 2 3 4 5 24) I would like for us to have a healthier lifestyle. 0 1 2 3 4 5 25) I would like my spouse to take better care of himself. 0 1 2 3 4 5 For Problem ? Solving Task: 1. Pick an area to discuss. 2. Outline "his side" and "her side" of the disagreement. Tell your spouse what it is exactly that you disagree with. 3. Find or work towards a resolution or compromise that you both can agree upon. 57 Areas of Disagreement ? Husband Read through the list below and indicate whether any of these issues are an area in your marriage that you might like changed. This list was compiled from responses from other marital couples when asked to make a list of potential areas of disagreement in a marriage. Read each issue and indicate whether it is a problem or not a problem for your marriage and if it is a problem, whether it is a small or serious problem. Circle a 0 if it is not an issue at the present time. Circle 1 if it is only a small problem and 5 if it is a major problem. We?d like you to consider these issues as they pertain to the past few months. 1) I would like us to talk to each other more. 0 1 2 3 4 5 2) I would like us to have more independence in this marriage. 0 1 2 3 4 5 3) I would like it if we were more organized. 0 1 2 3 4 5 4) I would like it if my spouse spent more time with me. 0 1 2 3 4 5 5) Our problems center on doing household chores. 0 1 2 3 4 5 6) I would like my spouse?s relationships with our families to improve. 0 1 2 3 4 5 7) I would like us to go to church, mosque, or synagogue together. 0 1 2 3 4 5 Not an issue Small problem Minor problem Moderate problem Somewhat major problem Major problem 0 1 2 3 4 5 58 8) I would like us to have more fun together. 0 1 2 3 4 5 9) I would like to have fewer problems with my jealousy. 0 1 2 3 4 5 10) I would like to have fewer problems with my partner?s jealousy. 0 1 2 3 4 5 11) I would like us to have more friends in common. 0 1 2 3 4 5 12) I would like to be consulted on important decisions. 0 1 2 3 4 5 13) I would like my partner to show more physical affection toward me. 0 1 2 3 4 5 14) I want us to go out on more ?dates? together. 0 1 2 3 4 5 15) I would like my partner to watch less television and talk to me instead. 0 1 2 3 4 5 16) I want us to make love more often. 0 1 2 3 4 5 17) I want more help with the finances. 0 1 2 3 4 5 18) I would want to receive more appreciation for what I do. 0 1 2 3 4 5 19) I would like for us to have fewer problems with in-laws. 0 1 2 3 4 5 59 20) I would like us to agree more about saving money. 0 1 2 3 4 5 21) I don?t feel that my partner listens to me when I am upset. 0 1 2 3 4 5 22) I don?t feel supported in this marriage. 0 1 2 3 4 5 23) I would like for us to take more trips together. 0 1 2 3 4 5 24) I would like for us to have a healthier lifestyle. 0 1 2 3 4 5 25) I would like my spouse to take better care of herself. 0 1 2 3 4 5 For Problem ? Solving Task: 1. Pick an area to discuss. 2. Outline "his side" and "her side" of the disagreement. Tell your spouse what it is exactly that you disagree with. 3. Find or work towards a resolution or compromise that you both can agree upon. 60 Appendix C-Coding System Coding system. The following pages are the code book and coding sheets used to behaviorally code the conflict interactions. 1. Individual Positive Affect -- AFFECT CODE This scale assesses the degree to which each partner responds positively towards the other. Unlike the fun/enjoyment scale which considers the overall level experienced by the dyad, this scale looks at the individual's expression of positive affect towards or in response to the other partner's behaviors.. Dimensions include laughing, smiling, vocalizations, and signs of affection (e.g., pats, kisses, etc.). Interest is NOT coded here, rather we are looking for an overall affective state. Two scores are given; one for the wife and one for the husband. 1. Extremely Uncharacteristic: The individual expresses no positive affect towards or in response to the other partner. He/she does not smile or laugh and does not seem to enjoy the interaction. 2. Highly Uncharacteristic: The individual displays minimal positive affect (perhaps 1 brief display). For example, he/she may smile briefly in response to a task related behavior but the affect lacks intensity and frequency. 3. Minimally Characteristic: There may be some sign of positive affect, perhaps an occasional smile and laugh, although the individual would not be described as affectively expressive. For the most part however, the individual's attitude towards the other would not be described as positive, but rather as affectively cool. 4. Somewhat Characteristic: The individual expresses low-level enjoyment or positive affect towards the other although these displays are not intense or prolonged. There are frequent and 61 somewhat prolonged lapses in the individual?s positive expressions. Just as often as there are positive displays, there are also pauses. (many small smiles) 5. Moderately Characteristic: Frequent displays of positive affect are evident. There are several incidences of laughing, smiling, or pleasure. The individual seems really be enjoying the interaction and appears comfortable expressing enjoyment to the other partner. There are a few brief lulls (rather than lapses) in positive affect. Several smiles and laughs ? the typical couple that seems to enjoy this interaction. 