POST-TRAUMATIC GROWTH ACROSS PARTNERS AND IN RELATIONSHIPS By Mariah Faith Purol A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Psychology – Doctor of Philosophy 2024 ABSTRACT Multiple theories have proposed the possibility of post-traumatic growth (PTG)—positive personality change occurring after an especially negative life experience (Jayawickreme & Blackie, 2014). Much PTG work documents the importance of close others in the expression and magnitude of this phenomenon. Romantic partners, in particular, appear to play an especially important role in PTG, sometimes facilitating/hindering growth and experiencing growth themselves—even when they do not directly experience a negative event. In three studies, each examining longitudinal samples from different countries (i.e., the United States, the Netherlands, and Switzerland), I examined trajectories of post-traumatic growth (i.e., increases in extraversion, conscientiousness, openness to experience, and agreeableness; decreases in neuroticism) among individuals and their romantic partners. Each study used growth curve modeling to parse apart patterns of personality change, determine if experiencing ostensibly negative life events (directly or vicariously) influences these patterns, and examine the role of potentially influential relationship characteristics. I found that, while individuals’ own negative life events and the negative life events of their partners were occasionally associated with positive personality change, this was relatively rare, and the effect sizes of these potentially impactful life events were relatively small. The relationship variables examined in this series of studies (i.e., support, relationship satisfaction, responsiveness, and closeness) were largely unassociated with adaptive personality trait change, although, when examined as outcomes, I found some evidence of PTG on the dyadic level (i.e., relationships improving after a negative life event). The final study also modeled trajectories of both self and observer (i.e., partner) reported personality change, finding that individuals’ perceptions of their personality change after a negative event varied slightly from their partners’ perceptions of their change. ACKNOWLEDGEMENTS Study 1 uses HRS panel data, supported by the National Institute on Aging (NIA U01AG009740) and the Social Security Administration. Study 2 uses LISS panel data, which were collected by CentERdata (Tilburg University, The Netherlands) through its MESS project funded by the Netherlands Organization for Scientific Research. Study 3 uses CouPers panel data, collected by Alexander Grob, Robert Philip Burriss, Rebekka Weidmann, Janina Larissa Bühler, Jenna Wünsche, Fabienne Fend, Sabrina Brunner, and Rahel Hütten and funded by the Swiss National Science Foundation (SNSF, project number 162697). I have so many people to thank, all of whom deserve pages and pages of acknowledgements. Thank you to the members of my committee: William Chopik, Richard Lucas, Kevin Hoff, and Adrian Blow, for donating their time and perspectives to this project (and many others). Extra thank-yous to my advisor, William Chopik. With the time and care you took to guide me through my time as a graduate student, you have opened doors for me that would have otherwise been closed forever. I will never have the words to thank you; I am forever proud to be your student. And, perhaps, the biggest thank-yous to my network of close others who kept me happy and human throughout my PhD. Thank you to my dearest friends, Samantha DeNooyer, Erica Wilson, and Sydney Steele, and to my endlessly supportive parents, Patti and Bernie Purol. And, of course, thank you to my favorite close other, Blaise Nugent. May our next chapter be as sweet as our last; I can’t wait to be your wife. iii TABLE OF CONTENTS CHAPTER 1: INTRODUCTION ....................................................................................................1 CHAPTER 2: STUDY 1 ................................................................................................................19 CHAPTER 3: STUDY 2 ................................................................................................................42 CHAPTER 4: STUDY 3 ................................................................................................................60 CHAPTER 5: CONCLUSIONS AND FUTURE DIRECTIONS.................................................85 REFERENCES..............................................................................................................................94 APPENDIX A: TABLES FOR CHAPTER 2..............................................................................107 APPENDIX B: TABLES FOR CHAPTER 3..............................................................................141 APPENDIX C: TABLES FOR CHAPTER 4..............................................................................173 APPENDIX D: SUPPLEMENTARY GROWTH MIXTURE RESULTS AND TABLES........223 iv CHAPTER 1: INTRODUCTION Anecdotal and theoretical accounts of post-traumatic growth (PTG) suggest the possibility of positive personal changes (i.e., increases in extraversion, agreeableness, conscientiousness, openness; decreases in neuroticism) reported by those who experience a strong negative life event (Infurna & Jayawickreme, 2019; Jayawickreme & Blackie, 2014; Jayawickreme et al., 2021). Close others, including romantic partners, are often implicated in this growth process, whether as active contributors or as beneficiaries of vicarious growth (Barre et al., 2023; Canevello et al., 2016; Schroevers et al., 2010; Zwahlen et al., 2010). In this dissertation, I proposed three studies to examine the incidence of PTG (e.g., partners’ trajectories of PTG), how these trajectories compare to those of individuals who directly experience trauma themselves, and which relationship-level factors might influence this growth. Each of the three proposed studies used pre-existing longitudinal panel data with couples (two of which were nationally representative). The final study also explored the possibility that partner perceptions of an individual’s PTG may vary from an individual’s perceptions of their own PTG. Observer reports offer further nuance to the role of romantic partners in the phenomenon of PTG by assessing whether PTG or positive personality growth after adversity is “detectable” by close others. Taken together, this suite of studies seeks to offer new insight into the role of romantic partners in PTG. Negative Life Events And Individuals’ Personality Change Inevitably, most people experience difficult, adverse, and/or painful challenges in their lives (i.e., negative life events). Some work on the topic suggests that over half of college-aged students (and some estimates suggest upwards of 84% of students) have already experienced at least one impactful negative life event in their lives (Smyth et al., 2008), such as the death of a 1 loved one, a physical or sexual assault, or a health scare. But do these experiences somehow change the individuals who live through them? Specifically, do negative life events have a lasting impact on an individual’s personality? To understand if and how individuals change in response to adversity specifically, it is important to characterize the nature of personality changes in the context of life events more broadly. Some theorists have posited that personality may change on the level of the broad, Big Five personality traits (i.e., extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience). In a review of the subject, Bleidorn, Hopwood & Lucas (2018) found that many negative life events have been at least somewhat linked to changes in Big Five personality traits, whether it be increases or decreases in ostensibly positive traits following an event. However, for most life events featured, there was no clear consensus on exactly whether and how these events change personality. The magnitude of trait changes, the direction of the effects, and even the specific traits implicated often varied across studies in ways that are not immediately straightforward (Bleidorn et al., 2018). For example, when examining divorce, Costa and colleagues (2000) found that men became less conscientious and more neurotic after divorce, while women became more extraverted and open. Other studies found post-divorce increases in agreeableness and conscientiousness for both men and women (Specht et al., 2011b). Yet other work found post-divorce decreases in extraversion specifically for both men and women (Allemand et al., 2015). How, then, does personality change after a life event like divorce? As Bleidorn and colleagues (2018) acknowledge, when examining the literature on personality change and divorce, the findings are unclear and, occasionally, incompatible with each other. Despite heterogeneity in the findings, the review highlighted a couple of life events that were at least more consistently associated with personality change across studies: the 2 transition to a first romantic relationship (i.e., it is associated with decreases in neuroticism and, occasionally, increases in extraversion) and the transition from high school to college or work (i.e., it is associated with increases in conscientiousness). Notably, these transitions are typically considered ostensibly positive life events. Negative life events, such as widowhood or unemployment, demonstrated fewer consistent effects on personality change, and their influence often varied based on demographic factors such as gender. All told, the literature concerning negative life events spurring Big Five personality change is a little murky. Others have theorized that, after adversity, it is more likely that, at least for some people, socially valued and positive characteristics might increase—signaling a form of resilience or flourishing (i.e., “what doesn’t kill you makes you stronger”, Nietzsche & Levy, 1909). This sort of positive personality change is often referred to in the literature as post-traumatic growth (PTG; Jayawickreme & Blackie, 2014). In this framework, negative life events are conceptualized as catalysts for positive personal growth. The most commonly used measure in studies of PTG, the Post-Traumatic Growth Inventory (PTGI; Tedeschi & Calhoun, 1996), identifies five areas of growth that individuals report after an adverse life event: 1) new possibilities (in which individuals find new interests, callings, or opportunities after a time of crisis); 2) relating to others (in which individuals experience greater closeness and compassion for others after a time of crisis); 3) personal strength (in which individuals are more confident in their strengths and feel more competent after a time of crisis); 4) spiritual change: in which individuals feel more connected to a religious faith or spirit after a time of crisis; and, lastly, 5) appreciation of life: in which individuals have a newfound appreciation for their lives and may shift their priorities in life after a time of crisis. However, in reviews of normative personality changes after life events, there is a great 3 deal of heterogeneity in whether personality changes in consistent ways. Thus, when frameworks began to emerge hypothesizing that personality can change in positive ways following adversity, some researchers advocated for the idea, while others were strongly skeptical about how common or possible PTG was. Indeed, both methodological and conceptual challenges have limited the work on successfully demonstrating PTG. For example, PTG (or any type of adjustment-related indicator) is often assessed only once (i.e., cross-sectionally) and after an event has already occurred. Further, the PTG literature has relied heavily on asking people to cognitively reflect on how they have grown in response to a negative experience. Others have noted that reflecting on change and negative events often requires a great deal of introspection on the part of participants to quantify their growth and how much of it is attributable to an adverse event (Tennen & Affleck, 1998). In a review of the work on this topic, Jayawickreme and colleagues (2021) summarize how other, closely-related conceptualizations of growth after trauma (e.g., benefit-finding, psychological well-being, changes in life narratives, etc.; Helgeson et al., 2006; Joseph & Linley, 2005; Pals & McAdams, 2004) have characterized post-crisis change. The authors discuss changes in character strengths—moral personality traits—as one potential avenue to think about and demonstrate post-traumatic growth. There is compelling evidence that individuals’ character strengths do, in fact, change after tragedy. For example, a study examining American participants’ character strengths before and after the September 11th terrorist attacks found post- attack increases in gratitude, hope, kindness, leadership, love, spirituality, and teamwork (Peterson & Seligman, 2003). Interestingly, when measured again almost a year after the attacks, these character strengths were still elevated, although to a somewhat lesser degree (Peterson & Seligman, 2003). However, as Jayawickreme and colleagues point out, not all challenging events 4 are related to changes in character strengths, either. For example, health crises are inconsistently related to positive changes in character strengths (Gander et al., 2020). Similarly, in an examination of newly deployed U.S. soldiers, soldiers were largely stable in their character strengths across the deployment cycle, changing very little over time and only being negligibly related to traumatic combat experiences (Chopik et al., 2021). Mangelsdorf and colleagues (2019) summarized much of the research on this topic in a meta-analysis of PTG and its counterpart, post-ecstatic growth (in which individuals experience positive personality change after positive life events). The project constituted a thorough review of over 150 studies, both longitudinal and post-hoc (i.e., in which personality change was reflected upon after the event). The research team examined several PTG-related outcomes— many of which suggested positive personality change after trauma. Although few studies included in the meta-analysis examined changes in personal strengths after negative events (with only one longitudinal study explicitly measuring strengths), the effect sizes depicting change were positive and stable over time (i.e., longitudinal window; all ds > .25). Other significant positive outcomes included increases in environmental mastery, autonomy, and self-esteem (Mangelsdorf et al., 2019). But people were heterogeneous in the exact characteristics that tended to change: for example, although some indicators improved after trauma experiences, others, such as a propensity for spiritual thoughts and experiences, were relatively unaffected by trauma experiences. Further, personal growth (an indicator suggesting psychological well-being) decreased after a negative event (although there was only one study in the meta-analysis that examined this outcome, Mangelsdorf et al., 2019). Clearly, there is no consensus surrounding personality change after a negative life event, whether measured as PTG or as a change in Big 5 traits or character strengths. However, at least some of this ambiguity might be attributable to a 5 lack of longitudinal, prospective data and applying methods that help identify whether subgroups within a population experience PTG (Jayawickreme et al., 2021). Further, at least some of this variation in how people change following life events can be explained by how individuals interpret the life events they experience. After all, not all crises (and our interpretations of them) are created equal. As further discussed by Jayawickreme and colleagues, how traumatic or challenging a life event is judged to be likely relies on a unique interaction between the event and a person. Perhaps negative life events produce less consistent changes in personality (Bleidorn et al., 2018) because of this complex interaction: a divorce can prompt increases in extraversion or a bout of personal growth in one person, but it may debilitate another person individually and socially. A positive life event, like the start of a new relationship, may incite more universal effects. Recently, there has been a call to account for this kind of person-event interaction in studies of PTG to more accurately characterize the nature of post-traumatic personality change (Jayawickreme & Blackie, 2014; Jayawickreme et al., 2021). One answer to this call comes in the form of changing the methodological approaches for studying life events—measuring the perceived impact of a negative life event instead of the mere presence or absence of the event. Recently, Luhmann and colleagues developed and published the Event Characteristics Scale (ECS; Luhmann et al., 2020). The ECS measures characteristics of major life events in hopes of better explaining why psychological outcomes of life events can vary in their strength, direction, and duration. Specifically, the scale asks participants to describe a life event in terms of 9 perceived characteristics (each creating their own subscale): 1) valence, 2) impact, 3) predictability, 4) challenge, 5) emotional significance, 6) change in world views, 7) change in social status, 8) external control, and 9) extraordinariness. In the validation of this scale, Luhmann et al. found that, when examining the influence of life events on subjective well- 6 being, these perceived characteristics significantly accounted for individual differences in well- being trajectories after a life event. Further, this predictive effect remained after controlling for pre-existing personality traits, age, and gender as covariates. Specifically, the characteristics of valence and challenge were associated with levels of both retrospective and prospective subjective well-being: participants who perceived a life event as more negative or challenging reported lower well-being over time; participants who perceived a life event as changing their worldview reported higher well-being over time (Luhmann et al., 2020). Indeed, the perception of a negative life event may play a large role in if and how a person experiences PTG. Negative Life Events and Close Others Of course, negative life events—and individuals’ experiences of them—do not occur in a vacuum. Often, when faced with challenges, individuals seek out social support from others (Taylor, 2011). In this way, close relationships, too, resemble a characteristic of the life event and may ultimately play a role in how individuals cope with or grow from trauma. Much research finds that social support plays an important buffering role in times of stress and that partner support, in particular, improves individuals’ outcomes (e.g., lessens stress or anxiety). This effect has been documented across a variety of stressful contexts, including throughout pregnancy, amidst serious medical diagnoses, and when breaking addictions, such as smoking (de Jong Gierveld & Van Tilburg, 1987; Mermelstein et al., 1986; Racine et al., 2019; Rini et al., 2006; Talley et al., 2010). Other work suggests that partner support may go beyond merely reducing the negative effects of stressful experiences; partner support may influence how individuals experience 7 positive change after a negative event (i.e., post-traumatic growth).1 Although work on partner and relationship characteristics affecting PTG is relatively rare, and often only examines relatively small groups of participants, it nevertheless offers some evidence that partners’ social support may facilitate an individual’s experience of PTG. This work could also be useful in explaining the heterogeneous findings seen in the life events literature. For example, in one recent study of growth among Korean women experiencing pregnancy loss, Yoon and colleagues (2022) found that partner support moderated the association between grief and PTG; in those with high partner support, more grief was related to higher PTG. However, in those with low partner support, grief was not as closely linked with PTG (Yoon et al., 2022). A similar conclusion was found in a 2013 study of stem cell transplant survivors. Specifically, social support from a spouse, particularly instrumental spousal support (i.e., assisting with tangible needs), positively predicted PTG (Nenova et al., 2013). Collectively, these and other nascent studies emphasize the positive influence that partners can have on individuals experiencing challenges and adverse circumstances. In the current dissertation, I focused on the reverse relationship: how an individual’s trauma or adversity can impact their partner or their relationship as a whole. Worth noting, this phenomenon, too, has received some attention in the literature. However, like work on individual personality change after trauma, the work done on this topic finds occasionally heterogeneous and contradictory results. On one hand, stress or trauma experienced by one person does occasionally affect the 1 Worth noting, even studies that critique the prospective influence of support on adjustment following negative life events acknowledge the role of support in encouraging better adjustment during the adaptation period in improving outcomes for individuals (Lucas & Chopik, 2021). 8 psychological functioning of their partner. This is most often evident in literature examining cross-over (from one person to another) or spill-over effects (from one context to another), such as vicarious experiences or work-life management. For example, after controlling for acts of discrimination experienced by individuals, acts of discrimination against a partner were negatively associated with an individual’s self-rated health and positively associated with depressive symptoms (Wofford et al., 2019). These effects were explained (i.e., mediated) by the negative effects that these discrimination experiences have on relationship functioning. More generally, burdens experienced in one domain of life (or by a close other) might also “spill over” into close relationships, and individuals may find themselves indirectly experiencing a partner’s stress (e.g., financial or job strain creating relationship difficulties; Norling & Chopik, 2020; Trail & Karney, 2012). Yet other work has found that the declining cognitive health of an individual can lead to poorer outcomes (i.e., increased loneliness) for their partners (Leggett et al., 2020). Such findings align with broader work on vicarious or secondary trauma. Vicarious or secondary trauma occurs when one incurs negative impacts of being close to someone who has experienced trauma, whether in a professional or personal setting, despite not personally experiencing the trauma. Sometimes described as the “contagion” of trauma (Gill-Emerson, 2015), a handful of studies have documented the negative effects (i.e., increased psychological distress) that can come from being close to someone who has experienced trauma (Gill-Emerson, 2015; Huggard et al., 2017; Smith et al., 2014). Other research on social networks suggests a comparable “spreading” of negative emotions and health problems primarily through people’s interactions with one another (Christakis & Fowler, 2007; Rosenquist et al., 2011). Of course, the person directly navigating the trauma often has the most immediate and severe consequences. But it may be surprising that those who are close to traumatized individuals experience negative 9 “ripple” effects in the wake of that trauma. Indeed, negative emotions are often found to be “contagious” (Hancock et al., 2008; Hill et al., 2010; Kimura et al., 2008; Kramer et al., 2014; Prochazkova & Kret, 2017), especially among those who are very close to one another (Mazzuca et al., 2019). However, trauma is not always associated with poorer outcomes for partners and relationships. In some cases, enduring a stressful event with a partner is associated with positive or resilient relationship outcomes. A longitudinal examination of couples found that after surviving a natural disaster, Hurricane Harvey, newlywed couples actually experienced a temporary boost in relationship satisfaction (Williamson et al., 2021). Other longitudinal work has found that partners of individuals experiencing a negative health event (e.g., a cancer diagnosis) demonstrate post-traumatic growth, often along with the diagnosed individual (Schroevers et al., 2010; Thornton & Perez, 2006; Weiss, 2004; Zwahlen et al., 2010). Some of this work suggests that partners’ experiences of PTG in these scenarios are directly related. Partners’ levels of PTG are often associated with one another (Hodges et al., 2005; Weiss, 2002, 2004; Zwahlen et al., 2010). Further, in cross-sectional examinations of both patients and their partners, a partner’s PTG predicts an individual’s own PTG over and above other growth- relevant variables (such as social/marital support, depth of commitment, and the intensity of the traumatic stressor; Weiss, 2004). Some of this research suggests that people’s responses to their partner’s adversity might be attributable to how their partner frames and experiences the adversity (thus setting the stage for partners to feed off or model positive growth seen in the individual experiencing personal PTG). Partners, then, appear to be susceptible not only to the negative ripple effects of an individual’s trauma but also to positive changes or growth. However, this work is often limited to examining PTG with respect to well-being or specific 10 growth indicators and rarely examined with respect to personality traits. Although there is some evidence for heterogeneous outcomes for partners, why are couples’ post-traumatic outcomes linked in these ways in the first place? Several mechanisms can help partially explain the link between partners’ shared experiences of PTG. Some researchers suggest that the mechanisms responsible for the shared experience of growth and positive emotion between partners after trauma may simply be the same broad mechanisms responsible for shared psychological distress (i.e., emotional contagion; Prochazkova & Kret, 2017; Zwahlen et al., 2010). Put simply, close others may experience more positive or negative emotions simply by interacting with partners who experience positive or negative emotions more frequently. Indeed, recent literature suggests that those who work closely with traumatized individuals, despite not being directly traumatized themselves, experience both vicarious growth and vicarious trauma (Barre et al., 2023). Other researchers suggest that there are more nuanced, couple-specific factors that influence how partners grow in response to each other’s trauma. For example, some work suggests that vicarious PTG relies on the resiliency—the ability to maintain well-being in the face of adversity (Herrman et al., 2011)—of the partner who is not directly experiencing the challenge. Partners with higher levels of resiliency are more likely to experience PTG (as are their partners; Zhang et al., 2021). Perhaps having a partner who is especially emotionally stable or positive in challenging times (both hallmarks of resiliency) provides an environment that is more conducive to personal growth. Other work points to the importance of partner responsiveness: partners’ ability to “understand, value, and support each other in fulfilling important personal needs and goals” (Reis & Clark, 2013). Work by Canevello and colleagues (2016) suggests that there is no direct link between individual and partner PTG at all. Rather, it is more likely that an individual’s PTG 11 leads them to become a more responsive partner, which, in turn, facilitates the PTG of their partner (Canevello et al., 2016). In this process, living through a challenge may prompt individuals to alter their priorities, shifting focus to caring for and validating their partner. This increased responsiveness is perceived by partners (Canevello & Crocker, 2010) and may prompt growth in many ways—perhaps encouraging trauma-specific disclosure and fostering cognitive processing (Calhoun & Tedeschi, 2014), reminding partners of positive coping techniques and strengths (Calhoun & Tedeschi, 2014; McMillen, 2004), or simply serving as a peer model for growth (Canevello et al., 2016; McMillen, 2004). Other models offer potential explanations for how negative life events may influence relationships as a whole. The Vulnerability Stress Adaptation Model (Karney & Bradbury, 1995), for example, offers a broad framework for relationship satisfaction that incorporates negative life events. In this model, the quality of a relationship is directly influenced by a couple’s ability to adapt to stressors (i.e., adaptive processes). This ability depends on each individual’s particular vulnerabilities and the external stressors that they, as a couple, might be exposed to. In this model, negative life events, and how successfully couples can cope with them, are central to a couple’s relationship satisfaction and stability. This model is especially helpful when attempting to explain potentially counterintuitive findings on relationship satisfaction and negative life events—such as the boost in satisfaction found after Hurricane Harvey (Williamson et al., 2021). Perhaps couple members who are, individually, less vulnerable (e.g., more resilient, more likely to use positive coping styles), are better able to cope with stressful events, resulting in successful adaptive processes (for them, their partner, and their relationship). Perhaps being able to successfully overcome hardship as a couple then boosts confidence and satisfaction in the relationship (e.g., “We can survive anything together”, “I’m 12 glad I have my partner when things get hard”). Taken as a whole, the literature on negative life events and close others offers at least some speculative evidence that individual outcomes after negative life events, both good and bad, are influenced by the actions of partners. Inversely, an individual’s adversity can also impact their partner or their relationship. However, again, it is not clear how common this experience is and whether it is seen in broader personality traits over time. The State of the Literature & Methodological Considerations While the current literature offers evidence that negative life events can potentially change individuals, partners, and relationships, the strength of this evidence is somewhat murky. As mentioned by Bleidorn, Hopwood & Lucas (2018), it is relatively unclear when and how we can expect negative life events to alter people and their relationships. At least some of this ambiguity comes from methodological complications that have come to characterize the PTG literature (Jayawickreme & Blackie, 2014; Jayawickreme et al., 2021). In a broad review of the state of PTG literature, Jayawickreme et al. (2021) point out several serious limitations to the methods typically used to answer questions about PTG. Reliance on cross-sectional data is one such limitation. In typical PTG studies, researchers often ask participants to recount a time that they endured a challenge and retroactively determine how they may have changed in response to this challenge. Of course, participants’ memories are not completely reliable, and adverse memories may be even more difficult to accurately reflect upon (Sachschal et al., 2019; Van der Kolk & Fisler, 1995). Of course, only longitudinal data can be used to truly answer questions about changes over time. To determine if people and their partners genuinely experience positive change after a traumatic event, Jayawickreme and colleagues (2021) argue that researchers need data on individuals’ personality characteristics both before and after an adverse event (sometimes 13 for many years after, as change can be slow to unfold; Schroevers et al., 2010). In this dissertation, I used prospective personality data from before to after adverse life events. Jayawickreme et al. (2021) also advocate for researchers to be more intentional about the samples that they use, both in cultural variation (e.g., encouraging the use of nationally representative data and non-WEIRD samples) and in sample size (i.e., prospectively getting a large enough sample of people who have experienced a given event). It is understandably difficult to gather a nationally representative sample that is well-powered enough to quantitatively compare outcomes across people. To some degree, it is difficult to predict when people experience any particular life event, as many events have at least some element of randomness. One potential solution to this problem, Jayawickreme and colleagues suggest, is for researchers to take advantage of pre-existing longitudinal studies. While these datasets come with limitations of their own (i.e., they are often not designed to explicitly measure PTG), they can offer a well-powered data pool with nationally representative and prospective personality data. In this dissertation, I leveraged three large prospective studies from three different countries to assess how common PTG is. Finally, when focusing on the topic of PTG in the context of close relationships, relatively few studies have examined the dyad as the unit of analysis. This may be surprising when considering that some examinations of PTG describe it as a process that cannot happen in isolation, envisioning growth as something that happens through conversations with close others (Cordova et al., 2001; Schroevers et al., 2010) and being dependent on partner support (Yoon et al., 2022). Some authors have explicitly advocated for PTG to be modeled as a couple-level factor (Ávila et al., 2017). However, to date, there have been very few longitudinal studies that model PTG as a dyadic process. One of the few exceptions, in a sample of couples in which one 14 member was a veteran of the Yom Kippur War, found patterns largely consistent with the cross- sectional literature (Lahav et al., 2017). But even this study found some conflicting results— wives (non-veterans) PTG experiences predicted their husbands’ PTG over time, but husbands’ PTG was largely unrelated to their wives’ PTG over time. However, PTG was a mixed bag. Although wives’ PTG predicted more PTG in their husbands over time, it also predicted more PTSD and worse relationship quality over time. As this study demonstrates, considering dyadic outcomes in examinations of PTG can help uncover important nuances that have not yet been thoroughly examined. In this dissertation, I used dyadic data—personality data from both partners—to examine whether adverse life events experienced either personally or vicariously are associated with positive personality change. Further, following work suggesting that positive relationship characteristics are essential for PTG, I examined whether relationship quality predicts the occurrence of PTG. Finally, an important limitation of the work on PTG is that it relies exclusively on individuals providing self-reports of their life events and their personality characteristics. However, a great deal of personality research is dedicated to how personality affects important life outcomes, such as those found in close relationships (Roberts et al., 2007). There is also a practical limitation that repeatedly reflecting on one’s own personality and how it might change in response to life events might also affect whether PTG is detected. In fact, such an idea—that PTG is gained through active reflection—is a foundational idea in narrative approaches to personality that posit people make sense of their life histories in ways that aid in emotion regulation and maximize well-being (McAdams et al., 2001; Syed & McLean, 2022). However, an important additional piece of information is whether personality change or growth is “observable” by close others. Although observer ratings have a long history in the field of 15 personality psychology (Paulhus & Vazire, 2007; Vazire, 2010), there are not as many examinations of observer ratings of personality change over longer periods (McCrae, 1993; Oltmanns et al., 2020; Schwaba et al., 2022; Watson & Humrichouse, 2006). In this dissertation, I supplement the traditional examination of PTG in individuals by examining whether it is also detectable or “seen” by romantic partners. The Current Studies The current dissertation examined patterns of personality change in three samples of couples—Study 1 was a 12-year longitudinal study conducted in the United States, Study 2 was a 14-year longitudinal study conducted in the Netherlands, and Study 3 was a 2-year study conducted in Switzerland but comprised of couples from Switzerland, Germany, and Austria. The studies sought to address the aforementioned methodological limitations and answer the following research questions: RQ1: Do people exhibit positive personality change (PTG) after their partner experiences a negative life event (Studies 1-3)? RQ2: If so, how do these changes compare to the PTG of the individual who experienced the negative life event themselves (Studies 1-3)? RQ3: Do relationship characteristics (e.g., closeness, satisfaction) predict PTG for individuals and their partners (Studies 1-3)? RQ4: Do partners perceive PTG in individuals who experience a negative life event (Study 3)? As discussed above, longitudinal data, as well as large and representative samples, are especially important when answering questions about PTG. For this reason, the studies proposed to address these questions make use of three separate longitudinal panel studies. Each of these datasets (i.e., the HRS, LISS, and CouPers datasets) contained relevant variables, such as 16 personality traits, life event histories, and relationship quality indicators, measured over several time points. Overview of the Dissertation Studies Three studies assessed the prevalence and predictors of PTG in three samples of romantic partners followed over time. Study 1 examined how experiencing a life event, both directly (as the primary individual the life event impacts) and indirectly (as the partner of an individual who is directly impacted) influences the trajectory of personality change. This study used data from the Health and Retirement Study (HRS, an English-language survey), a nationally representative panel of older adults in the United States. This study also examined how spousal support and spousal strain may further impact growth trajectories through their direct and interactive (with life event occurrence) effects on personality change trajectories. Of course, this is not the only relationship-relevant variable that may influence growth, and in Study 2, an additional variable—relationship satisfaction—was integrated while I revisited this question using a nationally representative sample of the Dutch population (the Longitudinal Internet studies in the Social Sciences [LISS]; a Dutch-language survey). In Study 2, I determined if the trajectories of growth found in Study 1 replicated and whether relationship satisfaction affected personality change trajectories in a similar way as support and strain in Study 1. Study 3 tested additional relationship-relevant variables, including support, responsiveness, and closeness, in a panel study of couples from Switzerland, Germany, and Austria (the Processes in Romantic Relationships and Their Impact on Relationship and Personal Outcomes [CouPers] study; a German-language survey). In addition to replicating the trajectories 17 found in the first two studies, Study 3 examined whether self-reported trajectories of growth differ from partner-reported trajectories on the same individuals. Comparing these informants helped determine if partners have a fundamentally different view of PTG than individuals. Because life events are experienced differently by each individual (i.e., what is distressing for one individual may be pleasant for another), Study 3 also allowed me to examine life events that were explicitly rated as negative by participants (the other studies only feature checklists of life events). In total, this set of studies allowed me to address each of the research questions listed above and contribute to a growing body of work on the role of romantic partners in PTG. 18 CHAPTER 2: STUDY 1 Do people experience positive personality growth if their partner experiences a negative life event? Study 1 tested this possibility. The current literature on PTG within the context of close relationships suggests that partners can play important roles in the lives of those experiencing trauma, whether it be through buffering negative emotions or taking an active role in promoting growth. But which relationship-level mechanisms, specifically, are implicated in this process? Much of the work about partners and PTG centers around the support that partners provide in challenging times. Thus, Study 1 aimed to examine how partner support provision impacts individuals’ trajectories of change after a challenging event. Partner Support in PTG One theory suggests that support provision within close relationships makes PTG possible. Multiple forms of partner support (i.e., emotional and instrumental support) are correlated with and predict PTG (Nenova et al., 2013; Schroevers et al., 2010). The literature on partners and PTG suggests that the effect sizes of partner support rival or surpass that of other growth-relevant variables, such as characteristics of the trauma (Nenova et al., 2013; Schroevers et al., 2010), other (e.g., religious) support, and grief intensity (Yoon et al., 2022). For example, in a longitudinal examination of cancer survivors and their partners, Schroevers et al. (2010) found that partner support was associated with cancer survivors finding more “silver linings” in their illness (e.g., “I appreciate life more because of my illness”, “My illness strengthened my relationships with others”). Specifically, these effects of emotional support were seen in the short weeks and months following the diagnosis, a challenging time for many individuals and couples. Impressively, this association remained significant over longer periods—up to eight years after diagnosis (Schroevers et al., 2010). 19 In discussing this finding, the authors provide an important, albeit simple, explanation: merely talking about the adverse event with a close other provided opportunities for positive reflection about adversity. These conversations can facilitate positive reappraisal and coping skills, ultimately contributing to positive growth for the affected individual (Cordova et al., 2001; Luszczynska et al., 2005; Schulz & Mohamed, 2004). Other work examining the outcomes of distressed stem cell transplant survivors echoes these findings: instrumental support—the tangible assistance that partners provide (e.g., doing chores, running errands)—uniquely predicts PTG: having a reliable partner to take care of logistics alleviates daily stressors, giving the directly impacted individual the mental and emotional space to grow (Nenova et al., 2013). Occasionally, support from partners might not directly affect the likelihood of PTG, but, instead, might alleviate some of the hindrances to PTG. In a study of women experiencing pregnancy loss, for example, partner support altered the damaging effects that grief had on PTG (Yoon et al., 2022). Surprisingly, for those who had especially supportive partners, more grief was related to more PTG. Taken together, whether it is instrumental or emotional forms of support, this research suggests that partner support has the potential to facilitate PTG, although there is some uncertainty about what kinds of support are most important (and when) in this process—and why. Interestingly, while emotional and instrumental support measure two distinct types of support and are independently associated with PTG, there is some evidence of synergistic effects. Specifically, the best well-being outcomes, for both the provider and the recipient of support, come from instrumental support that is also emotionally engaged (Morelli et al., 2015). With this in mind, the prosocial instrumental support provided by spouses, who assumedly care for their partner with a great deal of emotional investment, may be especially effective in 20 creating an environment for PTG. Partner Strain in PTG As Nenova et al. (2013) argue, perhaps partner support is so appreciated because it eliminates stressors or strain that would otherwise be too onerous to handle for individuals on their own. However, when partners are unable to successfully cope with a stressful situation, they may become sources of additional stress. While little work has been done to explicitly examine the role of partner strain on PTG, there is some work to suggest that partners who are especially stressed may impede the coping of someone directly impacted by a crisis. In general, spousal strain, the feeling that a relationship carries many hassles or demands, is related to many negative outcomes, such as increased substance use (Brazeau & Lewis, 2021), loneliness (Saenz, 2021), and negative affect (DeLongis et al., 2004). This strain may be especially relevant in the context of relationships in which one or both partners are experiencing adversity, as partners of those who experience adversity often report intense feelings of stress or burden throughout the experience (Fredman et al., 2014; Greene et al., 2014; Verhaeghe et al., 2005). In a review of families of those with traumatic brain injuries, Verhaeghe (2005) found that partners are especially vulnerable to feelings of stress and strain after crisis, and, importantly, partners’ ability to successfully cope is an important factor in the recovery of the individual with the injury. If a partner is unable to cope successfully, then, they may hinder both the recovery and the growth of the directly affected individual. In Study 1, I examined trajectories of personality change in the face of life events experienced by people and their partners. I also examined the moderating role of support and strain in the effect that these life events had on trajectories. 21 Participants Method In Study 1, data was sourced from the Health and Retirement Study (HRS). The HRS, a longitudinal panel study administered by the University of Michigan, has surveyed a representative sample of approximately 20,000 Americans 50 years of age and older every two years since 1992. Every other wave (i.e., every four years), they are provided with an extended self-report questionnaire. From 2006 to 2020 (every four years, resulting in 4 total waves2), respondents and their spouses received a self-report psychosocial questionnaire that covers six broad areas: 1) subjective well-being; 2) lifestyle and experience of stress; 3) quality of social ties; 4) personality traits; 5) work-related beliefs; and 6) self-related beliefs3. Although response rates varied across wave and cohort, when considering all eligible respondents together, response rates for individuals were relatively high (i.e., all response rates were between 61.8-87.7%). Because the research questions I have proposed exist solely within the context of romantic relationships, I only used data from partnered participants. This resulted in a sample of 6,820 opposite-gender couples (N=13,640; 50% male, 50% female) with at least one assessment of personality. Participant age ranged from 25 to 67 (Mage = 62.16, SD = 10.27) and had an average education of 12.83 years (although this varied from less than 1 year of education to 17 years, SD = 3.18 years). As the dataset is nationally representative, participants were mostly White (71.4%), followed by Hispanic (12.7%), Black (12.3%), and other races (3.5%). Partners had 2 In 2006, to ease participant burden, they randomly assigned one-half of the total sample the self-report questionnaire. In 2008, the other random half received the self-report questionnaire. Thus, two cohorts of participants were formed, Cohort 1 (assessed in 2006, 2010, 2014, and 2018) and Cohort 2 (assessed in 2008, 2012, 2016, and 2020). However, given the random splitting of the sample and the equidistant waves, they were combined into one larger sample, which is how many HRS users have used the data. 3 Each of the following measures was sourced from this psychosocial questionnaire. 22 been together for an average of 32 years (although this, too, varied from less than 1 year to 70.5 years (SD = 16.14 years). Because household assets were measured with such a specific degree of fidelity (i.e., including a broad array of assets like income, social security, and property), I was able to operationalize wealth as the difference between assets and debts (M = $548,328.77, SD = $1,308,019.91). Measures Life Events In the biannual waves (i.e., every two years from 2006-2020), participants were asked to indicate if they had experienced a particular life event since the last time they took the survey. Because the research questions I proposed exist solely within the context of romantic relationships, I only examined non-relationship-related life events (i.e., I excluded life events like marriages, bereavements, or divorces/separations from my analyses). Some life events were near universal (e.g., a negative change in health was experienced by almost 96% of the sample), while others were relatively rare (e.g., only 4.7% of the sample experienced unemployment; see frequencies of all life events in Table 1). Personality In each self-report wave (i.e., every four years), Big Five personality traits were assessed with the Midlife Development Inventory (MIDI; Lachman & Weaver, 1997). Participants were given a list of 26 adjectives that assessed levels of neuroticism (e.g., “moody”, “nervous”), conscientiousness (“organized”, “responsible”), extraversion (e.g., “outgoing”, “talkative”), openness (e.g., “creative”, “imaginative”), and agreeableness (e.g., “warm”, “helpful”) and then indicated how much each adjective described them on a scale ranging from 1 (not at all) to 4 (a lot; all αs > .87). Means of each trait, as well as correlations between traits, are displayed in 23 Table 2. Spousal Support and Strain In each self-report wave (i.e., every four years), each participant was asked three questions about the support derived from their relationship with their spouse (e.g., “How much do they really understand the way you feel about things?”) and four questions about the strain derived from their relationship (e.g., “How much do they let you down when you are counting on them?”). Participants responded to each question on a scale ranging from 1 (a lot) to 4 (not at all4). On average, participants reported feeling generally supported by their spouse (Msupport = 3.55, SD = .479) and did not feel especially strained (Mstrain = 1.90, SD = .540). Spousal support (α = .86) and spousal strain (α = .86) were reliable and moderately stable over time. Analytic Plan To determine if and how life events experienced directly or vicariously impact the trajectory of personality, I employed a series of dyadic growth curve models in the context of multi-level modeling (see Kenny et al., 2006; one set of models for each Big 5 personality trait). I coded time such that the first wave was time zero. I estimated random intercepts and random slopes of time. In each model, I included actor and partner effects for all reported life events5 that were not relationship-related (as my analyses required intact couples). Therefore, each GCM modeled actor and partner effects of the following life events as main effects: new chronic illness, negative health changes, positive health changes, death of a parent, new job, retirement, and unemployment. Each GCM also included a series of control variables commonly 4 Strain responses were recoded so that higher levels indicated more strain. 5 As highlighted in Table 3, some life events (i.e., moving, having a child, or experiencing the loss of a child) were highly (or perfectly) correlated, indicating that partners almost always experienced these life events together. For these life events, only the actor effects were modeled. 24 acknowledged as sources of variance in personality and close relationships research (Bernerth & Aguinis, 2016): age, gender6, race, education, wealth, and relationship length7. Lastly, these models also included a series of life event by slope interactions (such that a significant interaction indicated that the occurrence of that life event has impacted the trajectory of a given trait). Upon a significant interaction of an actor or partner life event with the slope of any personality trait, I completed an additional GCM. In these models, I included 1) the main effect of spousal support8, 2) two-way interactions between support and all life events, 3) the two-way interaction between support and slope, and 4) three-way interactions between any significant actor/partner life event, slope, and support (such that a significant interaction indicates that the impact of that life event on slope depends on the level of support). For example, if only the actor effect of chronic illness was significant, the only new variables I added to the second GCM were 1) the main effect of spousal support, 2) interactions between support and life events, 3) a Support*Slope interaction, and 4) a Slope*Actor’s chronic illness* Support interaction. Because of the complexity of these models, I adjusted the alpha level to .01 for a more conservative p- value to protect against false positive effects. I originally planned to also model the moderating effects of strain on these effects as well. The motivation behind this plan was that support may engender personality growth and strain might hinder personality growth. However, upon running these analyses, I found the 6 Gender was effect-coded so that women = -1 and men = 1. 7 Wealth (i.e., household assets) and relationship length were log-transformed for these analyses. 8 Support was centered on invariant, time 1 support for these analyses. 25 results for strain to be largely redundant with the effects seen for support. Thus, in the interest of parsimony, I report only the results for support, although I can provide the results for strain upon request. Nevertheless, following the recommendations from the committee, I also completed a life events analysis with support and strain as outcomes as well to show the effects of life events on these relational indicators outside the context of personality traits. Below, I report the results of each model, organized by construct. Results Agreeableness Growth curve model 1: life events only The results of this model are presented in Table 4. The slope of agreeableness was not significant, indicating that agreeableness did not significantly change over time (b = .02, p = .144). Three life events demonstrated significant main effects on agreeableness. Those higher in agreeableness were more likely to get a new job (b =.05, p < .001), while those who were lower in agreeableness were more likely to experience a new chronic illness (b =-.02, p = .007) and unemployment (b =-.05, p = .007). Several control variables were also significant; people of color (b =-.02, p = .020) and those with longer relationships (b =-.11, p = .027) were slightly less agreeable, while women (b = -.14, p < .001) and those who had more education were slightly more agreeable (b = .01, p < .001) There were also significant interactions between four life events and the slope of agreeableness. The actor effects of both negative health changes (b =-.03, p =.031) and positive health changes (b =-.01, p =.015) resulted in steeper negative slopes, indicating decreases in agreeableness over time. Those who experienced the death of their own parent (b = .10, p <.001) or whose partner began a new job (b = .01, p =.032) had stronger positive slopes for 26 agreeableness, indicating increases in agreeableness over time. Growth curve model 2: life events and partner support As described in the analytic plan, I completed a secondary growth curve model that included spousal support and three-way interaction terms to examine how support altered personality changes associated with life events (i.e., support x life event x time). However, following the recommendation of my committee, I ran a reduced model that restricted the tests to life events that were associated with personality change in the previous analysis. In other words, I examined support’s moderating effect on only the life events that exerted an influence on personality change over time. In modeling these three-way interactions, I also modeled the constituent two-way interactions and I discussed them below as well. The results of this growth curve are reported in Table 5. The main effect of spousal support was significant, such that those with more support were more agreeable (b = .28, p < .001). There were also three significant interactions between support and life events. Among those who had not experienced the death of a child, those who had more support were more agreeable. While this was still true among those who had experienced the death of a child, the gap between high and low-support participants in agreeableness was smaller (b = -.04, p < .001). A similar interaction occurred for those who had experienced a negative health event. Among those who had not experienced a negative health event, those who had more support were more agreeable. While this was still true among those who had experienced a negative health event, the gap between high and low-support participants in agreeableness was smaller (b = -.195, p = .004). Lastly, among those whose partner had not retired, those who had more support were more agreeable. While this was still true among those who had a partner retire, the gap between high and low-support participants in agreeableness was larger (b = .028, p = .004). 27 However, there were no significant three-way interactions between the slope, life events, and support. Conscientiousness Growth curve model 1: life events only The results of this model are presented in Table 6. The slope of conscientiousness significant and positive, indicating that participants increased slightly in conscientiousness over time (b = .04, p = .008). Three life events demonstrated significant main effects on conscientiousness. Those who experienced a new chronic illness (b = -.04), as well as those who had partners who experienced a new chronic illness (b = -.04), were lower in conscientiousness (ps < .001). Those who got a new job were higher in conscientiousness (b = .06, p < .001). Four control variables also demonstrated significant main effects on conscientiousness. Women (b = - .06, p < .001), as well as those with more wealth (b = .03, p < .001) and education (b = .02, p < .001) tended to be more conscientious on average, and older adults tended to be less conscientious (b = -.01, p < .001). Lastly, there were several significant interactions between life events and the slope. Life events that were associated with a steeper positive slope (indicating steeper gains in conscientiousness over time) included the actor and partner effects of a parent’s death (b = .01, p < .001; b = .01, p = .019 respectively) and the partner effect of getting a new job (b = .01, p < .001). Life events that were associated with a steeper negative slope (indicating shallower increases in conscientiousness over time) included the death of a child (b = -.01, p = .002), and the actor and partner effects of a new chronic illness (b = -.01, p <.001; b = -.01, p = .017, respectively). The actor effect of a negative health change was also associated with a steeper negative slope (indicating decreases in conscientiousness over time; b = -.05, p = .002). 28 Growth curve model 2: life events and partner support As described above, I completed a second model that included the support variable and support interaction terms. The results of this model are outlined in Table 7. While the main effect of support was significant, such that people with higher spousal support were more conscientious on average (b = .19, p = .008), there were no significant two- or three-way interactions between support, the slope, and a given life event. Extraversion Growth curve model 1: life events only The results of this model are displayed in Table 8. The slope of extraversion was not significant, indicating that extraversion did not significantly change over time (b = .02, p = .101). Five life events demonstrated significant main effects on extraversion. Those who began a new job (b = .11, p = .001) or retired (b = .02, p = .020) tended to be higher in extraversion, while those who had developed a new chronic illness (b = -.03, p = .001), had a partner retire (b = -.03, p = .010) or had a partner become unemployed (b = -.05, p = .039) tended to be lower in extraversion. Every control variable was associated with extraversion; older adults and those with a longer relationship duration tended to be lower in extraversion (b = -.002, p = .003; b = - .02, p = .003, respectively) while women (b = -.06, p < .001), people of color (b = .06, p < .001), those with more education (b = .01, p < .001), and those with more wealth (b = .02, p < .001) tended to be higher in extraversion. Lastly, four life events demonstrated significant interactions with slope. Those who had experienced a new chronic illness or negative health change tended to have steeper negative slopes, indicating drops in extraversion over time (b = -.01, p < .001; b = -.04, p = .008, respectively). Those who experienced the death of their own parent (b = .02, p < .001) or had a 29 partner who began a new job (b = .01, p = .008) tended to have steeper positive slopes, indicating increases in extraversion over time. Growth curve model 2: life events and partner support As described above, I completed a second model that included the support variable and support interaction terms. The results of this model are outlined in Table 9. While the main effect of support was significant, such that people with higher spousal support were more extraverted on average (b = .22, p = .009), there were no significant two or three-way interactions between support, the slope, and a given life event. Neuroticism Growth curve model 1: life events only The results of this model are displayed in Table 10. The slope of neuroticism was significant and negative, indicating that neuroticism decreased slightly over time (b = -.08, p < .001). Seven life events demonstrated significant main effects on neuroticism; those who directly or vicariously experienced a new chronic illness (b = .07, p < .001; b = .03, p = .012, respectively), as well as those who directly or vicariously experienced unemployment (b = .08, p = .001; b = .05, p = .040, respectively) tended to be higher in neuroticism on average. Those who began a new job (b = -.06, p < .001), experienced the death of a child (b = -.04, p < .001), or had a partner who retired (b = -.02, p = .026) were lower in neuroticism. Several control variables also demonstrated main effects; older adults (b = -.01, p < .001), people of color (b = -.10, p < .001), more educated people (b = -.02, p < .001), and wealthier people (b = -.03, p < .001) all tended to be less neurotic. Women tended to be slightly more neurotic on average (b = -.06, p < .001). Lastly, there were a series of significant interactions between life event and the slope. 30 Those who had experienced the birth of a child (b = -.01, p = .039), a positive health change (b = -.01, p = .008), the death of a parent (b = -.01, p < .001), or direct (b = -.02, p < .001) or vicarious (b = -.02, p = .002) unemployment tended to have steeper negative slopes, indicating more dramatic declines in neuroticism over time. Meanwhile, those who experienced a new chronic illness (b = .01, p < .001) or a direct (b = .04, p = .005) or vicarious (b = .02, p = .036) negative change in health tended to have more positive slopes, indicating more shallow decreases in neuroticism over time. Growth curve model 2: life events and partner support As described above, I completed a second model that included the support variable and support interaction terms. The results of this model are outlined in Table 11. While the main effect of support was significant, such that people with higher spousal support were less neurotic on average (b = -.34, p < .001), there were no significant two or three-way interactions between support, the slope, and a given life event. Openness Growth curve model 1: life events only The results of this model are displayed in Table 12. The slope of openness was not significant, indicating that openness did not significantly change over time (b = .01, p = .359). Three life events demonstrated significant main effects of openness. Those who moved (b = .02, p = .023), started a new job (b = .07, p < .001), or became unemployed (b = .06, p = .004) tended to be higher in openness. Most of the control variables also demonstrated significant main effects: older adults (b = -.004, p < .001) and those who had been married for longer (b = -.04, p < .001) tended to be less open. People of color (b = .05, p < .001), those with more education (b = .04, p < .001), and those with more wealth (b = .03, p < .001) tended to be higher in openness. 31 Lastly, there were several significant interactions between the slope and life events. Those who had experienced a negative health change (b = -.03, p = .033) or a partner’s new chronic illness (b = -.01, p = .026) tended to have more negative slopes, indicating decreases in openness over time. Meanwhile, those who experienced the direct (b = .01, p < .001) or vicarious (b = .01, p = .040) passing of a parent, as well as a direct (b = .01, p = .038) or vicarious (b = .01, p < .001) acquisition of a new job tended to have more positive slopes, indicating increases in openness over time. Growth curve model 2: life events and partner support As described above, I completed a second model that included the support variable and support interaction terms. The results of this model are outlined in Table 13. The main effect of support was significant, such that people with higher spousal support were more open on average (b = .30, p <.001). The interaction between slope and support was significant and positive, indicating that those with higher spousal support reported increases in openness over time (b = .08, p =.005). There was also a significant two-way interaction between one life event and support: among those who had not experienced a negative health change, those who had more support were more open. While this was still true among those who had experienced a negative health change, the gap between high and low support participants in openness was smaller (b = -.20, p = .009). Lastly, there was a significant three-way interaction between the slope of openness, an actor’s negative health change, and support. To decompose this interaction, I estimated the two- way support X slope effect among those who experienced a negative health change versus not. Among those who did not experience a negative health change, the interaction between slope and 32 support was significant and positive (b = .14, p = .003), indicating that those higher in support tended to experience increases in openness over time. Among those who did experience a negative health change, the interaction between slope and satisfaction was not significant (b = - .002, p = .665), indicating that those higher in support did not experience significant gains in openness over time if they had also experienced this life event. Spousal Support and Strain In addition to the GCMs of personality traits, I also completed separate GCMs for spousal support and spousal strain as outcomes (reported in Tables 14 and 15, respectively). Each of these models were identical to the trait models above; support or strain was merely swapped for as the dependent variables. Each of these models estimated intercepts and slopes, main effects of the same actor/partner life events, main effects of the same control variables, and the same series of life event by slope interactions (such that a significant interaction indicates that the experience of that life event altered the trajectory of support or strain). Support The results of this model are outlined in Table 14. The slope of support was significant and positive, suggesting that participants felt more support over time (b = .07, p < .001). Many life events were associated with less spousal support, including the death of a child (b = -.05, p = .002), directly (b = -.05, p < .001) or vicariously (b = -.02, p = .022) experiencing a new chronic illness, a partners’ positive health change (b = .-.04, p = .004), and direct (b = -.06, p = .005) or vicarious (b = -.07, p = .002) unemployment. Men (b = .01, p < .001) and those with more wealth (b = .02, p < .001) and education (b = .004, p = .022) tended to feel more supported, while people of color tended to feel less supported (b = -.05, p < .001). Lastly, two life events interacted with the slope. Those who had a partner experience a 33 negative health change tended to have more negative slopes, indicating shallower gains in support over time (b = -.03, p = .009). Those who started a new job tended to have more positive slopes, indicating steeper gains in support over time (b = .01, p = .029). Strain The results of this model are outlined in Table 15. The slope of strain was significant and negative, indicating that participants felt less strain in their relationships over time (b = -.07, p = .001). Five life events demonstrated positive main effects of strain; those who directly (b = .06, p < .001) or vicariously (b = .04, p < .001) experienced a new chronic illness, directly (b = .08, p = .002) or vicariously (b = .10, p < .001) experienced unemployment, or had a partner experience a positive health change (b = .05, p = .004) all tended to report higher strain. People of color (b = .11, p < .001), women (b = -.04, p < .001), and those in longer-lasting relationships (b = .03, p < .001) also tended to report higher strain, while older people (b = -.002, p = .004) and wealthier people (b = -.02, p < .001) tended to report lower strain. Lastly, there were two significant interactions between life events and the slope of strain. Those who experienced the death of a child tended to have more negative slopes, indicating steeper drops in strain over time (b = -.01, p = .040). Those who had a partner experience a negative health change tended to have more positive slopes, indicating shallower decreases in strain over time (b = .05, p < .001). Growth mixture modeling Lastly, as introduced in the proposal of this dissertation, I completed a series of growth mixture models for each personality trait in an attempt to identify and predict latent classes of personality change. The analytic approach and results of these analyses are detailed in Appendix 34 D.9 Discussion In Study 1, I examined if and how individuals’ own life events, as well as the life events of their partners, influenced the trajectory of their personality changes. Additionally, I examined how spousal support was implicated in this process: does having an especially supportive spouse facilitate more adaptive responses to negative life events? Lastly, I also examined trajectories of support and strain to determine if experiencing a given life event influenced relationship functioning. All these questions were answered using HRS data, a representative sample of older adults in the United States. Patterns of PTG As discussed in the introduction, despite being an inherently longitudinal phenomenon, PTG is rarely examined with longitudinal data. Even rarer still is an examination that uses large and representative longitudinal data. With this in mind, my first goal in this study was to simply look for any evidence of PTG. This sort of evidence could mostly clearly be found in the interactions between life events and the slope of personality changes in each model. In an adaptive response, supportive of the theory of PTG, an ostensibly negative life event would have a positive interaction with adaptive traits (i.e., agreeableness, conscientiousness, extraversion, or openness) and a negative interaction with maladaptive traits (i.e., neuroticism). This would reflect a relationship in which participants are actively “growing” in positive ways when exposed to a particular life event. As 9 These results are only available for the HRS sample (Study 1). As communicated to the committee, these models could not be completed in the other two studies due to interpretability issues (i.e., dominance of 1-class solutions). 35 summarized in Table 16, this was occasionally the case. Some negative life events, such as events that made participants less healthy (i.e., negative health changes and new chronic illnesses), were consistently related to maladaptive changes in personality (i.e., decreases in agreeableness, conscientiousness, extraversion, and openness and increases in neuroticism). For these types of events, there was no evidence of PTG. However, other ostensibly negative life events, such as the death of a parent, were consistently related to adaptive changes in personality (i.e., increases in agreeableness, conscientiousness, extraversion, and openness and decreases in neuroticism). This offers some evidence of positive change in the wake of tragedy or trauma. Perhaps this effect, however, is unique to the sample: older adults. Perhaps, at this life stage, when a parent’s death is far more normative and can sometimes come with an alleviation of emotional, physical, or financial burden (Bialon & Coke, 2012; Johnson, 2007), personal growth is more feasible. Surprisingly, there was also some evidence of PTG when examining unemployment (which was coded independently of retirement), a life event often found to be especially negative (Bleidorn et al., 2018; Lucas et al., 2004). If individuals or their partners had experienced unemployment, they tended to decline more steeply in neuroticism. There are many possible explanations for this somewhat counterintuitive finding. One lies in the sample: perhaps when an older adult loses their job and a return to work feels unlikely (Kanfer & Bufton, 2015), unemployment serves as a sort of proxy for retirement (eliminating worries about work or finding another job). Alternatively, perhaps those who decline more steeply in neuroticism begin to like work less or burn out quicker (Bianchi, 2018; McCann, 2018), ultimately culminating in being let go from a job. Or, perhaps, this is simply a demonstration of PTG, whereby a typically negative life event is met with personal coping and reframing skills, ultimately leading to growth. 36 Other life events that were more ambiguous, such as getting a new job, were also related to adaptive outcomes (i.e., steeper increases in agreeableness, conscientiousness, extraversion, and openness). Getting a new job, especially in older adulthood, is likely ambiguous on both the between and within person level (where each individual may have mixed feelings and different individuals may feel very different about it; Kanfer & Bufton, 2015). While this pattern does not offer strong support either for or against PTG, it does suggest that starting a new job in late life may be connected to positive personality change. Actors and Partners A second goal of this study was to discover if participants experienced vicarious growth: positive personality change explained by a partner’s experience rather than their own. In general, individuals’ own experiences were far more influential than those of their partners; only about a third of the significant effects detailed in Table 18 are partner effects. A smaller subset of those effects, yet, are reflective of positive vicarious growth. The strongest partner life event by slope interaction can be attributed to a partner’s new job (which was associated with steeper inclines in agreeableness, conscientiousness, extraversion, and openness). However, as discussed above, a new job is not typically considered to be a negative or potentially traumatic life event, and this interaction ultimately offers little support for the theory of vicarious PTG. Only two stereotypically negative life events experienced by a partner were significant: unemployment and the death of a parent. Those whose partners had lost a job saw steeper drops in neuroticism over time; those whose partners had lost a parent saw steeper increases in conscientiousness and openness over time. Interestingly, each of these partner effects was always accompanied by the same type of actor effect (e.g., a significant partner effect of unemployment on neuroticism was always accompanied by a significant actor effect of unemployment on neuroticism). These 37 findings would fit well within a framework that posits vicarious PTG as quite literally living your own experience via someone else. Perhaps individuals are only (or best) capable of growing through a partner’s life event when they have grown through the same event themselves. Or, perhaps, these life events simply exert the most influence in a dyad, and their effect is more likely to be felt across partners. Support and Strain Lastly, this study attempted to uncover the role of partner support in PTG. Some theories have suggested that partner support should be influential in PTG; those with more supportive partners may be more capable of positive personality change. However, this study did not find evidence to support that claim. While spousal support was a significant predictor of each personality trait (such that those who reported receiving more support were higher in agreeableness, conscientiousness, extraversion, and openness and lower in neuroticism), there was only a single significant three-way interaction between life event, slope, and support (i.e., between an actor’s negative health change, support, and openness), although this was not consistent with a pattern of PTG. Overall, however, how much support a participant received did not influence how a given life event impacted personality trajectories. The same was true for spousal strain. This finding may be surprising when considering the wealth of literature that implicates partner support in the adaptive processes of coping, reframing, and growth (Cordova et al., 2001; Luszczynska et al., 2005; Nenova et al., 2013; Schroevers et al., 2010; Schulz & Mohamed, 2004; Yoon et al., 2022). These effects could certainly be interpreted as null results. Perhaps life events or spousal support are relatively unimportant in personality change, especially when considering the heritable nature of personality and its tendency to remain so stable over time 38 (Bleidorn et al., 2022; Vukasović & Bratko, 2015). However, some recent work suggests that “invisible” support (i.e., support that is not detected and labeled as support by the receiver) may be the type of support that is more commonly linked to these sorts of positive outcomes (like adaptation and goal achievement; Girme et al., 2018; Girme et al., 2013). Perhaps invisible support, which cannot be captured with self-report measures of received support, is more likely to impact these kinds of trajectories. When examining support as an outcome, one typically negative life event—a partner’s negative health change—interacted with slope (such that those who had experienced this saw shallower increases in spousal support). This is an intuitive finding; experiencing these things might make it more difficult for partners to provide support, ultimately leading to lowered levels of felt support. However, this effect is not aligned with patterns of dyadic PTG—where a negative event improves the relationship or its functioning in some way. When examining strain as an outcome, the same life event (a partner’s negative health change) interacted with slope (such that those with a partner who experienced a negative health change saw shallower declines in spousal strain). Again, experiencing a negative health change might make it more difficult for partners to provide support, ultimately leading to lowered levels of felt support—creating a pattern that is not reflective of dyadic PTG. However, this study did uncover a small piece of evidence for the experience of dyadic PTG: those who experienced the death of a child event saw steeper decreases in strain. Work on child bereavement suggests that relational outcomes for parents can vary dramatically after this life event (Albuquerque et al., 2016). If interpreted as support for dyadic PTG, this effect suggests that couples who experience the death of a child may take special care not to be a strain on their partner in a particularly vulnerable time (Schwab, 1998). Alternatively, if child bereavement is the result of a prolonged 39 illness or a process associated with a large amount of stress, these declines could be a function of there being initially very high levels of strain at the start of the study among these couples. Overall, Study 1 found some evidence for PTG. Traditionally negative life events (i.e., the death of a parent and unemployment) were occasionally related to positive personality changes (i.e., adaptive changes in agreeableness, conscientiousness, extraversion, neuroticism, and openness). However, it was more common for negative life events to be associated with maladaptive change. Spousal support, while consistently related to mean levels of traits, appeared to be relatively uninfluential in adaptive changes in personality traits. There was some evidence of dyadic PTG though—those who experienced the death of a child saw steeper decreases in spousal strain. In evaluating the results of Study 1 more holistically, it is worth reflecting on the strength of evidence that life events affect personality development among couples. Specifically, the effects of life events (experienced both personally and vicariously) on personality development were a bit underwhelming in terms of their magnitude and how close some of the p-values were close to .05 (see Benjamin et al., 2018). Given the complexity of some of these models, it would have been reasonable to impose a more conservative alpha correction than the one I did. If I had done so, some of the effects reported above may not have reached statistical significance. More broadly, some of the more conservative takeaways from Study 1 are that personality is fairly stable, mean levels are relatively slow to change on average, and some negative life events may influence (improving or harming) relationship outcomes. In evaluating the design and results of Study 1, Study 2 had the opportunity to improve on several limitations. Specifically, Study 2 examines some of these same questions but in a large sample of couples from the Netherlands. Additionally, the samples in Study 1 and Study 2 40 offer a sample that is more age diverse (rather than the older adult sample used in Study 1). Further, given that the demographic composition of Study 2 is different from that of Study 1 (i.e., younger, with more assessment points to see whether a longer time frame is necessary to observe personality changes, and from a country whose social safety net might alter the effects of life events on psychological development), I had the opportunity to revisit this question of PTG more robustly. 41 CHAPTER 3: STUDY 2 In Study 1, I examined trajectories of personality change in a nationally representative sample of U.S. older adults. In Study 2, I attempted to replicate these trajectories with a new nationally representative sample followed over a longer period with shorter intervals, albeit participants from Study 2 came from the Netherlands. In addition, I examined how a different relationship-relevant factor, relationship satisfaction, was related to these trajectories of growth. Relationship Satisfaction After Trauma Perhaps unsurprisingly, adverse and traumatic events are typically related to decreases in relationship satisfaction (Bakhurst et al., 2018; Fayed et al., 2021; Vanbergen et al., 2020). Some work finds that, once an individual has experienced an adverse event, it can take intense intervention to even maintain relationship satisfaction, but relationships may never ultimately bounce back to pre-adversity levels (Fayed et al., 2021). Overall, stress is damaging to relationships and marital satisfaction (Randall & Bodenmann, 2009; Randall & Bodenmann, 2017), and negative life experiences are often characterized as extremely stressful. In a seminal review of stress and close relationships, Randall & Bodenmann (2009) interpret stress in a relational context, finding that stress’ impact on relationship satisfaction depends on the locus of control (i.e., external: from outside of the relationship vs. internal: from within the relationship), intensity (i.e., major: critical experiences requiring adaptation vs. minor: everyday hassles), and duration (i.e., acute: temporary, lasting only a few days vs. chronic: lasting serval months or more) of the stressor. Within this framework, most negative or traumatic life events can be considered external, major stressors (whether they are considered temporary or long-lasting varies a bit). It may seem intuitive to assume that these types of stressors, which are relatively uncontrollable and require long-term adaptation, have the strongest negative effect on 42 relationship satisfaction. However, Randall & Bodenmann (2009) found that external, minor, and chronic stressors had the strongest negative impact on relationship satisfaction. These findings fit well within the authors’ “stress-divorce model,” which posits that small, external, daily hassles contribute to stress spillover (as briefly discussed in the introduction), ultimately undermining relationship satisfaction. So, while negative or traumatic life events are typically bad for relationships, they may not be as damaging as the everyday struggles faced by most couples. There are some couples who, perhaps counterintuitively, report higher relationship satisfaction after enduring a major adverse event. As discussed in the introduction, Williamson et al. (2021) found that newlyweds who experienced Hurricane Harvey together reported increased relationship satisfaction immediately after the event. Other work highlights the existence of remarkably resilient couples in the face of major stressors. For example, a small study of parents in Finland found that, when comparing couples who had children during the COVID-19 pandemic to couples who had children during non-pandemic times (i.e., 2015), pandemic couples reported being just as happy in their relationships (Isokääntä et al., 2023). The same is true of romantic couples more generally during the COVID-19 pandemic (Williamson, 2020). In a qualitative examination of couples raising children with autism spectrum disorder, many couples indicated that a sense of “being in it together” (Sim et al., 2019) helped them maintain relationship satisfaction—a sentiment that may keep the satisfaction of many couples afloat in challenging times and in the face of profoundly stressful events. What makes some couples so resilient or even grow/flourish in response to negative life events? Two possible mechanisms include self-pruning—identifying and decreasing negative traits—and sacrificing with satisfaction—the genuine desire to self-sacrifice for the well-being of a partner or relationship. Aydogan & Dincer (2020) found a direct relationship between a 43 couple’s resilience after a negative life event and their ability to self-prune: couples who were better able to self-prune were more resilient. This relationship was partially mediated by a couple members’ satisfaction with sacrificing: partners who participated in self-pruning more often were also more satisfied with sacrificing, and, in turn, more resilient (Aydogan & Dincer, 2020). With all of this in mind, it is possible that, while negative life events are not typically predictive of increased relationship satisfaction, couple members most likely to grow might be uniquely satisfied with their relationships when faced with challenges. Relationship Satisfaction and PTG In some cases, relationship satisfaction may also serve as an indicator of PTG. For example, in a recent examination of couples who recently had a premature birth, Okay & Güler (2021) found that relationship satisfaction and PTG were positively correlated—parents who were happier in their relationships also reported higher PTG. These parents were also less stressed, depressed, and anxious. Interestingly, parents who described themselves as more emotionally dependent during this time were also higher in PTG, perhaps emphasizing the need for partner support in challenging times. Other work finds that couples who see personal and relationship growth after negative events are somewhat atypical. In a cluster analysis of parents whose children underwent stem cell transplants, Riva et al. (2014) found that parents tended to fall into one of four categories: 1) a low distress and low PTG group (~20% of the sample), 2) a high distress group (~15% of the sample), 3) a low distress/some PTG group (~39% of the sample), and 4) a high PTG group (~25% of the sample). Relationship satisfaction was highest for parents in the high PTG cluster than others (i.e., those with low and high distress). Qualitative analyses of military couples echo the notion that, while not every couple member finds themselves or their relationship growing after adversity, a small subgroup of 44 couples may. In these interviews, Wick & Goff (2014) found that post-deployment, couples tended to fall into two subgroups: those with high relationship satisfaction and those with low relationship satisfaction. Couples who reported positive relationship functioning (e.g., effective communication, good conflict management, and partner support) also reported higher relationship satisfaction and higher levels of PTG. So, while increases in relationship satisfaction and PTG may not be a universal experience after trauma, some literature suggests that a subgroup of couples does indeed experience positive post-traumatic outcomes, and that, for these couples, relationship satisfaction is often connected to the experience of PTG. Participants Method In this study, data was sourced from the Longitudinal Internet Studies for the Social Sciences (LISS) panel, which followed a representative, probability household sample of the Dutch population from 2008 to 2022, selected from the Netherlands’ population register (those without Internet or computer access were provided access to these resources). This online panel follows roughly 5,000 households (~ 7,500 individuals) over time, with each participant completing monthly questionnaires, as well as a core series of online questionnaires once a year. This dataset has a relatively low attrition rate from year-to-year (~10%); with 80% of household members participating in the surveys and roughly half of the dataset identifying as married (i.e., 45.7%, 45.3%, and 44.8% in 2008, 2009 and 2010). This resulted in a sample of 3,481 opposite-sex couples (N= 6,962, 50% male, 50% female). The sample was predominately (i.e., 97.4%) Dutch. As outlined in my proposal, I had initially intended to include race/ethnicity as a control variable. However, with only 2.6% of this sample being non-Dutch, I ultimately did not have enough diversity to examine the effects of 45 ethnicity and excluded this variable. Participant age ranged from 18 to 99 years old, with an average age of 48.33 (SD = 14.53). Participants had been in a relationship with their partners for an average of 15.72 years, although this ranged from under 1 year to 63 years (SD = 15.77). 39.8% of the sample indicated that their highest level of education was high school or less, while 25.5% attended a vocational or junior college, 24.4% attended college, and 10.2% attended university. Measures Life events Throughout the study, panel members were asked if they had experienced a series of life events. As in Study 1, because my research questions relate solely to partnered people, I only examined non-relationship-related life events. Unlike the HRS sample, no life event was nearly universal; the most experienced event was a negative change in health, reported by 43.3% of the sample (see frequencies of all life events in Table 17). The rarest life event by far was getting a first job, experienced by only .3% of the sample. In my proposal, I mentioned excluding events that were experienced by a very small subset of the sample. Thus, I excluded getting a first job from the following analyses. Personality In all waves, the Big Five personality traits were assessed with the 50-item International Personality Item Pool (IPIP) version of the Big Five Inventory (Goldberg, 1992). Participants were asked to indicate how accurate they found descriptor sentences of themselves (e.g., for extraversion, “I am the life of the party”). Responses were measured on a 5-point Likert scale ranging from 1 (very inaccurate) to 5 (very accurate; average α across waves > .76 for all traits). Means of each trait, as well as correlations between traits, are displayed in Table 18. Traits were 46 extremely stable over time. Relationship satisfaction In all waves, relationship satisfaction was measured with a single item: “How satisfied are you with your current relationship?” The answer categories ranged from 0 (entirely dissatisfied) to 10 (entirely satisfied, αs > .96). In general, participants were very satisfied with their relationships (Msat = 8.35, SD = 1.35). Analysis Plan The analysis plan for Study 2 follows the same structure as Study 1. For each trait, I completed a series of dyadic growth curve models in the context of multi-level modeling (see Kenny et al., 2006). Each GCM included: 1) a series of actor and partner life events10 (i.e., the birth of a child, death of a child, death of a parent, a negative health change, a positive health change, a new chronic illness, retirement, and unemployment), 2) a series of control variables (i.e., age, gender, relationship length, and level of education), and 3) a series of life event by slope interactions. Life events that had significant interactions with slope were carried through to a second GCM, which added the following predictors to the model: 1) these three-way life event*Slope*Relationship satisfaction interactions11, 2) two-way interactions between relationship satisfaction and each life event, 3) the two-way interaction between relationship satisfaction and slope, and 4) the main effect of relationship satisfaction. Because of the complexity of these expanded models with several two- and three-way interactions, I adjusted the alpha level to .01 for a more conservative p-value to protect against false positive effects. 10 As highlighted in Table 19, some life events (i.e., having a child or experiencing the loss of a child) were highly (or perfectly) correlated, indicating that partners almost always experienced these life events together. For these life events, only the actor effects were modeled. 11 Relationship satisfaction was centered on the invariant time 1 mean for these analyses. 