6. Highly Characteristic: Extensive positive affect is shown that is both frequent and intense. The individual is enjoying the interaction and expresses it through frequent laughs, smile, etc. There may be brief periods where no or minimal positive affect is shown, but smiling and laughter quickly resumes. Extremely positive with intensity. 7. Extremely Characteristic: Positive affect is consistently and continuously demonstrated and is both frequent and intense. The individual is thoroughly enjoying the interaction and laughs and smiles throughout. There are no noticeable delays or lapses in positive affect. 2. Individual Negative Affect -- AFFECT CODE Unlike the irritation/antagonism scale which looks at the couple as a unit, this scale considers the individual's expression of negative affect towards the other. (Note: only code if directed towards the other!) Look for frowning, rolling eyes, anger, averted gazes, etc. Two scores are given; one for the wife and one for the husband. Consider the context of the comments - whether hostile or antagonistic as well as the tone. Also consider how the comment affects the partner. (Conflict might reflect this score as well) 62 1. Extremely Uncharacteristic: The individual expresses no negative affect. 2. Highly Uncharacteristic: Negative affect is minimal. There may be one instance of frowning or rolling of the eyes, for example, but it is extremely brief and lacking intensity. 3. Minimally Characteristic: One mild expression of negative affect such as a marked frown, or two brief displays that are neither intense or prolonged (e.g., the individual might say, "No, you never help with feeding" while frowning. 4. Somewhat Characteristic: Negative affect is expressed periodically but does not disrupt the flow of the interaction. There may be clear signs such as frowning or angry facial expressions, but they are not particularly intense. 5. Moderately Characteristic: Some signs of negative affect are expressed towards the other and they may be mildly intense. For example, there may be frowning, averted gazes, and looks of disapproval. The individual appears displeased but not necessarily hostile. Although more intense than a 4, the negative affect still does not disrupt the interaction. 6. Highly Characteristic: Frequent signs of negative affect (e.g. a combination of negative expressions) are displayed towards the other. The individual seems irritated or angry, but the interaction is not as disrupted as in a 7. 7. Extremely Characteristic: Displays of negative affect are extreme. The individual appears markedly angry or sad towards the other. There is intense and frequent frowning, angry facial expressions, etc. that affect the flow of the interaction. The individual's overall tone towards the other is very negative. 3. Problem Solving Skills- CONTENT CODE 63 Problem-solving skills refer to an individual?s ability to define a problem and work toward a mutually satisfactory solution for the problem. Ratings are assigned based on a person?s ability to try to solve the problem, not on whether or not the problem was actually solved. Examples include: Recognizing the problem exists within the dyad, Describing/Defining the problem positively or neutrally without resorting to blaming partner, Clearly expressing wishes and desired outcome to be reached, Contributing to problem discussion effectively and keeping the conversation on task, Proposing positive plans or a solution designed to solve the problem, Negotiating, compromising, and/or working with his or her partner to come to a mutually agreeable conclusion, Making a commitment to take action towards the problem, and Suggesting a hypothetical plan(s) to solve the problem. They will almost always have the same score if they are both contributing 1. Extremely Uncharacteristic: Spouse makes no attempt to solve the problem; may make mention to the problem yet changes the topic 2. Highly Uncharacteristic: Spouse makes almost no attempts to solve a problem; displays few or weak signs of involvement in terms of solving the problem. 3. Minimally Characteristic: Spouse seems generally interested in trying to solve the problem and shows some signs of trying to come to a solution (This includes jumping from topic to topic and minimally discussing each ex: just mentioning each problem and what each spouse?s score was and one simple statement ?Well, we?re not going to fix that?. Not really offering a solution) 4. Somewhat Characteristic: Spouse spends some time making some attempts to solve the problem, but spends an equal amount of time during the interaction not making attempts to solve 64 a problem. (This includes jumping from topic to topic and somewhat discussing brief solutions to each) Ex) We should throw some things out to get more organized? a 4 is distinguished from a 3 in that they recognize that it is a problem and offer a brief solution (we need to go on more dates) and it doesn?t have to be for every problem listed 5. Moderately Characteristic: Spouse demonstrates notable problem solving skills throughout a large portion of the interaction. This includes discussing solutions to multiple topics in greater detail than a 4. This also includes working through solutions to one problem in fairly detailed terms. EX: So, we need to go on more dates?let?s go out to eat this Friday?yeah let?s do it every Friday?..sounds great. 