47 Results Agreeableness Growth curve model 1: life events only The results of this model are reported in Table 20. The slope was significant and negative, indicating that participants generally decreased in agreeableness over time (although this effect was very small; b = -.01, p < .001). Two life events had main effects of agreeableness: those who experienced a negative health change or new chronic illness tended to be more agreeable on average (b = .04, p = .026; b = .04, p = .025, respectively). Older participants (b = .004, p < .001), women (b = -.17, p < .001), and those with more education (b = .03, p < .001) also tended to be higher in agreeableness. Two life events interacted with the slope (although each of these effects were small and close to p = .05): those who had experienced the birth of a child tended to have more positive slopes, indicating shallower decreases in agreeableness over time (b = .01, p = .041); those who had experienced the death of a child tended to have more negative slopes, indicating more dramatic decreases in agreeableness over time (b = -.01, p = .044). Growth curve model 2: life events and relationship satisfaction As described above, I completed a second model that included relationship satisfaction and its associated interaction terms. The results of this model are displayed in Table 21. The main effect of relationship satisfaction was positive and significant, indicating that those high in agreeableness also tended to be satisfied with their relationship (b = .04, p < .001). Somewhat surprisingly, the interaction between satisfaction and the slope was significant and negative, indicating that those higher in relationship satisfaction tended to experience steeper decreases in agreeableness over time (although this effect was very small, b = -.002, p < .001). Only one other 48 interaction with support was significant: the 3-way interaction between the slope, death of a child, and satisfaction (b = -.007, p < .001). To decompose this interaction, I estimated the two-way relationship satisfaction x slope interaction among those who experienced child bereavement versus those who did not. Among those who did not experience the death of a child, the interaction between slope and satisfaction was significant and negative (b = -.002, p < .001), indicating that those higher in relationship satisfaction tended to experience slightly steeper decreases in agreeableness over time. Among those who did experience the death of a child, the interaction between slope and satisfaction was also significant and negative, although slightly stronger (b = -.008, p < .001), indicating that those higher in relationship satisfaction tended to experience even steeper decreases in agreeableness over time if they had also experienced this life event. Conscientiousness Growth curve model 1: life events only The results of this model are reported in Table 22. The slope was not significant, indicating that participants did not significantly change in conscientiousness over time (b = .00, p =.800). Three life events demonstrated main effects of conscientiousness: those who experienced a positive health change tended to be higher in conscientiousness (b = .07, p < .001), while those who retired (b = -.05, p = .011) or had a partner experience the death of a parent (b = -.04, p = .037) tended to be lower in conscientiousness. Older adults (b = .003, p =.001), women (b = -.04, p < .001), those in longer relationships (b = .002, p = .022), and those with more education (b = .02, p < .001) also tended to be higher in conscientiousness. Lastly, four life events interacted with the slope. Those who had a partner experience the death of a parent tended to have more positive slopes, indicating increases in conscientiousness 49 over time (although this effect was very small; b = .003, p = .049). Those who had experienced a negative health change (b = -.004, p = .037), a new chronic illness (b = -.003, p = .048), or the retirement of a partner (b = -.01, p < .001) tended to have more negative slopes, indicating decreases in conscientiousness over time. Growth curve model 2: life events and relationship satisfaction As described above, I completed a second model that included relationship satisfaction and its associated interaction terms. The results of this model are displayed in Table 23. The main effect of relationship satisfaction was positive and significant, indicating that those high in conscientiousness also tended to be satisfied with their relationship (b = .04 p < .001). However, there were no interactions between satisfaction, the slope, and life events. Extraversion Growth curve model 1: life events only The results of this model are reported in Table 24. The slope was negative and significant, indicating that participants slightly decreased in extraversion over time (b = -.01, p = .014). Three life events had main effects on extraversion. Those who experienced unemployment tended to be more extraverted on average (b = .10, p = .004), while those who directly (b = -.05, p = .005) or vicariously (b = -.04, p = .035) experienced a new chronic illness tended to be lower in extraversion. Those with more education also tended to be higher in extraversion (b = .03, p < .001). Lastly, there were two significant interactions between life events and the slope. Those who experienced a positive health change tended to have more positive slopes, indicating shallower declines in extraversion over time (b = .01, p = .001). Those who had experienced a negative health change (b = -.01, p = .002) tended to have more negative slopes, indicating 50 steeper decreases in extraversion over time. Growth curve model 2: life events and relationship satisfaction As described above, I completed a second model that included relationship satisfaction and its associated interaction terms. The results of this model are displayed in Table 25. The main effect of relationship satisfaction was positive and significant, indicating that those high in extraversion also tended to be satisfied with their relationship (b = .05 p < .001). However, there were no interactions between satisfaction, the slope, and life events. Neuroticism Growth curve model 1: life events only The results of this model are reported in Table 26. The slope was negative and significant, indicating that participants slightly decreased in neuroticism over time (b = -.02, p < .001). Three life events had main effects on neuroticism. Those who had experienced a new chronic illness (b = .11, p < .001) tended to be more neurotic on average, while those who experienced a positive health change (b = -.08, p = .001) or a partner’s retirement (b = -.05, p = .036) tended to be less neurotic. Those with longer relationships (b = -.002, p = .020) and more education (b = -.05, p < .001) also tended to be less neurotic, while women tended to be more neurotic (b = -.12, p < .001). Lastly, there were four significant interactions between life events and the slope. Those who experienced a negative health change (b = .01, p < .001) or a new chronic illness (b = .01, p = .001) tended to have steeper positive slopes, indicating shallower declines in neuroticism over time. Those who had experienced a positive health change (b = -.01, p = .028) or unemployment (b = -.01, p = .006) tended to have steeper negative slopes, indicating steeper decreases in neuroticism over time. 51 Growth curve model 2: life events and relationship satisfaction As described above, I completed a second model that included relationship satisfaction and its associated interaction terms. The results of this model are displayed in Table 27. The main effect of relationship satisfaction was negative and significant, indicating that those low in neuroticism also tended to be satisfied with their relationship (b = -.08 p < .001). The interaction between satisfaction and the slope was significant and positive, indicating that those higher in relationship satisfaction tended to experience more shallow declines in neuroticism over time (although this effect was near zero, b = .004, p = .009). There was also a significant two-way interaction between childbirth and relationship satisfaction (b = .08, p =.002): among those who had not experienced the birth of a child, those who were satisfied in their relationship tended to be less neurotic. However, among those who had experienced the birth of a child, those high and low in relationship satisfaction were similarly high in neuroticism. However, there were no significant three-way interactions between the slope, life events, and relationship satisfaction. Openness Growth curve model 1: life events only The results of this model are reported in Table 28. The slope was negative and significant, indicating that participants slightly decreased in openness over time (b = -.01, p = .002). Four life events had main effects on openness. Those who had experienced a positive health change (b = .05, p = .003), a partner's retirement (b = .05, p = .012), or unemployment (b = .08, p < .001) tended to be higher in openness on average. Those who experienced a partner’s negative health change (b = -.03 p = .035) tended to be lower in openness. Men (b = .03, p < 52 .001) and those with more education (b = .10, p < .001) tended to be higher in openness, while those in longer relationships tended to be lower in openness (b = -.004, p < .001). There were three significant interactions between life events and the slope. Those who experienced a negative health change (b = -.01, p = .012) tended to have more negative slopes, indicating steeper decreases in openness over time. Those who had experienced the birth of a child (b = .01, p = .003) or a positive health change (b = .01, p < .001) tended to have more positive slopes, indicating less dramatic declines in openness over time. Growth curve model 2: life events and relationship satisfaction As described above, I completed a second model that included relationship satisfaction and its associated interaction terms. The results of this model are displayed in Table 29. In this model, neither the main effect of relationship satisfaction nor any of its interactions were significantly associated with openness. Relationship Satisfaction Growth curve model: life events only The results of this model are reported in Table 30. The slope was negative and significant, indicating that participants slightly decreased in their relationship satisfaction over time (b = -.05, p < .001). Three life events had main effects of relationship satisfaction. Those who experienced a new chronic illness (b = -.11, p = .002) or a partner’s unemployment (b = - .15, p = .011) tended to be less satisfied with their relationships, while those who retired felt more satisfied with their relationships (b = .20, p < .001). There were four significant interactions between life events and the slope of relationship satisfaction. Those who experienced the birth of a child (b = -.02, p = .006) tended to decline more dramatically in relationship satisfaction over time. Those who experienced the direct (b = 53 .01, p = .007) or vicarious (b = .02, p = .005) death of a parent, as well as those whose partners retired (b = .01, p = .002) tended to have more positive slopes, indicating less dramatic declines in relationship satisfaction over time. Discussion Patterns of PTG As in Study 1, the first goal in this study was to simply look for any evidence of PTG. Again, evidence could mostly clearly be found in the interactions between life events and changes in personality. In an adaptive response, supportive of the theory of PTG, an ostensibly negative life event would have an interaction (i.e., promoting linear growth) with changes in adaptive traits (i.e., agreeableness, conscientiousness, extraversion, or openness) and a negative interaction (i.e., promoting more dramatic declines) with changes in maladaptive traits (i.e., neuroticism). This would reflect a relationship in which participants are actively “growing” in positive ways when exposed to a particular life event. As summarized in Table 31, this was very rarely the case. As in Study 1, negative life events that impacted participant health (i.e., negative health changes and new chronic illnesses) were consistently related to maladaptive changes in personality (i.e., decreases in conscientiousness, extraversion, and openness and increases in neuroticism). For these types of events, there was no evidence of PTG. The same is true of positive life events (i.e., positive health changes) which were consistently related to adaptive responses (i.e., increases in extraversion and openness, decreases in neuroticism). No evidence of PTG is found in these patterns. However, two ostensibly negative life events were occasionally related to adaptive changes in personality: the death of a partner’s parent (i.e., increases in agreeableness) and unemployment (decreases in neuroticism). Interestingly, these negative life events indicated 54 similarly adaptive change in Study 1, although the effects found here were less consistent (i.e., only influencing a single trait). In discussing Study 1, I framed these results in the context of the sample: perhaps, for older adults, the death of a parent is normative, and perhaps the loss of a job serves as a proxy for retirement at later stages of life. However, this sample was notably younger than the HRS sample (i.e., Mage = 48.3 in LISS vs 62.2 in HRS) and likely in a different life stage (e.g., middle adulthood vs. late adulthood). Certainly, the death of a parent in middle-to-late adulthood is still a normative event that may be accompanied by changes in (or even reductions in) emotional, physical, or financial burdens. However, in this sample, losing a job is less likely to be a retirement proxy. As mentioned in Study 1, perhaps those who decline more steeply in neuroticism begin to like work less or burn out quicker (Bianchi, 2018; McCann, 2018), ultimately culminating in being let go from (or quitting) a job. Alternatively, this could be interpreted as a demonstration of PTG, whereby a typically negative life event does indeed lead to growth. Some work in sociology examines job loss as a “status passage”—an ongoing process in which one reevaluates goals or identity (Ezzy, 1993). Perhaps, in this state of reevaluation, individuals are given space to re-prioritize what they want in their work life. Indeed, this may include shifting away from things that bring them anxiety (like struggling in a stressful job), ultimately reducing neuroticism. Of course, the ability to relax or self-expand after unemployment is a privilege, typically only afforded by those with stronger financial resources and without significant financial strain (like the strain of having children; Backhans & Hemmingsson, 2011). The size of this effect, in both Study 1 and Study 2, is very small, suggesting that, if this were the case, there were very few people privileged enough to participate in this growth. 55 Actors and Partners A second goal of this study was to discover if participants experienced vicarious growth: positive personality change explained by a partner’s experience rather than their own. In this study, there was only one significant interaction between slope and a partner’s life event—the death of their parent. This result, too, was found in Study 1: participants with partners who had lost a parent had steeper increases in conscientiousness. Perhaps this reflects the uptick in responsibilities that occur after the death of a parent or parent figure. Alternatively, it could be the case that when an individual experiences the loss of a parent, their partner participates in more conscientious behaviors to give them space to grieve and heal, alleviating daily responsibilities and logistics (as noted by Nenova et al., 2013). Interestingly, and unlike Study 1, the actor effect of a parent’s death was not significant. Perhaps because the death of a partner’s parent includes more psychological distance, the negative event feels less proximal, creating more favorable grounds for growth. Relationship Satisfaction Lastly, this study attempted to uncover the role of relationship satisfaction in possibly cultivating PTG. Some theories suggest that high relationship satisfaction can serve as an indicator of PTG, with especially happy couples reporting more PTG (Okay & Güler, 2021; Purol & Chopik, 2024; Riva et al., 2014), or negative life events making couples feel closer or happier together (Williamson, 2020; Williamson et al., 2021). Largely, this study did not find much evidence to support that claim. The main effect of relationship satisfaction was significant for all traits except openness, with more satisfied participants reporting higher agreeableness, conscientiousness, and extraversion, and lower neuroticism. There were also two interactions between satisfaction and the slope: those happier with their relationships tended to have 56 shallower declines in neuroticism over time but decreased more dramatically in agreeableness over time—although these effects were both very small (both bs < |.005|). The only life event to have a significant interaction with relationship satisfaction was the birth of a child: participants who were particularly happy in their relationships and had a new child tended to be as high in neuroticism as those who were lower in relationship satisfaction. Although this particular analysis was not designed for causal explanations (i.e., this was a two-way interaction pooled across waves), this finding does align with work suggesting that the birth of a child can increase stress, worry, and anxiety, even among happy couples (components of neuroticism; Bleidorn et al., 2016; van Scheppingen et al., 2018). Overall, however, relationship satisfaction was most commonly associated with mean levels of personality traits, rather than modulating how personality changed over time; it was rarely associated with adaptive or positive change. When examining relationship satisfaction as an outcome, two negative life events had positive impacts on the slope of relationship satisfaction: the death of one’s own or of a partner’s parent. As discussed in the previous sections, this life event was one of the few that was also associated with positive personality change across the Big 5 traits in Studies 1 and 2. There’s little literature that offers explanations for why a parent’s death, in adulthood, would improve relationship outcomes in their children’s relationship, specifically. If this is to be interpreted as dyadic PTG—improved outcomes on the couple level after a negative event—the PTG literature might suggest that couples who experienced this life event were able to successfully cope and grow, perhaps relying on positive relationship qualities (e.g., effective communication, good conflict management, and partner support; Wick & Nelson Goff, 2014). Summary Overall, Study 2 found only a small amount of evidence for PTG. Two effects consistent 57 with PTG from Study 1 were replicated in Study 2: actor’s unemployment and the death of a partner’s parent were both associated with positive personality changes (i.e., decreases in neuroticism and increases in conscientiousness, respectively). In general, when compared to Study 1, Study 2 found fewer actor and partner life events that were implicated in the process of personality change. Relationship satisfaction appeared to play little role in adaptive personality change, although it tended to be associated with positive traits on the mean level. There was some evidence that relationship satisfaction improved after couples experienced the death of a parent (regardless of which couple member lost the parent), which could be interpreted as evidence for dyadic PTG. Holistically, however, Study 2 found that evidence for PTG was relatively rare. One of the largest limitations to both Study 1 and Study 2 is that study personnel simply measured the experience (i.e., yes or no) of a life event rather than the perceived impact of that event. Study 3 improves upon this limitation, using only life events that participants reported as explicitly negative or distressing—a more direct assessment of predictions made from the PTG literature. Additionally, the first two studies were also limited in that they focused exclusively on self-reports and perceptions of personality. In thinking about PTG and personality change within the context of close relationships, a natural question is whether personality changes in response to life events are observable across partners. Knowing whether life events change partner reports of personality can shed some light on the impact that these events could have on relationship functioning. For example, if an individual perceives that their partner is growing in ostensibly positive ways after a life event (either their own or one that happens to an individual), presumably this might also be evidence of PTG in that it could enhance relationship functioning (by cultivating more positive personality traits in partners). Thus, Study 3 moves past self- 58 reports, examining how participants see their partner changing as they experience these life events. Study 3 also examined some of the same questions as Studies 1 and 2 (but with even more fine-grained assessments of a few months apart from each other), as well as improving upon these limitations, in a new sample of Swiss, German, and Austrian couples. 59 CHAPTER 4: STUDY 3 Why might close others be so central in an individual’s experience of PTG? Although there have been a few demonstrations of PTG being possible in the context of relationships, In Study 3, I hope to advance the study of this phenomenon by examining three major relationship- relevant mechanisms that might facilitate growth within and across people—support, partner responsiveness, and closeness. Additionally, I hope to overcome two methodological limitations that are often present in previous work on PTG: implicit categorization of certain life events as negative and reliance on self-report measures of personality. Relationship-relevant Mechanisms As discussed in Study 1, partner support is one of the most commonly proposed mechanisms in the discussion of PTG within the context of close relationships. Partner support, both emotional and instrumental, has been connected to many positive relationships outcomes, including higher PTG (Cordova et al., 2001; Nenova et al., 2013; Schroevers et al., 2010; Yoon et al., 2022). However, other mechanisms, such as partner responsiveness, offer alternative explanations for the relationship between partners and PTG. Responsiveness—a partner's ability to demonstrate that they understand and value an individual's needs—has been long-linked to relationship and life satisfaction (Reis, 2012; Reis & Clark, 2013; Selcuk et al., 2016). Indeed, in the field of relationship science, responsiveness is often considered the cornerstone on which all positive relationships are built (Reis, 2012). Although having a responsive partner is good for relationships when neither partner is experiencing a crisis, it may have extra benefits during challenging times. Responsiveness to mutual disclosures is thought to be one of the primary mechanisms that builds intimacy between 60 individuals over time and is why some dyads transition from strangers to intimate partners (Reis & Shaver, 1988). This is especially salient in the context of adversity. Being sensitive to and understanding of a partner’s needs during a time when they may feel especially vulnerable is important immediately after experiencing adversity—perhaps occasionally more important than providing instrumental support (Dagan et al., 2014). In an examination of patients with colorectal cancer and their partners, partners’ emotional responsiveness (e.g., understanding and validation) was more important in predicting patients’ depressive symptoms than other, more pragmatic dimensions (i.e., their partners’ caring behavior; Dagan et al., 2014). Other work has found that perceived partner responsiveness is associated with fewer reports of physical pain in veterans (this same relationship was not found for non-veteran partners; O'Neill et al., 2020). While responsiveness appears to be an important factor in maintaining well-being in stressful situations, others suggest that partner responsiveness is also key in understanding PTG (Canevello & Crocker, 2010; Canevello et al., 2016). Canevello and colleagues suggest that the correlation between an individual’s PTG and their partner’s PTG is not due to a direct causal link. Rather, they propose a pathway where an individual’s PTG leads them to become a more responsive partner, thereby ultimately facilitating the PTG in their partner (Canevello et al., 2016). In these studies, some of the key components predicting growth across partners are whether romantic partners adopt compassionate goals to better understand and listen to their partners (Jiang et al., 2022). These feelings of compassion after adversity might be necessary preconditions for even believing positive outcomes are possible (Canevello & Crocker, 2011). In this process, living through a challenge may prompt individuals to shift their priorities, offering more focus on the care and validation of their partner. This increased responsiveness may prompt growth in many ways—perhaps reminding partners of positive coping techniques and strengths 61 (Calhoun & Tedeschi, 2014; McMillen, 2004), encouraging trauma-specific disclosure and fostering cognitive processing (Calhoun & Tedeschi, 2014), or serving as a peer model for growth (Canevello et al., 2016; McMillen, 2004). The relationship between individual and partner PTG is further complicated by a sense of shared identity that partners often report (i.e., a shared sense of self that overlaps; Aron & Aron, 1996). Feeling close to others has been connected to PTG in the literature as those who report greater closeness (with a particular close other or with people in general) tend to report higher PTG (Baník et al., 2022; Hall et al., 2010). This may not be surprising considering that greater closeness or connectedness with others has been conceptualized as a feature of PTG itself (i.e., closeness sometimes also serves as a measure of PTG; Cann et al., 2010; Tedeschi & Calhoun, 1996). For partners who are vicariously experiencing trauma or adversity, closeness is also likely important. Negative emotions commonly experienced after adversity, like stress and depression, are often considered “contagious” in that they can be jointly experienced by people sharing a social network (Hancock et al., 2008; Hill et al., 2010; Kimura et al., 2008; Kramer et al., 2014; Prochazkova & Kret, 2017). This is especially the case among people who are very close (Mazzuca et al., 2019). There is a large body of literature on the vicarious adversity experienced by those close to a directly affected person but ultimately not directly impacted by it, such as healthcare workers and caretakers (Baird & Kracen, 2006; McNeillie & Rose, 2021). Indeed, working closely with individuals navigating adversity can have negative psychological consequences for others in their network, and gerontology and healthcare scientists often study caregiver burden as a salient example of vicarious adversity. However, some work suggests that these individuals also experience vicarious growth when in the presence or social network of someone experiencing adversity. One meta-analysis on the subject posits that, in merely 62 witnessing clients’ resilience and growth, therapists may experience personal growth and development themselves (McNeillie & Rose, 2021). Because emotional contagion is often strong in romantic couples (Mazzuca et al., 2019), it is possible that couple members higher in closeness report growth when their partner does. However, again, most of the evidence for this phenomenon comes from retrospective and introspective assessments of growth, rather than prospective measures of psychological characteristics. Methodological Limitations: Life Event Perceptions and Self-reports of Personality In addition to the care provided by partners, how an individual perceives a life event is also important for PTG trajectories. What is a devastating event for one person may be a joyous event for another (e.g., an unexpected pregnancy). Luhmann et al. (2020) found that valence— perceiving an event as negative or positive—is integral for understanding trajectories of growth: participants who perceived life events as more negative or challenging reported lower well-being over time after those events. However, although the perception of a negative life event may play a large role in whether a person experiences PTG, it is rarely examined as a predictor of prospective personality change in this context (although see recent work by Haehner et al., 2022). In the previous studies in this dissertation, life event valence was not explicitly measured and was left to be assumed (e.g., I assume that unemployment is a negative life event and categorized it as such; positive health changes are likely positive). Of course, this may not always be the case as some life events may be a little more ambiguous in terms if they are considered a blessing or a curse (Rakhshani et al., 2022). The participant’s perspective on a life event, as Luhmann et al. (2020) capture with the Event Characteristics Scale, can more accurately characterize the impact and features of life events. The data used in Study 3 also directly assessed participants’ perceptions of the valence of life events. 63 Past work has also overlooked at least one way in which romantic partners can further the study of PTG. Namely, partners may not solely serve as facilitators or beneficiaries of growth, but as observers of growth. Partners, who typically spend a great deal of time with the directly impacted individual, can offer unique insight that may help determine when—or, even, if— individuals do indeed change after adversity. Some have argued that PTG is merely a positive illusion that individuals use after the fact to cope with adversity (Infurna & Jayawickreme, 2019; Kunz et al., 2019; Sumalla et al., 2009). Perhaps the point of view of someone close to an individual who has experienced adversity—but is not the individual themselves or has not experienced the adversity directly— may be valuable in determining if individuals have undergone personality change. In other words, people who experience the life event themselves may be “too close” to the event to meaningfully introspect about variation in their personalities. Romantic partners are often close enough to the impacted individual to detect meaningful changes in personality and may have enough psychological distance from the event to provide a more objective measure of change. People are certainly not completely objective observers of their partner’s traits (Purol & Chopik, 2022), and they tend to evaluate each other more positively than others might. However, individuals are indeed capable of at least somewhat accurately identifying their partner’s personality traits (Fletcher, 2015; Fletcher & Kerr, 2010; Neff & Karney, 2005; Purol & Chopik, 2022), especially when the trait they are reporting on has some relative consensus or a set of criterion behaviors (e.g., people generally agree on what makes someone physically attractive or extroverted; Bashour, 2006; Eisenthal et al., 2006). Long-term partners are also capable of providing observer ratings of personality change over longer periods, something that is relatively rare in the personality change literature (McCrae, 1993; Oltmanns et al., 2020; Schwaba et al., 2022; Watson & Humrichouse, 2006). With all of 64 this in mind, partner reports of PTG may help determine if individuals’ perceptions of their own PTG vary from the perceptions of those around them. In this study, I examined whether these relationship mechanisms might explain personality change in response to life events and whether personality changes are observable across romantic partners. Method Participants In this study, data was sourced from The CouPers Study (Couples and Personality; Processes in Romantic Relationships and Their Impact on Relationship and Personal Outcomes), an online study funded by the Swiss National Science Foundation (SNSF) which tracked couples for four waves over two years (with the first two follow-up waves occurring 4-6 months after the previous ones, and the final wave 10-12 months after the last). This sample afforded 482 opposite-gender couples (after filtering for couples who had been together for the entire duration of the study; N= 964, 50% male, 50% female). Participant age ranged from 18-81, with the average age being 39.35 (SD = 17.72). On average, couples had been together for 8.5 years, although this ranged from under 1 year to 67 years (SD = 10.79). Most participants reported having education beyond high school (43.7% reported attending a university, 6.3% reported attending a technical school), although many people in the sample reported high school (31.8%) or levels below high school (18%) as their highest level of education. Most of the sample (35.2%) reported making the equivalent12 of $0-22,743 annually, followed by $23,880-45486 (22.9%), and, then, no income (13.7%); 2.5% of the sample reported making over $136,459 (the highest income category listed on the closed-response item). 12 Converted from Swiss francs and rounded to the nearest US dollar. 65 Measures Life events While this panel study did not measure prospective life events, it did retrospectively capture some negative experiences that overlap with much of the life event literature. In Wave 4 (i.e., the final wave), participants were asked if they had experienced a series of life events during the study duration, and, if they had, how meaningful (where 1 = very negative and 5 = very positive) and distressing (where 0 = not all at and 10 = very much) each event was. Life events rated as negative (i.e., 4 and higher on the meaningfulness scale) or distressing (6 and higher on the distress scale) were included in the following analyses. As in the first two studies, I only examined non-relationship-status-related life events. These events include the birth of a child, the birth of a grandchild, graduation, retirement, unemployment, change in career, moving residences, children moving out of the house, the onset of a health problem, loss (of a non- partner close other), miscarriage, and abortion. This study also included a free-response option, where participants could choose to disclose another significant change in their life not captured by the preceding list of life events. The frequency of life events is displayed in Table 32. Unlike the previous frequency tables, this is only the prevalence of life events that participants found either negative or distressing (e.g., a frequency of 0 for the life event of “having a grandchild” does not indicate that no one in the sample had a grandchild, rather, it indicates that no one in the sample had a grandchild and identified that event as negative or distressing). The most common life event reported here was the death of a close other, experienced and indicated as negative by 24.3% of the sample. Some life events, when experienced, were almost never indicated as negative or distressing (i.e., having a grandchild, retiring, becoming an empty-nester, or having an abortion; all frequencies < 1%). As in previous studies, these extremely rare life events were 66 excluded from the following analyses. Correlations between life events within a couple are displayed in Table 33. Personality In all waves, personality traits were assessed with the Big Five Inventory (BFI; John & Srivastava, 1999). This 45-item assessment asked participants to indicate how much they felt certain personality descriptors applied to them (e.g., for extraversion, “I see myself as someone who is full of energy”) on a scale of 1 (disagree strongly) to 5 (agree strongly; α across waves >.85 for all traits). Means of each trait, as well as correlations between traits, are displayed in Table 34. Traits were extremely stable over time. Partner Description of Personality13 In all waves, participants were also asked about the Big Five traits of their partner. For this assessment, the short form of the BFI was used (John & Srivastava, 1999). For this 21-item measure, partners were asked how much they felt certain personality descriptors applied to their partner (e.g. for extraversion, “He/she is outgoing, sociable”) on a scale of 1 (disagree strongly) to 5 (agree strongly; α across waves >.81 for all traits). Means of and correlations between each trait are displayed in Table 35. Traits were extremely stable over time. Correlations between self and partner-reported personality are reported in Table 36. Relationship-level factors Bivariate correlations among the Big Five traits and these relationship-level factors can be seen in Tables 34 and 35. 13 Two variables capture the first measurement of partner report of personality: one measured on the first day of Wave 1 data collection, and one measured on the last day of Wave 1 data collection (14 days later). All other waves have single measurements. For the sake of consistency, the first day’s measurements are used for Wave 1 in the following analyses. 67 Responsiveness. In all waves, perceived partner responsiveness was measured with a scale by Laurenceau et al. (2005). This 6-item scale asked participants to indicate how responsive their partner was that day (e.g., “Today I felt validated by my partner”) on a scale from 1 (very little) to 5 (a great deal; α = .91). In general, participants felt their partners were very responsive (M = 3.95, SD = .65). Received support. In all waves, received support was measured with a scale by Shrout et al. (2006). This scale includes 4 items, 2 of which asked if they received emotional support from their partner that day and 2 of which asked if they received practical support from their partner that day (recoded as 1= yes, 0= no). These items were averaged over time to create a variable that represented the proportion of days participants reported felt support. Because practical and emotional support were consistently correlated across waves (all rs between practical and emotional support in each wave were significant and greater than r = .61), I combined these variables into a single support variable. On average, participants reported feeling some type of support approximately 36% of the time across the 14-day duration of each wave of the study (or for about 5 of the 14 days). Interpersonal closeness. In all waves, interpersonal closeness was assessed with the Inclusion of Other in Self Scale (Aron et al., 1992). This visual scale consists of a series of seven images: two circles (one labeled “self” and one labeled “partner”), which begin as separate from one other (1) and slowly get closer until they are almost completely overlapping in the final image (7). Participants are asked to choose the picture that best describes their relationship with their partner, with higher values (i.e., pictures in which the circles share more overlap) indicating more interpersonal closeness (α = .65). In general, participants felt very close to their partners, although there was a considerable amount of variance in this rating (M = 6.30, SD = 3.17). 68 Analysis Plan The analysis plan for Study 3 follows the same structure as Studies 1 and 2. For each trait, I completed a series of dyadic growth curve models in the context of multi-level modeling (see Kenny et al., 2006). Each GCM included: 1) a series of actor and partner life events (i.e., the birth of a child, graduation, a change in job, moving, the death of a close other, miscarriage, and “other”—the free-response option), 2) a series of control variables (i.e., gender, age, education level, income, and relationship length) and 3) a series of life event by slope interactions. Life events that had significant interactions with slope were carried through to a second series of GCMs. These analyses added three relationship variables14 to the model: support, responsiveness, and closeness (each relationship variable was modeled independently, resulting in three total follow-up analyses). In these follow-up models, the following predictors were added to the model: 1) these three-way life event by slope by relationship variable interactions, 2) two-way interactions between relationship variables and each life event, 2) the two-way interaction between relationship variables and the slope, and 4) the main effect of a given relationship variable. Because of the complexity of these expanded models, I adjusted the alpha level to .01 for a more conservative p-value to protect against false positive effects. As in Studies 1 and 2, I also completed CGM for each relationship variable (i.e., support, responsiveness, and closeness) as an outcome. Given that these variables were correlated at small to moderate levels, they were examined as separate predictors (i.e., not combined). These models each included 1) the main effect of the same series of life events, 2) the same series of control variables, and 3) a series of slope * life event interactions. 14 Each relationship variable was centered on the invariant, time 1 mean of that variable. 69 Finally, I re-ran the first set of GCMs on each partner report of personality for each trait. These models were identical to the models containing self report of personality, although the interpretation is slightly different. In these models, a significant interaction between an actor’s life event and slope indicates that when an individual experiences a particular life event, they perceive their partner’s personality changing in a particular way. Similarly, a significant interaction between a partner’s life event and slope indicates that when a partner experiences a particular life event, individuals perceive their partner’s personality changing in a particular way (i.e., the outcome is the person’s rating of their partner’s personality). Results Agreeableness Growth curve model 1: life events only The results of this model are reported in Table 37. The slope was significant and negative, indicating that participants generally decreased in agreeableness over time (although this effect was very small; b = -.01, p = .017). One life event had a main effect of agreeableness: those who experienced a move tended to be more agreeable on average (b = .14, p = .011). Those who had been in a relationship for longer tended to be lower in agreeableness (b = -.005, p = .002). Only one life event interacted with the slope (although this effect was small and p = .05): those with a partner who had experienced the death of a close other tended to have more negative slopes, indicating steeper decreases in agreeableness over time (b = -.02, p = .050). Although this effect was on the edge of significance, I carried it through to the next series of analyses. Growth curve model 2: life events and support The results of this model are reported in Table 38. Neither the main effect of support nor 70 any of its interactions with life events and the slope were significant. Growth curve model 3: life events and responsiveness The results of this model are reported in Table 39. Neither the main effect of responsiveness nor any of its interactions with life events and the slope were significant. Growth curve model 4: life events and closeness The results of this model are reported in Table 40. Neither the main effect of closeness nor any of its interactions with life events and the slope were significant. Conscientiousness Growth curve model 1: life events only The results of this model are reported in Table 41. The slope was significant and negative, indicating that participants generally decreased in conscientiousness over time (although this effect was very small; b = -.01, p = .034). Only one control variable demonstrated a significant main effect on conscientiousness: older people tended to be slightly less conscientious (b = -.003, p = .022). However, no life event had a significant impact on the intercept or slope of conscientiousness. Thus, as a result, I did not examine the moderating effect of any of the relationship variables on life event-induced personality changes (because there were not any significant effects). Extraversion Growth curve model 1: life events only The results of this model are reported in Table 42. The slope was not significant, indicating that participants did not significantly change in extraversion over time (b = .001, p = .918). Two control variables had a significant main effect on extraversion: women and those with higher incomes tended to be higher in extraversion (b = -.14, p = .007; b = .05, p = .026, 71 respectively). However, no life event had a significant impact on the intercept or slope of extraversion. Thus, as a result, I did not examine the moderating effect of any of the relationship variables on life event-induced personality changes (because there were not any significant effects). Neuroticism Growth curve model 1: life events only For this model, I experienced some convergence problems, likely caused by the high stability in neuroticism and the uneven distribution of some of the life events. In diagnosing the source of the issues, the culprit was the low degree of variance in the slopes for men and women. I was able to get the models to run by removing these random effects and their covariances across partners. This also occurred for partner-reported agreeableness (see below). The results of this model are reported in Table 43. The slope was not significant, indicating that participants generally did not significantly change in neuroticism over time (b = - .018, p = .223). One life event had a significant main effect on neuroticism: those who had a partner experience the death of a close other were slightly lower in neuroticism (b = -.122, p = .049). Women tended to be higher in neuroticism (b = -.219, p < .001), while older people and those with higher incomes tended to be lower in neuroticism (b = -.01, p = .029; b = -.03, p = .008, respectively). No life event had a significant impact on the slope of neuroticism. Thus, as a result, I did not examine the moderating effect of any of the relationship variables on life event- induced personality changes (because there were not any significant effects). Openness Growth curve model 1: life events only The results of this model are reported in Table 44. The slope was not significant, 72 indicating that participants generally did not change in openness over time (b = -.004, p = .406). Only one control variable demonstrated a significant main effect on openness; more educated people in the sample tended to be slightly more open (b = .02, p = .038). However, no life event had a significant main effect on openness. There was one significant interaction between life event and the slope of openness. Namely, those who had a miscarriage tended to have more positive slopes, indicating increases in openness over time (b = .08, p = .009). Growth curve model 2: life events and support The results of this model are reported in Table 45. Neither the main effect of support nor any of its interactions with life events and the slope were significant. Growth curve model 3: life events and responsiveness The results of this model are reported in Table 46. Neither the main effect of responsiveness nor any of its interactions with life events and the slope were significant. Growth curve model 4: life events and closeness The results of this model are reported in Table 47. Neither the main effect of closeness nor any of its interactions with life events and the slope were significant. Support Growth curve model 1: life events only The results of this model are reported in Table 48. The slope was significant and negative, indicating that partner support gradually decreased over time (b = -.05, p < .001). Women (b = -.11, p < .001), older participants (b = .01, p = .002), and those with more education (b = .03, p = .001) all reported higher levels of received support. While no life event indicated a main effect of support, two life events did interact significantly with the slope of support. Those who had a partner experience a negative health event (b = .03, p = .023) or the death of a close 73 other (b = .02, p = .031) tended to have a more positive slope, indicating a shallower decrease in support over time. Responsiveness Growth curve model 1: life events only The results of this model are reported in Table 49. The slope was not significant, indicating that felt responsiveness did not change significantly over time (b = -.01, p = .126). One life event demonstrated a main effect of responsiveness: those who had a partner who experienced a negative health event felt as though their partner was less responsive (b = -.22, p = .042). One life event demonstrated an interaction with slope: those who experienced a negative health event themselves tended to have steeper positive slopes (b = .05, p < .001), indicating an increase in felt responsiveness over time. Closeness Growth curve model 1: life events only The results of this model are reported in Table 50. The slope was significant and negative, indicating that closeness gradually decreased over time (b = -.34, p = .002). Men tended to report higher levels of felt closeness (b = .17, p < .001), while older participants (b = - .02, p = .039) and those with more education (b = -.05, p = .048) tended to report lower levels of closeness. However, no life event was significantly related to the intercept or slope of closeness. Partner Reports: Agreeableness For this model, I again needed to adjust the model by removing some of the random effects to achieve convergence. The results of this model are reported in Table 51. The slope was not significant, indicating that participants generally saw their partners as not changing in agreeableness over time (b = -.01, p = .301). Five life events demonstrated main effects of 74 perceived agreeableness: those who had a partner experience a graduation (b = -.14, p = .012) or the death of a close other (b = -.12, p = .009) tended to see those partners as less agreeable. Those who had a partner experience a miscarriage (b = .35, p = .031) or other negative life event (b = .21, p = .006) tended to see those partners as more agreeable. Those who experienced a miscarriage themselves tended to see their partner as less agreeable (b = -.37, p = .016). There was one significant interaction between life event and the slope: those who experienced a miscarriage themselves tended to have steeper positive slopes of perceived agreeableness (b = .17, p = .014), indicating that they saw their partner as increasing in agreeableness over time. Partner Reports: Conscientiousness The results of this model are reported in Table 52. The slope was not significant, indicating that participants generally saw their partners as not changing in conscientiousness over time (b = .01, p = .143). Men tended to see their partners as slightly more conscientious (b = .15, p = .002). While no life event demonstrated main effects of perceived conscientiousness, one life event interacted with the slope. Those who had experienced a miscarriage themselves tended to have steeper positive slopes (b = .10, p = .143), indicating that they saw their partner increasing in conscientiousness over time. Partner Reports: Extraversion The results of this model are reported in Table 53. The slope was not significant, indicating that participants generally saw their partners as not changing in extraversion over time (b = .001, p = .586). Men tended to see their partners as slightly more extraverted (b = .16, p = .014). However, no life event had a main effect on perceived extraversion nor interacted with the slope. 75 Partner Reports: Neuroticism The results of this model are reported in Table 54. The slope was not significant, indicating that participants generally saw their partners as not changing in neuroticism over time (b = -.01, p = .497). Men (b = .37, p < .001) and those in longer relationships (b = .01, p = .045) tended to see their partners as more neurotic (although the 95% C.I. for relationship length had a lower bound of zero). Those with more education tended to see their partners as less neurotic (b = -.05, p = .042). However, no life event had a main effect on perceived extraversion nor interacted with the slope. Partner Reports: Openness The results of this model are reported in Table 55. The slope was not significant, indicating that participants generally saw their partners as not changing in openness over time (b = -.003 p = .674). Men (b = .22, p < .001) tended to see their partners as more open. However, no life event had a main effect on perceived extraversion nor interacted with the slope. Patterns of PTG Discussion As in Studies 1 and 2, the first goal in this study was to simply look for any evidence of PTG. Again, evidence could mostly clearly be found in the interactions between life events and changes in personality. In an adaptive response, supportive of the theory of PTG, an ostensibly negative life event would have a positive interaction (i.e., promoting linear growth) with changes in adaptive traits (i.e., agreeableness, conscientiousness, extraversion, or openness) and a negative interaction (i.e., promoting more dramatic declines) with changes in maladaptive traits (i.e., neuroticism). This would reflect a relationship in which participants are actively “growing” in positive ways when exposed to a particular life event. As summarized in Table 56, this was 76 rarely the case. This study only identified two life events as having a significant impact on slope: the death of a partner’s close other and an actor’s miscarriage. Interestingly, while Studies 1 and 2 found adaptive changes in response to a partner losing a presumably close other (i.e., a parent), this was not the case in Study 3, where participants reported steeper declines in agreeableness when their partner lost a close other. The only life event in Study 3 that was associated with adaptive changes was experiencing a miscarriage. Participants who experienced a miscarriage that they found negative or distressing tended to have positive slopes in openness, becoming more open over time. This is a pattern consistent with PTG: positive change after an explicitly negative event. Some previous work links positive post-miscarriage outcomes (like relationship satisfaction and healthy coping) to a sense of openness, although this is often conceptually broader than the openness implicated in Big Five personality traits (Hiefner, 2021; Kiełek-Rataj et al., 2020). The authors of this work suggest that openness is a key component of effective communication and coping within dyads. Perhaps in the face of this particular life event, couples must become more communicative to handle a shared sense of loss, ultimately allowing them to grow in openness. Aside from this effect, however, there were no patterns of change consistent with PTG. While Studies 1 and 2 found a handful of potentially influential life events implicated in both adaptive and maladaptive change, this study found only two life events were relevant. Although there are many explanations for this phenomenon (including demographic differences in the sample, like culture or age; Costa Jr et al., 2001; McCrae et al., 1999), it is also possible that differences between the studies are methodological or related to sampling variability. One explanation lies in the nature of the data. This dataset spanned the shortest amount of time (i.e., only 2 years to HRS’ 12 and LISS’ 14). And, although the sample size was still 77 considerably large, it also included the smallest sample of couples (i.e., 482 couples to HRS’ 6,820 and LISS’ 3,481). Even among the larger samples, the effect sizes for life events tended to be very, very small. It is possible that to detect such small effects—that are occurring for such stable variables—the data needs to capture a very large sample of couples over a very long period. Indeed, some traits did not even have enough variation to estimate the random effects. Another explanation lies in how the life events were coded within this sample. Unlike the previous two studies, this data included some qualitative characteristics of life events (i.e., its valence). Perhaps eliminating life events that were not considered negative or distressing influenced this pattern of results (or that life events impacted people even if they were not considered particularly distressing). For example, in Studies 1 and 2, I assumed that experiencing the death of a parent was a negative life event. However, while this life event may have prompted negative emotion, if participants had been asked if the passing of their parent was negative or distressing, as they were asked in Study 3, they may have indicated that it was not (especially in an older sample where death sometimes marks the end of a period of struggle or suffering). In that case, it would be less accurate to categorize that response as a pattern of PTG (as the life event was not inherently negative or stressful). Partner-reported patterns of PTG A secondary goal of this study was to look for any evidence of PTG within partner reports of personality change. Perhaps partners have unique perspectives on how individuals grow over time. Interestingly, partner reports of personality were slightly different than self- reports. For example, the slope of partner-reported personality was never significant, suggesting that while individuals saw their own personality as changing slightly, their partners saw it as not changing much at all. 78 When examining the relationship between life events and mean levels of personality traits, the only large discrepancies between self and other reports were in agreeableness. Most of these effects were partner effects: participants rated partners who had experienced a graduation or the death of a close other as less agreeable and partners who had experienced a miscarriage or other negative/distressing event as more agreeable. However, overall, participants and their partners appeared to have similar perspectives on when life events accompanied personality change: extremely rarely (see Table 56). Only one life event interacted significantly with the slope of partner-reported traits: the actor effect of miscarriage. Those who had a miscarriage reported that their partners became more agreeable and conscientious over time. The partner effect of miscarriage for these traits was not significant in the self-report models, suggesting that participants did not see themselves as becoming more agreeable and conscientious when facing miscarriage; this was a change only perceived by partners. Parts of this finding align well with previous work. Partners may indeed engage in more agreeable or conscientious behaviors when their partner is directly experiencing a negative life event (i.e., being especially kind or handling logistics; Nenova et al., 2013). However, why this is a change perceived only by partners and not the participants themselves is more of an open question. Perhaps people experiencing miscarriages become especially sensitive to the agreeable and conscientious behaviors of their partners as they search for cues of support (Pickett et al., 2004; Sejourne et al., 2010), leading them to report increases in these traits. Alternatively, perhaps participants are making these small changes to their personality outside of their awareness. Or, perhaps, participants are thinking about their personality as a whole when answering the self-report measure, and not of their personality in the context of their relationship (as their partner is likely doing; McCrae et al., 1998). 79 Actors and Partners Another goal of this study was to discover if participants experienced vicarious growth: positive personality change explained by a partner’s experience rather than their own. In this study, there was only one significant interaction between slope and a partner’s life event—the death of their parent. And, unlike the previous two studies, this was a maladaptive change; participants who had a partner experience the death of a close other experienced steeper declines in agreeableness. This pattern does not align with one of PTG. As discussed above, this could be due to Study 3’s inclusion of only negative or distressing life events. This sample was younger than the samples in the previous two studies (i.e., Mage 39.35 in CouPers vs.. 48.3 in LISS and 62.2 in HRS). In this sample, a parent’s death is certainly less normative and, possibly, more likely to be perceived negatively. Perhaps, when deaths are perceived as negative and distressing, they are less likely to lead to positive reframing, growth, or positive personality change. Relationship variables Lastly, this study attempted to uncover the role of three relationship-relevant variables in the cultivation of PTG. Support, responsiveness, and closeness have all been theoretically implicated in this process. Having a partner who is especially supportive or responsive may create more favorable grounds for growth, providing important resources for coping. Having a partner who you feel especially close to may make the experience of vicarious growth more likely, as the negative event is more proximal. However, this study found little support for any of these proposed processes. Support Neither the main effect of support nor any of its interactions with slope and life events 80 were significant. These results echo that of Study 1. How much support a participant reported receiving was unrelated to the trajectory of their personality. As in Study 1, there are two possible explanations for these types of effects. On one hand, they could certainly be interpreted as a null result: perhaps life events or spousal support are relatively unrelated to personality change, especially when considering the heritable and stable nature of personality (Bleidorn et al., 2022; Vukasović & Bratko, 2015). On the other hand, as discussed in Study 1, some work implicates “invisible” support (i.e., support that is not detected and labeled as support by the receiver) in positive outcomes (Girme et al., 2018; Girme et al., 2013). Perhaps invisible support is more likely to impact these kinds of trajectories. Support was also examined as an outcome. When examined in this way, there was some evidence for PTG on the dyadic level (i.e., where experiencing a negative event is good for the relationship as a whole). Those who had a partner experience a negative health event or the death of a close other tended to have a more positive slope of support, indicating a shallower decrease in support over time. This is interesting when considering literature that suggests that experiencing these things might make it more difficult for partners to provide support, ultimately leading to lowered levels of felt support (as it may have in Study 1). Perhaps these negative life events served as an especially salient cue for support, reminding both partners to engage in support behaviors. Alternatively, when one partner provided support for the other (in response to a negative life event), it may have begun a reciprocal chain of support, leading to more support behavior from both partners (much like the responsiveness chain theorized by Canevello et al., 2016). Responsiveness Neither the main effect of responsiveness nor any of its interactions with slope and life 81 events were significant. As discussed in the introduction to Study 3, responsiveness is central to many theories of relationship satisfaction and growth, especially surrounding growth after trauma (Dagan et al., 2014; Reis & Shaver, 1988). It is intuitive that responsiveness—the ability to identify and meet a partner’s needs—might be important in a post-crisis scenario. Some work has situated responsiveness as a necessary precondition for even believing positive outcomes are possible (Canevello & Crocker, 2011). One possible explanation for these results lies in the way that responsiveness was measured; participants were asked about how responsive they felt their partner had been that day. This is in contrast to the personality assessments, which occurred over longer intervals of time. While responsiveness was relatively stable across waves, it is possible that a trait measure of responsiveness, or a measure that specifically captured a partner’s responsiveness in a crisis context, would yield different results. Another interpretation is that responsiveness is less related to personality change, especially when considering, as mentioned earlier, the heritable and stable nature of personality (Bleidorn et al., 2022; Vukasović & Bratko, 2015). Perhaps having a close other anticipate and meet needs, while important for relationship outcomes, is simply not as influential in altering personality. Responsiveness was also examined as an outcome. When examined in this way, there was some evidence for PTG on the dyadic level (i.e., where experiencing a negative event is good for the relationship as a whole). Those who experienced a negative health event tended to have a more positive slope of responsiveness, indicating an increase in responsiveness over time. This finding fits well within the responsiveness literature, which would predict a rise in responsiveness to meet a new need (quite literally in response to a negative life event, Canevello et al., 2016). 82 Closeness As was the case for support and responsiveness, neither the main effect of closeness nor any of its interactions with slope and life events were significant. Those who felt especially close to their partners were not any more or less likely to experience vicarious growth (i.e., there were no three-way interactions between the slope, closeness, and a partner’s life event). Although this finding is contrary to theoretical models which suggest closeness would be important for vicarious growth (Hancock et al., 2008; Hill et al., 2010; Kimura et al., 2008; Kramer et al., 2014; Prochazkova & Kret, 2017), it aligns with the larger pattern of results in this study (where relationship variables tended to be unimportant in personality change processes) and in all three studies (where partner effects of life events were considerably rarer than actor effects of life events). If a partner’s life events are rarely important for an individual’s patterns of personality change, it is likely that other factors, such as the characteristics of the event itself (e.g., its salience, its impact, etc.), would be more likely to drive a significant partner effect. Closeness was also examined as an outcome. However, unlike support and responsiveness, closeness never interacted significantly with any life events. Factors like support and responsiveness likely play a more central role in coping or reframing, processes that are more likely to occur after a negative life event (and, perhaps, lead to relationship growth). Closeness, then, may be a more ancillary variable, related to relationship processes and outcomes broadly, but not in the specific context of trauma. Summary Overall, Study 3 found very little evidence for PTG. Only one life event, miscarriage, predicted adaptive personality change (i.e., increases in openness). In general, when compared to Studies 1 and 2, Study 3 found far fewer actor and partner life events that were implicated in the 83 process of personality change. This pattern was largely consistent over self and partner reports, although participants who experienced a miscarriage did tend to see their partners as becoming more agreeable and conscientious over time—a change partners did not see in themselves. Support, responsiveness, and closeness all appeared to play little (to no) role in adaptive personality change. However, there was some evidence that relationship outcomes (i.e., support and responsiveness) improved after couples experienced negative life events (i.e., either partners’ negative health event or the death of a partner’s parent), which could be interpreted as evidence for dyadic PTG. Overall, while Study 3 did find some evidence for dyadic PTG, this study found little support for PTG on the individual level. 84 CHAPTER 5: CONCLUSIONS AND FUTURE DIRECTIONS The PTG framework has been used as a way to characterize why some individuals and couples grow in the face of adversity. My approach leveraged insights from the study of life events potentially leading to personality change, theories about how negative life events might cultivate introspection and enhance relationship functioning, and addressed several methodological limitations that have plagued previous research. As discussed in the introduction of this dissertation, there is no clear consensus on when, how, or even if PTG exists. When conceptualized very broadly, as changes in Big 5 personality traits, increases in character strengths, gains in resilience, changes in narratives, or benefit- finding, there is some evidence that this phenomenon does exist (Bleidorn et al., 2018; Helgeson et al., 2006; Joseph & Linley, 2005; Pals & McAdams, 2004). The literature contains many reports of participant growth in the aftermath of trauma or negative life events (e.g., finding increases in qualities like gratitude, hope, kindness, leadership, and love after tragedies such as the September 11th terrorist attacks; Peterson & Seligman, 2003). However, these findings are far from consistent across the literature. Different studies find different results even when examining the same life event, occasionally producing incompatible results. For example, experiencing divorce is associated with both increases and decreases in conscientiousness, depending on the literature consulted (Costa et al., 2000; Specht et al., 2011a). In a meta-analysis of the topic, Mangelsdorf and colleagues (2019) found some evidence of what they called “genuine” PTG: positive, stable increases in personal strengths, autonomy, and self-esteem after negative events. However, these changes were very heterogeneous, suggesting that the average meta-analytic estimate might not necessarily characterize many of the samples included. Some indicators of well-being did have positive trajectories on average, suggesting an increase in 85 adaptive traits after a negative event. Other indicators tended not to change at all, while others decreased after negative events (Mangelsdorf et al., 2019). One avenue to exploring this heterogeneity lies in the social context of the life event, particularly whether it is experienced in the context of a close relationship or even vicariously. Much work suggests that close others, particularly romantic partners, may play an important role in growth after a negative life event. Indeed, many frameworks position partners as active contributors, hurdles, or beneficiaries to growth processes (e.g., the Vulnerability-Stress- Adaptation Model; Karney & Bradbury, 1995; Purol & Chopik, 2024). Some work suggests that partners provide the emotional and practical support required for individuals who have experienced a negative event to grow (e.g., having conversations that facilitate coping, offering logistic or task-basked assistance, responding to new needs, etc.; Calhoun & Tedeschi, 2014; Canevello & Crocker, 2010; Canevello et al., 2016; Jiang et al., 2022; McMillen, 2004). Yet other work focuses on the “emotional contagion” of trauma—how one partner’s trauma may influence one’s own outcomes, and those of their partner, for better or worse (Gill-Emerson, 2015). While work on spillover effects documents how partners’ negative experiences may impact individuals negatively (Hancock et al., 2008; Mazzuca et al., 2019), there is also some work to suggest that, when partners grow, individuals grow along with them (experiencing “vicarious growth”); when individuals report PTG, their partners are also more likely to report PTG (even when they are not directly experiencing the negative event; Hodges et al., 2005; Weiss, 2002; Zwahlen et al., 2010). Yet other work positions positive change in a relationship more deliberately as an outcome (instead of positive change within an individual), finding that negative life events are sometimes associated with increased relationship satisfaction or sense of closeness (Williamson et al., 2021) 86 However, most work on PTG, whether it is examined as an individual-level outcome or as a couple-level outcome, has suffered from a host of methodological limitations (Jayawickreme & Blackie, 2014; Jayawickreme et al., 2021). One of the most glaring limitations of this work is its reliance on cross-sectional data. In typical PTG studies, participants are asked to recall a time that they endured a challenge and retroactively determine how much they have changed (for the better) in response to that challenge. Participant memories, of course, are not completely reliable (especially when considering that accurately remembering traumatic events is difficult for many people; Sachschal et al., 2019; Van der Kolk & Fisler, 1995). Likewise, people also find it difficult to make such attributions of how they have changed in response to a traumatic event. For example, people may be motivated to report positive changes as a way of reducing cognitive dissonance (that something good must have come out of a bad situation). Also, it is a difficult assessment partially because it is cognitively taxing (e.g., quantifying how much they changed and then assigning a certain amount of that change to the event versus how they would have changed in the absence of the event). What is needed is a simpler, albeit more difficult, approach in which people are asked about their psychological characteristics prospectively. Unfortunately, few studies examine PTG with longitudinal data. Even fewer studies examine PTG longitudinally within the context of close relationships, despite, theoretically, partners playing such an important role in the process (Purol & Chopik, 2024). The small amount of work that has done this suggests that, although psychological changes can be related within couples, exactly how and any why partners’ personality changes are connected are very nuanced and requires more exploration (Lahav et al., 2017). There are other limitations to measuring PTG in this traditional way. In simply asking participants if a life event has occurred or not, researchers often make assumptions about the 87 characteristics of a given life event (e.g., that it was negative, impactful, etc.). However, the participant’s perception of the event (e.g., its valence, predictability, normativity, etc.) is important information to integrate into the analysis when trying to predict how they may change in response to it (Haehner et al., 2022; Rakhshani et al., 2022). This method of examining PTG also relies on self-report, largely ignoring if and when others’ perceptions of an individual’s personality change may vary from their own. Overall, when considering the methodological limitations typical of this work and the heterogeneity of the effects seen in the literature, there are many unanswered questions surrounding how individuals, partners, and relationships change in the wake of negative life events. Research Questions and Primary Conclusions In evaluating the literatures on the impact of life events on the potential for psychological change, the potential for individual and dyadic PTG, and the methodological limitations of both literatures, I examined three longitudinal data sets in which couples’ personalities were assessed over time and modeled as a function of life events experienced by both individuals and their partners. Specifically, this dissertation tried to answer the following research questions in three studies: RQ1: Do people exhibit positive personality change (post-traumatic growth; PTG) after their partner experiences a negative life event? This question was examined in all three studies. Study 1 found the most evidence of a partner’s negative life event occurring alongside adaptive change in personality (i.e., the death of a partner’s parent and a partner’s unemployment were related to adaptive changes in conscientiousness, neuroticism, and openness). Study 2 found only one negative life event that 88 followed a similar pattern; the death of a partner’s parent was related to adaptive changes in conscientiousness. Study 3 found no life events that followed this pattern. No single life event of a partner predicted positive personality change in all three studies. While these studies offer some evidence that this type of growth is possible—and does occur in some couples—the pattern is not consistent. RQ2: If so, how do these changes compare to the PTG of the individual who experienced the negative life event themselves (Studies 1-3)? This question was examined in all three studies. By far, an actor’s life events were more influential in changes in personality than partner effects. Again, Study 1 found the most evidence of an individual’s negative life event occurring alongside adaptive change in personality (i.e., the death of a parent and unemployment were related to adaptive changes in all 5 traits). Study 2 replicated one of these effects (i.e., unemployment was related to adaptive changes in neuroticism). Study 3 identified a new life event associated with positive change: experiencing a miscarriage was related to adaptive changes in openness. While these studies offer some evidence that this type of growth is possible—and is more commonly linked to one’s own experiences, rather than a partner’s—this effect, too, was inconsistent. It was more common for negative life events to be associated with maladaptive changes and for positive life events (e.g., positive changes in health, birth of a child) to be associated with adaptive changes. RQ3: Do relationship characteristics (e.g., closeness, satisfaction) predict PTG for individuals and their partners (Studies 1-3)? This question was examined in all three studies. The relationship characteristics examined here (i.e., support, relationship satisfaction, responsiveness, and closeness) were rarely associated with personality changes at all and never interacted with life events and slopes to 89 indicate adaptive change. However, when examined as outcome variables, these studies did find some evidence of dyadic PTG (relationships improving in response to a negative life event). In Study 1, those who experienced the death of a child saw adaptive changes in spousal strain. In Study 2, those who experienced the direct or vicarious death of a parent saw adaptive changes in relationship satisfaction. Finally, in Study 3, those who experienced a negative health event saw adaptive changes in responsiveness, while those who experienced a partner’s negative health event saw adaptive changes in support. These studies suggest that relationship-level variables may not be influential in an individual’s personality trajectories. However, when considered as outcomes in their own right, these patterns offer some evidence for the existence of dyadic PTG. RQ4: Do partners perceive PTG in individuals who experience a negative life event (Study 3)? This question was solely examined in Study 3. Overall, participants rarely perceived personality change in their partners at all; each slope of partner-reported personality was not significant. However, there was one life event that was associated with partner-reported personality change: those who experienced a miscarriage reported their partners becoming more agreeable and conscientious over time (a change partners did not report noticing in themselves). While participants rarely self-reported change associated with life events, the two life events that were associated with change (i.e., miscarriage and the death of a partner’s close other) were not detected as influential by partners. This study suggests that, if PTG does occur on the trait level, partners may have a different perspective of this growth than individuals do themselves. Limitations and Future Directions The three studies presented here had several strengths. Together, the studies were able to overcome many of the limitations that often plague studies of PTG. These studies made use of three independent samples from different cultures, two of which were nationally representative. 90 They also sampled a wide variety of life events, with at least one study capturing the valence of those life events. The three studies featured dyadic data, incorporating romantic partners into the PTG process. One of these studies also included partner perceptions of PTG, something that (to my knowledge), has not yet been done. Lastly, they used longitudinal data and models to answer longitudinal questions—something rarely done in this work, particularly in the relationships literature. Of course, however, these studies also had limitations. One such limitation was how personality was operationalized—as Big 5 personality traits. Much work on the Big 5 personality traits, as well as the studies in this dissertation themselves, emphasize the stability of traits. Put simply, they did not vary much over adulthood, which made it difficult to capture post-life-event change (Bleidorn et al., 2018; Bühler et al., 2023). Perhaps, then, traits are the wrong place to look for PTG, if it does, in fact, exist. Other levels of personality, such as an individual’s goals or their life narratives (i.e., the stories people tell about their own lives; McAdams, 1996) may be a more fruitful place to look for these changes. In the study of life narratives, reports of PTG-like- phenomena are relatively common, especially in Western cultures (Goodson, 2012). Narrative patterns like redemption sequences (where participants report overcoming bad situations to arrive at good ones) and self-improvement sequences (where participants report changing themselves for the better) capture effects that align with PTG and predict outcomes we would expect to be associated with PTG, like well-being (Bauer et al., 2019; McAdams, 1989; McAdams & McLean, 2013). Importantly, life narratives often implicate close others. The storyteller’s perception of other people and how they seek to build relationships are key themes in narrative storytelling (McAdams, 2005). Extracting reality (i.e., capturing “real” personality 91 change or “real” individual/dyadic PTG) from the retelling of a life is a challenging, but important, direction for future research (McAdams & McLean, 2013). Another limitation of these studies is their lack of qualitative information about these life events—how people perceived the event itself. As acknowledged in the discussion of the Life Event Characteristics Scale (Luhmann et al., 2020), several characteristics can impact how an event influences a person (i.e., its valence, impact, predictability, challenge, emotional significance, etc.; Haehner et al., 2022; Rakhshani et al., 2022). I studied mostly discrete events (e.g. if someone experienced unemployment or not). However, additional detail would offer more context. For example, was the period of unemployment temporary? Was it financially ruinous? Was it buffered by having a partner with a lucrative or stable job? While some of this information (i.e., valence) was available in the CouPers sample, across studies I could not capture 1) the full context of each life event or 2) the full array of events that may be influential for people and their partners. One future direction could be assessing the qualitative impact of life events that happen to people in our social network. Another potentially fruitful future direction lies in the differences between participant and partner perceptions of PTG. While the CouPers sample was an exploratory test of this question, it offered some evidence that perceptions of growth may vary between members of a couple. Although it would be difficult to test who has a more “accurate” view of personality change, partners might have a distance from their partner’s life events that help them evaluate it differently. Final Conclusion When introducing the topic of PTG in the introduction of this dissertation, I discussed the relative unreliability of this phenomenon (Infurna & Jayawickreme, 2019; Jayawickreme & 92 Blackie, 2014). Different work on the topic finds many mixed results, making it difficult to be confident about if the phenomenon exists, and, if so, under what conditions. The goal of this dissertation was to offer a stronger test of PTG than has historically existed and in a highly relevant context—that of close relationships. The findings detailed in this dissertation suggest that the occurrence of PTG is relatively rare and, when it does occur, the size of these effects s to be very small. And, while many of the relationship-relevant factors implicated in the process of growth did not have the anticipated effect, there was some evidence that relationships can improve alongside a negative event. While the results presented here are exploratory, future work exploring these results may help people identify and facilitate strengths in their relationships amidst challenging life experiences. 93 REFERENCES Albuquerque, S., Pereira, M., & Narciso, I. (2016). Couple’s relationship after the death of a child: A systematic review. Journal of Child and Family Studies, 25, 30-53. Allemand, M., Hill, P. L., & Lehmann, R. (2015). Divorce and personality development across middle adulthood. Personal Relationships, 22(1), 122-137. Aron, A., & Aron, E. N. (1996). Self and self-expansion in relationships. Knowledge structures in close relationships: A social psychological approach, 325-344. Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusion of other in the self scale and the structure of interpersonal closeness. Journal of personality and social psychology, 63(4), 596. Aydogan, D., & Dincer, D. (2020). Creating resilient marriage relationships: Self-pruning and the mediation role sacrifice with satisfaction. Current Psychology, 39(2), 500-510. https://doi.org/10.1007/s12144-019-00472-x Backhans, M. C., & Hemmingsson, T. (2011). Unemployment and mental health—who is (not) affected? European Journal of Public Health, 22(3), 429-433. https://doi.org/10.1093/eurpub/ckr059 Baird, K., & Kracen, A. C. (2006). Vicarious traumatization and secondary traumatic stress: A research synthesis. Counselling Psychology Quarterly, 19(2), 181-188. Bakhurst, M., McGuire, A., & Halford, W. K. (2018). Trauma symptoms, communication, and relationship satisfaction in military couples. Family process, 57(1), 241-252. Baník, G., Dědová, M., & Vargová, L. (2022). The Co-occurrence of Posttraumatic growth (PTG) and Posttraumatic stress disorder (PTSD) in cancer survivors from a network perspective. Barre, K., De Boer, S., & Guarnaccia, C. (2023). Vicarious trauma and posttraumatic growth among victim support professionals. Current Psychology, 1-14. Bashour, M. (2006). An objective system for measuring facial attractiveness. Plastic and Reconstructive Surgery, 118(3), 757-774. Bauer, J. J., Graham, L. E., Lauber, E. A., & Lynch, B. P. (2019). What growth sounds like: Redemption, self‐improvement, and eudaimonic growth across different life narratives in relation to well‐being. Journal of personality, 87(3), 546-565. Benjamin, D. J., Berger, J. O., Johannesson, M., Nosek, B. A., Wagenmakers, E.-J., Berk, R., . . . Camerer, C. (2018). Redefine statistical significance. Nature human behaviour, 2(1), 6- 10. 94 Bernerth, J. B., & Aguinis, H. (2016). A critical review and best‐practice recommendations for control variable usage. Personnel psychology, 69(1), 229-283. Bialon, L. N., & Coke, S. (2012). A study on caregiver burden: stressors, challenges, and possible solutions. American Journal of Hospice and Palliative Medicine®, 29(3), 210- 218. Bianchi, R. (2018). Burnout is more strongly linked to neuroticism than to work-contextualized factors. Psychiatry Research, 270, 901-905. Bleidorn, W., Buyukcan-Tetik, A., Schwaba, T., van Scheppingen, M. A., Denissen, J. J. A., & Finkenauer, C. (2016). Stability and Change in Self-Esteem During the Transition to Parenthood. Social Psychological and Personality Science, 7(6), 560-569. https://doi.org/10.1177/1948550616646428 Bleidorn, W., Hopwood, C. J., & Lucas, R. E. (2018). Life events and personality trait change. Journal of personality, 86(1), 83-96. Bleidorn, W., Schwaba, T., Zheng, A., Hopwood, C. J., Sosa, S. S., Roberts, B. W., & Briley, D. (2022). Personality stability and change: A meta-analysis of longitudinal studies. Psychological bulletin, 148(7-8), 588. Brazeau, H., & Lewis, N. A. (2021). Within-couple health behavior trajectories: The role of spousal support and strain. Health Psychology, 40(2), 125. Bühler, J. L., Orth, U., Bleidorn, W., Weber, E., Kretzschmar, A., Scheling, L., & Hopwood, C. J. (2023). Life events and personality change: A systematic review and meta-analysis. European Journal of Personality, 08902070231190219. Calhoun, L. G., & Tedeschi, R. G. (2014). Handbook of posttraumatic growth: Research and practice. Routledge. Canevello, A., & Crocker, J. (2010). Creating good relationships: responsiveness, relationship quality, and interpersonal goals. Journal of personality and social psychology, 99(1), 78. Canevello, A., & Crocker, J. (2011). Changing relationship growth belief: Intrapersonal and interpersonal consequences of compassionate goals. Personal Relationships, 18(3), 370- 391. Canevello, A., Michels, V., & Hilaire, N. (2016). Supporting close others’ growth after trauma: The role of responsiveness in romantic partners’ mutual posttraumatic growth. Psychological trauma: theory, research, practice, and policy, 8(3), 334. Cann, A., Calhoun, L. G., Tedeschi, R. G., Taku, K., Vishnevsky, T., Triplett, K. N., & Danhauer, S. C. (2010). A short form of the Posttraumatic Growth Inventory. Anxiety, Stress, & Coping, 23(2), 127-137. 95 Chopik, W. J., Kelley, W. L., Vie, L. L., Oh, J., Bonett, D. G., Lucas, R. E., & Seligman, M. E. P. (2021). Development of character strengths across the deployment cycle among U.S. Army soldiers. Journal of Personality, 89(1), 23-34. https://doi.org/10.1111/jopy.12564 Christakis, N. A., & Fowler, J. H. (2007). The spread of obesity in a large social network over 32 years. New England journal of medicine, 357(4), 370-379. Cordova, M. J., Cunningham, L. L., Carlson, C. R., & Andrykowski, M. A. (2001). Posttraumatic growth following breast cancer: a controlled comparison study. Health psychology, 20(3), 176. Costa Jr, P. T., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality traits across cultures: robust and surprising findings. Journal of personality and social psychology, 81(2), 322. Costa, P. T., Herbst, J. H., McCrae, R. R., & Siegler, I. C. (2000). Personality at midlife: Stability, intrinsic maturation, and response to life events. Assessment, 7(4), 365-378. Dagan, M., Sanderman, R., Hoff, C., Meijerink, W. J. H. J., Baas, P. C., Van Haastert, M., & Hagedoorn, M. (2014). The interplay between partners’ responsiveness and patients’ need for emotional expression in couples coping with cancer. Journal of Behavioral Medicine, 37(5), 828-838. https://doi.org/10.1007/s10865-013-9543-4 de Jong Gierveld, J., & Van Tilburg, T. G. (1987). The partner as source of social support in problem and non-problem situations. Journal of Social Behavior and Personality, 2(2), 191-200. DeLongis, A., Capreol, M., Holtzman, S., O'Brien, T., & Campbell, J. (2004). Social support and social strain among husbands and wives: a multilevel analysis. Journal of family psychology, 18(3), 470. Eisenthal, Y., Dror, G., & Ruppin, E. (2006). Facial attractiveness: Beauty and the machine. Neural Computation, 18(1), 119-142. Ezzy, D. (1993). Unemployment and mental health: a critical review. Social science & medicine, 37(1), 41-52. Fayed, C., Bartle-Haring, S., & Vanbergen, A. (2021). Interpersonal Trauma, Therapeutic Alliance and Relationship Satisfaction: A Dyadic Examination. Contemporary Family Therapy. https://doi.org/10.1007/s10591-021-09606-5 Fletcher, G. J. (2015). Accuracy and bias of judgments in romantic relationships. Current Directions in Psychological Science, 24(4), 292-297. Fletcher, G. J., & Kerr, P. S. (2010). Through the eyes of love: reality and illusion in intimate relationships. Psychological bulletin, 136(4), 627. 96 Frankfurt, S., Frazier, P., Syed, M., & Jung, K. R. (2016). Using group-based trajectory and growth mixture modeling to identify classes of change trajectories. The Counseling Psychologist, 44(5), 622-660. Fredman, S. J., Vorstenbosch, V., Wagner, A. C., Macdonald, A., & Monson, C. M. (2014). Partner accommodation in posttraumatic stress disorder: Initial testing of the Significant Others’ Responses to Trauma Scale (SORTS). Journal of Anxiety Disorders, 28(4), 372- 381. https://doi.org/10.1016/j.janxdis.2014.04.001 Gander, F., Hofmann, J., Proyer, R. T., & Ruch, W. (2020). Character strengths–Stability, change, and relationships with well-being changes. Applied Research in Quality of Life, 15(2), 349-367. Gill-Emerson, G. (2015). The Trauma Contagion. Trauma, 15(1). Girme, Y. U., Maniaci, M. R., Reis, H. T., McNulty, J. K., Carmichael, C. L., Gable, S. L., . . . Overall, N. C. (2018). Does support need to be seen? Daily invisible support promotes next day relationship well-being. Journal of Family Psychology, 32(7), 882. Girme, Y. U., Overall, N. C., & Simpson, J. A. (2013). When visibility matters: Short-term versus long-term costs and benefits of visible and invisible support. Personality and social psychology bulletin, 39(11), 1441-1454. Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure. Psychological assessment, 4(1), 26. Goodson, I. F. (2012). Developing narrative theory: Life histories and personal representation. Routledge. Greene, T., Lahav, Y., Bronstein, I., & Solomon, Z. (2014). The role of ex-POWs’ PTSD symptoms and trajectories in wives’ secondary traumatization. Journal of Family Psychology, 28(5), 666. Haehner, P., Rakhshani, A., Fassbender, I., Lucas, R. E., Donnellan, M. B., & Luhmann, M. (2022). Perception of major life events and personality trait change. European Journal of Personality, 08902070221107973. Hall, B. J., Hobfoll, S. E., Canetti, D., Johnson, R. J., Palmieri, P. A., & Galea, S. (2010). Exploring the association between posttraumatic growth and PTSD: a national study of Jews and Arabs following the 2006 Israeli-Hezbollah war. The Journal of nervous and mental disease, 198(3), 180-186. Hancock, J. T., Gee, K., Ciaccio, K., & Lin, J. M.-H. (2008). I'm sad you're sad: emotional contagion in CMC. Proceedings of the 2008 ACM conference on Computer supported cooperative work, 97 Helgeson, V. S., Reynolds, K. A., & Tomich, P. L. (2006). A meta-analytic review of benefit finding and growth. Journal of consulting and clinical psychology, 74(5), 797. Herrman, H., Stewart, D. E., Diaz-Granados, N., Berger, E. L., Jackson, B., & Yuen, T. (2011). What is resilience? The Canadian Journal of Psychiatry, 56(5), 258-265. Hiefner, A. R. (2021). Dyadic coping and couple resilience after miscarriage. Family Relations, 70(1), 59-76. Hill, A. L., Rand, D. G., Nowak, M. A., & Christakis, N. A. (2010). Emotions as infectious diseases in a large social network: the SISa model. Proceedings of the Royal Society B: Biological Sciences, 277(1701), 3827-3835. Hodges, L., Humphris, G., & Macfarlane, G. (2005). A meta-analytic investigation of the relationship between the psychological distress of cancer patients and their carers. Social science & medicine, 60(1), 1-12. Huggard, P., Law, J., & Newcombe, D. (2017). A systematic review exploring the presence of Vicarious Trauma, Compassion Fatigue, and Secondary Traumatic Stress in Alcohol and Other Drug Clinicians. Australasian Journal of Disaster & Trauma Studies, 21(2). Infurna, F. J., & Jayawickreme, E. (2019). Fixing the growth illusion: New directions for research in resilience and posttraumatic growth. Current Directions in Psychological Science, 28(2), 152-158. Infurna, F. J., & Luthar, S. S. (2016). Resilience to major life stressors is not as common as thought. Perspectives on Psychological Science, 11(2), 175-194. Isokääntä, S., Koivula, K., Kokki, H., Palokangas, S., Tavast, K., Toivonen, I., & Kokki, M. (2023). Psychiatric symptoms and couple satisfaction in parents of newborns before and during the COVID‐19 pandemic—A comparison of two prospective studies. Nursing Open, 10(4), 2667-2677. https://doi.org/10.1002/nop2.1475 Jayawickreme, E., & Blackie, L. E. (2014). Post–traumatic growth as positive personality change: evidence, controversies and future directions. European Journal of Personality, 28(4), 312-331. Jayawickreme, E., Infurna, F. J., Alajak, K., Blackie, L. E., Chopik, W. J., Chung, J. M., . . . Frazier, P. (2021). Post‐traumatic growth as positive personality change: Challenges, opportunities, and recommendations. Journal of personality, 89(1), 145-165. Jiang, T., Canevello, A., & Crocker, J. (2022). How Relationships Foster Growth: Compassionate Goals Predict Growth-Seeking Through Perceived Available Support Independent of Relationship Security. Personality and Social Psychology Bulletin, 01461672221080949. 98 John, O. P., & Srivastava, S. (1999). The Big-Five trait taxonomy: History, measurement, and theoretical perspectives. Johnson, R. W. (2007). The burden of caring for frail parents. Urban Institute. Joseph, S., & Linley, P. A. (2005). Positive adjustment to threatening events: An organismic valuing theory of growth through adversity. Review of general psychology, 9(3), 262-280. Jung, T., & Wickrama, K. A. (2008). An introduction to latent class growth analysis and growth mixture modeling. Social and personality psychology compass, 2(1), 302-317. Kanfer, R., & Bufton, G. (2015). Job loss, job search, and reemployment in later adulthood. Encyclopedia of Geropsychology, 1-11. 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(1), 3. Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). The analysis of dyadic data. In: New York, NY: Guilford Press. Kiełek-Rataj, E., Wendołowska, A., Kalus, A., & Czyżowska, D. (2020). Openness and communication effects on relationship satisfaction in women experiencing infertility or miscarriage: a dyadic approach. International journal of environmental research and public health, 17(16), 5721. Kimura, M., Daibo, I., & Yogo, M. (2008). The study of emotional contagion from the perspective of interpersonal relationships. Social Behavior and Personality: an international journal, 36(1), 27-42. Kramer, A. D., Guillory, J. E., & Hancock, J. T. (2014). Experimental evidence of massive-scale emotional contagion through social networks. Proceedings of the National Academy of Sciences, 111(24), 8788-8790. Kunz, S., Joseph, S., Geyh, S., & Peter, C. (2019). Perceived posttraumatic growth and depreciation after spinal cord injury: Actual or illusory? Health Psychology, 38(1), 53. Lachman, M. E., & Weaver, S. L. (1997). The Midlife Development Inventory (MIDI) personality scales: Scale construction and scoring. Waltham, MA: Brandeis University, 7, 1-9. Lahav, Y., Kanat-Maymon, Y., & Solomon, Z. (2017). Posttraumatic growth and dyadic adjustment among war veterans and their wives. Frontiers in psychology, 8, 1102. Laurenceau, J.-P., Barrett, L. F., & Rovine, M. J. (2005). The interpersonal process model of intimacy in marriage: a daily-diary and multilevel modeling approach. Journal of family psychology, 19(2), 314. 99 Leggett, A. N., Choi, H., Chopik, W. J., Liu, H., & Gonzalez, R. (2020). Early cognitive decline and its impact on Spouse’s loneliness. Research in human development, 17(1), 78-93. Loeys, T., Cook, W., De Smet, O., Wietzker, A., & Buysse, A. (2014). The actor–partner interdependence model for categorical dyadic data: A user‐friendly guide to GEE. Personal Relationships, 21(2), 225-241. Lucas, R. E., Clark, A. E., Georgellis, Y., & Diener, E. (2004). Unemployment alters the set point for life satisfaction. Psychological science, 15(1), 8-13. Luhmann, M., Fassbender, I., Alcock, M., & Haehner, P. (2020). A dimensional taxonomy of perceived characteristics of major life events. Journal of Personality and Social Psychology. Luszczynska, A., Mohamed, N. E., & Schwarzer, R. (2005). Self-efficacy and social support predict benefit finding 12 months after cancer surgery: The mediating role of coping strategies. Psychology, Health & Medicine, 10(4), 365-375. Mangelsdorf, J., Eid, M., & Luhmann, M. (2019). Does growth require suffering? A systematic review and meta-analysis on genuine posttraumatic and postecstatic growth. Psychological bulletin, 145(3), 302. Mazzuca, S., Kafetsios, K., Livi, S., & Presaghi, F. (2019). Emotion regulation and satisfaction in long-term marital relationships: The role of emotional contagion. Journal of Social and Personal Relationships, 36(9), 2880-2895. McAdams, D. P. (1989). The development of a narrative identity. In Personality psychology: Recent trends and emerging directions (pp. 160-174). Springer. McAdams, D. P. (1996). Personality, modernity, and the storied self: A contemporary framework for studying persons. Psychological inquiry, 7(4), 295-321. McAdams, D. P. (2005). Studying lives in time: A narrative approach. Advances in Life Course Research, 10, 237-258. McAdams, D. P., & McLean, K. C. (2013). Narrative identity. Current directions in psychological science, 22(3), 233-238. McAdams, D. P., Reynolds, J., Lewis, M., Patten, A. H., & Bowman, P. J. (2001). When bad things turn good and good things turn bad: Sequences of redemption and contamination in life narrative and their relation to psychosocial adaptation in midlife adults and in students. Personality and social psychology bulletin, 27(4), 474-485. McCann, S. J. (2018). US state resident big five personality and work satisfaction: The importance of neuroticism. Cross-Cultural Research, 52(2), 155-191. 100 McCrae, R. R. (1993). Moderated analyses of longitudinal personality stability. Journal of personality and social psychology, 65(3), 577. McCrae, R. R., Costa, P. T., de Lima, M. P., Simões, A., Ostendorf, F., Angleitner, A., . . . Barbaranelli, C. (1999). Age differences in personality across the adult life span: parallels in five cultures. Developmental psychology, 35(2), 466. McCrae, R. R., Stone, S. V., Fagan, P. J., & Costa Jr, P. T. (1998). Identifying causes of disagreement between self‐reports and spouse ratings of personality. Journal of personality, 66(3), 285-313. McMillen, J. C. (2004). Posttraumatic growth: What's it all about? Psychological Inquiry, 15(1), 48-52. McNeillie, N., & Rose, J. (2021). Vicarious trauma in therapists: A meta-ethnographic review. Behavioural and Cognitive Psychotherapy, 49(4), 426-440. Mermelstein, R., Cohen, S., Lichtenstein, E., Baer, J. S., & Kamarck, T. (1986). Social support and smoking cessation and maintenance. Journal of consulting and clinical psychology, 54(4), 447. Morelli, S. A., Lee, I. A., Arnn, M. E., & Zaki, J. (2015). Emotional and instrumental support provision interact to predict well-being. Emotion, 15(4), 484. Neff, L. A., & Karney, B. R. (2005). To know you is to love you: The implications of global adoration and specific accuracy for marital relationships. Journal of personality and social psychology, 88(3), 480. Nenova, M., DuHamel, K., Zemon, V., Rini, C., & Redd, W. H. (2013). Posttraumatic growth, social support, and social constraint in hematopoietic stem cell transplant survivors. Psycho‐oncology, 22(1), 195-202. Nietzsche, F., & Levy, O. (1909). The twilight of the idols (Vol. 17). Allen & Unwin. Norling, L. R., & Chopik, W. J. (2020). The association between coworker support and work- family interference: A test of work environment and burnout as mediators. Frontiers in psychology, 11, 819. O'Neill, A. S., Mohr, C. D., Bodner, T. E., & Hammer, L. B. (2020). Perceived partner responsiveness, pain, and sleep: A dyadic study of military-connected couples. Health Psychology, 39(12), 1089. Okay, İ., & Güler, Ç. Y. (2021). Relationship dynamics: the protective effect of relationship satisfaction and dependency on parents’ mental health after having a preterm baby. Health Psychology Report, 9(1). 101 Oltmanns, J. R., Jackson, J. J., & Oltmanns, T. F. (2020). Personality change: Longitudinal self- other agreement and convergence with retrospective-reports. Journal of Personality and Social Psychology, 118(5), 1065. Pals, J. L., & McAdams, D. P. (2004). The transformed self: A narrative understanding of posttraumatic growth. Psychological Inquiry, 65-69. Paulhus, D. L., & Vazire, S. (2007). The self-report method. Handbook of research methods in personality psychology, 1(2007), 224-239. Peterson, C., & Seligman, M. E. (2003). Character strengths before and after September 11. Psychological Science, 14(4), 381-384. Pickett, C. L., Gardner, W. L., & Knowles, M. (2004). Getting a cue: The need to belong and enhanced sensitivity to social cues. Personality and social psychology bulletin, 30(9), 1095-1107. Prochazkova, E., & Kret, M. E. (2017). Connecting minds and sharing emotions through mimicry: A neurocognitive model of emotional contagion. Neuroscience & Biobehavioral Reviews, 80, 99-114. Purol, M. F., & Chopik, W. J. (2022). Partner accuracy in humor perception and associations with relationship satisfaction. Personality and Social Psychology Bulletin, 01461672221080950. Purol, M. F., & Chopik, W. J. (2024). Do many hands make light work? The role of romantic partners and close relationships in post-traumatic growth. In. American Psychologist. Racine, N., Plamondon, A., Hentges, R., Tough, S., & Madigan, S. (2019). Dynamic and bidirectional associations between maternal stress, anxiety, and social support: the critical role of partner and family support. Journal of affective disorders, 252, 19-24. Rakhshani, A., Lucas, R. E., Donnellan, M. B., Fassbender, I., & Luhmann, M. (2022). Personality traits and perceptions of major life events. European Journal of Personality, 36(4), 683-703. Ram, N., & Grimm, K. J. (2009). Methods and measures: Growth mixture modeling: A method for identifying differences in longitudinal change among unobserved groups. International journal of behavioral development, 33(6), 565-576. Randall, A. K., & Bodenmann, G. (2009). The role of stress on close relationships and marital satisfaction. Clinical Psychology Review, 29(2), 105-115. https://doi.org/10.1016/j.cpr.2008.10.004 Randall, A. K., & Bodenmann, G. (2017). Stress and its associations with relationship satisfaction. Current opinion in psychology, 13, 96-106. 102 Reis, H. T. (2012). Perceived partner responsiveness as an organizing theme for the study of relationships and well-being. Reis, H. T., & Clark, M. S. (2013). Responsiveness. Reis, H. T., & Shaver, P. R. (1988). Intimacy as an interpersonal process. In S. Duck, D. F. Hay, S. E. Hobfoll, W. Ickes, & B. M. Montgomery (Eds.), Handbook of personal relationships: Theory, research and interventions (pp. 367-389). Wiley. Rini, C., Schetter, C. D., Hobel, C. J., Glynn, L. M., & Sandman, C. A. (2006). Effective social support: Antecedents and consequences of partner support during pregnancy. Personal Relationships, 13(2), 207-229. Riva, R., Forinder, U., Arvidson, J., Mellgren, K., Toporski, J., Winiarski, J., & Norberg, A. L. (2014). Patterns of psychological responses in parents of children that underwent stem cell transplantation. Psycho‐Oncology, 23(11), 1307-1313. Roberts, B. W., Kuncel, N. R., Shiner, R., Caspi, A., & Goldberg, L. R. (2007). The power of personality: The comparative validity of personality traits, socioeconomic status, and cognitive ability for predicting important life outcomes. Perspectives on Psychological science, 2(4), 313-345. Rosenquist, J. N., Fowler, J. H., & Christakis, N. A. (2011). Social network determinants of depression. Molecular psychiatry, 16(3), 273-281. Sachschal, J., Woodward, E., Wichelmann, J. M., Haag, K., & Ehlers, A. (2019). Differential effects of poor recall and memory disjointedness on trauma symptoms. Clinical Psychological Science, 7(5), 1032-1041. Saenz, J. L. (2021). Spousal Support, Spousal Strain, and Loneliness in Older Mexican Couples. The Journals of Gerontology: Series B, 76(4), e176-e186. https://doi.org/10.1093/geronb/gbaa194 Schroevers, M. J., Helgeson, V. S., Sanderman, R., & Ranchor, A. V. (2010). Type of social support matters for prediction of posttraumatic growth among cancer survivors. Psycho‐ Oncology, 19(1), 46-53. Schulz, U., & Mohamed, N. E. (2004). Turning the tide: Benefit finding after cancer surgery. Social science & medicine, 59(3), 653-662. Schwab, R. (1998). A child's death and divorce: Dispelling the myth. Death studies, 22(5), 445- 468. Schwaba, T., Bleidorn, W., Hopwood, C. J., Manuck, S. B., & Wright, A. G. (2022). Refining the maturity principle of personality development by examining facets, close others, and comaturation. Journal of Personality and Social Psychology. 103 Sejourne, N., Callahan, S., & Chabrol, H. (2010). Support following miscarriage: what women want. Journal of Reproductive and Infant Psychology, 28(4), 403-411. Selcuk, E., Gunaydin, G., Ong, A. D., & Almeida, D. M. (2016). Does Partner Responsiveness Predict Hedonic and Eudaimonic Well-being? A 10-Year Longitudinal Study. Journal of Marriage and Family, 78(2), 311-325. https://doi.org/10.1111/jomf.12272 Shrout, P. E., Herman, C. M., & Bolger, N. (2006). The costs and benefits of practical and emotional support on adjustment: A daily diary study of couples experiencing acute stress. Personal Relationships, 13(1), 115-134. Sim, A., Cordier, R., Vaz, S., & Falkmer, T. (2019). “We are in this together”: Experiences of relationship satisfaction in couples raising a child with autism spectrum disorder. Research in Autism Spectrum Disorders, 58, 39-51. Smith, K. V., & Ehlers, A. (2020). Cognitive predictors of grief trajectories in the first months of loss: A latent growth mixture model. Journal of consulting and clinical psychology, 88(2), 93. Smith, L. E., Bernal, D. R., Schwartz, B. S., Whitt, C. L., Christman, S. T., Donnelly, S., . . . Kish, J. (2014). Coping with vicarious trauma in the aftermath of a natural disaster. Journal of Multicultural Counseling and Development, 42(1), 2-12. Smyth, J. M., Hockemeyer, J. R., Heron, K. E., Wonderlich, S. A., & Pennebaker, J. W. (2008). Prevalence, type, disclosure, and severity of adverse life events in college students. Journal of American College Health, 57(1), 69-76. Specht, J., Egloff, B., & Schmukle, S. C. (2011a). Stability and change of personality across the life course: the impact of age and major life events on mean-level and rank-order stability of the Big Five. Journal of personality and social psychology, 101(4), 862. Specht, J., Egloff, B., & Schmukle, S. C. (2011b). Stability and change of personality across the life course: The impact of age and major life events on mean-level and rank-order stability of the Big Five. Journal of Personality and Social Psychology, 101(4), 862-882. https://doi.org/10.1037/a0024950 Sumalla, E. C., Ochoa, C., & Blanco, I. (2009). Posttraumatic growth in cancer: reality or illusion? Clinical psychology review, 29(1), 24-33. Syed, M., & McLean, K. C. (2022). Who gets to live the good life? Master narratives, identity, and well-being within a marginalizing society. Journal of Research in Personality, 100, 104285. Talley, A., Molix, L., Schlegel, R. J., & Bettencourt, A. (2010). The influence of breast cancer survivors' perceived partner social support and need satisfaction on depressive symptoms: a longitudinal analysis. Psychology and Health, 25(4), 433-449. 104 Taylor, S. E. (2011). Social support: A review. Tedeschi, R. G., & Calhoun, L. G. (1996). The Posttraumatic Growth Inventory: Measuring the positive legacy of trauma. Journal of traumatic stress, 9(3), 455-471. Tennen, H., & Affleck, G. (1998). Personality and transformation in me face of adversity. Posttraumatic growth: Positive changes in the aftermath of crisis, 65-98. Thornton, A. A., & Perez, M. A. (2006). Posttraumatic growth in prostate cancer survivors and their partners. Psycho-Oncology, 15(4), 285-296. https://doi.org/10.1002/pon.953 Trail, T. E., & Karney, B. R. (2012). What's (Not) Wrong With Low-Income Marriages. Journal of Marriage and Family, 74(3), 413-427. https://doi.org/10.1111/j.1741- 3737.2012.00977.x Van der Kolk, B. A., & Fisler, R. (1995). Dissociation and the fragmentary nature of traumatic memories: Overview and exploratory study. Journal of traumatic stress, 8, 505-525. van Scheppingen, M. A., Denissen, J. J. A., Chung, J. M., Tambs, K., & Bleidorn, W. (2018). Self-esteem and relationship satisfaction during the transition to motherhood. Journal of Personality and Social Psychology, 114(6), 973-991. https://doi.org/10.1037/pspp0000156 Vanbergen, A., Blalock, J., Bryant, A., Bortz, P., & Bartle-Haring, S. (2020). Couples and Trauma History: A Descriptive Overview of Interpersonal Trauma and Clinical Outcomes. Contemporary Family Therapy, 42(4), 335-345. https://doi.org/10.1007/s10591-020-09548-4 Vazire, S. (2010). Who knows what about a person? The self–other knowledge asymmetry (SOKA) model. Journal of personality and social psychology, 98(2), 281. Verhaeghe, S., Defloor, T., & Grypdonck, M. (2005). Stress and coping among families of patients with traumatic brain injury: a review of the literature. Journal of Clinical Nursing, 14(8), 1004-1012. https://doi.org/10.1111/j.1365-2702.2005.01126.x Vukasović, T., & Bratko, D. (2015). Heritability of personality: A meta-analysis of behavior genetic studies. Psychological bulletin, 141(4), 769. Watson, D., & Humrichouse, J. (2006). Personality development in emerging adulthood: integrating evidence from self-ratings and spouse ratings. Journal of Personality and Social Psychology, 91(5), 959. Weiss, T. (2002). Posttraumatic growth in women with breast cancer and their husbands: An intersubjective validation study. Journal of psychosocial Oncology, 20(2), 65-80. 105 Weiss, T. (2004). Correlates of posttraumatic growth in husbands of breast cancer survivors. Psycho-Oncology, 13(4), 260-268. https://doi.org/10.1002/pon.735 Wick, S., & Nelson Goff, B. S. (2014). A qualitative analysis of military couples with high and low trauma symptoms and relationship distress levels. Journal of Couple & Relationship Therapy, 13(1), 63-88. Williamson, H. C. (2020). Early effects of the COVID-19 pandemic on relationship satisfaction and attributions. Psychological Science, 31(12), 1479-1487. Williamson, H. C., Bradbury, T. N., & Karney, B. R. (2021). Experiencing a natural disaster temporarily boosts relationship satisfaction in newlywed couples. Psychological science, 32(11), 1709-1719. Wofford, N., Defever, A. M., & Chopik, W. J. (2019). The vicarious effects of discrimination: How partner experiences of discrimination affect individual health. Social psychological and personality science, 10(1), 121-130. Yoon, M.-S., Jeon, Y.-B., & Lee, S.-B. (2022). Grief, Partner Support, Posttraumatic Growth among Women with Pregnancy Loss in Korea. Journal of Loss and Trauma, 1-11. Zhang, X., Deng, X., Mo, Y., Li, Y., Song, X., & Li, H. (2021). Relationship between infertility- related stress and resilience with posttraumatic growth in infertile couples: gender differences and dyadic interaction. Human Reproduction, 36(7), 1862-1870. Zwahlen, D., Hagenbuch, N., Carley, M. I., Jenewein, J., & Buchi, S. (2010). Posttraumatic growth in cancer patients and partners—effects of role, gender and the dyad on couples' posttraumatic growth experience. Psycho‐oncology, 19(1), 12-20. Ávila, M., Coimbra, J. L., Park, C. L., & Matos, P. M. (2017). Attachment and posttraumatic growth after breast cancer: A dyadic approach. Psycho‐oncology, 26(11), 1929-1935. 106 APPENDIX A: TABLES FOR CHAPTER 2 Table 1. Frequency of life events; HRS sample. Life event Moved Birth of child Death of child New chronic illness Negative change in health Positive change in health Death of parent New job Retirement Unemployment Frequency (% of sample) 5126 (37.6%) 1035 (7.6%) 1515 (11.1%) 7885 (57.8%) 13050 (95.7%) 1506 (11%) 2992 (21.9%) 3771 (27.6%) 5011 (36.7%) 643 (4.7%) 107 Table 2. Descriptives of and correlations between traits and spousal support/strain; averaged across waves; HRS sample. C E N O Spousal support 1 A Mean SD 1 3.521 0.404 .463** 3.407 0.393 .577** 3.205 0.490 1.952 0.493 -.152** .444** 2.936 0.477 A C E N O Spousal support Spousal strain Note. ** Correlation is significant at the 0.01 level (2-tailed). Traits abbreviated to their first initial. .427** -.267** .484** 1.947 0.669 -.086** 1 -.234** -.250** .556** 3.502 0.613 -.130** -.047** -.208** -.089** .151** .262** .167** .119** .142** 1 1 -.535** 1 108 **Correlation is significant at the 0.01 level, *Correlation is significant at the 0.05 level. A indicates actor life event, P indicates partner life event. 109 Table 3. Biserial correlations between life events within couples; HRS sample. 123456789101112131415161718191. A: Moving12. P: Moving1.000**13. A: Birth of child.052**.052**14. P: Birth of child.050**.050**.851**15. A: Death of child.054**.054**.239**.215**16. P:Death of child.049**.049**.211**.239**.808**17. A: New chronic illness.054**.054**.027**.019**.068**.045**18. P: New chronic illness.047**.047**.015**.033**.036**.068**.093**19. A: Negative health change.018**.018**.030**0.008.042**0.007.155**.076**110. P: Negative health change0.0050.0050.005.041**-0.001.055**.062**.192**.201**111. A: Positive health change.048**.048**.057**.049**.040**.032**.046**.013*.020**0.009112. P: Positive health change.040**.040**.044**.052**.029**.041**.016**.048**.011*.022**.038**113. A: Parent dies0.0000.000.015**.011*-.025**-.029**.035**.027**.074**.058**-.015**-0.002114. P: Parent dies-0.008-0.0080.006.014**-.042**-.020**.024**.047**.050**.093**-0.006-.015**.129**115. A: New job.043**.043**.041**.037**-.015**-.014**.017**.013*.059**.036**-.012*-0.008.128**.119**116. P: New job.038**.038**.034**.046**-.019**-0.0050.010.029**.029**.081**-0.009-0.010.123**.137**.190**117. A: Retirement.048**.048**.017**-0.001.040**0.011.117**.068**.110**.048**.040**0.010.087**.072**.130**.055**118. P: Retirement.038**.038**-0.005.021**0.001.041**.072**.137**.055**.135**0.005.046**.085**.106**.063**.134**.190**119. A: Unemployment.031**.031**.032**.028**-0.010-.013*.025**.021**.028**.022**.015**.020**.065**.043**.205**.058**.082**.030**120. P: Unemployment.024**.024**.030**.038**-.016**-0.009.020**.029**.017**.035**.023**.012*.046**.069**.061**.210**.026**.087**.062**** Correlation is significant at the 0.01 level, *Correlation is significant at the 0.05 level. A indicates actor event, P indicates partner event. Table 3 (cont’d) 110 Table 3. Biserial correlations between life events within couples; HRS sample. 123456789101112131415161718191. A: Moving12. P: Moving1.000**13. A: Birth of child.052**.052**14. P: Birth of child.050**.050**.851**15. A: Death of child.054**.054**.239**.215**16. P:Death of child.049**.049**.211**.239**.808**17. A: New chronic illness.054**.054**.027**.019**.068**.045**18. P: New chronic illness.047**.047**.015**.033**.036**.068**.093**19. A: Negative health change.018**.018**.030**0.008.042**0.007.155**.076**110. P: Negative health change0.0050.0050.005.041**-0.001.055**.062**.192**.201**111. A: Positive health change.048**.048**.057**.049**.040**.032**.046**.013*.020**0.009112. P: Positive health change.040**.040**.044**.052**.029**.041**.016**.048**.011*.022**.038**113. A: Parent dies0.0000.000.015**.011*-.025**-.029**.035**.027**.074**.058**-.015**-0.002114. P: Parent dies-0.008-0.0080.006.014**-.042**-.020**.024**.047**.050**.093**-0.006-.015**.129**115. A: New job.043**.043**.041**.037**-.015**-.014**.017**.013*.059**.036**-.012*-0.008.128**.119**116. P: New job.038**.038**.034**.046**-.019**-0.0050.010.029**.029**.081**-0.009-0.010.123**.137**.190**117. A: Retirement.048**.048**.017**-0.001.040**0.011.117**.068**.110**.048**.040**0.010.087**.072**.130**.055**118. P: Retirement.038**.038**-0.005.021**0.001.041**.072**.137**.055**.135**0.005.046**.085**.106**.063**.134**.190**119. A: Unemployment.031**.031**.032**.028**-0.010-.013*.025**.021**.028**.022**.015**.020**.065**.043**.205**.058**.082**.030**120. P: Unemployment.024**.024**.030**.038**-.016**-0.009.020**.029**.017**.035**.023**.012*.046**.069**.061**.210**.026**.087**.062**** Correlation is significant at the 0.01 level, *Correlation is significant at the 0.05 level. A indicates actor event, P indicates partner event. Table 4. Linear growth curve model examining the effect of actor and partner life events on the slope of agreeableness; HRS sample. df b 8744.162 3.409 15420.550 0.019 5894.176 -0.008 t 58.208 1.462 -0.887 LB 3.295 -0.007 -0.025 p 0.000 0.144 0.375 SE 0.059 0.013 0.009 UB 3.524 0.045 0.010 0.000 0.011 -0.015 -0.009 -0.005 -0.010 -0.022 Intercept Slope Actor: Moving Actor: Birth of a child Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent dies Partner: Parent dies -0.002 Actor: New job 0.052 Partner: New job Actor: Retirement Partner: Retirement Actor: Unemployment Partner: Unemployment Age Gender PersonOfColor Education Wealth -0.051 0.000 -0.138 -0.024 0.012 0.000 -0.003 -0.012 0.003 0.002 0.011 0.005 0.018 7517.880 -0.009 0.992 -0.035 0.035 0.014 7267.828 0.781 0.435 -0.016 0.038 0.008 10909.284 -2.699 0.007 -0.038 -0.006 0.008 10968.408 -0.555 0.579 -0.021 0.011 0.035 13860.510 -0.300 0.764 -0.079 0.058 0.022 11949.289 -0.680 0.497 -0.057 0.028 0.013 10882.875 -0.672 0.501 -0.034 0.016 0.013 10526.573 0.373 0.710 -0.020 0.029 0.010 10091.308 -1.271 0.204 -0.032 0.007 0.010 0.009 10800.031 10699.549 -0.207 5.575 0.836 <.001 -0.021 0.034 0.017 0.070 0.009 10727.968 -0.322 0.747 -0.021 0.015 0.008 10411.163 0.360 0.719 -0.013 0.019 0.008 10999.478 0.221 0.825 -0.015 0.018 0.019 10788.423 0.550 0.583 -0.027 0.049 0.019 0.001 0.004 0.010 0.001 0.003 10339.470 9003.304 6431.481 8025.988 10992.963 6696.546 111 -2.689 -0.348 -36.667 -2.323 8.226 0.167 0.007 0.728 <.001 0.020 <.001 0.867 -0.089 -0.001 -0.145 -0.044 0.009 -0.005 -0.014 0.001 -0.130 -0.004 0.015 0.006 0.000 0.002 0.005 0.004 0.002 0.006 0.002 0.013 0.003 -0.001 -0.011 -0.003 -0.002 -0.002 -0.004 -0.027 Table 4 (cont’d) MarLength Slope* Actor: Moving Slope* Actor: Birth of a child Slope* Actor: death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change Slope* Actor: Parent dies Slope* Partner: Parent dies Slope* Actor: New job Slope* Partner: New job Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Note. Significant effects bolded, p < .05 -0.001 -0.007 -0.003 0.003 0.004 0.002 0.003 0.002 0.003 0.005 0.010 0.002 0.002 0.002 0.000 0.002 0.002 0.000 0.004 0.005 7312.221 3701.840 4592.456 3786.218 7202.782 7263.403 16968.540 7838.114 7123.923 6698.529 6233.762 6634.993 6645.412 6817.338 6622.909 6783.691 6832.372 6945.370 112 -2.208 0.027 -0.021 -0.001 -0.227 0.820 -0.004 0.003 -0.150 0.880 -0.009 0.007 -0.961 0.337 -0.009 0.003 -0.814 0.416 -0.006 0.002 -1.159 0.246 -0.006 0.002 -2.154 0.031 -0.052 -0.002 -0.738 0.461 -0.016 0.007 -2.431 0.015 -0.013 -0.001 -0.181 0.857 -0.007 0.005 4.652 <.001 0.006 0.014 1.549 0.121 -0.001 0.008 1.724 0.085 -0.001 0.008 2.147 0.032 0.000 0.009 0.165 0.869 -0.003 0.004 0.842 0.400 -0.002 0.006 -0.557 0.578 -0.011 0.006 -0.081 0.935 -0.009 0.009 Table 5. Linear growth curve model examining the effect of actor/partner life events and spousal support on the slope of agreeableness; HRS sample. Intercept Slope Actor: Moving Actor: Birth of a child Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent dies Partner: Parent dies t df SE b 3.380 55.843 0.061 8538.755 -0.001 0.015 12181.373 -0.044 -0.973 -0.009 0.009 5651.757 -0.148 -0.003 0.018 6973.108 0.538 0.014 6714.878 0.008 LB p 0.000 3.261 -0.030 0.965 -0.026 0.331 -0.038 0.882 -0.020 0.590 UB 3.499 0.029 0.009 0.033 0.035 -0.019 0.008 10651.022 -2.271 0.023 -0.035 -0.003 -0.006 0.008 10697.590 -0.739 0.460 -0.022 0.010 0.020 0.037 12649.054 0.541 0.588 -0.052 0.092 0.008 0.028 12625.642 0.289 0.773 -0.047 0.064 -0.005 0.013 10835.733 -0.353 0.724 -0.030 0.021 0.013 10399.947 0.869 0.011 -1.694 -0.017 0.010 9773.641 -0.003 0.010 10260.720 -0.273 0.385 0.090 0.785 -0.014 -0.036 -0.022 <.001 0.030 0.049 0.009 10314.991 5.193 Actor: New job -0.024 0.528 -0.006 0.009 10231.515 -0.631 Partner: New job -0.012 0.565 0.008 10224.915 0.575 0.005 Actor: Retirement -0.015 0.870 0.009 10533.458 0.163 0.001 Partner: Retirement -0.026 0.532 0.012 0.020 10440.267 0.624 Actor: Unemployment -0.081 0.023 -0.044 0.019 10038.443 -2.267 Partner: Unemployment 0.017 0.000 -0.001 0.987 0.001 8877.171 Age -38.399 <.001 -0.155 -0.148 0.004 6454.702 Gender 0.167 -0.014 0.010 7822.858 -0.034 -1.383 PersonOfColor <.001 0.008 0.001 10735.339 7.521 0.011 Education -0.007 0.581 -0.552 -0.002 0.003 6575.623 Wealth 0.025 -0.011 0.005 7136.633 -0.021 -2.240 MarLength <.001 0.140 0.071 14812.524 3.938 SpouseSupport 0.278 -0.008 0.259 0.009 23471.026 1.130 Actor: Moving*Support 0.011 Actor: Birth of a child*Support Actor: death of a child*Support Actor: New chronic illness*Support Partner: New chronic illness*Support -0.041 0.013 22456.899 -3.037 -0.002 0.010 23651.959 -0.170 0.018 23951.097 1.006 0.010 23998.072 0.176 -0.017 -0.067 -0.021 -0.017 0.018 0.002 0.315 0.002 0.865 0.861 113 0.036 0.003 0.017 0.067 0.012 0.021 0.018 0.051 -0.006 0.001 -0.140 0.006 0.014 0.004 -0.001 0.416 0.029 0.054 -0.015 0.017 0.021 0.002 0.008 0.028 0.011 22961.772 0.237 0.014 23382.402 2.045 -0.014 0.013 22890.776 -1.071 -0.012 0.011 22225.863 -1.064 -0.195 0.067 13717.548 -2.893 -0.006 0.028 15524.358 -0.221 -0.013 0.009 23275.875 -1.431 Table 5 (cont’d) Actor: Negative health change*Support Table 5 cont’d Partner: Negative health change*Support Actor: Positive health change*Support Partner: Positive health change*Support Actor: Parent dies*Support Partner: Parent dies*Support 0.011 24723.683 0.733 Actor: New job*Support -0.016 0.010 23958.695 -1.571 Partner: New job*Support Actor: Retirement*Support Partner: Retirement*Support Actor: Unemployment*Support -0.013 0.020 24787.131 -0.622 Partner: Unemployment*Support -0.002 0.020 23495.106 -0.096 0.024 15290.390 1.665 Slope*Support Slope* Actor: Moving -0.445 Slope* Actor: Birth of a child Slope* Actor: death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change 0.040 -0.001 0.002 3156.669 -0.016 0.013 15686.786 -1.218 -0.007 0.003 6642.746 -0.004 0.003 3102.449 -0.002 0.002 6645.628 -0.002 0.002 6551.781 -0.001 0.004 3948.902 -0.001 0.003 5974.579 0.010 24705.851 2.877 0.010 14481.256 0.720 -0.157 -2.172 -1.115 -0.743 -1.141 -0.180 0.007 0.028 0.004 -0.327 -0.063 0.825 -0.062 0.049 0.041 0.001 0.055 0.284 -0.040 0.012 0.288 -0.033 0.010 0.463 0.116 -0.014 -0.037 0.030 0.004 0.812 -0.018 0.023 0.152 -0.032 0.005 0.004 0.009 0.047 0.534 -0.052 0.027 0.924 0.096 0.656 -0.042 -0.007 -0.005 0.038 0.087 0.003 0.875 -0.009 0.008 0.265 -0.010 0.003 0.458 -0.006 0.003 0.254 -0.007 0.002 0.223 -0.042 0.010 0.471 -0.012 0.026 0.030 -0.014 -0.001 0.857 -0.007 0.006 114 0.000 0.009 0.001 Table 5 (cont’d) Slope* Actor: Parent dies Slope* Partner: Parent 0.003 dies Slope* Actor: New job 0.002 Slope* Partner: New job 0.004 Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Slope* Actor: Negative health change* Support Slope* Actor: Positive health change* Support Slope* Actor: Parent dies* Support Slope* Partner: New job* Support Note. Significant effects bolded, p < .01 0.000 0.003 0.003 0.002 5526.592 0.002 5677.266 0.002 5872.448 0.002 5837.316 0.002 5940.506 0.002 6002.202 3.783 <.001 0.004 0.013 1.457 0.928 1.685 0.145 0.353 0.092 -0.001 -0.002 -0.001 0.008 0.007 0.008 0.176 0.860 -0.004 0.004 0.360 0.719 -0.003 0.005 -0.602 0.547 -0.012 0.006 -0.350 0.726 -0.011 0.008 -0.003 0.005 5997.018 -0.002 0.005 5684.418 -0.044 0.024 15381.487 -1.852 0.005 12198.830 0.631 0.064 -0.091 0.003 0.528 -0.007 0.013 0.004 7908.704 0.004 8774.288 0.889 0.374 -0.004 0.011 0.107 0.914 -0.007 0.007 115 Table 6. Linear growth curve model examining the effect of actor and partner life events on the slope of conscientiousness; HRS sample. b 3.078 0.035 0.007 0.001 0.007 -0.044 -0.005 -0.037 -0.036 -0.007 Intercept Slope Actor: Moving Actor: Birth of a child Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive 0.000 health change Actor: Parent dies 0.002 Partner: Parent dies 0.008 0.057 Actor: New job 0.005 Partner: New job Actor: Retirement 0.000 Partner: Retirement 0.001 Actor: Unemployment Partner: Unemployment Age Gender PersonOfColor Education Wealth MarLength Slope* Actor: Moving Slope* Actor: Birth of a child -0.005 -0.005 -0.056 0.004 0.021 0.031 -0.008 -0.017 -0.001 0.001 SE 0.057 0.013 0.009 df 8558.268 15063.576 5758.811 t 53.688 2.645 0.758 p 0.000 0.008 0.449 LB 2.965 0.009 -0.010 UB 3.190 0.061 0.023 0.017 7285.560 -0.315 0.753 -0.040 0.029 0.013 7038.413 0.066 0.947 -0.026 0.027 0.008 11102.553 -4.504 <.001 -0.053 -0.021 0.008 11100.426 -4.351 <.001 -0.052 -0.020 0.035 13847.964 -1.272 0.203 -0.112 0.024 0.022 12284.580 0.335 0.738 -0.036 0.051 0.013 10916.722 -0.527 0.598 -0.031 0.018 0.013 0.010 0.010 0.009 0.009 0.008 0.008 10935.151 10450.561 10681.145 10739.390 10844.538 10547.688 10812.567 0.036 0.208 0.862 6.129 0.558 0.002 0.105 0.972 0.836 0.389 <.001 0.577 0.998 0.916 -0.024 -0.017 -0.011 0.039 -0.013 -0.016 -0.016 0.025 0.021 0.028 0.076 0.023 0.016 0.017 0.019 10604.804 -0.861 0.390 -0.054 0.021 0.019 0.001 0.004 0.010 0.001 0.003 0.005 10699.433 8883.248 6377.507 7828.781 10567.735 6635.127 7181.476 -0.242 -8.887 -14.595 0.419 14.676 11.467 -1.701 0.809 <.001 <.001 0.675 <.001 <.001 0.089 -0.042 -0.006 -0.063 -0.016 0.018 0.026 -0.018 0.033 -0.004 -0.048 0.024 0.024 0.037 0.001 0.002 3733.220 -0.522 0.601 -0.005 0.003 0.004 4620.678 0.234 0.815 -0.007 0.009 116 0.003 0.006 0.013 0.002 0.002 0.000 -0.008 -0.005 -0.003 -0.010 -0.045 Table 6 (cont’d) Slope* Actor: death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change Slope* Actor: Parent dies Slope* Partner: Parent dies Slope* Actor: New job Slope* Partner: New job Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Note. Significant effects bolded, p < .05 0.004 0.012 0.005 0.000 0.007 0.004 0.002 0.001 0.003 0.002 0.002 0.002 0.005 0.002 0.002 0.005 0.002 0.003 0.003 3835.154 -3.100 0.002 -0.016 -0.004 7160.780 -3.690 <.001 -0.012 -0.004 7187.390 -2.395 0.017 -0.009 -0.001 16925.674 -3.585 <.001 -0.070 -0.020 7994.038 -0.438 0.661 -0.015 0.010 6984.929 0.072 0.943 -0.006 0.006 6678.576 1.068 0.286 -0.003 0.009 6150.006 5.478 <.001 0.008 0.017 6472.871 2.356 0.019 0.001 0.010 6535.511 1.620 0.105 -0.001 0.008 6710.552 3.296 <.001 0.003 0.012 6538.803 0.083 0.934 -0.004 0.004 6668.569 1.814 0.070 0.000 0.008 6595.397 0.329 0.742 -0.008 0.011 6844.617 0.302 0.762 -0.008 0.011 117 Table 7. Linear growth curve model examining the effect of actor/partner life events and spousal support on the slope of conscientiousness; HRS sample. 0.000 -0.005 -0.027 -0.035 -0.032 b 3.092 0.035 0.007 -0.012 -0.001 Intercept Slope Actor: Moving Actor: Birth of a child Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health 0.004 change 0.001 Actor: Parent dies 0.004 Partner: Parent dies 0.057 Actor: New job 0.002 Partner: New job 0.000 Actor: Retirement -0.002 Partner: Retirement -0.014 Actor: Unemployment Partner: Unemployment 0.002 -0.004 Age -0.064 Gender 0.010 PersonOfColor 0.020 Education 0.029 Wealth -0.008 MarLength 0.186 SpouseSupport Actor: Moving*Support -0.004 Actor: Birth of a child*Support Actor: death of a child*Support Actor: New chronic illness*Support -0.016 -0.003 0.008 SE 0.059 0.015 0.009 0.017 0.014 df t 52.474 8216.188 11997.803 2.339 0.789 5456.135 -0.677 6683.794 -0.056 6381.006 LB p 0.000 2.977 0.019 0.006 0.430 0.498 0.955 UB 3.208 0.064 -0.010 0.024 -0.046 0.022 -0.028 0.026 0.008 10672.096 -3.828 <.001 -0.048 -0.015 0.008 10661.216 -4.299 <.001 -0.052 -0.019 0.036 12561.208 -0.734 0.463 -0.098 0.045 0.028 12420.757 -0.017 0.987 -0.056 0.055 0.013 10647.728 -0.423 0.672 -0.031 0.020 0.013 0.010 0.010 0.009 0.009 0.008 0.008 0.019 0.019 0.001 0.004 0.010 0.001 0.003 0.005 0.070 0.024 0.768 10583.931 0.295 0.932 0.086 9979.935 0.652 10072.513 0.452 <.001 0.039 10209.350 6.070 0.852 10220.067 0.187 0.976 10162.331 0.030 0.854 -0.184 10242.106 0.454 -0.748 10100.429 0.925 10210.020 0.094 -8.634 8746.713 <.001 -16.467 <.001 6386.777 7532.818 0.324 0.986 10220.579 14.105 <.001 0.017 10.784 <.001 0.024 6461.349 0.110 -1.597 6930.480 0.008 0.048 13940.127 2.648 0.859 -0.177 14312.696 -0.021 0.029 -0.018 0.020 -0.015 0.024 0.075 -0.017 0.020 -0.016 0.017 -0.018 0.015 -0.052 0.023 -0.036 0.040 -0.003 -0.006 -0.057 -0.072 -0.010 0.030 0.023 0.035 -0.018 0.002 0.323 -0.051 0.042 0.009 24746.785 -0.367 0.714 -0.022 0.015 0.018 24818.449 0.448 0.654 -0.027 0.043 0.014 22524.931 -1.197 0.231 -0.043 0.010 118 0.014 0.012 -0.107 -0.016 -0.007 -0.003 -0.004 -0.008 Table 7 (cont’d) Partner: New chronic illness*Support Actor: Negative health change*Support Partner: Negative health change*Support Actor: Positive health change*Support Partner: Positive health change*Support Actor: Parent dies*Support Partner: Parent -0.022 dies*Support Actor: New job*Support 0.024 Partner: New job*Support Actor: Retirement*Support Partner: Retirement*Support Actor: Unemployment*Support 0.021 Partner: Unemployment*Support 0.006 -0.027 Slope*Support Slope* Actor: Moving -0.001 Slope* Actor: Birth of a child Slope* Actor: Death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change -0.008 -0.009 -0.001 -0.008 -0.037 -0.005 -0.008 0.001 0.002 0.010 24245.448 -0.334 0.739 -0.023 0.016 0.010 24495.039 -0.845 0.398 -0.028 0.011 0.067 13025.742 -1.590 0.112 -0.238 0.025 0.029 17333.848 0.493 0.622 -0.043 0.072 0.013 25831.262 0.925 0.355 -0.014 0.039 0.013 24477.735 -0.287 0.774 -0.030 0.022 0.011 0.011 -2.045 23689.897 25435.687 2.149 -0.044 0.041 0.032 0.002 -0.001 0.046 0.010 25432.512 -1.498 0.134 -0.036 0.005 0.011 24450.184 -0.651 0.515 -0.028 0.014 0.009 24911.118 -0.847 0.397 -0.027 0.011 0.010 26077.667 2.148 0.032 0.002 0.040 0.020 0.021 0.002 25720.643 0.278 -1.329 24884.199 -0.348 3296.430 0.781 0.184 0.728 -0.034 0.045 -0.067 0.013 -0.005 0.003 0.004 4097.061 -0.127 0.899 -0.009 0.008 0.003 3214.580 -2.689 0.007 -0.015 -0.002 0.002 6577.885 -3.546 <.001 -0.012 -0.003 0.002 6654.675 -2.247 0.025 -0.009 -0.001 0.013 15727.256 -2.784 0.005 -0.063 -0.011 0.010 14746.645 -0.774 0.439 -0.027 0.012 0.003 6616.857 0.228 0.820 -0.006 0.007 0.003 6026.497 0.649 0.516 -0.004 0.008 119 0.002 0.001 0.003 0.011 -0.001 Table 7 (cont’d) Slope* Actor: Parent dies Slope* Partner: Parent 0.005 dies Slope* Actor: New job 0.003 Slope* Partner: New job 0.006 Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Slope* Actor: death of a child*Support Slope* Actor: New chronic illness*Support Slope* Partner: New chronic illness*Support Slope* Actor: Negative health change*Support Slope* Actor: Parent dies*Support Slope* Partner: Parent dies*Support Slope* Partner: New job*Support Note. Significant effects bolded, p < .01 -0.005 -0.003 -0.006 0.003 0.002 0.006 0.006 0.000 0.002 0.002 0.002 0.002 0.002 0.005 0.005 0.005 0.004 0.004 0.024 0.004 0.004 0.004 5499.889 4.917 <.001 0.007 0.016 5655.753 5845.952 5813.242 1.945 1.281 2.637 0.052 0.000 0.200 0.008 0.002 0.009 -0.002 0.007 0.011 5924.586 -0.423 0.672 -0.005 0.003 5977.417 1.506 0.132 -0.001 0.007 5904.026 0.123 0.902 -0.009 0.010 5671.065 -0.017 0.987 -0.009 0.009 9279.889 1.243 0.214 -0.004 0.016 10224.139 -0.937 0.349 -0.010 0.004 10378.289 -1.699 0.089 -0.013 0.001 14545.328 0.269 0.788 -0.041 0.053 7986.491 0.796 0.426 -0.004 0.010 9101.497 0.503 0.615 -0.006 0.010 8980.067 -1.234 0.217 -0.012 0.003 120 Table 8. Linear growth curve model examining the effect of actor and partner life events on the slope of extraversion; HRS sample. 0.020 -0.014 -0.049 -0.032 -0.001 b 2.936 Intercept 0.025 Slope 0.005 Actor: Moving -0.003 Actor: Birth of a child Actor: death of a child 0.016 Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent dies Partner: Parent dies Actor: New job Partner: New job Actor: Retirement Partner: Retirement Actor: Unemployment Partner: -0.049 Unemployment -0.002 Age -0.054 Gender 0.056 PersonOfColor 0.011 Education 0.024 Wealth MarLength -0.018 Slope* Actor: Moving 0.000 Slope* Actor: Birth of a child Slope* Actor: death of a child Slope* Actor: New chronic illness 0.025 -0.018 -0.003 0.105 -0.003 0.024 -0.027 -0.013 -0.009 -0.005 0.002 SE df 0.069 8568.141 0.015 16085.119 0.010 5997.236 0.021 7552.907 0.016 7388.276 t 42.499 1.638 0.441 -0.135 0.989 LB p 0.000 2.801 0.101 0.659 0.892 0.323 UB 3.072 -0.005 0.055 -0.016 0.025 -0.045 0.039 -0.016 0.049 0.010 11615.253 -3.181 0.001 -0.051 -0.012 0.010 11604.649 -1.414 0.157 -0.034 0.005 0.041 14520.893 -1.201 0.230 -0.130 0.031 0.027 12871.653 -0.054 0.957 -0.054 0.051 0.015 11394.193 1.267 0.205 -0.011 0.050 0.016 11531.044 0.012 11034.542 0.012 11150.289 0.011 11234.882 0.011 11346.370 0.010 11101.738 0.010 11289.364 0.024 11046.597 0.024 11246.088 0.001 8790.717 0.005 6509.564 0.012 7916.069 0.002 10575.177 0.003 6698.399 0.006 7277.057 0.002 3793.845 1.594 -1.450 -0.252 9.165 -0.255 2.333 -2.571 -0.559 -0.006 0.055 0.111 -0.041 0.006 0.147 -0.027 0.021 0.801 0.128 <.001 0.083 -0.025 0.020 0.799 0.044 0.020 0.004 -0.047 0.010 -0.006 -0.059 0.033 0.576 0.039 -2.067 -2.923 0.003 -11.525 <.001 4.561 6.220 7.287 -2.937 0.207 -0.003 -0.095 -0.001 -0.003 -0.045 -0.064 0.081 <.001 0.032 0.014 <.001 0.007 0.031 <.001 0.018 -0.030 0.003 -0.006 -0.004 0.005 0.836 0.005 4667.986 0.355 0.723 -0.007 0.011 0.003 3889.053 -1.537 0.124 -0.012 0.001 0.002 7172.378 -3.934 <.001 -0.013 -0.004 121 0.015 17664.693 0.002 7183.956 -0.003 -0.039 -0.003 -0.003 Table 8 (cont’d) Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change 0.002 Slope* Actor: Parent dies Slope* Partner: Parent dies Slope* Actor: New job Slope* Partner: New job Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Note. Significant effects bolded, p <.05 0.001 0.018 0.002 0.003 0.003 0.006 0.004 0.000 0.007 7921.845 0.003 7003.217 0.003 6730.388 0.002 6204.940 0.002 6463.997 0.002 6545.264 0.002 6708.270 0.002 6577.002 0.002 6668.033 0.005 6608.621 0.005 6885.283 -1.145 0.252 -0.007 0.002 -2.646 0.008 -0.068 -0.010 -0.397 0.691 -0.016 0.011 -0.868 0.385 -0.010 0.004 0.532 0.595 -0.005 0.008 7.427 <.001 0.013 0.023 0.102 0.919 -0.005 0.005 1.166 0.244 -0.002 0.007 2.638 0.008 0.002 0.011 1.197 0.231 -0.002 0.007 1.668 0.095 -0.001 0.008 0.494 0.621 -0.007 0.012 0.125 0.901 -0.009 0.011 122 Table 9. Linear growth curve model examining the effect of actor/partner life events and spousal support on the slope of extraversion; HRS sample. t 41.975 SE b 0.071 2.961 0.017 0.032 0.001 0.010 -0.012 0.021 0.017 0.014 df 8439.862 12917.881 1.832 0.142 5834.214 -0.548 7093.695 0.826 6875.933 p 0.000 0.067 0.887 0.584 0.409 LB 2.822 -0.002 -0.019 -0.053 -0.019 UB 3.099 0.066 0.022 0.030 0.046 Intercept Slope Actor: Moving Actor: Birth of a child Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent dies Partner: Parent dies Actor: New job Partner: New job Actor: Retirement Partner: Retirement Actor: Unemployment Partner: Unemployment Age Gender PersonOfColor Education Wealth MarLength Support Slope*Support Actor: Moving*Support Actor: Birth of a child*Support Actor: death of a child*Support -0.025 0.010 11279.204 -2.476 0.013 -0.045 -0.005 -0.015 0.010 11258.404 -1.537 0.124 -0.035 0.004 -0.023 0.043 13285.698 -0.538 0.591 -0.108 0.061 -0.027 0.034 13290.421 -0.812 0.417 -0.093 0.039 0.021 0.016 11230.385 1.319 0.187 -0.010 0.051 0.033 0.016 -0.022 0.012 -0.006 0.012 0.104 0.011 -0.003 0.011 0.028 0.010 -0.031 0.010 -0.006 0.024 -0.038 0.024 -0.001 0.001 -0.067 0.005 0.012 0.066 0.002 0.010 0.021 0.003 -0.018 0.006 0.082 0.215 0.019 0.028 -0.004 0.011 0.034 11247.766 2.116 0.064 -1.855 10622.578 0.604 10663.439 -0.519 <.001 10789.045 9.046 0.818 -0.230 10813.454 0.006 10780.510 2.725 0.002 -3.025 10827.359 0.805 -0.247 10645.348 0.108 -1.609 10829.272 -2.364 8683.799 0.018 -13.924 <.001 6568.970 <.001 5.439 7707.260 <.001 10282.811 5.716 <.001 6579.696 6.477 0.002 -3.069 7104.166 0.009 14647.661 2.609 0.496 15000.594 0.681 0.700 -0.386 24852.319 0.002 -0.046 -0.030 0.081 -0.025 0.008 -0.051 -0.052 -0.084 -0.003 -0.076 0.042 0.007 0.015 -0.030 0.053 -0.036 -0.025 0.064 0.001 0.017 0.126 0.020 0.048 -0.011 0.040 0.008 0.000 -0.057 0.090 0.013 0.028 -0.007 0.376 0.074 0.017 0.032 0.020 24497.214 1.593 0.111 -0.007 0.072 -0.035 0.015 23308.195 -2.310 0.021 -0.065 -0.005 123 Table 9 (cont’d) Actor: New chronic illness*Support Partner: New chronic illness*Support Actor: Negative health change*Support Partner: Negative health change*Support Actor: Positive health change*Support Partner: Positive health change*Support Actor: Parent dies*Support Partner: Parent dies*Support Actor: New job*Support Partner: New job*Support Actor: Retirement*Support Partner: Retirement*Support Actor: Unemployment*Support Partner: Unemployment*Support Slope* Actor: Moving Slope* Actor: Birth of a child Slope* Actor: Death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change -0.001 0.011 24620.353 -0.121 0.904 -0.023 0.021 -0.001 0.011 25630.203 -0.118 0.906 -0.023 0.020 -0.139 0.079 13768.962 -1.767 0.077 -0.294 0.015 0.037 0.034 18595.859 1.084 0.278 -0.030 0.104 0.007 0.015 25473.152 0.461 0.645 -0.023 0.037 0.008 0.015 24323.724 0.552 0.581 -0.021 0.038 -0.024 0.012 23831.043 -1.975 0.048 -0.049 0.000 0.030 0.013 -0.002 0.012 25534.995 2.385 -0.168 25148.946 0.017 0.867 0.005 -0.025 0.055 0.021 -0.010 0.012 24414.179 -0.821 0.412 -0.033 0.014 -0.006 0.011 24938.257 -0.552 0.581 -0.027 0.015 0.023 0.011 25833.132 2.069 0.039 0.001 0.044 -0.007 0.023 25051.847 -0.302 0.763 -0.051 0.038 -0.003 0.023 0.002 0.000 24435.019 3303.902 -0.126 -0.026 0.900 0.979 -0.048 -0.005 0.043 0.005 -0.002 0.005 4107.260 -0.324 0.746 -0.011 0.008 -0.005 0.004 3256.025 -1.365 0.172 -0.012 0.002 -0.009 0.002 6662.845 -3.591 <.001 -0.013 -0.004 -0.003 0.002 6716.779 -1.258 0.208 -0.008 0.002 -0.033 0.016 16592.838 -2.093 0.036 -0.063 -0.002 -0.012 0.011 15241.364 -1.107 0.268 -0.035 0.010 -0.003 0.004 6671.999 -0.738 0.461 -0.010 0.004 0.003 0.004 6139.948 0.931 0.352 -0.004 0.010 124 0.003 0.003 0.003 0.003 0.002 0.002 0.005 0.016 0.002 Table 9 (cont’d) Slope* Actor: Parent dies Slope* Partner: Parent 0.000 dies Slope* Actor: New job 0.002 Slope* Partner: New job 0.005 Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Slope* Actor: New chronic illness* Support 0.003 Slope* Actor: Negative health change* Support Slope* Actor: Parent dies* Support Slope* Partner: New job* Support Note. Significant effects bolded, p <.01 0.002 0.004 0.003 0.003 0.004 0.004 -0.002 0.005 -0.027 0.028 0.004 5612.810 6.302 <.001 0.011 0.021 5724.899 5933.491 5899.454 -0.062 0.733 1.991 0.950 0.464 0.047 -0.005 -0.003 0.000 0.005 0.007 0.010 6022.226 1.052 0.293 -0.002 0.007 6056.820 1.367 0.172 -0.001 0.008 5979.546 0.555 0.579 -0.007 0.013 5794.325 -0.307 0.759 -0.012 0.009 10617.796 0.694 0.488 -0.005 0.011 15212.030 -0.967 0.334 -0.082 0.028 8326.872 1.001 0.317 -0.004 0.012 9246.719 0.417 0.676 -0.006 0.010 125 Table 10. Linear growth curve model examining the effect of actor and partner life events on the slope of neuroticism; HRS sample. Intercept Slope Actor: Moving Actor: Birth of a child Actor: Death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent dies Partner: Parent dies Actor: New job Partner: New job Actor: Retirement Partner: Retirement Actor: Unemployment Partner: Unemployment Age b 2.866 -0.077 0.002 -0.033 -0.038 p LB SE df 0.073 8604.615 0.017 15885.885 -4.629 <.001 0.170 0.011 5874.358 0.865 -1.501 0.133 0.022 7591.305 -2.209 0.027 0.017 7353.891 t 39.298 0.000 2.723 -0.109 -0.020 -0.077 -0.072 UB 3.009 -0.044 0.023 0.010 -0.004 0.072 0.010 11481.764 7.005 <.001 0.052 0.092 0.026 0.010 11457.719 2.522 0.012 0.006 0.046 0.017 0.043 14523.523 0.388 0.698 -0.068 0.101 0.040 0.028 13351.368 1.440 0.150 -0.014 0.094 0.009 0.016 11101.199 0.553 0.580 -0.022 0.040 -0.020 -0.005 0.005 -0.058 0.000 -0.009 -0.023 0.078 0.050 -0.006 0.016 11359.566 -1.241 0.215 0.012 10853.501 -0.441 0.659 0.012 10850.220 0.420 0.675 0.012 11064.623 -4.935 <.001 0.966 0.012 11246.006 0.042 0.010 11057.128 -0.830 0.407 0.010 11164.322 -2.222 0.026 0.024 10765.748 3.239 -0.051 -0.030 -0.019 -0.081 -0.022 -0.029 -0.044 0.001 0.031 0.024 11141.093 2.059 0.001 9060.621 0.040 0.002 -0.007 0.011 0.019 0.029 -0.035 0.023 0.012 -0.003 0.125 0.097 -0.004 Gender PersonOfColor -0.058 -0.097 0.005 6461.282 0.013 7964.355 -0.021 -0.026 0.007 -0.001 0.002 10743.785 0.003 6636.617 0.006 7199.970 0.003 3770.637 Education Wealth MarLength Slope* Actor: Moving Slope* Actor: Birth of a child Slope* Actor: death of a child -8.334 <.001 - 12.447 <.001 -7.478 <.001 - 11.498 <.001 -7.536 <.001 1.103 0.270 -0.385 0.700 -0.068 -0.122 -0.049 -0.071 -0.024 -0.033 -0.005 -0.006 -0.017 -0.019 0.020 0.004 -0.011 0.005 4703.535 -2.070 0.039 -0.021 -0.001 0.004 0.004 3910.723 1.108 0.268 -0.003 0.012 126 0.011 0.044 0.016 -0.001 -0.010 Table 10 (cont’d) Slope* Actor: New chronic illness Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change Slope* Actor: Parent dies Slope* Partner: Parent dies Slope* Actor: New job Slope* Partner: New job Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Note. Significant effects bolded, p <.05 -0.002 0.001 0.003 -0.002 -0.018 -0.010 -0.001 -0.007 -0.020 0.003 7481.776 0.002 7435.979 4.366 <.001 0.006 0.016 -0.349 0.727 -0.006 0.004 0.016 17559.317 2.791 0.005 0.013 0.074 0.008 8579.466 0.004 7132.998 0.004 7022.976 0.003 6444.600 0.003 6569.055 0.003 6770.628 0.003 6948.574 0.002 6835.637 0.002 6846.941 0.006 6680.582 0.006 7143.862 2.097 0.036 0.001 0.031 -2.654 0.008 -0.017 -0.003 -1.807 0.071 -0.014 0.001 -3.494 <.001 -0.015 -0.004 -0.731 0.465 0.509 0.611 0.341 0.951 -0.887 0.375 -0.007 -0.004 -0.003 -0.007 0.003 0.007 0.008 0.003 -0.389 0.697 -0.006 0.004 -3.526 <.001 -0.031 -0.009 -3.070 0.002 -0.029 -0.006 127 LB 0.885 0.463 0.302 0.064 0.022 0.007 0.016 0.084 0.042 0.048 0.043 0.095 -0.015 -0.095 -0.082 <.001 0.044 0.033 0.002 UB 3.061 -0.010 0.028 0.015 -0.011 0.010 11137.702 6.222 0.010 11108.161 2.126 0.045 13297.747 0.145 0.016 10905.445 1.032 -0.026 0.035 13866.189 -0.733 SE df b 0.075 8475.815 2.914 -0.047 0.019 12490.921 -2.473 0.622 0.007 0.011 5676.483 -1.290 -0.029 0.022 7064.004 -2.594 -0.045 0.017 6751.175 Table 11. Linear growth curve model examining the effect of actor/partner life events and spousal support on the slope of neuroticism; HRS sample. t