6. Highly Characteristic: Spouse demonstrates strong problem solving skills throughout all or nearly all the interaction. Must discuss one problem in detail and/or come up with some kind of plan ? really solid back and forth about a solution. 7. Extremely Characteristic: Spouse demonstrates exemplary problem solving skills. The whole of the interaction is dedicated to solving the problem constructively 4. Denial- CONTENT CODE . Denial is the active rejection of personal responsibility for the problem being discussed. Examples include: Making excuses for his or her role in the problem area, Acknowledging the problem but refusing to take any personal responsibility, Acknowledging the problem and entirely blaming his or her partner, Blames partner for blowing the problem out of proportion, Claims partner is imagining or making up the problem. (Probably in most cases will be a 1, 2, or 3 ? unless there is a ton of blatant denial throughout). 65 1. Extremely Uncharacteristic: Spouse demonstrates absolutely no denial; he/she is aware of, acknowledges and discusses the problem? a neutral reason for the problem does not move it off the anchor 2. Highly Uncharacteristic: Spouse displays almost no denial; may make a brief rebuttal or clarification ? Ex: Sounds like they?re giving a lame excuse for their part in the problem 3. Minimally Characteristic: Spouse displays some (multiple) weak or infrequent signs of denial Blatantly says something like ?That?s not my problem? at least once 4. Somewhat Characteristic: Over the course of the interaction, spouse spends about half of the time showing signs of denial. 5. Moderately Characteristic: Spouse displays denial throughout a large portion of the Interaction. 6. Highly Characteristic: Spouse demonstrates denial throughout almost the entire interaction; individual strongly demonstrates that he/she does not think there is a problem or that he/she has no role in problem 7. Extremely Characteristic: Spouse denies any awareness of the problem, responsibility for it and/or is unwilling to learn more about problem from partner; denial is exhibited throughout entire interaction 5. Dominance- CONTENT CODE . Dominance is the actual achievement of control or influence an individual exerts over his 66 or her partner during the interaction. Dominance may be identified through forceful, monopolizing, and/or coercive behaviors. Examples of dominance include: Directing the course of the conversation, Talking forcefully and/or taking charge, Commanding partner and partner complies, Talking more often than partner and/or not letting partner talk, Successfully interrupting partner and/or resisting partner?s interruptions, Starts or introduces problem discussion and/or closure of problem discussion abruptly, against partner?s wishes or without input or consent from partner, Forces partner to accept own opinions without reasons, Completely changes partner?s opinions, or Withholds contributions to conversations as a means of exerting control. **Does not apply to someone who is just more generally talkative . 1. Extremely Uncharacteristic: Spouse displays no signs of dominance in the time apportioned. Spouse either shares the ?floor? (i.e., speaking turn) with their partner equitably or is completely stifled and overrun by a more dominant partner. 2. Highly Uncharacteristic: Spouse displays almost no dominance; may demonstrate a characteristic of dominance, but has little effect on the direction and course of the discussion, a little bit of interrupting each other, must be a clear example of cutting the other person off (feels somewhat negative). Partner talks almost the entire time 3. Minimally Characteristic: Spouse displays some signs of dominance that are weak in intensity. 4. Somewhat Characteristic: Over the course of the interaction, Spouse spends about half of the time exerting dominance and half of the time sharing the floor with their spouse. 67 5. Moderately Characteristic: Spouse demonstrates dominance throughout a large portion of the interaction; has a significant effect on the interaction throughout their expression of dominance. 6. Highly Characteristic: Spouse demonstrates dominance throughout almost the entire interaction; rarely allows partner an opportunity to express his/herself. 7. Extremely Characteristic: Spouse demonstrates a remarkably intense level of dominance that is exhibited throughout the entire interaction 6. Sensitivity/Support -- COMBINED CODE Sensitivity refers to listening to the partner, perceiving and interpreting feelings and signals accurately, and responding appropriately. Consider the frequency, latency, and the appropriateness of response to the spouse. This code is more focused on the behaviors of the listener, but keep in mind it is still a dyadic code. At the highest point, quick, warm and sensitive responses are characteristic, but don't require personal expense. At the lowest point, coldness, rejection and ignoring are typical. Sensitivity/support needs to go beyond listening, as all couples are instructed to listen to one another ? try to consider what optimal responding is. 1. Extremely Uncharacteristic: There is little regard or consideration for the other. Expressed desires or comments of the other get no response, or a very delayed or a negative response, which may create distress. If one seems to enjoy creating distress in the other, score 1. 2. Highly Uncharacteristic: One sees occasional but rare positive responding. More often than not, they seem oblivious to each other's needs and comments, though they may very occasionally respond to very obvious signals in a neutral or occasionally inconsiderate or defensive manner. 68 3. Minimally Characteristic: Responsivity is generally low. Many comments go unheeded but very clear signs of distress or need would likely receive some response. Responses may be neutral, or appropriate but delayed. There is some "coolness" here. 4. Somewhat Characteristic: This spouse shows moderate responsivity and support. Comments and needs are responded to fairly often, sometimes neutrally and sometimes sensitively. There is nothing blatantly insensitive; however the spouse is not particularly sensitive either. 5. Moderately Characteristic: In the context of generally high responsivity and sensitivity, these partners may show brief occasions of insensitive disregard. When called for, sensitivity is more likely than not but is not a given. 6. Highly Characteristic: This spouse lacks the consistency or harmony of 7. They may be characteristically sensitive and responsive but lack fine-tuning. There may be infrequent and minor but noticeable lapses in responding or offering support. 7. Extremely Characteristic: This spouse is characteristically responsive, sensitive, and supportive. Each spouse is responsive and attentive to the desires and actions of the other, especially to dissatisfaction and distress. Needs and comments are responded to quickly and appropriately, but not at one's own personal expense. 7. Conflict- COMBINED CODE Conflict is an expressed struggle between two individuals with incompatible goals or opinions. The level of tension, hostility, disagreement, antagonism or negative affect an individual displays can identify conflict. Face displays tension, nervousness (includes eye contact, clenched jaw, eye twitches, nostrils flair, decreased or overly intense eye contact), Body 69 is tense, tight, Speaks in a negative voice ? impatient, angry, whining, cold or curt, Reacts with negative affect to own or partner?s negative affect. Examples may include: Judges and criticizes partner or people/things important to partner, Imposes own will on partner, is controlling, Demonstrates indifference and lack of commitment, Minimizes the value of partner?s contributions, Expressing rigidity in one?s willingness to listen to partner, Disagrees more often than agrees with partner, Makes negative interpretations/mind reads ? attributes negative feelings, attitudes, beliefs or motives to partner (e.g., ?You never wanted to go to my parents? house in the first place?), Makes negative overgeneralizations ? e.g., ?You always say that!!!? or ?You never ask me how my day went??, Antagonizes partner by using sarcasm, complaining in response to partner?s complaint, or commenting negatively on partner?s negative behavior and Appears to instigate more conflict. 1. Extremely Uncharacteristic: Spouse displays no affective or content signs of conflict 2. Highly Uncharacteristic: Spouse displays almost no signs of conflict 3. Minimally Characteristic: Spouse displays some signs of conflict that signs are weak and/or infrequent 4. Somewhat Characteristic: The spouse, over the course of the interaction, spends half of the time showing signs of conflict. 5. Moderately Characteristic: Spouse displays the characteristics of conflict throughout a large portion of the interaction, though these signs are inconsistent 6. Highly Characteristic: Spouse displays strong signs of conflict throughout almost the entire interaction 70 7. Extremely Characteristic: Spouse displays remarkably intense signs of conflict throughout the entire interaction 8. Balance/Reciprocity ? DYADIC CODE This scale assesses the relative contributions of each partner to the interaction. Included are dimensions such as control, turn-taking, and equity. **Only coded on wife?s sheet 1. Extremely Uncharacteristic: This couple's interaction is characterized by the dominance of one partner over the other. Each partner's contribution to the interaction is by no means equal and one partner is likely to control the interaction while rarely considering the other's perspective. One member may be so passive that she/he relinquishes power to the other. The couple appears to be in disequilibrium. 2. Highly Uncharacteristic: Turn-taking is minimal and although there may be attempts to include both spouses in the interaction, it is primarily one-sided. 3. Minimally Characteristic: Some turn-taking is present and each partner makes a contribution to the interaction. One partner may control the flow of the interaction, but there are a few attempts to listen to and solicit responses or opinions from the other partner. 4. Somewhat Characteristic: These spouses are fairly consistent in including the other partner, particularly through the solicitation of opinions and response. Control of the interaction may shift periodically but one partner is responsible for the progress of the interaction. 71 5. Moderately Characteristic: Both partners appear to be initiating conversations and responses although the complexity and length of one's responses may be greater than the other. Thus, although one partner appears more dominant, there is sharing of opinions and responses. 6. Highly Characteristic: Substantial balance is shown, including smooth turn-taking, sharing of control, and equal contributions to the interaction. Brief periods where one partner dominates the interaction may be present, but the balance is quickly restored. 7. Extremely Characteristic: This couple seems to be in complete synchrony. There is a readiness to share responsibility for the interaction and a willingness to listen to and include the other partner. Turn-taking is smooth and both partners contribute equally to the interaction without dominating. 72 HUSBAND Problem-Solving Task Coding Sheet Couple ID: _________ Coder: _________ Extremely Highly Minimally Somewhat Moderately Highly Extremely Uncharacteristic Uncharacteristic Characteristic Characteristic Characteristic Characteristic Characteristic 1 2 3 4 5 6 7 Affect Codes Positive Affect 1 2 3 4 5 6 7 Negative Affect 1 2 3 4 5 6 7 Content Codes ID Problem Yes No Problem Solving Skills 1 2 3 4 5 6 7 Denial 1 2 3 4 5 6 7 Dominance 1 2 3 4 5 6 7 Combined Codes Sensitivity/Support 1 2 3 4 5 6 7 Conflict 1 2 3 4 5 6 7 Communication skills 1 2 3 4 5 6 7 73 Wife Problem-Solving Task Coding Sheet Couple ID: _________ Coder: _________ Extremely Highly Minimally Somewhat Moderately Highly Extremely Uncharacteristic Uncharacteristic Characteristic Characteristic Characteristic Characteristic Characteristic 1 2 3 4 5 6 7 Affect Codes Positive Affect 1 2 3 4 5 6 7 Negative Affect 1 2 3 4 5 6 7 Content Codes ID Problem Yes No Problem Solving Skills 1 2 3 4 5 6 7 Denial 1 2 3 4 5 6 7 Dominance 1 2 3 4 5 6 7 Combined Codes Sensitivity/Support 1 2 3 4 5 6 7 Conflict 1 2 3 4 5 6 7 Communication skills 1 2 3 4 5 6 7 Balance/Reciprocity 1 2 3 4 5 6 7 74 Appendix D-Extended Literature Older Adults Older adults typically experience a host of unique health problems (Fried et al., 2001). In light of this, as couples age, at some point, one spouse often finds themselves taking on a care- giving role to their aging spouse (Connell, 1994). Like other developmental transitions, this transition to being a caregiver can be exceedingly stressful and damaging to the health and well- being of a person (Connell). Physically, Connell found that caregivers report being tired, exhausted, or fatigued, being under stress, having sleep problems, gaining or losing weight, having high blood pressure, having difficulty concentrating, and having limited opportunities for exercise. Additionally, one half of their sample reported having limitations to their daily activities due to health problems like arthritis, high blood pressure, and back troubles (Connell).Interestingly, the majority of the sample rated their own physical health as worse than that of their spouse for whom they are providing care (Connell). Mentally, caregivers report higher instances of depression, lowered morale and well-being, emotional exhaustion, anxiety, helplessness and interference with social and recreational pursuits in comparison to their same aged, non care-giving counterparts (Connell). Ho, Chan, Woo, Chon, and Sham (2009) found consistent results when comparing the health and well-being of caregivers with non-caregivers. The authors found that primary informal caregivers, when controlling for education and work status, report more doctor?s visits per year as well as greater weight loss. Caregivers were also more likely to report higher rates of depression worsening health from year to year as well as lower quality of life. Female caregivers were also found to be at a 1.5 to 2 fold risk for symptoms such as headaches, dizziness, heart palpitations, worsening memory, unstable emotions and trouble with sleep when compared to non-caregivers (Ho, Chan, Woo & Sham, 2009). 75 With these added stressors, there is an increased chance of marital conflict (Cutrona, 1996). The current study looks at a sample of couples who are still healthy enough to neither require care-giving nor be caregivers, but are on the cusp of taking on such a role. These are couples who have the stressors mentioned previously likely coming in their near future. By understanding what conflict management strategies satisfied older adult couples utilize at this time, practitioners can teach couples these successful conflict management strategies prior to stressful times when health problems and care-giving take center stage. Understanding successful conflict management strategies can potentially assist couples in getting through the conflicts that may arise in the future. This is especially pertinent considering our aging society and the idea that couples will spend more time both together in marriage and in care-giving roles.