Parameter p 38.761 <.001 2.766 Intercept -0.084 0.013 Slope -0.015 0.534 Actor: Moving -0.072 0.197 Actor: Birth of a child -0.079 0.009 Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent dies Partner: Parent dies Actor: New job Partner: New job Actor: Retirement Partner: Retirement Actor: Unemployment 0.069 Partner: Unemployment 0.037 Age Gender PersonOfColor Education Wealth MarLength Support Slope*Support Actor: Moving*Support Actor: Birth of a child*Support Actor: death of a child*Support Actor: New chronic illness*Support Partner: New chronic illness*Support -0.061 0.068 -0.025 0.915 0.590 -0.017 <.001 -0.083 -0.024 0.919 -0.032 0.289 -0.044 0.028 0.004 0.022 0.128 -0.011 <.001 -0.007 -0.005 0.001 8981.085 <.001 -0.055 -0.046 0.005 6507.191 -0.104 0.013 7728.123 <.001 -0.129 -0.020 0.002 10467.672 -10.952 <.001 -0.023 -7.169 <.001 -0.032 -0.025 0.003 6518.600 0.197 0.008 -0.004 1.290 0.006 7010.353 <.001 -0.524 -0.343 0.092 13649.008 -3.714 -0.121 0.102 -0.055 0.034 15329.410 -1.636 -0.031 0.532 -0.008 0.012 25849.534 -0.625 -0.029 0.016 11020.219 -1.825 -0.001 0.012 10408.264 -0.106 0.007 0.012 10362.587 0.539 -0.060 0.012 10619.984 -5.163 -0.001 0.012 10700.526 -0.101 -0.011 0.010 10721.179 -1.060 -0.023 0.010 10717.780 -2.194 0.024 10356.406 2.858 0.024 10699.541 1.521 -8.191 -9.595 -8.103 -0.013 0.012 26280.786 -1.020 0.017 25102.437 1.804 0.023 25143.686 2.458 0.012 25057.234 0.503 0.002 0.023 0.030 -0.037 0.022 0.009 -0.002 0.116 0.084 -0.004 -0.036 -0.079 -0.016 -0.018 0.021 -0.162 0.011 0.016 0.014 0.011 -0.003 -0.037 -0.018 0.102 0.065 0.012 0.031 0.057 0.031 0.006 0.071 0.308 0.615 128 0.003 0.040 0.006 0.028 0.001 0.017 0.176 0.014 24652.428 0.218 0.050 19095.048 0.549 0.012 26460.637 0.501 0.083 13062.063 2.126 0.013 25643.033 0.072 0.017 25495.882 0.977 0.017 25000.710 2.313 -0.007 0.012 25639.456 -0.567 0.014 26089.978 0.068 0.013 25999.228 0.383 Table 11 (cont’d) Actor: Negative health change*Support Partner: Negative health change*Support Actor: Positive health change*Support Partner: Positive health change*Support Actor: Parent dies*Support Partner: Parent dies*Support 0.001 Actor: New job*Support 0.005 Partner: New job*Support Actor: Retirement*Support Partner: Retirement*Support Actor: Unemployment*Support -0.006 0.025 26133.321 -0.237 Partner: Unemployment*Support -0.027 0.026 25878.468 -1.047 Slope* Actor: Moving 0.301 Slope* Actor: Birth of a child Slope* Actor: Death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change Slope* Actor: Parent dies Slope* Partner: Parent dies Slope* Actor: New job -0.003 0.003 5811.718 0.003 6126.560 0.001 -0.007 0.012 15682.181 -0.599 -0.003 0.003 6894.965 -0.008 0.004 6348.949 -0.007 0.004 6731.088 -0.011 0.006 4117.866 -0.009 0.003 5780.259 0.016 16279.723 2.356 0.004 3246.802 0.003 3291.744 0.003 6904.867 -1.139 0.427 -2.070 -1.777 -1.104 -2.012 -3.104 0.039 0.005 0.010 0.001 3.776 1.218 129 0.034 0.014 0.338 0.583 -0.071 0.126 0.021 0.006 0.074 0.328 -0.017 0.050 0.827 -0.024 0.030 0.946 0.702 -0.026 -0.021 0.028 0.031 0.942 -0.025 0.027 0.571 -0.030 0.017 0.617 -0.018 0.030 0.813 -0.056 0.044 0.295 0.764 -0.078 -0.004 0.024 0.006 0.044 -0.022 0.000 0.223 -0.003 0.013 <.001 0.005 0.015 0.269 -0.008 0.002 0.018 0.007 0.071 0.549 -0.031 0.016 0.076 -0.015 0.001 0.038 -0.016 0.000 0.002 -0.014 -0.003 0.255 0.670 -0.009 -0.004 0.002 0.007 0.003 6083.226 -0.002 0.003 6240.762 -0.001 0.003 6212.084 -0.020 0.006 6066.747 -0.016 0.006 6075.348 0.821 0.412 -0.003 0.008 -0.799 0.424 -0.007 0.003 -0.423 0.672 -0.006 0.004 -3.388 <.001 -0.031 -0.008 -2.755 0.006 -0.028 -0.005 -0.008 0.009 10669.008 -0.909 0.005 10788.711 2.008 0.030 14328.377 2.332 -0.013 0.019 20515.703 -0.664 -0.002 0.007 12495.063 -0.378 0.364 -0.025 0.009 0.045 0.000 0.018 0.020 0.011 0.128 0.506 -0.051 0.025 0.706 -0.015 0.010 -0.011 0.005 8461.659 -2.327 0.020 -0.020 -0.002 Table 11 (cont’d) Slope* Partner: New job 0.002 Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Slope* Actor: Birth of a child* Support Slope* Actor: New chronic illness* Support 0.009 Slope* Actor: Negative health change* Support Slope* Partner: Negative health change* Support Slope* Actor: Positive health change* Support Slope* Actor: Parent dies* Support Slope* Actor: Unemployment* Support Slope* Partner: Unemployment* Support Note. Significant effects bolded, p <.01 0.069 -0.009 0.009 11092.610 -1.017 0.309 -0.027 0.008 -0.004 0.009 8447.526 -0.431 0.667 -0.022 0.014 130 Table 12. Linear growth curve model examining the effect of actor and partner life events on the slope of openness; HRS sample. Intercept Slope Actor: Moving Actor: Birth of a child Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent dies Partner: Parent dies Actor: New job Partner: New job Actor: Retirement Partner: Retirement Actor: Unemployment Partner: Unemployment Age Gender PersonOfColor Education Wealth MarLength Slope* Actor: Moving Slope* Actor: Birth of a child b 2.412 0.014 0.024 SE df 0.068 8740.516 0.015 16153.439 0.917 2.282 0.010 5906.110 t 35.387 <.001 0.359 0.023 p UB LB 2.546 2.279 -0.016 0.043 0.044 0.003 -0.008 0.021 7732.337 -0.396 0.692 -0.049 0.033 -0.008 0.016 7617.119 -0.467 0.640 -0.039 0.024 -0.013 0.009 11320.580 -1.409 0.159 -0.032 0.005 -0.015 0.009 11293.418 -1.558 0.119 -0.033 0.004 -0.071 0.039 14393.654 -1.808 0.071 -0.149 0.006 -0.004 0.025 13229.658 -0.177 0.859 -0.054 0.045 0.012 0.015 10903.833 0.793 0.428 -0.017 0.040 0.025 -0.005 0.009 0.067 0.011 0.008 0.001 0.015 11177.766 1.675 0.011 10754.010 -0.409 0.011 10720.236 0.805 0.011 10981.275 6.242 0.011 11160.994 1.007 0.010 11042.514 0.886 0.010 11118.470 0.128 0.094 0.682 0.421 <.001 0.314 0.376 0.899 -0.004 0.053 -0.027 0.018 -0.013 0.031 0.088 0.046 -0.010 0.032 -0.010 0.027 -0.018 0.020 0.063 0.022 10593.805 2.845 0.004 0.020 0.107 -0.023 -0.004 -0.003 0.051 0.041 0.031 -0.036 -1.018 -6.937 -0.761 4.244 0.309 0.022 10994.734 <.001 0.001 9247.071 0.447 0.004 6463.218 <.001 0.012 8256.698 0.002 11093.334 24.600 <.001 <.001 0.003 6722.006 <.001 0.006 7373.982 9.393 -5.951 -0.066 0.021 -0.005 -0.003 -0.011 0.005 0.075 0.028 0.044 0.038 0.037 0.024 -0.024 -0.047 -0.002 0.002 3629.001 -1.109 0.267 -0.007 0.002 0.004 0.005 4493.624 0.868 0.386 -0.005 0.013 131 0.014 17559.016 0.007 8138.833 0.003 3750.669 0.002 7154.619 0.002 7113.719 -0.030 -0.006 -0.003 -0.006 -0.003 Table 12 (cont’d) Slope* Actor: death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness -0.005 Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change Slope* Actor: Parent dies Slope* Partner: Parent dies Slope* Actor: New job Slope* Partner: New job Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Note. Significant effects bolded, p < .05 -0.002 0.005 0.005 0.000 0.008 0.014 0.002 0.003 0.006 0.003 6867.801 0.003 6732.204 0.002 6191.113 0.002 6288.992 0.002 6483.349 0.002 6651.690 0.002 6569.633 0.002 6567.281 0.005 6385.820 0.005 6844.806 -0.896 0.370 -0.009 0.004 -1.445 0.148 -0.008 0.001 -2.227 0.026 -0.009 -0.001 -2.138 0.033 -0.058 -0.003 -0.921 0.357 -0.020 0.007 -1.664 0.096 -0.012 0.001 0.589 0.556 -0.005 0.009 6.068 <.001 0.010 0.019 2.055 0.040 0.000 0.010 2.093 0.036 0.000 0.010 3.311 <.001 0.003 0.013 0.070 0.944 -0.004 0.004 1.597 0.110 -0.001 0.008 1.137 0.256 -0.004 0.015 -0.471 0.638 -0.012 0.008 132 p LB 0.191 0.777 0.009 0.019 -0.131 0.032 -0.054 0.072 -0.010 0.049 -0.029 0.009 -0.033 0.004 -1.192 0.233 -1.519 0.129 -1.028 0.304 0.015 10735.319 1.308 0.032 13465.298 0.283 0.015 10880.547 2.016 -0.049 0.041 13136.410 -0.014 0.010 10979.613 -0.010 0.010 11013.278 UB 2.528 -0.027 0.041 0.044 -0.053 0.030 -0.046 0.019 t 33.764 <.001 2.250 0.694 0.024 0.003 Table 13. Linear growth curve model examining the effect of actor/partner life events and spousal support on the slope of openness; HRS sample. SE df b 0.071 8560.606 2.389 0.017 12721.026 0.394 0.007 2.261 0.024 0.011 5733.933 -0.559 0.576 -0.012 0.021 7226.077 -0.803 0.422 -0.013 0.017 7037.950 Intercept Slope Actor: Moving Actor: Birth of a child Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change 0.030 Actor: Parent dies -0.008 0.011 10363.019 Partner: Parent dies 0.009 Actor: New job 0.067 Partner: New job 0.009 Actor: Retirement 0.006 Partner: Retirement 0.000 Actor: Unemployment 0.068 Partner: Unemployment -0.994 0.320 -0.022 0.022 10585.623 Age -6.702 <.001 -0.005 -0.004 0.001 9207.091 Gender -0.020 -2.696 0.007 -0.012 0.004 6524.773 PersonOfColor 0.012 8114.479 0.059 <.001 0.035 4.872 Education 0.002 10905.864 24.142 <.001 0.037 0.041 Wealth 8.875 0.029 <.001 0.023 0.003 6643.328 MarLength -5.753 <.001 -0.046 -0.034 0.006 7245.025 <.001 0.146 Support 0.304 0.005 0.022 0.075 Slope*Support Actor: Moving*Support 0.000 Actor: Birth of a child*Support Actor: death of a child*Support Actor: New chronic illness*Support Partner: New chronic illness*Support 0.011 10276.348 0.753 0.011 10586.999 6.137 0.011 10671.930 0.858 0.010 10761.228 0.649 0.010 10719.511 0.022 10208.609 3.022 -0.683 0.494 0.451 <.001 0.045 0.391 0.517 -0.037 0.970 0.059 -0.030 0.015 -0.014 0.031 0.088 -0.012 0.030 -0.013 0.025 -0.019 0.019 0.111 -0.066 0.022 -0.003 -0.003 0.083 0.044 0.036 -0.023 0.461 0.128 -0.021 0.021 0.080 14781.551 3.782 0.027 14467.445 2.793 0.011 25607.478 -0.034 0.015 24171.223 -0.003 0.011 24828.157 0.011 25655.127 0.710 0.020 25154.963 1.687 -0.036 0.971 -2.224 0.026 -0.247 0.805 -0.006 0.074 -0.024 0.019 -0.014 0.029 0.044 0.001 0.003 0.024 -0.064 -0.004 0.008 0.034 0.092 0.478 133 0.018 0.018 0.016 -0.023 0.012 23981.524 -0.006 0.015 24706.203 -0.018 0.033 18082.211 -0.198 0.076 13118.565 -0.011 0.012 25277.364 -0.022 0.011 25112.085 Table 13 (cont’d) Actor: Negative health change*Support Partner: Negative health change*Support Actor: Positive health change*Support Partner: Positive health change*Support Actor: Parent dies*Support Partner: Parent dies*Support Actor: New job*Support 0.009 Partner: New job*Support Actor: Retirement*Support Partner: Retirement*Support Actor: Unemployment*Support -0.008 0.022 25264.533 Partner: Unemployment*Support -0.051 0.023 25091.479 Slope* Actor: Moving -0.003 0.002 3157.821 Slope* Actor: Birth of a child Slope* Actor: Death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change Slope* Actor: Parent dies Slope* Partner: Parent dies Slope* Actor: New job 0.003 5677.140 0.002 5933.337 -0.029 0.015 16148.892 -0.005 0.002 6699.107 -0.003 0.004 6535.192 -0.004 0.004 3139.880 -0.004 0.002 6678.863 0.004 6153.151 0.003 5620.256 0.005 3947.361 0.005 0.006 0.013 0.002 0.001 0.003 134 0.015 25878.897 1.172 0.241 -0.012 0.047 -2.608 0.009 -0.348 -0.049 -0.524 0.600 -0.083 0.048 -0.418 0.676 -0.036 0.023 -1.895 0.058 -0.047 0.001 0.012 24903.059 0.771 -0.903 0.366 0.441 -0.036 0.013 -0.014 0.032 0.012 24667.230 1.526 0.127 -0.005 0.041 -2.096 0.036 -0.043 -0.001 0.011 25970.970 1.437 0.151 -0.006 0.037 -0.360 0.719 -0.052 0.036 -2.204 0.028 -1.251 0.211 -0.096 -0.006 -0.008 0.002 0.509 0.611 -0.007 0.012 -1.121 0.262 -0.011 0.003 -1.542 0.123 -0.008 0.001 -2.061 0.039 -0.009 0.000 -1.928 0.054 -0.058 0.000 -0.816 0.414 -0.010 0.004 0.885 0.376 -0.004 0.010 5.211 <.001 0.008 0.018 1.946 2.278 0.052 0.000 0.023 0.001 0.010 0.010 0.011 15477.858 0.115 0.908 -0.020 0.023 0.002 5902.267 -0.001 0.002 6074.804 0.002 6052.074 0.005 5849.515 -0.003 0.005 5807.256 -0.004 0.004 10677.845 -0.076 0.027 14527.244 0.007 0.004 0.005 Table 13 (cont’d) Slope* Partner: New job Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Slope* Partner: New chronic illness*Support Slope* Actor: Negative health change*Support Slope* Actor: Parent dies*Support Slope* Partner: Parent dies*Support Slope* Actor: New job*Support Slope* Partner: New job*Support Note. Significant effects bolded, p < .01 0.002 0.004 0.005 0.004 8177.823 0.004 9207.096 0.004 9406.067 -0.003 0.004 9070.130 2.738 0.006 0.002 0.012 -0.619 0.536 -0.006 0.003 1.686 0.092 -0.001 0.008 0.930 0.352 -0.005 0.015 -0.648 0.517 -0.014 0.007 -1.014 0.311 -0.012 0.004 -2.817 0.005 -0.129 -0.023 1.198 0.231 -0.003 0.013 0.551 0.582 -0.006 0.011 0.944 0.345 -0.004 0.012 -0.811 0.417 -0.011 0.005 135 Table 14. Linear growth curve model examining the effect of actor and partner life events on the slope of spousal support; HRS sample. SE df b 0.077 853.313 3.334 0.071 0.021 100.227 -0.002 0.011 2605.684 0.022 3718.688 0.000 -0.053 0.017 4203.829 -0.052 0.010 3884.525 t 43.414 <.001 3.184 3.387 0.001 0.029 -0.181 0.856 -0.015 0.988 -3.043 0.002 -5.460 <.001 UB 3.485 0.112 -0.024 0.020 -0.044 0.043 -0.019 -0.087 -0.033 -0.071 LB p Intercept Slope Actor: Moving Actor: Birth of a child Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent dies Partner: Parent dies Actor: New job Partner: New job Actor: Retirement Partner: Retirement Actor: Unemployment Partner: Unemployment Age Gender PersonOfColor Education Wealth MarLength Slope* Actor: Moving Slope* Actor: Birth of a child Slope* Actor: death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness -0.022 0.009 4119.495 -2.294 0.022 -0.040 -0.003 -0.055 0.042 5579.256 -1.322 0.186 -0.137 0.027 0.020 0.034 1701.021 0.585 0.559 -0.047 0.087 -0.014 0.015 5582.163 -0.938 0.348 -0.043 0.015 0.009 6063.822 0.695 0.011 6613.657 0.038 0.011 6817.710 1.532 0.011 6719.719 1.193 -2.907 0.004 0.125 0.233 -0.138 0.891 0.970 -1.576 0.115 0.487 -2.830 0.005 -3.127 0.002 -0.444 0.657 -0.044 0.015 6280.303 0.017 0.013 -0.001 0.011 6238.048 0.000 -0.015 0.010 6818.871 0.007 -0.061 0.022 7059.351 -0.069 0.022 5564.011 0.001 1466.685 0.000 0.004 7104.476 25.362 <.001 0.090 0.097 -0.079 -0.054 0.013 2695.745 0.022 0.001 0.004 0.020 <.001 0.013 0.000 0.002 -0.073 -0.014 -0.005 0.040 -0.008 0.035 -0.022 0.019 -0.020 0.021 -0.034 0.004 -0.012 0.025 -0.019 -0.104 -0.026 -0.113 -0.002 0.001 0.105 -0.029 0.007 0.027 -0.013 0.012 -0.007 0.010 0.002 4109.214 2.286 0.004 1611.958 5.480 0.006 1568.813 0.004 37.020 -0.062 0.951 0.666 0.435 -4.220 <.001 0.014 0.008 56.535 1.766 0.083 -0.002 0.031 -0.006 0.006 33.411 -1.051 0.301 -0.019 0.006 -0.004 0.003 67.706 -1.338 0.185 -0.011 0.002 -0.007 0.003 87.561 -1.956 0.054 -0.013 0.000 136 0.005 70.646 -0.039 0.016 1064.481 -0.034 0.013 417.873 0.001 Table 14 (cont’d) Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change Slope* Actor: Parent dies Slope* Partner: Parent dies Slope* Actor: New job Slope* Partner: New job Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Note. Significant effects bolded, p < .05 0.011 0.005 70.669 0.002 0.004 50.822 0.004 0.004 76.705 0.002 0.004 59.045 0.008 0.002 0.004 61.175 -0.001 0.003 67.676 0.003 63.861 0.000 -0.002 0.008 69.012 0.008 46.503 -2.489 0.013 -0.070 -0.008 -2.644 0.009 -0.059 -0.009 0.225 0.822 -0.009 0.012 0.746 0.325 0.288 1.074 0.533 0.627 0.029 0.001 2.233 0.649 0.519 -0.230 0.819 0.983 0.022 -0.009 0.012 -0.003 0.011 -0.005 0.010 0.015 -0.005 0.009 -0.007 0.006 -0.006 0.007 -0.315 0.753 -0.017 0.013 1.396 0.169 -0.005 0.026 137 Table 15. Linear growth curve model examining the effect of actor and partner life events on the slope of spousal strain; HRS sample. Intercept Slope Actor: Moving Actor: Birth of a child Actor: death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent dies Partner: Parent dies Actor: New job Partner: New job Actor: Retirement Partner: Retirement Actor: Unemployment Partner: Unemployment Age Gender PersonOfColor Education Wealth MarLength Slope* Actor: Moving Slope* Actor: Birth of a child SE df 0.089 8378.755 b 2.043 -0.069 0.022 11624.833 0.024 0.014 5776.402 p t 22.990 <.001 -3.229 0.001 0.083 1.735 LB 1.869 -0.112 -0.003 UB 2.217 -0.027 0.050 0.049 0.027 7818.978 1.845 0.065 -0.003 0.101 0.000 0.021 7732.396 0.012 0.990 -0.040 0.041 0.056 0.011 10124.802 4.922 <.001 0.034 0.078 0.042 0.011 10097.522 3.745 <.001 0.020 0.064 -0.046 0.047 12999.722 -0.964 0.335 -0.138 0.047 0.037 0.038 13411.971 0.972 0.331 -0.038 0.112 0.027 0.018 9733.018 1.544 0.123 -0.007 0.062 0.018 9897.123 0.052 -0.010 0.014 9526.466 0.013 9334.594 0.011 0.013 9798.231 0.014 -0.005 0.013 9933.752 0.013 0.007 0.011 10143.887 1.160 0.011 10036.839 0.646 0.004 2.907 -0.736 0.462 0.409 0.826 0.281 1.078 -0.386 0.700 0.246 0.519 0.017 -0.037 -0.015 -0.011 -0.030 -0.009 -0.015 0.087 0.017 0.037 0.039 0.020 0.036 0.030 0.080 0.026 9252.873 3.025 0.002 0.028 0.132 0.027 9609.660 <.001 3.551 0.094 -2.916 0.004 -0.002 0.001 10391.367 -9.111 <.001 -0.041 0.004 6237.065 <.001 7.480 0.015 8848.846 0.114 0.002 0.339 0.002 11412.837 0.956 -3.626 <.001 -0.016 0.004 6692.463 <.001 3.635 0.008 7456.710 0.028 0.042 -0.004 -0.049 0.084 -0.002 -0.024 0.013 0.146 -0.001 -0.032 0.144 0.006 -0.007 0.043 -0.002 0.003 3385.180 -0.622 0.534 -0.008 0.004 0.001 0.007 4308.509 0.187 0.852 -0.012 0.014 138 -0.010 0.005 3488.106 0.003 7204.911 0.003 7024.668 -2.046 0.041 -0.020 0.000 0.657 0.511 -0.004 0.008 0.882 0.378 -0.003 0.009 0.017 16066.789 0.425 0.013 15285.410 3.694 0.671 -0.027 0.041 <.001 0.023 0.076 0.142 0.887 -0.009 0.010 0.308 0.758 -0.008 0.011 -1.073 0.283 -0.010 0.003 0.010 0.992 -0.006 0.006 -0.343 0.731 -0.007 0.005 0.350 0.727 -0.005 0.008 0.004 0.997 -0.006 0.006 1.328 0.184 -0.002 0.010 1.749 0.080 -0.001 0.025 0.012 0.991 -0.013 0.014 0.002 0.003 0.050 0.007 0.001 Table 15 (cont’d) Slope* Actor: death of a child Slope* Actor: New chronic illness Slope* Partner: New chronic illness Slope* Actor: Negative health change Slope* Partner: Negative health change Slope* Actor: Positive health change Slope* Partner: Positive health change Slope* Actor: Parent dies Slope* Partner: Parent dies Slope* Actor: New job Slope* Partner: New job Slope* Actor: Retirement Slope* Partner: Retirement Slope* Actor: Unemployment Slope* Partner: Unemployment Note. Significant effects bolded, p < .05 0.000 0.000 0.001 0.001 0.000 0.004 0.012 0.005 6629.248 0.005 6706.670 -0.004 0.003 6064.885 0.003 5846.611 -0.001 0.003 6326.614 0.003 6314.008 0.003 6583.715 0.003 6429.965 0.007 5929.736 0.007 6386.858 139 Table 16. A summary of which life events produced a (mal)adaptive response, organized by trait; HRS sample. Trait Agreeableness Maladaptive Actor: Negative health change Actor: Positive health change Adaptive Actor: Death of a parent Partner: New job Conscientiousness Actor: Death of a parent Partner: Death of a parent Partner: New job Actor/Partner: Death of a child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Extraversion Actor: Parent dying Partner: New job Actor: New chronic illness Actor: Negative health change Neuroticism Openness Actor/Partner: Birth of a child Actor: Positive health change Actor: Death of a parent Actor: Unemployment Partner: Unemployment Actor: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Death of a parent Partner: Death of a parent Actor: New job Partner: New job Partner: New chronic illness Actor: Negative health change Note. Type of response (i.e., adaptive or maladaptive) based on Event*Slope interactions. "Adaptive" responses include increases in all traits with the exception of neuroticism. "Maladaptive" responses include decreases in all traits with the exception of neuroticism. 140 APPENDIX B: TABLES FOR CHAPTER 3 Table 17. Frequency of life events; LISS sample. Life event Birth of child Death of a child Death of a parent Negative change in health Positive change in health New chronic illness Retirement Unemployment First job Frequency (% of sample) 632 (9.1%) 310 (4.5%) 995 (14.3%) 3014 (43.3%) 2880 (41.4%) 2334 (33.5%) 1834 (26.3%) 495 (7.1%) 23 (.3%) 141 Table 18. Descriptives of and correlations between traits and relationship satisfaction; averaged across waves; LISS sample. Agreeableness Conscientiousness Extraversion Neuroticism Openness Agreeableness Conscientiousness Extraversion Neuroticism Openness Relationship satisfaction ** Correlation is significant at the 0.01 level (2-tailed). Mean SD 3.861 0.510 .302** 3.767 0.507 3.245 0.652 .327** 2.490 0.684 -.077** .264** 3.469 0.499 .111** 1 .108** -.243** .253** .130** 1 .108** 1 -.240** .337** .109** 1 -.205** -.228** 1 .029** 142 Note. **Correlation is significant at the .01 level (2-tailed), *Correlation is significant at the 0.05 level (2-tailed). A indicates actor life event, P indicates partner life event. 143 Table 19. Biserial correlations between life events within couples; LISS sample. 12345678910111213141516171. A: Birth of a child12. P: Birth of a child1.000**13. A: Death of a child.024*.024*14. P: Death of a child.024*.024*1.000**15. A: Death of a parent0.0100.010.035**.035**16. P: Death of a parent0.0100.010.035**.035**.166**17. A: Negative health change.077**.077**.108**.108**.209**.152**18. P: Negative health change.077**.077**.108**.108**.152**.209**.337**19. A: Positive health change.076**.076**.080**.080**.213**.159**.668**.295**110. P: Positive health change.076**.076**.080**.080**.159**.213**.295**.668**.304**111. A: New chronic illness-0.007-0.007.074**.074**.164**.132**.389**.253**.325**.219**112. P: New chronic illness-0.007-0.007.074**.074**.132**.164**.253**.389**.219**.325**.317**113. A: Retirement-.125**-.125**.092**.092**-0.0030.009.161**.149**.120**.117**.296**.294**114. P: Retirement-.125**-.125**.092**.092**0.009-0.003.149**.161**.117**.120**.294**.296**.668**115. A: Unemployment.027*.027*0.0050.005.093**.090**.085**.083**.094**.084**.060**.047**-.033**-.037**116. P: Unemployment.027*.027*0.0050.005.090**.093**.083**.085**.084**.094**.047**.060**-.037**-.033**.124**117. A: First job.052**.052**0.0120.0120.019-0.0020.020-0.0150.023-0.0030.007-0.020-.034**-.029*-0.0060.004118. P: First job.052**.052**0.0120.012-0.0020.019-0.0150.020-0.0030.023-0.0200.007-.029*-.034**0.004-0.006-0.003Note. ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed). A indicates actor event, P indicates partner event. Table 19 (cont’d) 144 Table 19. Biserial correlations between life events within couples; LISS sample. 12345678910111213141516171. A: Birth of a child12. P: Birth of a child1.000**13. A: Death of a child.024*.024*14. P: Death of a child.024*.024*1.000**15. A: Death of a parent0.0100.010.035**.035**16. P: Death of a parent0.0100.010.035**.035**.166**17. A: Negative health change.077**.077**.108**.108**.209**.152**18. P: Negative health change.077**.077**.108**.108**.152**.209**.337**19. A: Positive health change.076**.076**.080**.080**.213**.159**.668**.295**110. P: Positive health change.076**.076**.080**.080**.159**.213**.295**.668**.304**111. A: New chronic illness-0.007-0.007.074**.074**.164**.132**.389**.253**.325**.219**112. P: New chronic illness-0.007-0.007.074**.074**.132**.164**.253**.389**.219**.325**.317**113. A: Retirement-.125**-.125**.092**.092**-0.0030.009.161**.149**.120**.117**.296**.294**114. P: Retirement-.125**-.125**.092**.092**0.009-0.003.149**.161**.117**.120**.294**.296**.668**115. A: Unemployment.027*.027*0.0050.005.093**.090**.085**.083**.094**.084**.060**.047**-.033**-.037**116. P: Unemployment.027*.027*0.0050.005.090**.093**.083**.085**.084**.094**.047**.060**-.037**-.033**.124**117. A: First job.052**.052**0.0120.0120.019-0.0020.020-0.0150.023-0.0030.007-0.020-.034**-.029*-0.0060.004118. P: First job.052**.052**0.0120.012-0.0020.019-0.0150.020-0.0030.023-0.0200.007-.029*-.034**0.004-0.006-0.003Note. ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed). A indicates actor event, P indicates partner event. Table 20. Linear growth curve model examining the effect of actor and partner life events on the slope of agreeableness; LISS sample. Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic illness Actor: Retired Partner: Retired Actor: Unemployed Partner: Unemployed Age b 3.524 -0.008 -0.026 -0.028 -0.001 0.026 t p df LB UB SE 3.603 0.040 4103.069 87.235 0.000 3.445 -4.121 <.001 -0.012 0.002 2222.506 -0.004 0.279 -1.083 -0.072 0.021 0.024 2448.555 0.315 -1.004 -0.083 0.027 0.028 2180.222 -0.033 0.030 0.927 0.016 4099.892 -0.092 -0.005 0.058 0.101 0.016 4232.074 1.641 0.039 0.017 5209.448 2.223 0.026 0.005 0.073 -0.019 0.017 5029.903 -1.125 0.261 -0.053 0.014 -0.004 0.017 5176.352 -0.221 0.825 -0.037 0.029 -0.014 0.017 4983.184 -0.846 0.397 -0.047 0.018 0.031 0.014 4563.679 2.240 0.025 0.004 0.057 0.019 -0.001 0.008 0.024 0.006 0.004 0.014 4594.124 1.389 0.019 4520.791 -0.045 0.018 4583.301 0.415 0.022 4332.348 1.078 0.022 4220.065 0.249 0.001 3606.324 4.856 0.165 0.964 0.678 0.281 0.804 <.001 0.003 -0.008 0.046 -0.037 0.035 -0.028 0.044 -0.020 0.068 -0.038 0.049 0.006 0.005 -0.171 -0.001 0.029 Gender Relationship length Education Slope*Actor: Birth of a child Slope*Actor: Death of a -0.006 child Slope*Actor: Parent death 0.001 Slope*Partner: Parent death Slope*Actor: Negative health change Slope*Partner: Negative health change Slope*Actor: Positive health change -0.002 0.002 0.001 0.000 - 30.600 <.001 0.006 3029.985 0.064 0.001 3164.951 -1.850 <.001 0.021 0.004 5041.401 7.143 -0.160 -0.182 -0.003 0.000 0.037 0.002 1493.901 2.044 0.041 0.000 0.010 -2.013 0.003 1342.926 0.001 2061.169 0.752 0.044 0.452 -0.011 0.000 -0.002 0.004 0.002 2074.179 0.415 0.678 -0.002 0.004 0.002 3100.691 -0.029 0.977 -0.004 0.004 0.002 2789.774 -0.864 0.388 -0.006 0.002 0.002 3093.124 0.956 0.339 -0.002 0.006 145 0.003 -0.001 Table 20 (cont’d) Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: New chronic illness Slope*Actor: Retired Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed Note. Significant effects bolded, p < .05 0.000 0.000 0.000 0.000 0.003 0.002 2775.219 1.354 0.176 -0.001 0.006 0.001 2399.273 -1.022 0.307 -0.004 0.001 0.001 2330.960 -0.223 0.002 1779.130 0.160 -0.287 0.002 1781.502 0.002 2065.845 0.145 0.002 2091.515 1.267 0.823 0.873 0.774 -0.003 0.002 -0.003 0.004 -0.004 0.003 0.885 -0.004 0.005 0.205 -0.002 0.007 146 Table 21. Linear growth curve model examining the effect of actor/partner life events and relationship satisfaction on the slope of agreeableness; LISS sample. p 0.000 0.053 0.166 0.506 0.875 0.208 SE 0.045 3331.959 79.183 -1.939 -1.385 -0.666 -0.157 0.017 3614.333 1.260 b 3.555 -0.004 0.002 1945.073 -0.035 0.026 2216.163 -0.019 0.029 1986.854 -0.003 0.016 3522.061 0.021 LB 3.467 -0.008 -0.086 -0.076 -0.035 -0.012 UB 3.643 0.000 0.015 0.038 0.030 0.054 df t Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic illness Actor: Retired Partner: Retired Actor: Unemployed Partner: Unemployed Age Gender Relationship length Education Relationship satisfaction Slope*Relationship satisfaction Actor: Birth of a child*Satisfaction Actor: Death of a child*Satisfaction Actor: Parent death*Satisfaction Partner: Parent death*Satisfaction Actor: Negative health change*Satisfaction 0.033 0.019 4496.768 1.793 0.073 -0.003 0.069 -0.015 0.018 4266.199 -0.841 0.400 -0.051 0.020 0.005 0.018 4442.356 0.273 0.785 -0.030 0.040 -0.018 0.018 4221.109 -1.016 0.309 -0.053 0.017 0.026 0.014 3910.531 1.828 0.068 -0.002 0.055 0.015 3915.244 1.130 -0.583 0.020 3859.731 0.033 0.024 3665.638 1.206 0.024 3653.640 0.713 0.001 2998.134 3.824 0.016 -0.012 0.020 3749.192 0.001 0.029 0.017 0.004 -0.172 0.006 2501.673 -0.001 0.001 2677.778 0.028 0.259 0.560 0.974 0.228 0.476 <.001 -27.743 <.001 0.235 -1.187 <.001 0.004 4080.434 6.258 -0.012 -0.050 -0.038 -0.018 -0.030 0.002 -0.185 -0.003 0.019 0.045 0.027 0.039 0.075 0.064 0.006 -0.160 0.001 0.037 0.044 0.009 4869.800 4.835 <.001 0.026 0.062 -0.002 0.001 1903.968 -3.383 <.001 -0.003 -0.001 -0.005 0.018 3531.100 -0.274 0.784 -0.040 0.030 0.010 0.020 3320.890 0.487 0.626 -0.029 0.048 0.004 0.012 3853.463 0.372 0.710 -0.019 0.027 -0.016 0.012 3991.983 -1.335 0.182 -0.039 0.007 -0.017 0.013 4160.949 -1.321 0.187 -0.042 0.008 147 0.013 3738.184 0.014 0.989 -0.025 0.025 0.012 4230.669 1.861 0.063 -0.001 0.047 -1.259 0.208 -0.040 0.009 -0.896 0.370 -0.035 0.013 -0.669 0.503 -0.027 0.013 -0.002 0.999 -0.020 0.020 -0.893 0.372 -0.036 0.013 -1.146 0.252 -0.049 0.013 0.004 0.023 0.000 0.000 0.010 4148.049 -0.016 0.013 4335.501 -0.007 0.010 4038.729 -0.011 0.012 3682.443 -0.011 0.012 4204.795 Table 21 (cont’d) Partner: Negative health change*Satisfaction Actor: Positive health change*Satisfaction Partner: Negative health change*Satisfaction Actor: New chronic illness*Satisfaction Partner: New chronic illness*Satisfaction Actor: Retired*Satisfaction Partner: Retired*Satisfaction Actor: Unemployed*Satisfaction -0.018 0.016 4014.364 Partner: Unemployed*Satisfaction 0.011 Slope*Actor: Birth of a child Slope*Actor: Death of a child Slope*Actor: Parent death Slope*Partner: Parent death Slope*Actor: Negative health change Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: New chronic illness Slope*Actor: Retired Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed -0.001 0.001 2150.273 0.002 1641.080 0.000 -0.001 0.002 1637.489 -0.002 0.001 2204.447 -0.002 0.002 2515.221 -0.003 0.003 1366.413 -0.001 0.002 1915.605 0.002 1935.156 0.002 0.000 0.001 0.000 0.000 0.002 0.016 3804.394 0.735 0.002 1432.074 1.779 0.463 -0.019 0.042 0.075 0.000 0.009 -1.176 0.240 -0.009 0.002 0.001 1930.996 0.200 0.841 -0.003 0.003 -0.060 0.952 -0.003 0.003 0.002 2764.968 0.000 1.000 -0.004 0.004 -1.224 0.221 -0.006 0.001 0.002 2774.476 0.592 0.002 2517.403 1.186 0.554 -0.003 0.005 0.236 -0.002 0.006 -1.548 0.122 -0.005 0.001 -0.470 -0.041 -0.672 0.638 0.967 0.501 -0.004 -0.003 -0.005 0.002 0.003 0.002 -0.504 0.614 -0.005 0.003 0.002 1933.849 0.851 0.395 -0.002 0.006 148 Table 21 (cont’d) Slope*Actor: Birth of a child*Satisfaction Slope*Actor: Death of a child*Satisfaction Note. Significant effects bolded, p < .01 0.001 0.002 2115.871 0.703 0.482 -0.002 0.004 -0.007 0.002 2271.314 -3.802 <.001 -0.011 -0.004 149 Table 22. Linear growth curve model examining the effect of actor and partner life events on the slope of conscientiousness; LISS sample. SE t 0.043 3993.633 80.274 0.000 3.359 -0.003 0.002 2098.818 0.254 b 3.443 0.000 0.800 LB df p UB 3.527 0.004 0.025 2526.254 0.073 0.942 -0.047 0.051 0.017 4396.220 1.855 -1.506 0.132 0.064 -0.103 -0.002 0.013 0.064 -2.081 0.037 -0.069 -0.002 -0.026 0.980 -0.036 0.035 -0.161 0.872 -0.038 0.032 0.018 5462.969 3.963 0.017 5251.479 1.159 0.014 4832.654 0.167 <.001 0.035 0.105 0.247 -0.014 0.054 0.867 -0.026 0.031 0.070 0.020 0.002 0.002 0.000 0.018 5509.259 -0.003 0.018 5285.501 -0.036 0.017 4518.512 -0.045 0.030 2297.514 0.031 Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic -0.016 0.014 4823.938 illness -0.050 0.020 4741.878 Actor: Retired 0.031 Partner: Retired Actor: Unemployed 0.012 Partner: Unemployed 0.020 0.003 Age -0.040 0.006 3013.487 Gender 0.002 Relationship length 0.024 Education Slope*Actor: Birth of a child Slope*Actor: Death of a child Slope*Actor: Parent death Slope*Partner: Parent death Slope*Actor: Negative health change -0.004 0.002 3075.954 0.004 0.003 0.003 0.002 0.019 4592.958 1.594 0.024 4570.554 0.492 0.024 4578.550 0.827 0.001 3580.735 3.205 0.001 3146.717 2.285 0.004 4971.105 5.558 0.002 1437.683 1.747 0.002 1316.780 1.391 0.001 2053.209 1.292 0.001 2056.240 1.970 -0.045 -1.134 0.257 -0.089 -2.540 0.011 -0.007 0.111 -0.035 0.623 -0.027 0.408 0.001 0.001 -0.052 0.022 0.000 <.001 0.016 -6.641 <.001 0.012 -0.011 0.069 0.058 0.066 0.005 -0.028 0.003 0.032 0.081 0.000 0.008 0.164 -0.001 0.008 0.197 -0.001 0.005 0.049 0.000 0.006 -2.090 0.037 -0.008 0.000 150 0.000 0.001 0.004 Table 22 (cont’d) Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: New chronic illness Slope*Actor: Retired Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed Note. Significant effects bolded, p < .05 0.003 0.002 0.002 2730.857 -0.221 0.825 -0.004 0.003 0.002 3061.006 1.837 0.002 2721.662 0.440 0.066 0.000 0.007 0.660 -0.003 0.004 -0.003 0.001 2375.446 -0.001 0.001 2308.167 -0.002 0.002 1753.604 -0.006 0.002 1756.479 -1.981 0.048 -0.005 0.000 -0.497 0.619 -0.934 0.350 -0.003 -0.005 0.002 0.002 -3.522 <.001 -0.009 -0.003 0.002 2014.054 1.716 0.002 2087.830 1.087 0.086 0.000 0.007 0.277 -0.002 0.006 151 Table 23. Linear growth curve model examining the effect of actor/partner life events and relationship satisfaction on the slope of conscientiousness; LISS sample. p LB Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic illness Actor: Retired Partner: Retired Actor: Unemployed Partner: Unemployed Age Gender Relationship length Education Relationship satisfaction Slope*Relationship satisfaction Actor: Birth of a child*Satisfaction Actor: Death of a child*Satisfaction Actor: Parent death*Satisfaction Partner: Parent death*Satisfaction Actor: Negative health change*Satisfaction Partner: Negative health change*Satisfaction SE b 0.048 3.463 0.002 0.002 -0.018 0.027 -0.034 0.031 0.027 0.018 -0.031 0.018 t df 3221.515 72.530 0.000 3.370 1821.390 1.254 2229.994 2030.962 3688.785 1.534 3771.478 0.210 -0.671 0.502 -1.094 0.274 0.125 -1.736 0.083 UB 3.557 -0.001 0.006 -0.071 0.035 -0.094 0.027 -0.008 0.062 -0.067 0.004 -0.010 0.020 4618.651 -0.534 0.593 -0.049 0.028 -0.006 0.019 4381.789 -0.297 0.767 -0.043 0.032 0.074 0.019 4564.048 3.905 <.001 0.037 0.111 0.024 0.000 0.019 0.015 4361.685 1.268 4058.039 0.205 -0.015 0.988 -0.013 0.061 -0.030 0.030 -0.012 0.015 -0.050 0.021 0.021 0.021 0.025 0.039 0.026 0.023 0.003 0.001 -0.043 0.007 0.001 0.002 0.005 0.026 0.011 0.037 4014.717 3878.083 3775.235 1.012 3792.177 1.525 3836.397 0.893 2954.216 2.486 2459.088 2645.970 1.979 3973.841 5.405 5678.368 3.391 -0.783 0.434 -2.348 0.019 0.311 0.127 0.372 0.013 0.001 -0.056 0.048 0.000 <.001 0.016 <.001 0.016 -0.042 0.018 -0.008 -0.092 -0.020 0.062 -0.011 0.088 -0.027 0.073 0.005 -0.030 0.004 0.035 0.058 -6.403 <.001 -0.003 0.001 2598.173 -2.201 0.028 -0.005 0.000 -0.029 0.018 3867.474 -1.558 0.119 -0.065 0.007 -0.007 0.020 3552.392 -0.325 0.745 -0.047 0.033 -0.021 0.013 3938.970 -1.637 0.102 -0.046 0.004 -0.018 0.013 3737.284 -1.381 0.167 -0.043 0.007 -0.013 0.014 4582.375 -0.887 0.375 -0.041 0.015 -0.001 0.014 4454.308 -0.064 0.949 -0.028 0.026 152 0.000 0.005 0.003 0.002 0.003 0.022 Table 23 (cont’d) Actor: Positive health change*Satisfaction Partner: Negative health change*Satisfaction Actor: New chronic illness*Satisfaction Partner: New chronic illness*Satisfaction Actor: Retired*Satisfaction Partner: Retired*Satisfaction Actor: Unemployed*Satisfaction Partner: Unemployed*Satisfaction Slope*Actor: Birth of a child Slope*Actor: Death of a 0.003 child Slope*Actor: Parent death 0.001 Slope*Partner: Parent death 0.002 Slope*Actor: Negative health change Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: New chronic illness Slope*Actor: Retired Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed Slope*Actor: Parent death*Satisfaction Slope*Actor: Negative health change*Satisfaction Slope*Actor: New chronic illness*Satisfaction 0.000 0.004 0.001 0.001 0.001 0.001 0.014 4406.389 0.217 0.828 -0.024 0.029 0.013 4398.322 1.667 0.096 -0.004 0.049 0.012 3964.451 0.221 0.825 -0.020 0.025 0.009 0.011 -0.013 0.014 4156.589 0.783 3731.431 0.434 -0.952 0.341 -0.013 0.031 -0.040 0.014 0.014 3679.014 -0.002 0.999 -0.027 0.027 0.017 3959.494 0.294 0.769 -0.028 0.038 -0.022 0.017 3979.678 -1.302 0.193 -0.056 0.011 0.002 1362.119 1.052 0.293 -0.002 0.007 0.003 0.001 0.001 1270.156 1.354 1859.511 1.014 1854.577 1.539 0.176 0.311 0.124 -0.002 0.008 -0.001 0.004 -0.001 0.005 -0.005 0.002 2656.140 -2.299 0.022 -0.009 -0.001 0.002 2381.776 -0.207 0.836 -0.004 0.003 0.002 2655.336 1.876 0.061 0.000 0.008 0.002 2389.576 0.624 0.532 -0.003 0.005 -0.004 0.001 0.000 0.001 -0.002 0.002 -0.006 0.002 0.002 0.002 2115.297 -2.772 0.006 -0.007 -0.001 2061.326 1604.649 1602.610 1803.880 0.798 -0.186 0.852 -1.265 0.206 -3.327 <.001 0.425 -0.003 0.002 -0.005 0.001 -0.002 -0.009 -0.002 0.006 0.002 1878.397 0.382 0.702 -0.003 0.005 -0.002 0.001 1866.711 -1.901 0.058 -0.004 0.000 0.001 2244.767 1.253 0.210 -0.001 0.004 0.001 2107.928 0.882 0.378 -0.001 0.003 153 Table 23 (cont’d) Slope*Partner: Retired*Satisfaction Note. Significant effects bolded, p < .01 -0.001 0.001 1892.625 -1.058 0.290 -0.003 0.001 154 Table 24. Linear growth curve model examining the effect of actor and partner life events on the slope of extraversion; LISS sample. b 3.179 -0.005 0.002 2079.933 SE t 0.054 3808.969 58.846 0.000 3.073 -0.009 -2.463 0.014 LB df p UB 3.285 -0.001 Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic illness Actor: Retired Partner: Retired Actor: Unemployed Partner: Unemployed Age Gender Relationship length Education Slope*Actor: Birth of a child Slope*Actor: Death of a child Slope*Actor: Parent death Slope*Partner: Parent death Slope*Actor: Negative health change -0.020 0.031 2595.317 -0.630 0.529 -0.081 0.042 0.047 0.002 0.004 0.037 2383.884 1.262 0.023 4643.882 0.075 0.023 4761.499 0.180 0.207 0.940 0.857 -0.026 0.120 -0.043 0.046 -0.041 0.049 0.021 0.024 5650.881 0.876 0.381 -0.026 0.067 0.021 0.023 5457.487 0.900 0.368 -0.025 0.067 0.029 0.023 5609.752 1.244 0.213 -0.017 0.074 -0.023 0.023 5421.683 -1.021 0.307 -0.068 0.022 -0.054 0.019 4984.233 -2.797 0.005 -0.091 -0.016 0.026 4571.165 1.880 0.032 4842.305 2.880 -0.041 0.019 5010.780 -0.022 0.027 4655.037 0.049 0.091 -0.012 0.032 4806.884 -0.001 0.001 3428.050 0.010 0.000 0.031 0.008 3004.145 1.234 0.001 3107.120 0.131 0.006 4679.518 5.585 -2.109 0.035 -0.824 0.410 0.060 0.004 0.029 -0.381 0.703 -0.827 0.408 0.217 0.896 <.001 0.020 -0.078 -0.003 -0.074 0.030 -0.002 0.100 0.153 -0.074 0.050 -0.003 0.001 -0.006 0.027 -0.002 0.002 0.042 -0.001 0.002 1384.819 -0.474 0.636 -0.006 0.004 0.000 0.003 1259.738 0.165 0.869 -0.005 0.006 0.000 0.002 2101.219 0.157 0.875 -0.003 0.003 -0.001 0.002 2106.374 -0.508 0.611 -0.004 0.002 -0.007 0.002 3153.234 -3.110 0.002 -0.012 -0.003 155 0.002 0.007 Table 24 (cont’d) Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: New chronic illness Slope*Actor: Retired 0.001 Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed Note. Significant effects bolded, p < .05 0.001 0.003 0.002 2808.360 0.708 0.002 3141.444 3.222 0.479 -0.003 0.006 0.001 0.003 0.012 -0.001 0.002 2794.861 -0.002 0.002 2433.932 -0.609 0.543 -0.005 0.003 -1.334 0.182 -0.005 0.001 -0.002 0.002 2364.788 0.002 1773.934 0.431 -1.471 0.141 0.667 -0.005 0.001 -0.003 0.005 0.002 1777.704 1.338 0.002 2081.541 0.462 0.181 -0.001 0.006 0.644 -0.003 0.006 -0.001 0.002 2128.998 -0.570 0.569 -0.006 0.003 156 Table 25. Linear growth curve model examining the effect of actor/partner life events and relationship satisfaction on the slope of extraversion, LISS sample. Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic illness Actor: Retired Partner: Retired Actor: Unemployed Partner: Unemployed Age Gender Relationship length Education Relationship satisfaction Slope*Relationship satisfaction Actor: Birth of a child*Satisfaction Actor: Death of a child*Satisfaction Actor: Parent death*Satisfaction Partner: Parent death*Satisfaction Actor: Negative health change*Satisfaction df SE b 3.144 0.061 3096.119 -0.004 0.002 1813.488 -0.023 0.035 2283.667 0.051 0.039 2091.443 0.005 0.024 3857.075 0.011 0.024 3935.957 t 51.397 -1.815 -0.677 1.288 0.229 0.465 LB p 0.000 3.024 -0.008 0.070 -0.091 0.499 -0.027 0.198 -0.041 0.819 -0.036 0.642 UB 3.264 0.000 0.044 0.128 0.052 0.059 0.028 0.026 4692.137 1.078 0.281 -0.023 0.078 0.025 0.026 4459.758 0.971 0.331 -0.025 0.075 0.034 0.025 4651.102 1.374 0.170 -0.015 0.084 -0.023 0.025 4442.496 -0.916 0.360 -0.071 0.026 -0.065 0.021 4143.733 -3.136 0.002 -0.105 -0.024 -0.048 0.021 4136.321 -0.031 0.029 3817.173 0.042 0.028 3770.308 0.084 0.034 3959.336 0.008 0.035 3991.667 -0.001 0.001 2852.792 0.008 0.009 2490.161 0.001 0.001 2623.958 0.034 0.006 3803.163 -2.315 -1.090 1.494 2.472 0.243 -0.613 0.794 0.688 5.455 -0.089 0.021 -0.088 0.276 0.135 -0.013 0.013 0.017 -0.060 0.808 -0.004 0.540 0.427 -0.011 -0.001 0.491 <.001 0.022 -0.007 0.025 0.098 0.151 0.076 0.002 0.026 0.003 0.046 0.054 0.013 5264.143 3.996 <.001 0.027 0.080 0.000 0.001 2530.882 -0.162 0.872 -0.003 0.002 -0.059 0.024 3787.942 -2.414 0.016 -0.107 -0.011 0.041 0.027 3443.422 1.529 0.126 -0.012 0.093 0.009 0.017 4110.483 0.524 0.600 -0.025 0.043 -0.004 0.017 4211.064 -0.238 0.812 -0.038 0.030 -0.009 0.020 4455.358 -0.476 0.634 -0.048 0.029 157 Table 25 (cont’d) Partner: Negative health change*Satisfaction Actor: Positive health change*Satisfaction Partner: Negative health change*Satisfaction Actor: New chronic illness*Satisfaction Partner: New chronic illness*Satisfaction Actor: Retired*Satisfaction Partner: Retired*Satisfaction Actor: Unemployed*Satisfactio n Partner: Unemployed*Satisfactio n Slope*Actor: Birth of a child Slope*Actor: Death of a child Slope*Actor: Parent death Slope*Partner: Parent death Slope*Actor: Negative health change Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: New chronic illness Slope*Actor: Retired Slope*Partner: Retired Slope*Actor: Unemployed -0.015 0.018 4430.603 -0.817 0.414 -0.051 0.021 -0.005 0.019 4560.990 -0.246 0.805 -0.042 0.033 0.011 0.018 4361.842 0.617 0.537 -0.024 0.046 0.006 0.015 4246.361 0.393 0.694 -0.024 0.035 -0.021 0.015 4311.244 -1.376 0.169 -0.051 0.009 -0.021 0.019 3971.374 -1.138 0.255 -0.058 0.015 0.002 0.018 3929.787 0.086 0.931 -0.035 0.038 -0.016 0.023 4177.244 -0.677 0.498 -0.061 0.030 0.039 0.023 4105.729 1.720 0.086 -0.006 0.084 -0.002 0.003 1322.584 -0.943 0.346 -0.008 0.003 0.002 0.003 1221.549 0.689 0.491 -0.004 0.008 0.000 0.002 1937.842 0.199 0.843 -0.003 0.004 -0.001 0.002 1941.021 -0.781 0.435 -0.005 0.002 -0.008 0.002 2763.285 -3.251 0.001 -0.013 -0.003 0.002 0.002 2493.511 0.927 0.354 -0.002 0.007 0.007 0.002 2763.788 3.183 0.001 0.003 0.012 -0.002 0.002 2498.677 -0.909 0.363 -0.006 0.002 -0.002 0.002 2201.902 -1.452 0.147 -0.006 0.001 -0.002 0.002 2151.009 0.001 0.002 1622.432 0.002 0.002 1622.165 -1.306 0.608 1.105 0.192 0.543 0.269 -0.005 -0.003 -0.002 0.001 0.005 0.006 -0.001 0.002 1901.960 -0.590 0.555 -0.006 0.003 158 Table 25 (cont’d) Slope*Partner: Unemployed Slope*Actor: Negative health change*Satisfaction Slope*Actor: Positive health change*Satisfaction Note. Significant effects bolded, p < .01 -0.002 0.002 1947.214 0.000 0.002 2449.542 -0.002 0.002 2683.945 -0.607 0.544 -0.006 0.003 0.178 0.859 -0.003 0.004 -1.295 0.195 -0.006 0.001 159 Table 26. Linear growth curve model examining the effect of actor and partner life events on the slope of neuroticism; LISS sample. Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic illness Actor: Retired Partner: Retired Actor: Unemployed Partner: Unemployed Age Gender Relationship length Education Slope*Actor: Birth of a child Slope*Actor: Death of a child Slope*Actor: Parent death Slope*Partner: Parent death Slope*Actor: Negative health change df b 2.736 -0.018 0.003 2049.960 SE 0.057 4073.509 47.790 -7.155 t p 0.000 <.001 LB 2.624 -0.023 UB 2.849 -0.013 0.053 0.034 2512.962 1.577 0.115 -0.013 0.119 -0.018 0.040 2279.923 -0.047 0.022 4260.109 -0.447 -2.114 0.655 0.035 -0.096 -0.090 0.060 -0.003 0.008 0.022 4375.267 0.352 0.725 -0.036 0.052 0.018 0.024 5371.929 0.753 0.451 -0.029 0.065 0.022 0.024 5143.879 0.926 0.354 -0.024 0.068 -0.075 0.023 5321.090 -3.200 0.001 -0.121 -0.029 -0.028 0.023 5113.276 -1.216 0.224 -0.073 0.017 0.109 0.019 4716.773 5.728 <.001 0.071 0.146 0.027 -0.013 0.026 4718.548 -0.053 0.025 4507.330 0.011 0.019 4691.511 1.401 -0.497 -2.098 0.031 4400.130 0.340 0.161 0.619 0.036 0.734 0.016 -0.001 0.001 3645.190 -0.122 0.008 2978.163 -0.002 0.001 3175.061 -0.050 0.006 5026.188 0.613 0.031 4453.819 0.506 -0.525 0.600 -15.725 <.001 0.020 -2.322 <.001 -8.773 -0.011 -0.064 -0.102 -0.050 -0.045 -0.003 -0.137 -0.005 -0.061 0.064 0.038 -0.003 0.072 0.077 0.002 -0.107 0.000 -0.039 0.000 0.003 1368.962 -0.016 0.987 -0.006 0.006 0.001 0.004 1241.882 0.179 0.858 -0.006 0.008 0.000 0.002 2036.831 -0.135 0.893 -0.004 0.004 -0.003 0.002 2047.719 -1.650 0.099 -0.007 0.001 0.011 0.003 3069.820 3.840 <.001 0.005 0.016 160 0.003 2737.322 1.054 0.292 -0.002 0.008 0.003 Table 26 (cont’d) Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: Change in chronic condition Slope*Actor: Retired Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed Note. Significant effects bolded, p < .05 0.006 0.003 0.000 -0.006 0.003 3054.962 -0.001 0.003 2724.304 0.002 2369.718 3.218 -2.201 0.028 -0.011 -0.001 -0.402 0.687 -0.006 0.004 0.002 2303.172 1.767 0.001 0.002 0.010 0.077 0.000 0.007 -0.001 0.002 1734.073 -0.001 0.002 1736.374 -0.008 0.003 2018.885 0.003 2080.149 -0.445 0.656 -0.005 0.003 -0.352 0.725 -0.005 0.004 -2.760 0.006 -0.013 -0.002 -0.145 0.885 -0.006 0.005 161 Table 27. Linear growth curve model examining the effect of actor/partner life events and relationship satisfaction on the slope of neuroticism; LISS sample. Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic illness Actor: Retired Partner: Retired Actor: Unemployed Partner: Unemployed Age Gender Relationship length Education Relationship satisfaction Slope*Relationship satisfaction Actor: Birth of a child*Satisfaction Actor: Death of a child*Satisfaction Actor: Parent death*Satisfaction Partner: Parent death*Satisfaction Actor: Negative health change*Satisfaction t p df LB b 2.732 -0.018 0.003 1780.715 0.056 -0.020 0.041 2020.115 -0.051 0.023 3605.232 -0.007 0.023 3686.564 SE 0.063 3285.402 43.394 0.000 2.608 -0.023 -0.014 -0.100 -0.096 -0.053 -6.561 <.001 0.117 -0.487 0.626 -2.204 0.028 -0.287 0.774 0.036 2228.160 1.568 UB 2.855 -0.012 0.127 0.060 -0.006 0.039 0.016 0.026 4552.574 0.609 0.543 -0.035 0.066 0.028 0.025 4311.526 1.112 0.266 -0.021 0.077 -0.080 0.025 4494.304 -3.238 0.001 -0.129 -0.032 -0.034 0.025 4291.975 -1.395 0.163 -0.082 0.014 0.106 0.020 3995.850 5.251 <.001 0.066 0.145 0.020 3936.688 1.025 0.021 0.000 0.028 3866.397 -0.057 0.027 3706.492 -0.002 0.033 3687.168 -0.018 0.034 3763.736 0.001 0.306 -0.008 0.994 -2.100 0.036 -0.072 0.943 -0.545 0.586 0.653 0.001 2997.901 0.449 -0.114 0.009 2424.321 -0.004 0.001 2669.939 -0.057 0.006 4000.060 -0.082 0.014 5825.202 - 13.163 <.001 -2.974 0.003 -9.090 <.001 -5.689 <.001 -0.019 -0.055 -0.109 -0.067 -0.084 -0.002 -0.131 -0.006 -0.069 -0.111 0.060 0.054 -0.004 0.062 0.047 0.003 -0.097 -0.001 -0.044 -0.054 0.004 0.002 2608.468 2.633 0.009 0.001 0.007 0.076 0.024 3884.317 3.144 0.002 0.029 0.123 -0.015 0.027 3596.730 -0.549 0.583 -0.067 0.038 -0.005 0.017 3800.856 -0.313 0.754 -0.038 0.027 -0.005 0.017 3908.126 -0.288 0.773 -0.037 0.028 0.013 0.019 4237.041 0.682 0.495 -0.025 0.051 162 0.020 0.002 0.008 0.010 Table 27 (cont’d) Partner: Negative health change*Satisfaction Actor: Positive health change*Satisfaction Partner: Negative health change*Satisfaction Actor: New chronic illness*Satisfaction Partner: New chronic illness*Satisfaction Actor: Retired*Satisfaction Partner: Retired*Satisfaction Actor: Unemployed*Satisfaction 0.013 Partner: Unemployed*Satisfaction 0.013 Slope*Actor: Birth of a child Slope*Actor: Death of a child Slope*Actor: Parent death Slope*Partner: Parent death Slope*Actor: Negative health change Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: New chronic illness Slope*Actor: Retired Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed 0.009 0.007 0.004 0.000 0.018 4417.026 0.585 0.558 -0.025 0.045 0.019 4359.758 0.414 0.679 -0.029 0.045 -0.022 0.017 4367.819 -0.014 0.015 3896.637 -0.006 0.015 4054.097 -1.250 0.211 -0.056 0.012 -0.917 0.359 -0.043 0.016 -0.422 0.673 -0.035 0.022 0.018 3597.470 1.136 0.256 -0.015 0.055 0.018 3613.326 0.107 0.915 -0.033 0.036 0.022 3512.444 0.582 0.561 -0.031 0.057 0.022 3923.322 0.569 0.570 -0.031 0.056 -0.001 0.003 1307.434 -0.001 0.004 1207.919 -0.002 0.002 1887.347 -0.003 0.002 1892.530 -0.362 0.718 -0.008 0.005 -0.359 0.720 -0.794 0.427 -0.008 -0.006 0.006 0.002 -1.351 0.177 -0.007 0.001 0.003 2703.980 3.056 0.002 0.003 0.015 0.003 2439.664 1.586 0.113 -0.001 0.010 -0.005 0.003 2701.509 -0.002 0.003 2443.270 -1.658 0.097 -0.010 0.001 -0.802 0.422 -0.007 0.003 0.002 2159.781 3.487 <.001 0.003 0.011 0.002 2104.947 1.568 0.003 -0.002 0.002 1636.655 -0.001 0.002 1636.205 0.117 -0.696 0.487 -0.347 0.729 -0.001 -0.006 -0.005 0.007 0.003 0.004 -0.007 0.003 1850.405 -2.274 0.023 -0.012 -0.001 0.003 1910.473 -0.064 0.949 -0.006 0.006 163 Table 27 (cont’d) Slope*Actor: Negative health change*Satisfaction Slope*Actor: Positive health change*Satisfaction Slope*Actor: New chronic illness*Satisfaction Slope*Actor: Unemployed*Satisfaction Note. Significant effects bolded, p < .01 0.001 -0.002 0.002 2381.684 -0.769 0.442 -0.006 0.003 0.002 2645.092 0.557 0.578 -0.003 0.005 -0.002 0.001 2187.186 -0.002 0.002 1896.226 -1.111 0.267 -0.004 0.001 -0.779 0.436 -0.006 0.002 164 Table 28. Linear growth curve model examining the effect of actor and partner life events on the slope of openness; LISS sample. SE df b 0.040 4005.758 3.113 -0.005 0.002 1886.688 t 77.487 0.000 3.035 -0.009 -3.175 0.002 UB 3.192 -0.002 LB p 0.049 0.000 0.008 0.013 4844.699 0.017 5303.164 0.016 5472.820 -0.034 0.016 5265.728 -0.014 0.023 2543.819 -0.023 0.017 5520.066 -0.028 0.028 2323.851 0.028 0.016 4415.444 -0.023 0.016 4541.299 Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic 0.012 0.013 4850.982 illness -0.028 0.018 4736.763 Actor: Retired 0.018 4619.259 0.045 Partner: Retired 0.022 4602.224 Actor: Unemployed 0.081 0.022 4586.476 Partner: Unemployed 0.021 0.001 3608.985 0.001 Age 0.034 Gender 0.006 2989.965 -0.004 0.001 3161.364 Relationship length 0.004 5014.269 0.103 Education Slope*Actor: Birth of a child Slope*Actor: Death of a child Slope*Actor: Parent death Slope*Partner: Parent death Slope*Actor: Negative health change -0.001 0.002 1199.670 -0.005 0.002 2815.633 0.001 1869.190 0.001 1863.408 0.002 1308.376 0.006 0.002 0.002 165 -0.604 0.546 -0.060 0.032 -1.000 0.317 1.778 0.075 -1.473 0.141 -0.082 -0.003 -0.054 0.027 0.059 0.008 -1.362 0.173 -0.056 0.010 0.480 0.631 -0.024 0.040 2.984 0.003 0.017 0.081 -2.104 0.035 -0.066 -0.002 -0.012 0.991 -0.026 0.026 -0.014 0.887 0.375 -0.064 -1.526 0.127 0.012 0.010 2.503 <.001 0.038 3.678 -0.023 0.352 0.931 -0.001 0.475 0.714 <.001 0.023 6.173 -5.055 <.001 -0.005 25.364 <.001 0.095 0.038 0.008 0.080 0.124 0.064 0.002 0.045 -0.002 0.110 2.947 0.003 0.002 0.010 -0.253 0.800 -0.005 0.004 1.446 0.148 -0.001 0.004 1.277 0.202 -0.001 0.004 -2.520 0.012 -0.008 -0.001 -0.001 0.002 2462.433 0.002 2786.719 0.002 2463.752 0.000 0.007 Table 28 (cont’d) Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: Change in chronic condition Slope*Actor: Retired Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed Note. Significant effects bolded, p < .05 0.000 0.002 -0.001 0.001 2094.888 -0.001 0.001 1617.399 0.001 1619.160 0.002 1792.421 -0.001 0.002 1911.602 -0.002 0.001 2164.462 -0.869 0.385 -0.005 0.002 3.856 <.001 0.003 0.010 0.162 0.871 -0.003 0.003 -1.811 0.070 -0.005 0.000 -0.427 0.669 -0.844 0.399 -0.003 -0.004 0.002 0.002 -0.116 0.908 -0.003 0.003 0.990 0.322 -0.002 0.005 -0.278 0.781 -0.004 0.003 166 Table 29. Linear growth curve model examining the effect of actor/partner life events and relationship satisfaction on the slope of openness; LISS sample. Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic illness Actor: Retired Partner: Retired Actor: Unemployed Partner: Unemployed Age Gender Relationship length Education Relationship satisfaction Slope*Relationship satisfaction Actor: Birth of a child*Satisfaction Actor: Death of a child*Satisfaction Actor: Parent death*Satisfaction Partner: Parent death*Satisfaction Actor: Negative health change*Satisfaction SE df b 0.045 3256.502 3.099 -0.004 0.002 1679.448 -0.021 0.026 2252.004 -0.027 0.029 2059.032 0.021 0.016 3728.448 -0.018 0.017 3811.911 t 68.538 -2.152 -0.821 -0.920 1.258 -1.044 LB p 0.000 3.010 -0.007 0.032 -0.071 0.412 -0.084 0.357 -0.012 0.208 -0.050 0.297 UB 3.188 0.000 0.029 0.030 0.053 0.015 -0.020 0.018 4658.831 -1.103 0.270 -0.056 0.016 0.012 0.018 4415.604 0.672 0.502 -0.023 0.047 0.058 0.018 4600.412 3.281 0.001 0.023 0.092 -0.029 0.017 4389.352 -1.637 0.102 -0.063 0.006 -0.008 0.014 4095.126 -0.549 0.583 -0.036 0.020 0.014 4061.917 0.019 -0.033 0.020 3923.643 0.019 3844.218 0.038 0.024 3833.666 0.093 0.024 3873.411 0.025 0.001 3007.915 0.000 0.030 0.006 2458.331 -0.003 0.001 2677.443 0.004 4051.515 0.104 0.010 5612.861 0.011 1.331 -1.651 1.950 3.938 1.057 0.470 4.813 -4.076 23.148 1.134 -0.009 0.183 0.099 -0.072 0.051 0.000 <.001 0.047 -0.022 0.291 0.638 -0.002 <.001 0.018 -0.005 <.001 <.001 0.095 -0.008 0.257 0.047 0.006 0.076 0.139 0.072 0.003 0.042 -0.002 0.113 0.031 -0.002 0.001 2163.155 -1.709 0.088 -0.003 0.000 -0.035 0.018 3724.229 -1.992 0.046 -0.070 -0.001 -0.004 0.019 3611.940 -0.205 0.838 -0.042 0.034 0.020 0.012 3882.162 1.645 0.100 -0.004 0.043 -0.009 0.012 3997.747 -0.781 0.435 -0.033 0.014 0.002 0.014 4327.769 0.116 0.908 -0.025 0.028 167 0.024 0.007 0.000 Table 29 (cont’d) Partner: Negative health change*Satisfaction Actor: Positive health change*Satisfaction Partner: Negative health change*Satisfaction Actor: New chronic illness*Satisfaction Partner: New chronic illness*Satisfaction Actor: Retired*Satisfaction Partner: Retired*Satisfaction Actor: Unemployed*Satisfaction Partner: Unemployed*Satisfaction Slope*Actor: Birth of a child Slope*Actor: Death of a 0.000 child Slope*Actor: Parent death 0.001 Slope*Partner: Parent death Slope*Actor: Negative health change Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: New chronic illness Slope*Actor: Retired Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed 0.007 0.000 0.000 0.005 0.002 -0.006 0.013 4416.971 -0.009 0.013 4444.084 -0.441 0.659 -0.031 0.019 -0.635 0.526 -0.035 0.018 0.013 4352.756 1.946 0.052 0.000 0.049 0.010 4127.356 0.649 0.516 -0.014 0.027 -0.008 0.011 4124.210 -0.787 0.431 -0.029 0.012 0.013 3676.096 -0.014 0.989 -0.025 0.025 -0.014 0.013 3668.848 -0.025 0.016 3930.707 -0.001 0.016 3914.626 -1.114 0.265 -0.039 0.011 -1.548 0.122 -0.056 0.007 -0.045 0.964 -0.032 0.030 0.002 1275.814 2.400 0.017 0.001 0.009 0.002 1182.235 0.001 1727.049 0.132 1.059 0.895 0.290 -0.004 -0.001 0.005 0.004 0.001 1718.241 1.627 0.104 0.000 0.005 -0.005 0.002 2453.847 -0.001 0.002 2186.021 -2.669 0.008 -0.009 -0.001 -0.767 0.443 -0.005 0.002 0.002 2446.428 3.699 <.001 0.003 0.010 0.002 2200.286 -0.051 0.959 -0.003 0.003 -0.003 0.001 1961.702 0.000 0.001 1904.658 -0.001 0.001 1501.616 0.001 1497.593 0.000 -2.394 0.017 -0.005 -0.001 -0.336 -0.868 -0.296 0.737 0.386 0.767 -0.003 -0.004 -0.003 0.002 0.002 0.002 0.002 1649.174 0.192 0.848 -0.003 0.004 -0.002 0.002 1754.192 -0.966 0.334 -0.006 0.002 168 0.001 Table 29 (cont’d) Slope*Actor: Birth of a child* Satisfaction Slope*Actor: Negative health change* Satisfaction Slope*Actor: Positive health change* Satisfaction Note. Significant effects bolded, p < .01 0.001 0.000 0.001 1887.467 0.001 2334.616 0.001 2488.003 0.682 0.495 -0.002 0.004 0.944 0.345 -0.001 0.004 -0.284 0.776 -0.003 0.002 169 Table 30. Linear growth curve model examining the effect of actor and partner life events on the slope of relationship satisfaction; LISS sample. b 8.027 -0.051 SE 0.119 4796.371 67.730 0.000 -7.620 <.001 0.007 1964.540 LB 7.794 -0.064 UB 8.259 -0.038 df p t Intercept Slope Actor: Birth of a child Actor: Death of a child Actor: Parent death Partner: Parent death Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Negative health change Actor: New chronic illness Partner: New chronic illness Actor: Retired Partner: Retired Actor: Unemployed Partner: Unemployed Age Gender Relationship length Education Slope*Actor: Birth of a child Slope*Actor: Death of a child Slope*Actor: Parent death Slope*Partner: Parent death Slope*Actor: Negative health change 0.017 0.075 2276.167 0.227 0.820 -0.130 0.164 -0.087 -0.062 -0.041 0.089 2059.665 0.042 3132.272 0.043 3234.499 -0.976 0.329 -1.487 0.137 -0.969 0.333 -0.261 -0.145 -0.125 0.088 0.020 0.042 0.027 0.046 4060.381 0.578 0.563 -0.064 0.117 -0.058 0.046 3926.137 -1.265 0.206 -0.148 0.032 0.075 0.045 4024.930 1.669 0.095 -0.013 0.163 0.041 0.045 3900.816 0.922 0.357 -0.046 0.129 -0.112 0.036 3611.684 -3.113 0.002 -0.183 -0.042 -0.012 0.200 0.076 -0.101 -0.152 0.007 0.046 0.000 -0.037 0.036 3605.015 0.047 4575.961 4.286 0.045 4385.006 1.683 0.060 3254.435 -0.338 0.736 <.001 0.092 -1.686 0.092 0.060 3297.874 0.003 4489.410 2.853 0.011 2904.083 4.114 0.002 3383.404 0.181 0.010 5048.260 -2.538 0.011 0.004 <.001 0.857 -3.523 <.001 -0.083 0.108 -0.012 -0.218 -0.269 0.002 0.024 -0.004 -0.057 0.059 0.291 0.164 0.016 -0.035 0.013 0.068 0.005 -0.016 -0.023 0.008 1316.489 -2.747 0.006 -0.040 -0.007 -0.014 0.010 1213.857 -1.401 0.161 -0.034 0.006 0.013 0.005 1894.268 2.694 0.007 0.004 0.023 0.015 0.005 1941.521 3.012 0.003 0.005 0.025 -0.001 0.007 2912.684 -0.195 0.845 -0.014 0.012 170 0.012 0.008 -0.009 Table 30 (cont’d) Slope*Partner: Negative health change Slope*Actor: Positive health change Slope*Partner: Negative health change Slope*Actor: New chronic illness Slope*Partner: New chronic illness 0.007 Slope*Actor: Retired 0.007 Slope*Partner: Retired Slope*Actor: Unemployed Slope*Partner: Unemployed Note. Significant effects bolded, p < .05 0.001 0.013 0.012 0.000 0.006 2622.982 -1.456 0.145 -0.022 0.003 0.006 2866.760 1.824 0.006 2598.446 1.340 0.068 -0.001 0.024 0.180 -0.004 0.020 0.005 2273.729 -0.101 0.920 -0.009 0.009 0.005 2212.020 1.495 0.005 1931.753 1.480 0.005 1930.765 2.400 0.007 1925.162 1.897 0.007 1985.900 0.082 0.135 0.139 -0.002 -0.002 0.016 0.017 0.016 0.002 0.022 0.058 0.000 0.027 0.935 -0.013 0.015 171 Table 31. A summary of which life events produced a (mal)adaptive response, organized by trait; LISS sample. Trait Agreeableness Actor/Partner: Death of a child Actor/Partner: Birth of a child Maladaptive Adaptive Conscientiousness Partner: Parent dying Extraversion Neuroticism Openness Actor: Positive health change Actor: Positive health change Actor: Unemployment Actor/Partner: Birth of a child Actor: Positive health change Actor: New chronic illness Actor: Negative health change Partner: Retirement Actor: Negative health change Actor: New chronic illness Actor: Negative health change Actor: Negative health change Note. Type of response (i.e., adaptive or maladaptive) based on Event*Slope interactions. "Adaptive" responses include steeper increases in all traits with the exception of neuroticism. "Maladaptive" responses include steeper decreases in all traits with the exception of neuroticism. 172 APPENDIX C: TABLES FOR CHAPTER 4 Table 32. Frequency of life events; CouPers sample. Life event Birth of child Birth of a grandchild Graduation Retirement Unemployment Change in job Moving Becoming an empty nester Negative health event Death of a close other Miscarriage Abortion Other Note. This table reflects only the prevalence of life events that participants found either negative or distressing. Frequency (% of sample) 33 (3.4%) 0 (0%) 146 (15.1%) 2 (.2%) 8 (.8%) 165 (17.1%) 161 (16.7%) 5 (.5%) 143 (14.8%) 234 (24.3%) 25 (2.6%) 5 (.5%) 51 (5.3%) 173 Note. **Correlation is significant at the 0.01 level (2-tailed), *Correlation is significant at the 0.05 level (2-tailed). A indicates actor life event, P indicates partner life event. 174 1234567891011121314151. A: Birth of a child12. P: Birth of a child.379**13. A: Graduation-0.036-0.05114. P: Graduation-0.051-0.036.237**15. A: Change in job-0.017-0.003.107**0.00816. P: Change in job-0.003-0.0170.008.107**0.02717. A: Moving.070*0.028.182**.084**.166**.073*18. P: Moving0.028.070*.084**.184**.073*.167**.324**19. A: Negative health event0.0390.054-0.055-.063*.065*0.0220.0470.048110. P: Negative health event0.0540.039-.063*-0.0550.022.064*0.0480.0480.027111. A: Death of a close other-0.042-0.0550.0220.0400.0240.035-0.014-0.013.098**.085**112. P: Death of a close other-0.055-0.0420.0400.0210.0350.023-0.013-0.013.085**.097**.333**113. A: Miscarriage.106**.072*-0.052-0.0520.0390.039-0.0250.008.086**.069*.070*.070*114. P: Miscarriage.072*.106**-0.052-0.0520.0390.0390.008-0.025.069*.086**.070*.070*.684**115. A: Other-0.023-0.0460.0590.012.097**-0.0140.0550.033-0.0110.047-0.018-0.0180.0410.014116. P: Other-0.046-0.0230.0120.059-0.014.097**0.0330.0560.047-0.011-0.018-0.0180.0140.041.123**Table 33. Biserial correlations between life events within couples; CouPers sample. Note.** Correlation is significant at the 0.01 level (2-tailed); * Correlation is significant at the 0.05 level (2-tailed). A indicates actor event, P indicates partner event. Table 33 (cont’d) 175 1234567891011121314151. A: Birth of a child12. P: Birth of a child.379**13. A: Graduation-0.036-0.05114. P: Graduation-0.051-0.036.237**15. A: Change in job-0.017-0.003.107**0.00816. P: Change in job-0.003-0.0170.008.107**0.02717. A: Moving.070*0.028.182**.084**.166**.073*18. P: Moving0.028.070*.084**.184**.073*.167**.324**19. A: Negative health event0.0390.054-0.055-.063*.065*0.0220.0470.048110. P: Negative health event0.0540.039-.063*-0.0550.022.064*0.0480.0480.027111. A: Death of a close other-0.042-0.0550.0220.0400.0240.035-0.014-0.013.098**.085**112. P: Death of a close other-0.055-0.0420.0400.0210.0350.023-0.013-0.013.085**.097**.333**113. A: Miscarriage.106**.072*-0.052-0.0520.0390.039-0.0250.008.086**.069*.070*.070*114. P: Miscarriage.072*.106**-0.052-0.0520.0390.0390.008-0.025.069*.086**.070*.070*.684**115. A: Other-0.023-0.0460.0590.012.097**-0.0140.0550.033-0.0110.047-0.018-0.0180.0410.014116. P: Other-0.046-0.0230.0120.059-0.014.097**0.0330.0560.047-0.011-0.018-0.0180.0140.041.123**Table 33. Biserial correlations between life events within couples; CouPers sample. Note.** Correlation is significant at the 0.01 level (2-tailed); * Correlation is significant at the 0.05 level (2-tailed). A indicates actor event, P indicates partner event. Table 34. Descriptives of and correlations between self-reported traits and relationship variables; averaged across waves; CouPers sample. E C O N Support Mean A 3.129 C 3.438 E 3.479 N 2.797 O 3.564 0.360 Support Responsiveness 3.953 Closeness 6.300 Note. * Correlation is significant at the 0.05 level (2-tailed);** Correlation is significant at the 0.01 level (2-tailed). Traits abbreviated with their first letter. A 1 .227** .109** .199** .130** 0.013 -.047** .075** SD 0.331 0.319 0.774 0.790 0.451 0.408 0.650 3.171 1 -.034* .138** .136** -0.024 .068** .065** 1 -.227** .258** .117** .138** 0.002 1 .132** 1 .142** -0.001 1 -0.009 .052** -.173** 0.006 1 .171** .225** .107** Responsiveness 176 Table 35. Descriptives of and correlations between partner-reported traits and relationship variables; averaged across waves; CouPers sample. E C O N Support Mean 3.056 A 3.884 C 3.679 E 2.781 N 3.840 O Support 0.360 Responsiveness 3.953 6.300 Closeness Note. * Correlation is significant at the 0.05 level (2-tailed);** Correlation is significant at the 0.01 level (2-tailed). Traits abbreviated with their first letter. A 1 -.053** -.106** -0.023 -.046* 0.019 -.041* -0.014 1 .200** -.053** .245** .072** .210** -0.007 SD 0.521 0.839 1.065 1.009 0.850 0.408 0.650 3.171 1 -.173** .325** -0.004 .182** 0.024 1 .071** -.129** -.207** .040* 1 .047** .223** 0.006 1 .225** .107** 1 .171** Responsiveness 177 Table 36. Correlations between partner-reported and self-reported personality, averaged across waves; CouPers sample. C E N O PR A PR C PR E PR N A 1 .227** .109** .199** .130** .146** -.072** -.087** .083** -.040* A C E N O PR A PR C PR E PR N PR O Note. ** Correlation is significant at the 0.01 level. * Correlation is significant at the 0.05 level. Traits indicated by their first initial. PR indicates a partner-reported trait. 1 -0.009 .038* .093** -.251** 0.009 -.132** 0.002 -0.032 -.192** .121** -.213** 1 -.227** 1 .258** 0.032 .109** -.079** -.119** 0.013 1 -.034* .138** .136** .047** 0.001 -.081** .096** -0.027 1 -.053** -.106** -0.023 -.046* 1 -.173** 1 .325** 1 .200** -.053** .245** .071** 178 Table 37. Linear growth curve model examining the effect of actor and partner life events on the slope of agreeableness; CouPers sample. Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child SE b 3.225 0.081 -0.010 0.004 df 191.808 134.195 p t 39.704 <.001 -2.412 0.017 LB 3.065 -0.019 UB 3.385 -0.002 -0.215 0.135 175.190 -1.587 0.114 -0.481 0.052 0.130 0.135 176.743 0.956 0.340 -0.138 0.397 -0.042 0.051 251.050 -0.818 0.414 -0.142 0.059 -0.064 0.051 248.736 -1.265 0.207 -0.165 0.036 0.032 0.044 244.272 0.732 0.465 -0.054 0.118 -0.019 0.043 0.142 0.055 -0.043 0.056 240.959 226.767 221.726 -0.454 0.650 0.011 2.565 -0.779 0.437 -0.104 0.033 -0.153 0.065 0.251 0.066 -0.039 0.049 232.125 -0.804 0.422 -0.135 0.057 0.064 0.048 227.715 1.318 0.189 -0.031 0.159 -0.007 0.042 242.274 -0.160 0.873 -0.089 0.076 -0.038 0.042 238.637 -0.900 0.369 -0.120 0.045 -0.034 0.141 221.200 -0.245 0.807 -0.311 0.243 -0.192 0.138 -0.040 0.066 -0.006 0.069 -0.026 0.015 0.000 0.002 -0.005 0.007 -0.005 0.008 211.564 248.272 249.188 141.962 150.164 250.958 222.351 -1.394 0.165 -0.603 0.547 -0.087 0.931 -1.752 0.082 -0.163 0.871 -0.648 0.517 -0.630 0.530 -0.463 -0.169 -0.141 -0.056 -0.004 -0.019 -0.020 0.079 0.090 0.129 0.003 0.003 0.010 0.011 -0.005 0.002 140.979 -2.519 0.013 -0.009 -0.001 0.006 0.026 171.622 0.232 0.817 -0.045 0.057 0.003 0.026 171.991 0.117 0.907 -0.048 0.054 179 0.005 0.010 -0.010 0.008 0.004 0.009 -0.005 0.007 0.002 0.009 0.000 0.010 0.006 0.009 Table 37 (cont’d) Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 -0.022 0.012 0.020 0.025 0.010 0.009 0.036 0.025 -0.005 0.012 -0.015 0.008 0.003 0.008 257.803 257.656 259.376 259.177 216.832 219.041 240.244 244.815 236.994 235.123 225.859 223.796 252.221 255.353 0.204 0.839 -0.015 0.019 0.488 0.626 -0.013 0.022 -1.310 0.191 -0.025 0.005 -0.665 0.507 -0.020 0.010 0.488 0.626 -0.015 0.025 -0.031 0.975 -0.021 0.020 0.760 0.448 -0.010 0.023 1.210 0.227 -0.007 0.027 0.363 0.717 -0.012 0.018 -1.970 0.050 -0.030 0.000 0.784 0.434 -0.030 0.069 1.469 0.143 -0.012 0.085 -1.861 0.064 -0.045 0.001 -0.405 0.686 -0.029 0.019 180 Table 38. Linear growth curve model examining the effect of actor/partner life events and support on the slope of agreeableness; CouPers sample. df b 3.238 -0.010 0.005 169.412 SE 0.083 198.682 39.069 -2.067 t LB p <.001 3.074 -0.019 0.040 UB 3.401 0.000 -0.160 0.146 206.624 -1.100 0.273 -0.447 0.127 0.102 -0.044 0.053 279.171 -0.074 0.055 307.252 0.143 190.780 0.714 -0.833 -1.347 0.045 262.022 0.557 -0.010 0.044 260.530 0.107 -0.008 0.060 269.635 -0.219 0.059 252.323 1.814 -0.128 0.476 0.405 0.179 0.578 -0.180 -0.149 -0.182 -0.063 0.827 0.071 0.898 -0.097 -0.009 -0.127 0.385 0.060 0.034 0.113 0.077 0.223 0.111 -0.061 0.050 246.226 -1.219 0.224 -0.160 0.038 0.050 241.407 0.925 0.356 -0.052 0.144 -0.005 0.043 253.346 -0.111 0.912 -0.090 0.081 -0.025 0.043 249.056 -0.140 0.161 331.033 -0.239 0.206 595.132 -0.062 0.067 250.876 0.016 -0.025 0.016 147.883 0.002 150.203 0.000 -0.005 0.008 252.604 -0.007 0.008 227.665 -0.005 0.002 136.439 0.004 -0.580 -0.871 -1.162 -0.924 0.070 251.777 0.228 -1.608 -0.132 -0.625 -0.828 -2.648 0.034 862.873 0.124 0.562 0.384 0.246 0.356 0.820 0.110 0.895 0.532 0.408 0.009 0.901 -0.110 -0.457 -0.644 -0.193 -0.123 -0.056 -0.004 -0.020 -0.022 -0.009 -0.062 0.060 0.176 0.165 0.070 0.155 0.006 0.003 0.010 0.009 -0.001 0.071 0.028 214.797 0.697 0.487 -0.036 0.075 -0.004 0.027 189.489 -0.141 0.888 -0.057 0.050 0.046 Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job 0.025 Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Support Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation 0.001 0.004 0.019 0.009 268.592 0.382 0.702 -0.015 0.022 0.009 270.991 0.065 0.948 -0.018 0.019 181 0.005 0.008 0.003 0.003 0.005 0.006 Table 38 (cont’d) Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Actor: Birth of a child*Support Partner: Birth of a child*Support Actor: Graduation*Support 0.032 Partner: Graduation*Support Actor: Change in job*Support Partner: Change in job*Support Actor: Moving*Support Partner: Moving*Support Actor: Negative health event*Support 0.084 0.443 0.017 0.173 -0.010 0.008 265.310 -0.005 0.008 279.506 -1.211 0.227 -0.025 0.006 -0.627 0.531 -0.021 0.011 0.010 213.641 0.307 0.759 -0.017 0.024 0.011 229.799 0.451 0.652 -0.016 0.026 0.009 250.017 0.714 0.476 -0.011 0.024 0.009 247.146 0.937 0.350 -0.009 0.026 0.008 259.624 0.384 0.701 -0.013 0.019 -0.014 0.008 242.311 -1.741 0.083 -0.030 0.002 0.027 305.989 0.169 0.866 -0.049 0.059 0.033 -0.022 0.012 247.982 0.028 355.263 1.169 -1.799 0.243 0.073 -0.023 -0.046 0.089 0.002 -0.003 0.013 261.832 -0.260 0.795 -0.029 0.022 0.380 455.658 1.166 0.244 -0.303 1.188 -0.237 0.368 445.845 -0.644 0.520 -0.960 0.486 0.083 865.493 0.388 0.698 -0.131 0.195 -0.062 0.099 871.535 -0.627 0.531 -0.256 0.132 0.061 926.932 0.274 0.784 -0.103 0.136 0.065 893.203 1.295 0.196 -0.043 0.212 -0.172 0.103 729.128 -1.674 0.095 -0.373 0.030 0.109 803.185 1.586 0.113 -0.041 0.388 -0.009 0.071 777.556 -0.124 0.902 -0.147 0.130 182 0.070 Table 38 (cont’d) Partner: Negative health event*Support Actor: Death of a close other*Support Partner: Death of a close other*Support Actor: Miscarriage*Suppor t Partner: Miscarriage*Suppor t Actor: Other*Support Partner: Other*Support Slope*Partner: Death of a close other*Support Note. Significant effects bolded, p < .01 -0.481 0.340 710.520 -0.105 0.516 624.055 -0.058 0.094 870.218 -0.051 0.089 703.436 -0.007 0.016 396.438 -0.077 0.072 825.080 -0.016 0.071 910.738 -1.081 0.280 -0.218 0.063 -0.225 0.822 -0.155 0.123 0.066 934.785 1.048 0.295 -0.061 0.200 -1.413 0.158 -1.149 0.187 -0.203 0.839 -1.118 0.908 -0.612 0.541 -0.242 0.127 -0.577 0.564 -0.226 0.123 -0.420 0.675 -0.038 0.024 183 t p df LB Table 39. Linear growth curve model examining the effect of actor/partner life events and responsiveness on the slope of agreeableness; CouPers sample. b 3.218 -0.011 0.004 135.498 -0.271 0.155 189.421 0.205 -0.020 0.052 259.005 -0.046 0.052 252.552 0.031 -0.022 0.043 245.329 0.135 -0.059 0.057 222.076 -0.022 0.050 239.664 0.062 UB SE 3.378 0.081 186.513 39.623 <.001 3.057 -0.019 -2.485 -0.002 -0.577 0.036 -1.743 -0.092 0.501 0.150 193.199 1.361 -0.122 0.082 -0.386 -0.147 0.056 -0.885 -0.055 0.117 0.044 242.464 0.710 -0.108 0.063 -0.518 0.247 0.057 230.157 2.378 -0.171 0.053 -1.045 -0.120 0.075 -0.451 -0.037 0.161 0.050 250.073 1.242 0.014 0.083 0.175 0.700 0.377 0.478 0.605 0.018 0.023 0.297 0.652 0.215 Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Responsiveness Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other -0.021 0.043 238.488 -0.042 0.043 235.085 -0.490 -0.995 0.625 0.321 -0.105 0.063 -0.126 0.041 -0.025 0.151 236.833 -0.257 0.151 212.458 -0.039 0.066 243.807 0.006 -0.029 0.016 147.446 0.000 0.002 149.096 -0.004 0.007 246.511 -0.003 0.008 218.350 -0.005 0.002 139.697 -0.022 0.024 753.018 -0.004 0.027 204.239 -0.163 -1.708 -0.587 0.070 250.288 0.087 -1.837 -0.292 -0.481 -0.439 -2.462 -0.918 -0.141 0.871 0.089 0.558 0.931 0.068 0.771 0.631 0.661 0.015 0.359 0.888 -0.323 0.273 -0.554 0.040 -0.169 0.092 -0.131 0.143 -0.060 0.002 -0.004 0.003 -0.018 0.011 -0.019 0.012 -0.008 -0.001 -0.070 0.025 -0.057 0.050 0.013 0.027 198.976 0.481 0.631 -0.040 0.066 0.006 0.001 -0.012 0.008 257.728 -0.008 0.008 263.102 0.009 257.676 0.652 0.009 270.134 0.128 -1.573 -1.062 0.515 0.899 0.117 0.289 -0.012 0.023 -0.017 0.019 -0.027 0.003 -0.023 0.007 0.006 0.001 0.006 0.010 213.904 0.592 0.010 212.324 0.074 0.009 248.754 0.728 0.555 0.941 0.467 -0.014 0.026 -0.019 0.021 -0.011 0.023 0.010 0.009 247.978 1.134 0.258 -0.007 0.027 0.003 0.008 232.880 0.340 0.734 -0.013 0.018 184 0.024 0.137 Table 39 (cont’d) Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Actor: Birth of a child*Responsiveness Partner: Birth of a child*Responsiveness Actor: Graduation*Responsiveness Partner: Graduation*Responsiveness Actor: Change in job*Responsiveness Partner: Change in job*Responsiveness Actor: Moving*Responsiveness Partner: Moving*Responsiveness Actor: Negative health event*Responsiveness Partner: Negative health event*Responsiveness Actor: Death of a close other*Responsiveness Partner: Death of a close other*Responsiveness Actor: Miscarriage*Responsivenes s Partner: Miscarriage*Responsivenes s Actor: Other*Responsiveness Partner: Other*Responsiveness Slope*Partner: Death of a close other*Responsiveness Note. Significant effects bolded, p < .01 0.044 0.094 0.112 0.038 0.001 -0.014 0.008 230.825 -1.779 0.077 -0.028 0.001 0.031 0.041 -0.021 0.012 248.194 -0.003 0.012 249.658 0.227 0.027 261.300 1.165 0.025 222.722 1.653 -1.761 -0.280 0.182 541.744 1.252 0.245 0.100 0.079 0.780 0.211 -0.021 0.083 -0.008 0.090 -0.044 0.002 -0.028 0.021 -0.129 0.584 -0.017 0.173 558.809 -0.100 0.921 -0.358 0.323 0.057 893.773 2.418 0.016 0.026 0.248 0.066 931.510 0.359 0.719 -0.106 0.154 -0.042 0.048 929.511 -0.030 0.044 939.311 -0.097 0.069 832.799 -0.870 0.384 -0.137 0.053 -0.686 0.493 -0.116 0.056 -1.403 0.161 -0.232 0.039 0.070 812.079 0.546 0.585 -0.100 0.176 0.053 803.070 2.104 0.036 0.008 0.217 -0.017 0.048 915.526 -0.102 0.044 880.106 -0.367 0.714 -0.111 0.076 -2.346 0.019 -0.188 -0.017 0.045 892.731 0.962 0.336 -0.045 0.132 -0.287 0.211 670.510 -1.360 0.174 -0.702 0.128 0.146 708.688 0.004 0.997 -0.287 0.288 0.074 926.281 1.266 0.206 -0.052 0.239 -0.149 0.076 675.457 -0.001 0.010 276.153 -1.948 0.052 -0.299 0.001 -0.092 0.927 -0.020 0.019 185 Table 40. Linear growth curve model examining the effect of actor/partner life events and closeness on the slope of agreeableness; CouPers sample. b SE df Intercept 0.025 0.077 717.279 -0.009 0.066 675.126 -0.075 0.078 713.232 3.228 0.083 208.624 -0.009 0.005 157.116 -0.225 0.181 413.264 Slope Actor: Birth of a child Partner: Birth of a 0.185 433.802 0.032 child Actor: Graduation 0.005 0.070 639.318 Partner: Graduation -0.091 0.073 689.488 Actor: Change in job 0.060 604.680 0.049 Partner: Change in job 0.055 0.059 620.844 Actor: Moving 0.083 682.239 0.082 Partner: Moving -0.140 0.082 626.422 Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Closeness Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job -0.100 0.087 765.094 -0.103 0.186 568.504 -0.116 0.180 506.878 -0.065 0.099 683.225 0.049 0.114 808.571 -0.026 0.015 143.020 0.000 0.002 149.745 -0.004 0.007 250.256 -0.004 0.008 221.068 -0.005 0.002 139.827 0.005 530.550 0.002 -0.020 0.036 420.623 -0.010 0.009 363.157 0.037 442.430 0.010 320.735 0.010 328.626 0.010 0.003 0.001 186 p t 38.72 5 <.001 -1.966 0.051 -1.239 0.216 0.862 0.174 0.074 0.941 -1.246 0.213 0.419 0.809 0.348 0.939 0.325 0.985 -1.705 0.089 LB UB 3.063 -0.018 -0.581 -0.331 -0.131 -0.235 -0.069 -0.060 -0.082 -0.302 3.392 0.000 0.132 0.395 0.142 0.052 0.167 0.171 0.246 0.021 -0.959 0.338 -0.229 0.079 0.331 0.741 -0.125 0.176 -0.135 0.892 -0.138 0.120 -1.156 0.248 -0.552 0.581 -0.642 0.521 -0.658 0.511 0.433 0.665 -1.749 0.082 -0.127 0.899 -0.582 0.561 -0.481 0.631 -2.545 0.012 0.651 0.452 -0.270 -0.468 -0.470 -0.261 -0.174 -0.056 -0.004 -0.019 -0.019 -0.009 -0.008 0.070 0.262 0.238 0.130 0.273 0.003 0.003 0.010 0.012 -0.001 0.012 0.273 0.785 -0.063 0.083 -0.565 0.573 -0.092 0.051 0.343 0.731 -0.016 0.023 0.112 0.911 -0.019 0.021 -1.093 0.275 -0.028 0.008 0.002 0.007 0.009 Table 40 (cont’d) Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Actor: Birth of a child*Closeness Partner: Birth of a child*Closeness Actor: Graduation*Closeness 0.009 Partner: Graduation*Closeness Actor: Change in job*Closeness Partner: Change in job*Closeness Actor: Moving*Closeness Partner: Moving*Closeness Actor: Negative health event*Closeness Partner: Negative health event*Closeness Actor: Death of a close other*Closeness Partner: Death of a close other*Closeness 0.002 0.015 0.002 0.000 0.005 0.009 349.139 -0.001 0.011 267.980 -0.009 0.012 280.112 0.524 0.601 -0.104 0.918 -0.013 -0.024 0.022 0.021 -0.807 0.420 -0.032 0.013 0.009 288.553 0.916 0.360 -0.010 0.027 0.009 286.271 0.778 0.437 -0.011 0.026 0.009 275.333 0.204 0.838 -0.015 0.019 -0.039 0.024 545.002 -0.005 0.034 357.138 0.053 0.029 279.472 -0.024 0.013 276.712 -0.003 0.013 276.874 -1.631 0.103 -0.086 0.008 -0.151 0.880 -0.073 0.062 1.842 0.067 -1.813 0.071 -0.260 0.795 -0.004 -0.050 -0.030 0.111 0.002 0.023 0.026 442.112 0.087 0.931 -0.050 0.054 -0.027 0.027 446.347 -1.016 0.310 -0.080 0.026 0.010 801.295 0.905 0.366 -0.010 0.027 -0.004 0.010 751.053 -0.425 0.671 -0.025 0.016 0.008 768.497 0.297 0.766 -0.013 0.018 0.008 814.518 1.835 0.067 -0.001 0.031 -0.012 0.012 671.390 -0.019 0.012 654.710 -0.004 0.011 730.625 -0.007 0.010 788.835 -1.012 0.312 -0.036 0.012 -1.609 0.108 -0.043 0.004 -0.382 0.702 -0.026 0.017 -0.687 0.493 -0.028 0.013 0.010 604.816 -0.048 0.962 -0.019 0.018 -0.011 0.013 670.294 -0.810 0.418 -0.037 0.015 187 -0.023 0.027 731.549 0.025 583.428 -0.006 0.015 683.950 Table 40 (cont’d) Actor: Miscarriage*Closeness Partner: Miscarriage*Closeness 0.022 Actor: Other*Closeness Partner: Other*Closeness Slope*Partner: Death of a close other*Closeness Note. Significant effects bolded, p < .01 0.011 -0.004 0.004 564.731 0.017 732.492 -0.838 0.402 -0.076 0.030 0.893 0.372 -0.027 0.071 -0.390 0.696 -0.035 0.023 0.661 0.509 -0.022 0.044 -1.115 0.265 -0.012 0.003 188 Table 41. Linear growth curve model examining the effect of actor and partner life events on the slope of conscientiousness; CouPers sample. Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income b 3.483 -0.008 SE df 0.073 168.327 0.004 129.532 t 47.930 -2.144 p <.001 0.034 LB 3.339 -0.016 UB 3.626 -0.001 0.096 0.153 151.099 0.624 0.534 -0.207 0.399 -0.249 0.152 147.989 -1.637 0.104 -0.550 0.052 0.037 0.051 238.502 0.735 0.463 -0.063 0.137 -0.009 0.051 243.986 -0.173 0.862 -0.110 0.092 0.004 0.043 240.813 0.093 0.926 -0.080 0.088 0.027 -0.012 0.042 249.383 0.059 178.450 0.636 -0.210 0.525 0.834 -0.056 -0.128 0.110 0.104 0.019 0.060 194.011 0.322 0.748 -0.099 0.138 -0.058 0.047 208.489 -1.248 0.214 -0.150 0.034 0.044 0.049 250.150 0.898 0.370 -0.052 0.140 -0.053 0.044 198.496 -1.210 0.228 -0.140 0.033 -0.031 0.045 208.238 -0.695 0.488 -0.119 0.057 -0.179 0.134 202.293 -1.333 0.184 -0.443 0.086 0.070 0.092 0.025 0.016 -0.003 0.011 -0.006 0.142 231.599 0.065 210.651 0.070 239.833 0.018 136.589 0.001 139.755 0.007 208.908 0.007 180.924 0.495 1.412 0.361 0.906 -2.312 1.536 -0.896 0.621 0.159 0.718 0.366 0.022 0.126 0.371 -0.210 -0.037 -0.113 -0.019 -0.006 -0.003 -0.021 0.351 0.221 0.164 0.051 0.000 0.025 0.008 189 0.002 132.200 0.027 161.141 0.027 151.688 0.009 232.216 0.009 239.744 0.007 242.443 0.007 243.850 0.010 183.274 0.011 195.868 0.003 0.006 0.011 0.003 -0.013 -0.004 -0.002 -0.014 Table 41 (cont’d) Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: 0.000 Other Note. Significant effects bolded, p < .05 -0.004 -0.008 -0.003 0.011 0.004 0.003 0.023 0.020 0.008 207.351 0.024 207.116 0.025 223.374 0.012 217.880 0.013 236.785 0.008 203.038 0.008 214.471 0.009 251.980 -1.525 0.130 -0.006 0.001 -0.516 0.607 -0.068 0.040 0.118 0.906 -0.050 0.056 1.203 0.230 -0.007 0.028 -0.399 0.691 -0.021 0.014 0.345 0.731 -0.012 0.017 -1.712 0.088 -0.027 0.002 0.570 0.569 -0.015 0.026 1.072 0.285 -0.010 0.032 0.507 0.613 -0.012 0.020 -0.303 0.762 -0.020 0.014 -0.534 0.594 -0.020 0.011 -0.961 0.338 -0.023 0.008 0.833 0.406 -0.027 0.068 0.905 0.366 -0.027 0.073 0.285 0.776 -0.020 0.026 -0.020 0.984 -0.025 0.024 190 Table 42. Linear growth curve model examining the effect of actor and partner life events on the slope of extraversion; CouPers sample. b 3.255 0.001 0.047 0.067 -0.808 -0.092 -0.104 0.132 -0.006 Intercept Slope Actor: Birth of a child Partner: Birth of a 0.598 child Actor: Graduation 0.093 Partner: Graduation 0.005 Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation 0.059 0.012 -0.064 -0.135 0.006 -0.010 0.046 -0.007 0.079 0.228 -0.003 -0.053 -0.018 0.140 0.049 SE 0.207 0.006 df 154.328 139.157 p t 15.723 <.001 0.918 0.103 LB 2.846 -0.010 UB 3.664 0.012 0.427 90.508 -1.892 0.062 -1.656 0.040 0.428 0.145 0.144 158.099 222.204 244.464 1.397 0.646 0.033 0.164 0.519 0.974 -0.247 -0.192 -0.280 1.443 0.378 0.289 0.122 205.188 0.390 0.697 -0.193 0.288 0.119 0.168 0.168 221.973 196.835 192.406 -0.875 0.383 0.788 0.432 -0.034 0.973 -0.340 -0.199 -0.336 0.131 0.463 0.325 0.137 220.614 -0.671 0.503 -0.361 0.178 0.135 159.976 0.499 0.619 -0.199 0.333 0.125 200.775 1.121 0.264 -0.106 0.386 0.124 0.398 202.024 189.394 0.636 0.573 0.525 0.567 0.386 0.190 0.197 0.049 0.004 0.020 0.020 0.005 182.436 224.819 196.777 138.679 141.048 200.325 184.260 132.565 0.880 0.151 0.062 0.950 -0.323 0.747 -2.758 0.007 0.162 1.406 -0.493 0.623 2.243 0.026 -1.573 0.118 -0.166 -0.557 -0.704 -0.362 -0.452 -0.232 -0.002 -0.049 0.006 -0.016 0.325 1.013 0.821 0.386 0.325 -0.038 0.014 0.030 0.086 0.002 0.033 208.534 1.488 0.138 -0.016 0.114 0.032 210.738 -1.644 0.102 -0.117 0.011 0.011 302.205 -1.590 0.113 -0.040 0.004 0.011 291.065 -0.303 0.762 -0.026 0.019 191 0.010 0.013 0.013 0.011 0.010 0.010 0.004 0.014 -0.003 -0.008 Table 42 (cont’d) Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 -0.007 -0.010 0.034 0.015 -0.009 -0.016 0.012 0.001 0.020 0.010 0.011 0.010 0.031 0.016 277.917 -0.795 0.427 -0.027 0.011 282.936 -0.274 0.784 -0.022 0.016 262.805 0.755 0.451 -0.015 0.035 274.846 0.330 0.742 -0.022 0.031 269.762 1.374 0.170 -0.006 0.035 279.733 0.057 0.954 -0.022 0.023 269.224 -0.921 0.358 -0.028 0.010 275.806 1.249 0.213 -0.007 0.032 283.031 -0.535 0.593 -0.076 0.044 227.160 280.497 -0.208 0.836 -0.703 0.483 -0.074 -0.040 0.060 0.019 299.210 1.251 0.212 -0.012 0.052 192 Table 43. Linear growth curve model examining the effect of actor and partner life events on the slope of neuroticism; CouPers sample. 0.067 0.087 -0.002 0.092 -0.026 -0.026 0.228 -0.064 -0.128 0.057 b 3.006 Intercept Slope -0.018 Actor: Birth of a child -0.202 Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job -0.122 -0.130 -0.220 0.111 0.148 -0.219 -0.005 -0.002 -0.030 0.001 -0.004 -0.024 0.036 0.015 0.026 df SE 0.114 766.141 0.015 527.871 0.206 687.235 t 26.410 -1.220 -0.982 1.101 0.207 705.453 -0.871 0.074 1048.261 0.074 1022.853 -1.745 0.063 1049.531 0.909 0.062 1026.829 1.497 -0.307 0.083 892.642 -0.320 0.082 842.794 p <.001 0.223 0.326 0.271 0.384 0.081 0.364 0.135 0.759 0.749 LB 2.783 -0.048 -0.607 -0.179 -0.209 -0.273 -0.066 -0.029 -0.189 -0.188 UB 3.229 0.011 0.202 0.635 0.081 0.016 0.181 0.213 0.138 0.135 0.071 1013.862 1.226 0.220 -0.052 0.226 0.069 911.596 0.959 0.338 -0.070 0.203 0.062 963.196 -0.039 0.969 -0.124 0.120 0.062 922.329 -1.973 0.205 954.239 -0.636 -1.129 0.195 853.915 0.097 1038.972 1.154 0.100 1006.338 1.490 -9.365 0.023 588.193 -2.186 0.002 609.870 -0.196 0.011 1004.305 -2.670 0.011 848.184 0.484 0.003 574.302 0.049 0.525 0.259 0.249 0.136 <.001 0.029 0.844 0.008 0.629 -0.243 -0.533 -0.602 -0.078 -0.047 -0.265 -0.010 -0.023 -0.052 -0.004 -0.001 0.272 0.162 0.301 0.344 -0.173 -0.001 0.019 -0.008 0.006 0.092 675.469 -0.261 0.795 -0.204 0.156 0.092 695.820 0.278 0.781 -0.156 0.207 0.031 1032.412 1.142 0.254 -0.026 0.097 0.031 1017.054 -0.138 0.890 -0.066 0.057 0.027 1050.402 0.542 0.588 -0.038 0.068 193 -0.013 Table 43 (cont’d) Slope*Partner: Change in job -0.009 Slope*Actor: Moving 0.001 Slope*Partner: Moving Slope*Actor: Negative health event 0.003 Slope*Partner: Negative health event 0.004 Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 0.080 -0.015 0.010 -0.014 0.024 0.020 0.027 1029.039 0.037 890.267 0.036 845.886 -0.325 0.030 0.745 0.976 -0.061 -0.071 0.043 0.073 -0.362 0.718 -0.085 0.058 0.031 1019.312 0.086 0.931 -0.058 0.063 0.030 912.795 0.028 958.857 0.027 919.384 0.090 950.122 0.148 0.882 -0.054 0.063 0.878 0.380 -0.030 0.079 -0.513 0.608 -0.067 0.039 0.224 0.823 -0.157 0.198 0.086 841.255 0.930 -0.363 0.042 1044.447 0.044 1002.343 0.227 0.353 0.717 0.821 -0.089 -0.099 -0.076 0.248 0.068 0.096 194 Table 44. Linear growth curve model examining the effect of actor and partner life events on the slope of openness; CouPers sample. b 3.346 -0.004 -0.106 -0.031 -0.004 -0.091 0.000 0.015 Intercept Slope Actor: Birth of a child Partner: Birth of a -0.014 child 0.053 Actor: Graduation Partner: Graduation 0.111 Actor: Change in job 0.067 Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation -0.067 -0.226 0.172 0.005 0.040 -0.019 0.004 0.023 -0.022 -0.005 -0.043 -0.007 -0.009 0.097 0.046 df SE 0.117 181.754 0.005 129.436 p t 28.503 <.001 -0.833 0.406 LB 3.114 -0.014 UB 3.577 0.006 0.227 162.810 -0.468 0.641 -0.554 0.342 0.227 162.777 0.079 254.113 0.079 253.456 0.066 254.870 -0.060 0.952 0.500 0.676 0.159 1.413 0.312 1.013 0.065 255.490 0.089 203.211 0.090 204.661 -1.390 0.166 -0.001 0.999 0.872 0.162 -0.461 -0.102 -0.044 -0.064 -0.219 -0.176 -0.163 0.434 0.208 0.266 0.198 0.038 0.176 0.192 0.074 233.621 -0.049 0.961 -0.149 0.141 0.074 240.484 -0.424 0.672 -0.177 0.114 0.067 222.323 1.445 0.150 -0.035 0.229 0.067 222.548 0.213 221.826 0.212 220.267 0.102 238.088 0.107 248.114 0.025 143.387 0.002 147.958 0.011 232.246 0.012 199.568 0.003 139.807 -1.001 0.318 -1.062 0.289 0.419 0.809 0.958 0.053 0.374 0.709 -0.746 0.457 0.114 1.592 2.089 0.038 -1.882 0.061 -1.914 0.058 -0.200 -0.647 -0.247 -0.195 -0.171 -0.069 -0.001 0.001 -0.045 -0.010 0.065 0.194 0.590 0.206 0.251 0.031 0.009 0.045 0.001 0.000 0.031 166.919 -1.412 0.160 -0.104 0.017 0.031 163.645 1.491 0.138 -0.015 0.106 0.010 253.385 -0.881 0.379 -0.029 0.011 0.010 255.635 -0.667 0.505 -0.027 0.014 195 0.013 0.006 0.005 -0.005 -0.017 Table 44 (cont’d) Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 -0.010 0.015 -0.028 -0.005 -0.011 0.075 0.000 0.009 255.595 0.009 259.582 0.012 203.560 0.012 216.387 0.010 228.680 0.010 254.792 0.009 226.634 0.009 230.485 0.029 215.070 0.029 231.645 0.014 240.608 0.015 255.685 -1.940 0.053 -0.035 0.000 -0.517 0.605 -0.022 0.013 0.429 0.668 -0.018 0.029 1.101 0.272 -0.011 0.037 0.662 0.508 -0.013 0.026 -0.523 0.601 -0.025 0.015 -1.193 0.234 -0.029 0.007 -0.046 0.963 -0.018 0.017 2.627 0.009 0.019 0.131 -0.340 0.734 1.081 0.281 -1.888 0.060 -0.068 -0.012 -0.056 0.048 0.042 0.001 196 Table 45. Linear growth curve model examining the effect of actor/partner life events and support on the slope of openness; CouPers sample. 0.086 -0.025 -0.009 b 3.395 Intercept -0.003 Slope Actor: Birth of a child -0.135 Partner: Birth of a child 0.008 0.069 Actor: Graduation 0.110 Partner: Graduation 0.074 Actor: Change in job -0.081 Partner: Change in job -0.017 Actor: Moving Partner: Moving 0.026 Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Support Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job Slope*Partner: Change in job -0.063 -0.428 0.115 -0.023 0.038 -0.016 0.004 0.019 -0.022 -0.005 0.011 -0.053 -0.008 -0.006 -0.012 -0.005 0.055 SE 0.119 0.005 0.233 0.231 0.080 0.082 0.067 0.066 0.092 0.094 t p df 186.588 28.487 <.001 -0.563 0.574 159.214 -0.579 0.563 179.591 0.974 172.505 0.033 0.389 270.300 0.862 0.180 288.607 1.343 0.271 263.527 1.104 -1.217 0.225 265.674 -0.189 0.850 220.766 0.780 232.110 0.280 LB 3.160 -0.014 -0.595 -0.449 -0.089 -0.051 -0.058 -0.211 -0.199 -0.158 UB 3.630 0.008 0.325 0.464 0.227 0.271 0.207 0.050 0.164 0.211 0.075 243.631 -0.115 0.909 -0.157 0.140 0.075 246.206 -0.327 0.744 -0.172 0.123 0.068 230.307 1.271 0.205 -0.048 0.220 0.068 0.234 0.278 0.103 0.108 0.026 0.002 0.011 0.012 0.003 0.040 228.168 303.091 516.526 0.412 240.470 248.963 0.354 147.055 149.055 1.474 230.614 1.656 204.091 136.725 817.600 0.265 -0.929 0.354 -1.827 0.069 0.680 -0.223 0.824 0.723 -0.630 0.529 0.142 0.099 -1.920 0.056 -1.804 0.073 0.791 -0.197 -0.890 -0.432 -0.225 -0.175 -0.067 -0.001 -0.004 -0.045 -0.010 -0.069 0.071 0.033 0.662 0.179 0.251 0.034 0.008 0.041 0.001 0.000 0.090 0.034 202.660 -1.586 0.114 -0.119 0.013 0.032 172.386 1.701 0.091 -0.009 0.118 0.011 252.168 -0.783 0.434 -0.030 0.013 0.011 269.482 -0.512 0.609 -0.028 0.016 0.009 265.103 -1.237 0.217 -0.030 0.007 0.009 271.413 -0.551 0.582 -0.024 0.013 197 0.001 0.013 0.389 -0.002 -0.302 -0.003 -0.010 -0.023 0.014 -0.020 Table 45 (cont’d) Slope*Actor: Moving 0.000 Slope*Partner: Moving 0.018 Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Actor: Birth of a child*Support Partner: Birth of a child*Support Actor: Graduation*Support Partner: Graduation*Support Actor: Change in job*Support Partner: Change in -0.004 job*Support Actor: Moving*Support -0.186 Partner: Moving*Support Actor: Negative health event*Support Partner: Negative health event*Support Actor: Death of a close other*Support Partner: Death of a close other*Support Actor: Miscarriage*Support Partner: Miscarriage*Support Actor: Other*Support -0.387 0.107 -0.125 -0.045 -0.049 -0.033 -0.824 0.150 0.070 0.081 0.134 0.012 0.013 196.054 0.020 237.368 1.378 0.984 0.170 -0.024 -0.008 0.025 0.044 0.010 231.698 0.096 0.924 -0.019 0.021 0.011 256.619 -0.289 0.773 -0.024 0.018 0.010 249.710 -1.064 0.288 -0.030 0.009 0.009 231.766 -0.176 0.861 -0.020 0.017 0.041 317.655 0.324 0.746 -0.067 0.094 0.035 0.014 0.015 369.169 226.943 0.991 260.123 -0.654 0.514 0.322 -1.307 0.192 -0.091 -0.014 -0.050 0.045 0.042 0.010 0.431 406.574 -0.701 0.484 -1.149 0.545 0.417 392.206 0.934 0.351 -0.430 1.208 0.098 805.508 0.710 0.478 -0.123 0.262 0.120 869.874 -0.376 0.707 -0.280 0.190 0.071 818.728 1.136 0.256 -0.059 0.221 0.079 0.121 875.825 643.826 -0.054 0.957 -1.537 0.125 -0.159 -0.424 0.150 0.052 0.132 761.634 1.134 0.257 -0.110 0.410 0.082 718.704 -1.523 0.128 -0.287 0.036 0.088 896.010 1.519 0.129 -0.039 0.308 0.084 855.839 -0.580 0.562 -0.214 0.116 0.080 852.142 -0.413 0.680 -0.189 0.123 0.394 672.784 -2.092 0.037 -1.597 -0.050 0.615 0.114 589.432 835.662 0.941 -0.629 0.530 0.347 -1.596 -0.116 0.822 0.331 198 Table 45 (cont’d) Partner: Other*Support 0.143 Slope*Actor: Miscarriage*Support Note. Significant effects bolded, p < .01 -0.208 0.171 0.112 813.736 1.277 0.202 -0.077 0.362 421.131 -1.218 0.224 -0.544 0.128 199 Table 46. Linear growth curve model examining the effect of actor/partner life events and responsiveness on the slope of openness; CouPers sample. b SE df Intercept -0.002 0.076 252.472 0.067 220.682 3.351 0.117 175.653 Slope -0.002 0.005 126.756 Actor: Birth of a child -0.100 0.249 182.805 Partner: Birth of a child 0.242 182.532 0.001 Actor: Graduation 0.079 256.879 0.070 Partner: Graduation 0.128 0.079 251.753 Actor: Change in job 0.067 250.578 0.062 Partner: Change in job -0.093 0.065 254.151 Actor: Moving -0.024 0.090 207.005 Partner: Moving 0.091 204.104 0.005 Actor: Negative health event 0.000 0.075 239.568 Partner: Negative health event Actor: Death of a close other 0.083 Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Responsiveness Slope*Actor: Birth of a child Slope*Partner: Birth of a 0.035 202.465 0.020 child Slope*Actor: Graduation -0.007 0.010 250.946 Slope*Partner: Graduation -0.003 0.011 267.370 Slope*Actor: Change in job -0.021 0.009 254.365 Slope*Partner: Change in job -0.003 0.009 263.717 Slope*Actor: Moving 0.012 198.662 Slope*Partner: Moving 0.012 206.853 Slope*Actor: Negative health event -0.068 0.067 220.158 -0.330 0.228 239.012 0.226 218.505 0.147 0.014 0.102 235.135 0.108 246.694 0.028 -0.015 0.026 145.992 0.004 0.002 144.574 0.011 223.545 0.024 -0.021 0.012 193.931 -0.005 0.003 136.856 0.030 821.742 0.035 -0.009 0.037 223.253 0.010 240.090 0.006 0.010 0.005 200 p t 28.64 3 <.001 -0.403 0.688 -0.401 0.689 0.996 0.005 0.378 0.884 1.619 0.107 0.355 0.927 -1.423 0.156 -0.263 0.793 0.954 0.058 -0.002 0.998 LB UB 3.120 3.582 -0.012 0.008 -0.591 0.392 -0.476 0.478 -0.086 0.226 -0.028 0.283 -0.069 0.193 -0.222 0.036 -0.202 0.154 -0.173 0.184 -0.147 0.147 -0.028 0.978 0.218 1.236 -0.151 0.147 -0.049 0.215 -1.004 0.316 -1.443 0.150 0.515 0.652 0.135 0.893 0.795 0.261 -0.572 0.568 1.471 0.144 0.032 2.152 -1.809 0.072 -1.821 0.071 0.237 1.183 -0.241 0.810 0.562 0.581 -0.630 0.529 -0.299 0.765 -2.335 0.020 -0.304 0.761 0.616 0.503 0.413 0.821 -0.200 0.065 -0.780 0.120 -0.298 0.593 -0.187 0.215 -0.184 0.241 -0.065 0.036 -0.001 0.008 0.046 0.002 -0.044 0.002 -0.010 0.000 -0.023 0.094 -0.083 0.065 -0.049 0.090 -0.027 0.014 -0.024 0.018 -0.039 -0.003 -0.020 0.015 -0.018 0.030 -0.014 0.034 0.452 0.652 -0.015 0.024 0.085 0.469 0.059 Table 46 (cont’d) Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Actor: Birth of a child*Responsiveness Partner: Birth of a child*Responsiveness Actor: Graduation*Responsiveness Partner: Graduation*Responsiveness Actor: Change in job*Responsiveness Partner: Change in job*Responsiveness Actor: Moving*Responsiveness Partner: Moving*Responsiveness Actor: Negative health event*Responsiveness Partner: Negative health event*Responsiveness Actor: Death of a close other*Responsiveness Partner: Death of a close other*Responsiveness Actor: Miscarriage*Responsiveness 0.260 Partner: Miscarriage*Responsiveness Actor: Other*Responsiveness Partner: Other*Responsiveness Slope*Actor: Miscarriage*Responsiveness 0.032 0.015 0.046 0.005 0.121 -0.003 0.010 256.961 -0.012 0.009 222.650 -0.001 0.009 223.428 0.032 241.776 0.067 -0.005 0.030 229.769 0.013 0.014 235.771 -0.024 0.015 251.671 -0.448 0.231 597.064 -0.310 0.757 -0.024 0.017 -1.320 0.188 -0.030 0.006 -0.086 0.932 0.038 2.088 -0.168 0.867 0.904 0.367 -1.608 0.109 -0.019 0.017 0.131 0.004 -0.064 0.054 -0.015 0.040 -0.053 0.005 -1.939 0.053 -0.901 0.006 0.220 569.602 2.133 0.033 0.037 0.901 0.070 929.014 1.219 0.223 -0.052 0.223 0.082 949.172 0.718 0.473 -0.102 0.220 -0.035 0.060 951.786 -0.583 0.560 -0.152 0.082 0.054 942.180 2.247 0.025 0.015 0.226 0.083 784.506 0.055 0.956 -0.159 0.168 -0.048 0.087 818.805 -0.027 0.064 755.839 -0.558 0.577 -0.219 0.122 -0.416 0.678 -0.153 0.100 0.058 911.124 0.785 0.433 -0.069 0.160 -0.072 0.053 835.026 -0.026 0.055 850.410 -1.368 0.172 -0.176 0.031 -0.474 0.636 -0.134 0.082 0.263 702.765 0.991 0.322 -0.255 0.776 -0.310 0.184 733.476 -1.690 0.091 -0.671 0.050 0.090 896.317 0.165 0.869 -0.162 0.192 -0.006 0.093 673.456 -0.066 0.947 -0.188 0.176 0.036 179.123 0.882 0.379 -0.039 0.103 201 Table 46 (cont’d) Note. Significant effects bolded, p < .01 202 p Table 47. Linear growth curve model examining the effect of actor/partner life events and closeness on the slope of openness; CouPers sample. SE b 3.321 0.120 -0.007 0.005 0.268 0.058 -0.224 0.269 0.096 0.115 0.100 0.153 0.130 0.083 -0.026 0.081 0.115 0.002 0.115 0.083 t 27.683 <.001 -1.409 0.161 0.829 0.217 -0.833 0.406 0.230 1.201 0.128 1.526 1.572 0.116 -0.323 0.747 0.987 0.016 0.471 0.722 UB LB 3.084 3.558 -0.018 0.003 -0.469 0.585 -0.753 0.305 -0.073 0.304 -0.044 0.350 -0.032 0.292 -0.185 0.133 -0.225 0.228 -0.143 0.309 df 192.565 153.912 284.995 289.152 503.200 556.361 514.489 520.796 468.367 450.707 -0.050 0.103 608.297 -0.488 0.626 -0.252 0.152 -0.128 0.103 631.434 -1.247 0.213 -0.330 0.074 0.122 0.090 539.798 1.364 0.173 -0.054 0.299 -0.098 0.088 0.471 0.387 0.254 0.171 -0.064 0.135 0.062 0.151 -0.018 0.026 0.002 0.003 0.023 0.011 -0.022 0.012 -0.005 0.003 -0.010 0.006 515.509 718.274 409.016 567.513 667.233 143.327 148.242 231.687 199.393 139.285 503.942 -1.108 0.268 0.411 0.822 0.501 0.674 -0.476 0.634 0.415 0.678 -0.696 0.488 0.190 1.318 2.033 0.043 -1.851 0.066 -1.790 0.076 -1.725 0.085 -0.271 0.075 -0.537 1.311 -0.329 0.672 -0.330 0.201 -0.233 0.358 -0.068 0.033 -0.002 0.008 0.001 0.045 -0.044 0.001 -0.010 0.001 -0.022 0.001 -0.013 0.042 429.716 -0.308 0.758 -0.097 0.070 0.008 0.042 407.519 0.185 0.853 -0.074 0.089 -0.004 0.011 302.762 -0.371 0.711 -0.026 0.018 -0.002 0.012 320.937 -0.131 0.896 -0.025 0.022 -0.013 0.010 332.946 -1.299 0.195 -0.034 0.007 203 Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Closeness Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job Table 47 (cont’d) Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Actor: Birth of a child*Closeness Partner: Birth of a child*Closeness Actor: Graduation*Closeness Partner: Graduation*Closeness Actor: Change in job*Closeness Partner: Change in job*Closeness Actor: Moving*Closeness Partner: Moving*Closeness Actor: Negative health event*Closeness Partner: Negative health event*Closeness Actor: Death of a close other*Closeness Partner: Death of a close other*Closeness Actor: Miscarriage*Closeness -0.001 0.010 0.013 0.003 0.013 0.021 350.942 237.661 266.931 -0.118 0.906 0.813 0.237 0.114 1.585 -0.021 0.019 -0.022 0.028 -0.005 0.047 0.006 0.011 262.841 0.568 0.571 -0.015 0.027 -0.010 0.011 296.768 -0.948 0.344 -0.032 0.011 -0.007 0.010 247.314 -0.683 0.495 -0.026 0.013 -0.002 0.010 256.466 -0.210 0.834 -0.021 0.017 0.253 0.144 623.112 1.758 0.079 -0.030 0.535 -0.013 0.034 0.015 0.006 -0.026 0.016 296.577 242.309 281.750 -0.373 0.710 0.702 0.383 -1.657 0.099 -0.080 0.054 -0.024 0.035 -0.057 0.005 0.019 0.030 433.502 0.634 0.527 -0.040 0.078 -0.032 0.031 421.224 -1.052 0.294 -0.092 0.028 0.015 0.011 777.436 1.323 0.186 -0.007 0.036 0.009 0.012 751.880 0.714 0.475 -0.015 0.033 0.014 0.009 756.362 1.463 0.144 -0.005 0.032 0.014 0.010 812.388 1.448 0.148 -0.005 0.033 0.000 0.014 630.459 -0.032 0.975 -0.028 0.027 0.014 0.014 644.523 1.019 0.309 -0.013 0.042 -0.007 0.013 634.685 -0.586 0.558 -0.032 0.017 -0.017 0.012 776.696 -1.357 0.175 -0.041 0.007 0.004 0.011 603.585 0.394 0.694 -0.018 0.027 -0.005 0.011 583.809 -0.512 0.609 -0.026 0.016 0.111 0.071 558.737 1.567 0.118 -0.028 0.250 204 Table 47 (cont’d) Partner: Miscarriage*Closeness Actor: Other*Closeness Partner: Other*Closeness Slope*Actor: Miscarriage*Closeness Note. Significant effects bolded, p < .01 -0.004 0.030 -0.015 0.017 0.019 0.023 0.022 0.005 591.854 685.957 -0.128 0.899 -0.845 0.398 -0.062 0.054 -0.049 0.019 706.422 0.239 0.811 -0.033 0.043 645.440 0.981 0.327 -0.022 0.067 205 Table 48. Linear growth curve model examining the effect of actor and partner life events on the slope of support; CouPers sample. Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation b -0.110 -0.051 SE df 0.105 207.602 0.006 128.504 p t -1.046 0.297 -7.959 <.001 LB -0.316 -0.064 UB 0.097 -0.038 -0.010 0.140 221.737 -0.072 0.943 -0.287 0.267 0.083 -0.022 -0.029 0.141 227.725 0.064 209.571 0.063 207.398 0.585 0.559 -0.353 0.724 -0.460 0.646 -0.196 -0.148 -0.154 0.361 0.103 0.096 0.039 0.056 195.840 0.700 0.485 -0.071 0.149 -0.011 -0.016 0.050 0.055 189.582 0.063 247.019 0.063 236.439 -0.208 0.836 -0.249 0.803 0.422 0.805 -0.120 -0.140 -0.073 0.097 0.109 0.174 -0.052 0.063 192.539 -0.837 0.404 -0.176 0.071 0.012 0.061 182.470 0.194 0.846 -0.109 0.133 0.016 0.049 242.258 0.329 0.742 -0.080 0.113 0.037 -0.240 -0.198 0.079 0.148 -0.106 0.007 0.031 -0.011 -0.001 0.049 236.159 0.181 185.973 0.174 175.479 0.081 211.586 0.084 216.388 0.014 142.776 0.002 165.928 0.008 224.257 0.009 237.840 0.003 157.850 0.446 0.763 -1.325 0.187 -1.136 0.258 0.335 0.966 1.769 0.078 -7.419 <.001 0.002 3.204 3.721 <.001 -1.187 0.236 -0.277 0.782 -0.059 -0.597 -0.541 -0.082 -0.017 -0.134 0.003 0.015 -0.029 -0.006 0.134 0.117 0.146 0.239 0.313 -0.078 0.012 0.048 0.007 0.004 -0.018 0.032 167.258 -0.558 0.578 -0.081 0.045 0.015 0.032 178.603 0.472 0.638 -0.049 0.079 0.003 0.012 229.522 0.225 0.823 -0.021 0.026 0.010 0.012 237.522 0.789 0.431 -0.014 0.033 206 0.012 236.476 0.010 225.562 0.010 222.237 0.013 232.906 0.013 204.584 0.018 0.013 0.013 0.003 -0.014 Table 48 (cont’d) Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 -0.015 0.015 -0.017 -0.004 0.022 0.026 0.005 0.012 198.999 0.010 237.240 0.010 225.475 0.036 228.542 0.032 180.860 0.016 244.598 0.017 220.428 1.244 0.215 -0.008 0.033 0.291 0.771 -0.017 0.023 0.972 0.332 -0.013 0.039 -1.040 0.300 -0.039 0.012 1.475 0.141 -0.006 0.042 2.284 0.023 0.004 0.049 -1.673 0.096 -0.037 0.003 2.177 0.031 0.002 0.042 -0.120 0.904 -0.075 0.066 -0.474 0.636 0.354 0.928 -0.079 -0.017 0.049 0.047 0.277 0.782 -0.028 0.037 207 Table 49. Linear growth curve model examining the effect of actor and partner life events on the slope of responsiveness; CouPers sample. Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job b 4.148 -0.014 -0.331 0.075 -0.046 0.040 0.034 -0.018 0.045 0.010 SE 0.170 0.009 0.232 0.233 0.109 0.108 0.096 0.095 0.106 0.106 df 210.366 133.786 238.693 242.957 204.840 202.927 189.163 183.634 242.704 235.540 p t 24.355 <.001 0.126 -1.541 0.155 -1.425 UB LB 3.813 4.484 -0.031 0.004 -0.789 0.127 0.320 -0.423 0.368 0.351 -0.186 0.425 0.094 0.749 0.673 0.713 0.726 0.853 0.672 0.926 -0.384 0.534 -0.260 0.168 -0.174 0.254 -0.155 0.222 -0.204 0.169 -0.164 0.254 -0.199 0.219 -0.124 0.107 181.531 -1.161 0.247 -0.336 0.087 -0.217 0.106 177.006 -2.044 0.042 -0.427 -0.007 0.076 0.083 232.997 0.916 0.360 -0.088 0.240 0.059 -0.114 -0.217 -0.025 -0.071 -0.014 0.004 -0.025 -0.015 -0.004 0.083 0.310 0.305 0.139 0.144 0.021 0.004 0.013 0.015 0.004 230.147 178.596 171.374 199.987 209.104 144.118 164.900 208.284 226.910 158.111 0.710 -0.369 -0.711 -0.182 -0.495 -0.641 1.011 -1.915 -1.062 -1.030 0.478 0.712 0.478 0.856 0.621 0.522 0.314 0.057 0.289 0.305 -0.105 0.223 -0.726 0.497 -0.818 0.385 -0.300 0.249 -0.355 0.212 -0.056 0.029 -0.004 0.011 -0.050 0.001 -0.044 0.013 -0.013 0.004 -0.014 0.039 199.863 -0.350 0.727 -0.091 0.063 0.019 0.039 193.737 0.491 0.624 -0.057 0.095 0.000 0.015 234.420 -0.015 0.988 -0.030 0.030 -0.010 0.016 242.186 -0.612 0.541 -0.040 0.021 -0.015 0.014 216.666 -1.085 0.279 -0.041 0.012 208 0.014 0.016 0.003 -0.011 -0.001 Table 49 (cont’d) Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event 0.054 Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 0.013 0.005 0.001 -0.002 -0.026 0.020 0.014 0.017 0.015 0.016 0.013 0.013 0.043 0.046 0.020 0.022 220.340 231.866 -0.782 -0.056 0.435 0.955 -0.037 0.016 -0.033 0.031 246.543 0.191 0.849 -0.030 0.036 194.159 3.568 <.001 0.024 0.083 224.251 -1.618 0.107 -0.057 0.006 242.728 1.132 0.259 -0.011 0.040 249.734 -0.133 0.894 -0.027 0.023 188.537 0.459 0.647 -0.066 0.106 205.821 220.590 241.833 0.291 0.225 0.047 0.772 0.822 0.963 -0.077 0.103 -0.036 0.045 -0.042 0.044 209 Table 50. Linear growth curve model examining the effect of actor and partner life events on the slope of closeness; CouPers sample. b 5.997 -0.338 0.139 0.257 -0.122 -0.341 -0.487 0.383 0.140 0.094 Intercept Slope Actor: Birth of a child Partner: Birth of a child 0.532 Actor: Graduation 0.250 Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation 0.576 0.431 0.040 0.168 -0.016 -0.051 -0.003 -0.104 -0.299 -0.077 0.170 0.016 0.977 0.009 SE 0.348 0.063 df 10.381 5.570 t 17.254 -5.336 p <.001 0.002 LB 5.227 -0.496 UB 6.768 -0.180 0.461 6.570 0.558 0.596 -0.847 1.361 0.467 0.228 5.904 10.106 1.139 1.096 0.299 0.299 -0.616 -0.257 1.680 0.756 0.227 4.509 0.613 0.570 -0.463 0.741 0.204 4.895 -0.600 0.575 -0.650 0.405 0.202 0.220 0.216 8.434 20.452 13.047 1.901 0.636 0.437 0.092 0.532 0.669 -0.078 -0.319 -0.371 0.844 0.599 0.560 0.232 11.183 -1.468 0.170 -0.851 0.169 0.224 3.362 -2.176 0.108 -1.159 0.184 0.173 13.409 0.984 0.343 -0.202 0.542 0.171 5.762 0.054 0.959 -0.414 0.433 0.673 1.119 1.452 0.365 -5.691 7.645 0.636 0.297 0.300 0.043 0.008 0.026 0.029 0.906 1.191 1.450 11.727 0.132 1.366 268.403 3.947 -2.128 42.596 -1.989 266.427 -0.093 155.323 0.511 0.173 0.912 <.001 0.039 0.048 0.926 -5.009 -0.218 -2.034 0.084 -0.032 -0.102 -0.059 6.162 1.080 2.113 0.252 -0.001 -0.001 0.054 0.009 46.214 1.823 0.075 -0.002 0.033 0.188 26.705 -1.588 0.124 -0.686 0.087 0.187 1.059 -0.555 0.673 -2.190 1.982 0.094 3.421 -0.823 0.464 -0.355 0.201 210 0.121 0.054 0.095 0.085 0.097 0.092 0.094 0.086 0.091 -0.117 -0.165 -0.233 Table 50 (cont’d) Slope*Partner: Graduation Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 -0.470 -0.263 -0.010 0.129 0.073 0.285 0.073 0.280 0.126 0.099 0.020 0.035 0.094 0.054 1.967 -1.250 0.340 -0.528 0.293 3.419 -2.729 0.062 -0.488 0.021 3.014 -1.937 0.148 -0.436 0.106 1.265 1.042 0.456 -0.620 0.810 3.597 0.585 0.593 -0.214 0.322 3.319 1.247 0.293 -0.172 0.415 2.904 0.546 0.624 -0.266 0.374 55.534 -0.137 0.891 -0.156 0.136 11.911 0.481 0.639 -0.124 0.194 0.259 -1.679 0.624 -9862.438 9861.497 16.988 -0.922 0.370 -0.864 0.338 3.384 0.162 0.880 -0.356 0.397 1.614 0.731 0.556 -0.610 0.799 211 Table 51. Linear growth curve model examining the effect of actor and partner life events on the slope of partner-reported agreeableness; CouPers sample. 0.048 -0.026 -0.070 b 3.148 -0.011 -0.244 0.189 -0.008 -0.140 0.100 -0.040 -0.022 0.046 Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a 0.100 child Slope*Actor: Graduation 0.016 Slope*Partner: Graduation Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving -0.121 -0.368 0.351 -0.042 0.205 0.024 0.000 -0.010 0.007 -0.003 -0.014 0.018 -0.012 -0.040 0.018 SE 0.084 0.011 0.170 0.168 0.055 0.056 0.047 0.047 0.062 0.062 df 653.668 490.658 673.450 649.361 872.754 876.112 881.732 877.534 733.003 716.384 t 37.490 -1.035 -1.437 1.127 -0.138 -2.516 2.124 -0.849 -0.351 0.751 p <.001 0.301 0.151 0.260 0.890 0.012 0.034 0.396 0.726 0.453 LB 2.983 -0.033 -0.577 -0.140 -0.115 -0.249 0.008 -0.131 -0.144 -0.075 UB 3.313 0.010 0.089 0.518 0.100 -0.031 0.193 0.052 0.100 0.168 0.052 832.172 0.931 0.352 -0.053 0.149 0.053 761.032 -1.327 0.185 -0.174 0.034 0.046 791.283 -0.571 0.568 -0.116 0.064 0.046 0.152 0.162 0.071 0.074 0.018 0.002 0.008 0.008 0.002 785.435 769.875 693.692 834.720 847.064 539.952 597.164 825.563 788.836 553.408 -2.616 -2.422 2.162 -0.593 2.775 1.335 0.114 -1.250 0.795 -1.741 0.009 0.016 0.031 0.553 0.006 0.183 0.909 0.212 0.427 0.082 -0.211 -0.666 0.032 -0.182 0.060 -0.011 -0.003 -0.026 -0.010 -0.007 -0.030 -0.070 0.669 0.098 0.351 0.060 0.004 0.006 0.023 0.000 0.069 580.982 -0.573 0.567 -0.176 0.097 0.071 0.023 592.282 857.953 1.409 0.694 0.159 0.488 -0.040 -0.029 0.240 0.061 0.024 865.155 -0.510 0.610 -0.059 0.034 0.021 882.021 0.859 0.390 -0.023 0.058 0.020 0.028 882.235 740.282 -0.721 0.634 0.471 0.526 -0.054 -0.037 0.025 0.073 212 0.028 0.023 0.023 0.016 -0.023 -0.025 Table 51 (cont’d) Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 -0.047 0.039 0.015 -0.002 0.008 0.165 0.020 0.020 0.067 0.077 0.032 0.033 733.300 -0.908 0.364 -0.080 0.029 849.621 -1.008 0.314 -0.067 0.022 788.216 0.729 0.466 -0.028 0.061 788.875 0.408 0.684 -0.032 0.048 781.789 -0.103 0.918 -0.042 0.038 695.301 2.475 0.014 0.034 0.296 658.337 840.089 840.297 -0.613 1.213 0.445 0.540 0.226 0.656 -0.199 -0.024 -0.051 0.104 0.102 0.080 213 Table 52. Linear growth curve model examining the effect of actor and partner life events on the slope of partner-reported conscientiousness; CouPers sample. Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job b 3.649 0.012 0.070 -0.185 -0.149 -0.238 -0.046 0.029 0.106 0.019 SE df 0.207 174.703 0.008 131.408 0.400 141.761 0.405 148.172 0.140 237.723 0.139 235.724 0.118 243.522 0.115 238.406 0.162 195.691 0.159 178.757 p LB t 17.668 <.001 3.241 0.143 1.474 0.861 0.175 0.648 -0.457 0.289 -1.064 0.088 -1.714 0.695 -0.393 0.802 0.252 0.512 0.657 0.903 0.122 UB 4.056 -0.004 0.028 -0.721 0.861 -0.985 0.615 -0.424 0.127 -0.513 0.036 -0.279 0.186 -0.198 0.256 -0.213 0.426 -0.294 0.333 -0.162 0.136 253.288 -1.190 0.235 -0.429 0.106 -0.062 0.126 204.213 -0.494 0.622 -0.311 0.187 0.149 0.120 209.053 1.249 0.213 -0.086 0.385 0.033 -0.119 0.527 -0.206 -0.157 0.150 0.005 0.023 0.006 0.001 0.118 195.438 0.398 237.310 0.361 185.834 0.186 240.401 0.190 231.793 0.048 138.176 0.004 142.979 0.020 221.352 0.020 183.680 0.005 133.192 0.281 -0.298 1.457 -1.109 -0.828 3.131 1.106 1.166 0.303 0.220 0.779 0.766 0.147 0.269 0.409 0.002 0.055 0.271 0.245 0.762 0.826 -0.200 0.266 -0.902 0.665 -0.186 1.240 -0.571 0.160 -0.532 0.217 0.244 -0.004 0.013 -0.016 0.063 -0.034 0.046 -0.008 0.010 0.002 0.047 122.825 0.037 0.970 -0.092 0.095 0.000 0.047 123.895 -0.008 0.994 -0.094 0.093 0.005 0.017 219.084 0.283 0.777 -0.029 0.038 -0.011 0.018 223.976 -0.612 0.541 -0.046 0.024 0.015 0.015 223.116 1.005 0.316 -0.014 0.044 214 -0.009 Table 52 (cont’d) Slope*Partner: Change 0.017 in job -0.035 Slope*Actor: Moving Slope*Partner: Moving 0.026 Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 -0.046 -0.033 -0.022 -0.010 -0.010 0.003 0.100 0.014 221.849 0.020 183.765 0.020 185.198 0.016 207.138 0.016 217.395 0.015 198.398 0.015 200.014 0.048 235.000 0.051 117.917 0.023 217.392 0.025 217.405 1.242 -1.776 1.324 0.215 0.077 0.187 -0.010 0.045 -0.075 0.004 -0.013 0.066 -0.535 0.594 -0.041 0.024 -0.612 0.541 -0.042 0.022 -0.693 0.489 -0.040 0.019 0.205 0.838 -0.026 0.033 2.068 0.040 0.005 0.194 -0.897 -1.439 -0.868 0.372 0.152 0.386 -0.147 0.055 -0.078 0.012 -0.071 0.027 215 p Table 53. Linear growth curve model examining the effect of actor and partner life events on the slope of partner-reported extraversion; CouPers sample. SE t b 0.275 13.888 <.001 3.820 0.586 0.546 0.012 0.006 0.477 0.713 0.570 0.407 -1.681 0.095 -0.955 0.568 -0.502 0.616 -0.096 0.190 -0.010 0.992 -0.002 0.192 0.845 0.196 0.159 0.031 0.433 0.068 0.665 0.156 -0.427 0.670 -0.094 0.221 0.073 1.805 0.218 0.393 df 172.502 104.093 158.268 158.830 235.438 226.272 245.106 238.172 190.686 175.940 LB 3.277 -0.017 -0.720 -2.077 -0.471 -0.380 -0.282 -0.240 -0.530 -0.037 UB 4.363 0.030 1.533 0.167 0.280 0.376 0.344 0.375 0.341 0.822 Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving -0.013 0.183 243.332 -0.072 0.942 -0.375 0.348 -0.239 0.175 210.731 -1.364 0.174 -0.584 0.106 -0.054 0.163 208.707 -0.332 0.740 -0.376 0.268 0.376 0.163 -0.004 0.635 0.490 0.035 -0.127 0.254 -0.187 0.256 0.065 0.161 0.001 0.006 -0.014 0.027 0.028 0.011 -0.008 0.006 201.738 141.352 194.091 235.783 221.802 143.644 142.487 220.394 181.516 132.229 2.301 0.022 -0.006 0.995 0.943 0.072 -0.498 0.619 -0.731 0.466 0.014 2.487 0.207 0.836 -0.531 0.596 0.680 0.413 -1.327 0.187 0.054 -1.259 -0.932 -0.628 -0.690 0.033 -0.010 -0.067 -0.043 -0.021 0.699 1.251 1.003 0.375 0.317 0.290 0.012 0.039 0.066 0.004 0.078 0.083 135.509 0.943 0.347 -0.085 0.241 -0.109 0.085 0.026 0.024 -0.006 0.025 128.062 174.670 165.819 -1.287 0.200 1.056 0.292 -0.226 0.822 -0.277 -0.022 -0.055 0.059 0.074 0.044 0.008 0.021 186.408 0.382 0.703 -0.033 0.049 0.020 0.012 -0.047 0.027 177.943 131.107 0.554 0.592 -1.764 0.080 -0.028 -0.101 0.052 0.006 216 -0.004 0.023 0.027 0.024 0.017 0.005 Table 53 (cont’d) Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: Miscarriage Slope*Actor: Other Slope*Partner: Other Note. Significant effects bolded, p < .05 -0.007 0.022 -0.012 0.077 0.043 0.065 -0.002 0.035 0.034 0.005 -0.001 0.022 133.981 0.178 0.859 -0.049 0.059 179.275 0.727 0.468 -0.030 0.064 172.234 -0.169 0.866 -0.050 0.042 158.404 -0.025 0.980 -0.043 0.042 153.834 133.812 -0.326 0.745 -0.161 0.872 -0.051 -0.164 0.037 0.139 171.114 201.318 156.560 0.664 0.508 -0.058 0.954 0.874 0.159 -0.085 -0.072 -0.061 0.171 0.068 0.072 217 Table 54. Linear growth curve model examining the effect of actor and partner life events on the slope of partner-reported neuroticism; CouPers sample. b 3.145 -0.008 0.183 0.084 0.008 -0.062 -0.063 -0.123 0.120 Intercept Slope Actor: Birth of a child Partner: Birth of a -0.070 child Actor: Graduation 0.214 Partner: Graduation 0.166 Actor: Change in job Partner: Change in job Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation 0.043 0.217 -0.041 0.368 -0.005 -0.046 -0.006 -0.027 0.456 -0.017 -0.203 -0.100 0.098 0.011 SE df 0.237 171.910 0.011 113.616 t 13.295 -0.682 p <.001 0.497 LB 2.678 -0.030 UB 3.612 0.015 0.476 151.191 -0.131 0.896 -1.002 0.878 0.474 146.302 0.159 233.311 0.162 246.736 -0.147 1.344 1.029 0.884 0.180 0.305 -1.006 -0.099 -0.152 0.867 0.526 0.485 0.134 238.274 0.624 0.533 -0.180 0.348 0.133 253.267 0.183 177.982 0.188 192.323 -0.473 -0.669 0.640 0.637 0.504 0.523 -0.326 -0.484 -0.250 0.200 0.239 0.490 0.147 206.417 0.051 0.959 -0.282 0.297 0.153 248.371 1.193 0.234 -0.119 0.484 0.137 200.958 -1.474 0.142 -0.474 0.068 0.140 213.299 0.422 200.279 -0.190 1.080 0.850 0.281 -0.302 -0.376 0.249 1.288 0.449 239.504 0.207 215.702 0.224 251.174 0.055 142.637 0.005 143.108 0.023 212.110 0.023 192.372 0.096 1.047 -0.184 6.730 -1.007 -2.048 -0.252 0.924 0.296 0.854 <.001 0.316 0.042 0.802 -0.841 -0.191 -0.482 0.260 -0.015 -0.091 -0.052 0.927 0.625 0.400 0.476 0.005 -0.002 0.040 0.005 133.382 2.020 0.045 0.000 0.021 0.087 157.831 1.128 0.261 -0.073 0.268 0.089 133.044 -1.131 0.260 -0.275 0.075 0.024 184.849 -0.692 0.490 -0.064 0.031 218 0.045 0.045 0.003 -0.047 -0.029 -0.027 Table 54 (cont’d) Slope*Partner: Graduation Slope*Actor: Change in job Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: 0.002 Miscarriage Slope*Actor: Other 0.028 Slope*Partner: Other Note. Significant effects bolded, p < .05 -0.031 -0.009 -0.028 0.015 0.028 0.026 200.589 0.021 211.079 0.021 197.474 0.029 148.187 0.031 161.708 0.024 193.617 0.024 185.451 0.023 180.504 0.023 178.688 0.069 212.981 0.072 153.773 0.034 180.829 0.036 191.574 1.740 0.083 -0.006 0.096 0.129 0.897 -0.039 0.045 -2.208 0.028 -0.089 -0.005 1.554 0.122 -0.012 0.103 -0.860 0.391 -0.088 0.035 -1.187 0.237 -0.076 0.019 1.176 0.241 -0.019 0.075 -1.256 0.211 -0.073 0.016 0.679 0.498 -0.029 0.060 -0.451 0.652 -0.168 0.105 0.030 0.835 0.976 0.405 -0.140 -0.039 0.144 0.095 -0.257 0.797 -0.081 0.063 219 Table 55. Linear growth curve model examining the effect of actor and partner life events on the slope of partner-reported openness; CouPers sample. t 16.983 <.001 -0.422 0.674 -0.343 0.732 SE b 3.471 0.204 -0.003 0.008 -0.126 0.367 df 162.325 116.685 143.175 LB 3.067 -0.020 -0.852 UB 3.875 0.013 0.600 p -0.335 0.373 0.113 0.134 0.133 0.076 -0.038 0.114 0.112 -0.126 0.154 0.148 0.092 -0.053 0.136 -0.075 0.122 149.661 217.154 204.620 220.475 209.044 198.596 170.219 -0.898 0.371 0.849 0.397 0.570 0.568 -0.329 0.743 0.545 0.586 -0.821 0.413 0.532 0.626 -1.071 -0.150 -0.187 -0.262 -0.159 -0.430 -0.199 0.402 0.377 0.339 0.187 0.281 0.177 0.384 240.929 -0.389 0.698 -0.320 0.215 185.579 -0.614 0.540 -0.317 0.166 0.113 203.621 0.070 0.944 -0.214 0.230 0.008 Intercept Slope Actor: Birth of a child Partner: Birth of a child Actor: Graduation Partner: Graduation Actor: Change in job Partner: Change in job 0.061 Actor: Moving Partner: Moving Actor: Negative health event Partner: Negative health event Actor: Death of a close other Partner: Death of a close other Actor: Miscarriage Partner: Miscarriage Actor: Other Partner: Other Gender Age Education Income Relationship length Slope*Actor: Birth of a child Slope*Partner: Birth of a child Slope*Actor: Graduation Slope*Partner: Graduation Slope*Actor: Change in job 0.110 0.182 0.391 0.433 -0.177 0.337 -0.284 0.181 -0.123 0.180 0.044 0.217 0.004 0.008 0.021 0.020 0.020 0.000 -0.004 0.005 -0.016 0.047 -0.040 0.047 -0.014 0.017 -0.020 0.018 -0.015 0.015 186.541 237.804 172.923 226.579 191.623 132.086 134.076 211.339 175.598 126.260 0.100 1.651 0.269 1.107 -0.526 0.599 -1.575 0.117 -0.687 0.493 <.001 4.876 0.059 1.901 1.059 0.291 -0.016 0.988 -0.796 0.427 -0.035 -0.338 -0.842 -0.640 -0.478 0.129 0.000 -0.018 -0.040 -0.013 0.399 1.204 0.488 0.071 0.231 0.305 0.017 0.059 0.039 0.006 121.279 -0.339 0.735 -0.108 0.076 124.868 -0.845 0.399 -0.133 0.053 207.981 -0.829 0.408 -0.048 0.020 214.573 -1.101 0.272 -0.055 0.016 213.741 -1.010 0.313 -0.044 0.014 220 0.036 0.026 0.017 0.020 0.014 0.000 0.015 0.021 Table 55 (cont’d) Slope*Partner: Change in job Slope*Actor: Moving Slope*Partner: Moving Slope*Actor: Negative health event Slope*Partner: Negative health event Slope*Actor: Death of a close other Slope*Partner: Death of a close other Slope*Actor: Miscarriage Slope*Partner: -0.047 0.043 Miscarriage 0.023 0.010 Slope*Actor: Other Slope*Partner: Other 0.023 0.024 Note. Significant effects bolded, p < .05 -0.002 0.015 -0.024 0.016 -0.012 0.047 0.014 0.016 204.945 194.572 0.924 0.357 -0.019 0.985 -0.016 -0.041 0.043 0.041 173.275 1.801 0.073 -0.003 0.075 217.497 1.460 0.146 -0.009 0.060 179.040 -1.524 0.129 -0.055 0.007 189.501 -0.141 0.888 -0.032 0.028 175.893 1.102 0.272 -0.013 0.045 191.715 -0.251 0.802 -0.105 0.081 154.707 202.973 187.365 -1.109 0.269 0.651 0.452 0.296 1.048 -0.132 -0.035 -0.022 0.037 0.055 0.070 221 Table 56. A summary of which life events produced a (mal)adaptive response, organized by trait; CouPers sample. Trait Adaptive Actor: miscarriage Agreeableness Conscientiousness Actor: miscarriage Maladaptive Partner: death of a close other Extraversion Neuroticism Openness Actor: miscarriage Note. Type of response (i.e., adaptive or maladaptive) based on Event*Slope interactions. "Adaptive" responses include steeper increases/stability in in all traits with the exception of neuroticism. "Maldaptive" responses include steeper decreases in all traits with the exception of neuroticism. Self-reported changes in black, partner-reported changes in red. 222 APPENDIX D: SUPPLEMENTARY GROWTH MIXTURE RESULTS AND TABLES Results: Growth Mixture Models To characterize changes in personality, I employed a variant of growth mixture modeling (GMM; Infurna & Luthar, 2016; Ram & Grimm, 2009). Growth mixture modeling combines latent growth models with mixture modeling to determine latent classes. This approach enabled me to model multiple latent classes characterizing trajectories of personality change within one sample. In other words, it allows for the possibility that one group of people might start relatively low on a trait (e.g., neuroticism) and then increase after an event. Likewise, another group might start relatively high on neuroticism and decrease after an event. Further, another group might start low and have relatively stable levels of neuroticism after an event happens. Permitting multiple growth trajectories is an advantage over other models that have been traditionally used to model longitudinal data (e.g., more general growth curve modeling) that assume the population under study changes in one homogenous way over time. I adopted this particular approach because it gives feedback about whether a certain percentage of the sample “grows” over time (i.e., experiences ostensibly positive personality change). All couples were included in the analyses. In order to ensure that various subgroups were allowed to vary in terms of their starting points and how they change over time (as well as avoid the over-estimation of resilience and growth sometimes implicated in fixed-effects approaches; Infurna & Jayawickreme, 2019; Infurna & Luthar, 2016), I allowed intercepts and slopes to vary randomly. I used maximum likelihood estimation with robust standard errors in Mplus to account for missing data. I tested 1-4 class solutions for each personality trait; the fit indices for each of these models are reported in Supplemental Table 1. As mentioned in my proposal, I considered AIC, BIC, consistency/interpretability of classes, and entropy for model selection (Frankfurt et 223 al., 2016; Jung & Wickrama, 2008; Ram & Grimm, 2009; Smith & Ehlers, 2020). Entropy levels approaching 1 indicate appropriate model selection (Jung & Wickrama, 2007). After identifying these latent classes, I used a multi-level logistic regression to predict membership in the “growth” class (if applicable). The multi-level component will account for the non-independence of couple members’ change outcomes (Loeys et al., 2014). Each logistic regression included actor life events, partner life events, spousal support, spousal strain, a set of interaction variables (spousal support X actor life event, spousal support X partner life event, spousal strain X actor life event, spousal strain X partner life event) and a set of control variables (i.e., actor’s age, gender, race/ethnicity, education, income, and relationship length). Life events were tested simultaneously in one model (rather than having a separate model for each life event). Modeling the main effects of support and strain (irrespective of life event occurrence) allowed me to estimate how much personality change is attributable to having a supportive (or strenuous) spouse. Importantly, interactions between support/strain and life events allow for a critical test of whether experiencing a life event in the presence of a supportive relationship engenders positive personality change/growth over time. Below, I report the results of each analysis, organized by personality trait. Additionally, as discussed in the dissertation proposal, I report results for a similar set of analyses where spousal support and spousal strain are positioned as outcome variables. Agreeableness Growth mixture model The fit indices for each class solution for agreeableness are displayed in Table 57. I followed the same process of model selection for each trait. First, I examined the number of participants grouped into each latent class. Occasionally, the GMM would identify a class that 224 consisted of a very small subsection of the sample (e.g., Class 3 in the 4-class solution for agreeableness contained only 3.8% of the sample). In order to ensure that subgroups were large enough to be meaningful, I set a cut-off limit of 5% (i.e., class solutions which produced subgroups of <5% would be modified). After identifying these potentially problematic subgroups, I compared AIC/BIC, entropy, and interpretability of classes across subgroups. Ultimately, I pursued the 2-class solution for agreeableness, as the 3 and 4 class solutions each produced subclasses with < 5% of the sample, did not vary dramatically in fit from the 2-class solution, and produced similar subclasses (i.e., one set of participants which started relatively high in agreeableness and remained stable/increased very slightly over time and another set which started slightly lower in agreeableness and decreased slightly over time). These classes were then recoded so that Class 1 represented a “resilient” or “growth” class (Class = 1), which saw slight gains in agreeableness over time (i=3.70, s=.007, p = .003), and a “non-resilient” class (Class =2), which saw slight decreases in agreeableness over time (i= 2.83, s=. -007, p < .001). Multilevel logistic regression This class membership was then used as the outcome variable in two multilevel logistic regressions, the first of which only included actor/ partner life events and control variables, and the second of which incorporated spousal support and strain. Model 1: Actor/partner life events. As highlighted in Table 3, some life events (i.e., moving, having a child, or experiencing the loss of a child) were highly (or perfectly) correlated, indicating that partners always experienced these life events together. For these life events, only the actor effects were modeled. In addition to these effects, actor and partner effects for all other life events (i.e., new chronic illness, negative health changes, positive health changes, death of a 225 parent, new job, retirement, and unemployment), as well as a series of control variables (i.e., age, gender, race, education, wealth, and relationship length) were included as predictors of class membership. Results of this analysis are reported in Table 58. As the “resilient” or “growth” class was coded as Class 1, this group served as the reference group. Therefore, significant negative bs indicate that experiencing a given life event is predictive of a participant being assigned to the “resilient” class. Significant positive bs indicate that experiencing a given life event is predictive of a participant being assigned to the “non-resilient” class. These bs were then exponentiated and converted to odds ratios (where negative odds ratios indicate a higher chance of being assigned to the “resilient” class). Of the effects examined, two were predicative of being classified as “resilient” in agreeableness: the actor effect of a parent’s death (b= -.18, p = .003) and the partner effect of unemployment (b= -.30, p = .007). One effect was predictive of being classified as “non- resilient” in agreeableness: the actor effect of getting a new job (b= .21, p < .001). A handful of control variables were also significant; women and those who had been married for longer were more likely to be classified as “resilient” (b= -.58, p < .001; b= -.07, p = .024, respectively). Those with more education were more likely to be classified as “non-resilient” (b= .05, p < .001). Model 2: Actor/partner life events and spousal support/strain. Spousal support and spousal strain were then entered as predictors in the model described above. In addition, support and strain were included as interaction variables with life events that had predicted class membership (i.e., Actor: Parent died*Support, Actor: New job*Support. Partner: Unemployment*Support, Actor: Parent died*Strain, Actor: New job*Strain, Partner: 226 Unemployment*Strain). The results of this model are reported in Table 59. Of these, only the main effect of spousal support was significant; those who reported higher spousal support were more likely to be classified as “non-resilient” (b= .56, p < .001). Conscientiousness Growth mixture model The fit indices for each class solution for conscientiousness are displayed in Table 57. As was the case when examining agreeableness, the 3 and 4 class solutions each included subclasses that included < 5% of the sample. Interpretability of the classes suggested a 3-class solution (i.e., a group that starts low in conscientiousness and increases, another that starts higher and decreases, and a third that starts highest in conscientiousness and remains stable). However, the group that declined in conscientiousness only consisted of 4.5% of the sample. Again, the fit indices did not vary dramatically between the classes (especially between the 2 and 3 class solutions). Ultimately, I chose to pursue a 2-class solution. These classes were then recoded so that Class 1 represented a “resilient” or “growth” class (Class = 1), which remained stable in conscientiousness over time (i=3.49, s=.001, p = .101), and a “non-resilient” class (Class =2), which saw slight decreases in conscientiousness over time (i= 2.92, s=.-028, p < .001). Multilevel logistic regression Model 1: Actor/partner life events. I repeated the regression procedure outlined in above section (i.e., agreeableness). The results of this model are presented in Table 60. Of the effects examined, one predicative of being classified as “resilient” in conscientiousness: the actor effect of a new job (b= -.30, p < .001). Three effects were predictive of being classified as “non- resilient” in conscientiousness: the actor effect of a new chronic illness (b= .21, p < .001), the 227 partner effect of a new chronic illness (b= .20, p < .001), and the actor effect of retirement (b= .12, p < .04). Some control variables were also significant; those with more education and wealth were more likely to be classified as “resilient” in conscientiousness (both b= -.13, p < .001). Men and older participants were more likely to be classified as “non-resilient” (both b= .02, p < .001). Model 2: Actor/partner life events and spousal support/strain. As described above, spousal support, spousal strain, and interactions between any significant life events and support/strain were entered as predictors into the model. The results of this model are reported in Table 61. Of these, one main effect was significant; those with more spousal support were more likely to be classified as “resilient” (b= .52, p < .001). Additionally, there was an interaction of support and an actor’s chronic illness such that those who experienced the onset of a new illness and had higher support were more likely to be classified as “non-resilient” (although this effect was marginal, b= .24, p = .04). Extraversion Growth mixture model The fit indices for each class solution for extraversion are displayed in Table 57. Here, the 3 and 4 class solutions each produced subclasses that contained less than 1% of the sample. The AIC and BIC were not dramatically different between the 2 class and the 3 and 4 class solutions, and entropy was highest for the 2-class solution. Due to the small subclass sizes of the larger models and the fit indices, I ultimately pursued the 2-class solution. This resulted in one subclass (coded as Class 1, the “resilient” class) which started relatively high in extraversion and remained stable (i= 3.62, s=-.002, p = .125) and another subclass (coded as Class 2, the “non- resilient” class) which started at approximately the same level of extraversion, but declined relatively steeply (i= 3.20, s=-.109, p = .002). 228 Multilevel logistic regression Model 1: Actor/partner life events. I repeated the regression procedure outlined in above sections. The results of this model are reported in Table 62. Of the effects examined, none were predicative of being classified as “resilient” in extraversion. Three effects were predictive of being classified as “non-resilient” in extraversion: the actor effect of a new chronic illness (b= .41, p = .019), the partner effect of a new chronic illness (b= .48, p = .007), and the actor effect of a positive health change (b= .44, p = .032). Two control variables were also significant; those with more education were more likely to be classified as “resilient” in extraversion (b= -.10, p < .001). Those who had been in longer relationships were more likely to be classified as “non- resilient” (although this effect was marginal, b= .21, p = .049). Model 2: Actor/partner life events and spousal support/strain. I repeated the support and strain analyses as described in the previous sections. The results of this model are reported in Table 63. Of these, one main was effect significant; those with more spousal support were more likely to be classified as “resilient” (b= -.92, p < .001). Additionally, there was an interaction of support and an actor’s chronic illness such that those who experienced the onset of a new illness and had higher support were more likely to be classified as “non-resilient” (although this effect was marginal, b= .67, p = .04). Neuroticism Growth mixture model The fit indices for each class solution for neuroticism are displayed in Table 57. As with other traits, the 3 and 4 class solutions contained produced subclasses that contained < 5% of the sample. When examining interpretability of classes, the 3-class solution identified a subclass that had a higher intercept than the other 2 classes (i.e., 3.03 vs 1.93 and 2.47). The AIC, BIC, and 229 entropy also suggested that the 3-class solution was a better fit. However, this class contained <4% of the sample. In order to maintain the more accurate estimates afforded by the 3-class solution, I maintained this solution and modified to class membership so that the participants in this rare class would be reassigned to one of the other two groups (based on their next highest probability of membership). This resulted in two subgroups. The first (coded as Class 1, the “resilient” class) started slightly lower in neuroticism and remained stable (i= 1.92, s=-.007, p = .267) and another subclass (coded as Class 2, the “non-resilient” class) which started slightly higher in neuroticism and increased slightly over time (i= 2.47, s= .033 p = .007). Multilevel logistic regression Model 1: Actor/partner life events. I repeated the regression procedure outlined in above sections. The results of this model are reported in Table 64. Of the effects examined, two were predicative of being classified as “resilient” in neuroticism: the actor effect of a new job (b= -.18, p = .035) and the partner effect of retirement (b= -.21, p = .005). One effect was predictive of being classified as “non-resilient” in neuroticism: the actor effect of a new chronic illness (b= .45, p < .001). Five control variables were also significant; older people (b= -.03), women (b= -.19), people of color (b= -.47), those with more education (b= -.07), and those with more wealth (b= -.11) were all more likely to be classified as “resilient” in neuroticism (all ps < .001). Model 2: Actor/partner life events and spousal support/strain. I repeated the support and strain analyses as described in the previous sections. The results of this model are reported in Table 65. Of these, one main was effect significant; those with more spousal support were more likely to be classified as “resilient” (b= -.37, p = .017). No interactions were significant. 230 Openness Growth mixture model When attempting to find subclasses of change for openness, 2, 3, and 4 class solutions each generated subclasses with ≤ 1% of the sample. Because, once again, model fit did not vary dramatically (although entropy was higher for smaller class solutions), I ultimately decided a 1- class solution best fit this data. Of course, without distinct classes to predict, I could not complete the logistic regression. 231 Table 57. Fit indices for 1-4 class solutions of Big 5 personality traits; HRS sample. Trait Agreeableness Conscientiousness Extraversion Neuroticism Class Solution AIC BIC Entropy N per Class (% of sample) 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 37128.242 37195.923 NA 35279.532 35369.772 0.739 34589.901 34702.701 0.733 33641.722 33777.082 0.757 36198.007 36265.691 NA 35172.325 35262.571 0.656 34481.792 34594.599 0.686 34251.173 34386.541 0.731 43032.065 43099.751 NA 42658.059 42748.307 0.865 42044.575 42109.717 0.676 41883.04 42018.412 0.651 49873.922 49941.595 NA 49510.5 49600.731 0.655 49360.011 49472.799 0.662 232 13630 (100%) 2899 (21.3%) 10731( 78.7%) 2798 (20.5%) 507 (3.7%) 10325 (75.8%) 615 (4.5%) 3595 (26.4%) 523 (3.8%) 8897 (65.3%) 13636 (100%) 2315 (17.0%) 11321 (83.0%) 2089 (15.3%) 610 (4.5%) 10946 (80.3%) 95 (.01%) 2544 (18.7%) 502 (3.7%) 10495 (77.0%) 13639 (100%) 195 (1.4%) 13444 (98.6%) 3031 (22.2%) 193 (1.4%) 10415 (76.3%) 297 (2.2%) 2743 (20.1%) 224 (1.6%) 10375 (76.1%) 13619 (100%) 12488 (91.7%) 1131 (8.3%) 12288 (90.2%) Table 57 (cont’d) Openness 4 1 2 3 4 49366.008 49501.354 0.732 43503.456 43571.132 NA 43265.607 43355.841 0.915 43079.05 43191.842 0.867 43040.511 43175.862 0.886 531 (3.9%) 800 (5.9%) 12287 (90.2%) 0 (0%) 544 (4.0%) 788 (5.8%) 13623 (100%) 94 (.07%) 13529 (99.3%) 13368 (98.1%) 132 (1.0%) 123 (0.9%) 113 (0.8%) 132 (1.0%) 15 (0.1%) 13363 (98.1%) 233 Table 58. Multilevel logistic regression using actor and partner life events to predict membership in a "non-resilient" class for agreeableness; HRS sample. Intercept Moving Birth of child Death of child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent died Partner: Parent died Actor: New job Partner: New job Actor: Retired Partner: Retired Actor: Unemployment Partner: Unemployment Age Gender Person of color Education Wealth b Exponen- tiated b 0.730 - -0.022 -0.015 0.106 0.978 0.985 1.112 Odds of class membership (%) - -2.176% -1.489% 11.182% LB 0.098 -0.122 -0.226 -0.059 UB 1.361 0.079 0.196 0.271 SE 0.322 0.051 0.108 0.084 p 0.024 0.672 0.890 0.207 -0.101 0.904 -9.607% -0.196 -0.005 0.049 0.039 0.002 1.002 0.200% -0.092 0.097 0.048 0.960 0.161 1.175 17.468% -0.069 0.390 0.117 0.171 0.022 1.022 2.224% -0.222 0.267 0.125 0.859 -0.059 0.943 -5.729% -0.203 0.085 0.074 0.422 0.094 1.099 9.856% -0.056 0.245 0.077 0.220 -0.180 0.835 -16.473% -0.298 -0.062 0.060 0.003 -0.029 0.210 -0.044 0.030 0.005 0.971 1.234 0.957 1.030 1.005 -2.858% 23.368% -4.305% 3.045% 0.501% -0.143 0.097 -0.155 -0.071 -0.092 0.086 0.324 0.067 0.130 0.102 0.059 0.626 0.058 <.001 0.434 0.057 0.564 0.051 0.917 0.049 0.061 1.063 6.290% -0.167 0.288 0.116 0.603 -0.300 0.002 -0.584 0.044 0.050 0.002 0.741 1.002 0.558 1.045 1.051 1.002 -25.918% 0.200% -44.234% 4.498% 5.127% 0.200% 234 -0.516 -0.004 -0.631 -0.073 0.033 -0.031 -0.084 0.008 -0.537 0.162 0.066 0.035 0.007 0.110 0.551 0.003 0.000 0.024 0.060 0.458 0.009 <.001 0.896 0.017 Table 58 (cont’d) Length of marriage Note. "Resilient" class (i.e., those who increased in agreeableness) used as reference group. Negative odds of class membership indicate increased odds of class membership in this group. -6.667% -0.129 -0.069 -0.009 0.933 0.031 0.024 235 Table 59. Multilevel logistic regression using actor and partner life events and spousal support and strain to predict membership in a "non-resilient" class for agreeableness; HRS sample. Intercept Moving Birth of child Death of child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent died Partner: Parent died Actor: New job Partner: New job Actor: Retired Partner: Retired Actor: Unemployment Partner: Unemployment Age Gender Person of color Education Wealth Length of marriage Spousal support Spousal strain b 0.713 0.013 -0.061 0.188 Exponen- tiated b - 1.013 0.941 1.207 Odds of class membership (%) - 1.308% -5.918% 20.683% UB LB -0.119 1.545 -0.104 0.131 -0.342 0.220 -0.002 0.378 p SE 0.425 0.093 0.060 0.822 0.143 0.670 0.097 0.052 -0.022 0.978 -2.176% -0.139 0.096 0.060 0.716 0.070 1.073 7.251% -0.045 0.186 0.059 0.231 0.242 1.274 27.379% -0.020 0.504 0.134 0.070 -0.028 0.972 -2.761% -0.320 0.265 0.149 0.852 0.000 1.000 0.000% -0.175 0.176 0.089 0.996 0.174 1.190 19.006% -0.177 -0.029 0.217 -0.047 0.029 0.039 0.207 -0.290 0.004 -0.661 0.000 0.042 -0.024 -0.045 0.557 -0.027 -16.222% -2.858% 24.234% -4.591% 2.942% 3.977% 22.998% -25.174% 0.401% -48.367% -0.010% 4.289% -2.371% -4.400% 74.543% -2.664% 0.838 0.971 1.242 0.954 1.029 1.040 1.230 0.748 1.004 0.516 1.000 1.043 0.976 0.956 1.745 0.973 236 -0.011 0.360 - -0.329 0.024 -0.175 0.118 0.069 0.365 -0.192 0.097 -0.094 0.153 -0.079 0.157 -0.117 0.531 -0.593 0.013 -0.003 0.012 - 0.600 -0.722 -0.156 0.156 0.021 0.063 -0.065 0.018 -0.122 0.032 0.430 0.684 -0.145 0.090 0.095 0.065 0.078 0.023 0.075 0.701 0.075 0.004 0.074 0.520 0.063 0.641 0.060 0.517 0.165 0.210 0.155 0.060 0.004 0.259 0.031 0.000 0.080 0.999 0.011 <.001 0.021 0.264 0.039 0.250 0.065 0.000 0.060 0.646 0.868 0.978 -0.141 -0.022 -13.151% -0.394 0.111 Table 59 (cont’d) Actor: Parent died*Support Actor: New job*Support Partner: Unemployment*Support Actor: Parent died*Strain Actor: New job*Strain Partner: Unemployment*Strain Note. "Resilient" class (i.e., those who increased in agreeableness) used as reference group. Negative odds of class membership indicate increased odds of class membership in this group. -0.307 0.193 -0.093 0.395 0.128 0.655 0.124 0.224 -5.541% 16.300% -0.057 0.151 -0.774 0.315 -0.473 0.541 -0.279 0.235 0.945 1.163 0.259 0.896 0.278 0.408 0.131 0.867 0.129 0.273 -20.547% -2.176% 3.458% -0.230 1.035 0.034 0.795 237 Table 60. Multilevel logistic regression using actor and partner life events to predict membership in a "non-resilient" class for conscientiousness; HRS sample. b -0.440 -0.020 0.008 0.088 Exponen- tiated b - 0.980 1.008 1.092 Odds of class membership (%) - -1.980% 0.803% 9.199% LB -1.092 -0.128 -0.210 -0.078 UB SE p 0.213 0.333 0.089 0.055 0.227 0.112 0.254 0.085 0.187 0.722 0.942 0.298 0.214 1.239 23.862% 0.107 0.320 0.054 <.001 0.202 1.224 22.385% 0.096 0.307 0.054 <.001 0.061 1.063 6.290% -0.207 0.328 0.137 0.657 -0.069 0.933 -6.667% -0.328 0.191 0.132 0.603 0.001 1.001 0.100% -0.156 0.157 0.080 0.995 -0.017 0.983 -1.686% -0.176 0.141 0.081 0.829 -0.056 0.946 -5.446% -0.192 0.081 0.070 0.424 -0.073 0.930 -7.040% -0.204 0.059 0.067 0.278 -0.301 0.740 -25.992% -0.431 -0.171 0.066 <.001 0.019 1.019 1.918% -0.105 0.144 0.064 0.763 0.116 1.123 12.300% 0.006 0.226 0.056 0.038 0.031 1.031 3.149% -0.077 0.139 0.055 0.572 0.102 1.107 10.738% -0.155 0.359 0.131 0.437 -0.012 0.016 0.216 0.030 0.988 1.016 1.241 1.030 -1.193% 1.613% 24.110% 3.045% -0.269 0.009 0.165 -0.098 0.245 0.131 0.023 0.003 0.267 0.026 0.159 0.066 0.928 <.001 <.001 0.645 238 Intercept Moving Birth of child Death of child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent died Partner: Parent died Actor: New job Partner: New job Actor: Retired Partner: Retired Actor: Unemployment Partner: Unemployment Age Gender Person of color Table 60 (cont’d) 0.916 0.882 -0.088 -0.126 -8.424% -11.839% Education Wealth Length of marriage Note. “Resilient” class (i.e., those who remained stable in conscientiousness) used as reference group. Negative odds of class membership indicate increased odds of class membership in this group. -0.071 0.009 -0.094 0.016 -0.105 -0.159 0.118 0.033 5.548% -0.011 1.055 0.054 0.000 <.001 0.102 239 Table 61. Multilevel logistic regression using actor and partner life events and spousal support and strain to predict membership in a "non-resilient" class for conscientiousness; HRS sample. Intercept Moving Birth of child Death of child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent died Partner: Parent died Actor: New job Partner: New job Actor: Retired Partner: Retired Actor: Unemployment Partner: Unemployment Age Gender Person of color Education Wealth b -0.384 -0.039 0.018 0.081 Exponen- tiated b - 0.962 1.018 1.084 Odds of class membership (%) - -3.825% 1.816% 8.437% LB UB SE p -1.278 -0.167 -0.271 -0.109 0.509 0.456 0.089 0.065 0.306 0.147 0.271 0.097 0.399 0.550 0.905 0.406 0.143 1.154 15.373% 0.013 0.273 0.066 0.031 0.186 1.204 20.442% 0.058 0.314 0.065 0.004 0.032 1.033 3.252% -0.270 0.334 0.154 0.838 -0.114 0.892 -10.774% -0.409 0.180 0.150 0.447 0.042 1.043 4.289% -0.143 0.227 0.094 0.655 -0.039 0.962 -3.825% -0.226 0.149 0.096 0.687 -0.038 0.963 -3.729% -0.211 0.134 0.088 0.664 -0.065 -0.322 0.002 0.159 0.028 0.937 0.725 1.002 1.172 1.028 -6.293% -27.530% -0.229 -0.488 0.100 0.084 -0.156 0.085 0.200% 17.234% 2.840% -0.157 0.026 -0.103 0.162 0.081 0.293 0.068 0.160 0.067 0.440 <.001 0.978 0.019 0.673 0.102 1.107 10.738% -0.243 0.448 0.176 0.562 -0.210 0.017 0.288 -0.019 -0.082 -0.124 0.811 1.017 1.334 0.981 0.921 0.883 -18.942% 1.715% 33.376% -1.882% -7.873% -11.662% 240 -0.568 0.008 0.224 -0.186 -0.103 -0.167 0.148 0.183 0.025 0.004 0.351 0.033 0.148 0.085 -0.061 0.011 -0.082 0.022 0.251 <.001 0.000 0.826 <.001 <.001 0.237 0.019 0.031 0.011 1.267 1.019 1.031 -0.053 3.149% 1.918% 26.744% 0.595 1.221 0.463 0.115 0.115 0.043 -0.519 0.200 -0.759 -0.032 -40.488% 22.140% -0.279 0.123 0.432 0.119 Table 61 (cont’d) Length of marriage Spousal support Spousal strain Actor: New chronic illness*Support Partner: New chronic illness*Support Actor: New job*Support Actor: Retired*Support Actor: New chronic illness*Strain Partner: New chronic illness*Strain Actor: New job*Strain Actor: Retired*Strain Note. "Resilient" class (i.e., those who remained stable in conscientiousness) used as reference group. Negative odds of class membership indicate increased odds of class membership in this group. 0.250 0.113 0.285 0.108 0.078 0.132 0.149 0.107 0.140 0.107 0.156 0.140 0.244 0.115 -11.041% -16.389% -6.761% -5.918% 7.573% 2.840% -0.179 -0.437 -0.117 -0.061 -0.070 -0.270 -0.138 -0.206 -0.194 -0.280 -0.391 0.890 1.028 0.941 1.076 0.932 0.836 0.028 0.073 0.469 <.001 0.092 0.040 0.867 0.401 0.805 0.497 0.570 0.173 0.515 241 Table 62. Multilevel logistic regression using actor and partner life events to predict membership in a "non-resilient" class for extraversion; HRS sample. Odds of class membership (%) Exponen- tiated b UB SE p b -4.548 0.066 0.098 0.133 LB -6.408 -0.265 -0.496 -0.355 0.412 0.758 0.204 0.066 0.439 0.175 0.839 0.476 0.133 0.819 0.176 0.038 0.208 0.234 0.550 0.092 1.510 1.551 1.610 1.096 0.789 0.032 0.007 0.019 0.695 -0.682 -0.237 -0.366 9.636% 55.116% 60.962% 50.983% -21.101% <.001 0.697 0.746 0.592 0.225 -0.019 0.949 0.169 0.303 0.249 - 1.068 1.103 1.142 -2.688 0.396 0.692 0.621 - 6.823% 10.296% 14.225% Intercept Moving Birth of child Death of child Actor: New chronic illness Partner: New chronic illness Actor: Positive health change Partner: Positive health change Actor: Parent died Partner: Parent died Actor: New job Partner: New job Actor: Retired Partner: Retired Actor: Unemployment Partner: Unemployment Age Gender Person of color Education Wealth Length of marriage Note. "Resilient" class (i.e., those who remained stable in extraversion) used as reference group. Negative odds of class membership indicate increased odds of class membership in this group. Negative health change excluded as a life event for quasi-complete separation. 11.963% 1.308% 6.930% 34.447% -9.335% -7.226% 0.113 0.013 0.067 0.296 -0.098 -0.075 -0.626 -0.007 -0.095 -0.080 -0.147 -0.175 0.852 0.034 0.229 0.672 -0.048 0.026 1.120 1.013 1.069 1.344 0.907 0.928 0.377 0.010 0.083 0.192 0.025 0.051 2.122% 21.653% -12.278% -0.360 -0.143 -0.463 0.021 0.196 -0.131 1.021 1.217 0.877 0.402 0.534 0.201 0.194 0.173 0.169 25.232% -1.882% -0.167 -0.413 1.252 0.981 0.617 0.376 0.200 0.201 0.764 0.196 0.416 0.123 <.001 0.146 0.912 0.257 0.440 0.260 0.926 22.753% -7.226% -0.860 -0.075 0.852 0.049 0.296 0.928 1.228 0.104 0.410 0.001 0.227 0.205 0.401 0.711 242 Table 63. Multilevel logistic regression using actor and partner life events and spousal support and strain to predict membership in a "non-resilient" class for conscientiousness; HRS sample. Odds of class membership (%) - Exponen- tiated b UB LB SE p 0.967 1.198 1.270 -3.343% 19.842% 26.998% -1.278 -0.167 -0.271 -0.109 0.509 0.089 0.306 0.271 0.456 0.065 0.147 0.097 0.399 0.550 0.905 0.406 b -4.964 - -0.034 0.181 0.239 Intercept Moving Birth of child Death of child Actor: New chronic illness Partner: New chronic illness Actor: Positive health change Partner: Positive health change Actor: Parent died Partner: Parent died Actor: New job Partner: New job Actor: Retired Partner: Retired Actor: Unemployment Partner: Unemployment Age Gender Person of color Education Wealth Length of marriage Spousal support Spousal strain Actor: New chronic illness*Support 0.486 1.626 62.580% 0.013 0.273 0.066 0.031 0.495 1.640 64.050% 0.058 0.314 0.065 0.004 0.428 1.534 53.419% -0.143 0.227 0.094 0.655 -0.333 -0.428 0.392 -0.001 0.152 0.213 -0.168 0.717 0.652 1.480 0.999 1.164 1.237 0.845 -28.323% -34.819% -0.226 -0.211 0.149 0.134 0.096 0.088 47.994% -0.100% 16.416% 23.738% -15.465% -0.229 -0.488 -0.157 0.026 -0.103 0.100 -0.156 0.162 0.293 0.160 0.084 0.085 0.081 0.068 0.067 0.687 0.664 0.440 <.001 0.978 0.019 0.673 0.254 1.289 28.917% -0.243 0.448 0.176 0.562 0.070 0.024 0.049 0.321 -0.081 -0.092 0.126 -0.916 -0.022 1.073 1.024 1.050 1.379 0.922 0.912 1.134 0.400 0.978 7.251% 2.429% 5.022% 37.851% -7.781% -8.789% -0.568 0.008 0.224 -0.186 -0.103 -0.167 0.148 0.025 0.351 0.148 -0.061 -0.082 13.428% -59.988% -2.176% -0.053 -0.759 -0.032 0.115 -0.279 0.432 0.183 0.004 0.033 0.085 0.011 0.022 0.043 0.123 0.119 0.251 <.001 0.000 0.826 <.001 <.001 0.469 <.001 0.092 0.669 1.952 95.228% 0.011 0.463 0.115 0.040 243 0.867 0.244 0.115 0.760 0.406 1.501 2.138 -0.206 50.080% 113.828% Table 63 (cont’d) Partner: New chronic illness*Support Actor: Positive health change*Support Actor: New chronic illness*Strain Partner: New chronic illness*Strain Partner: Positive health change*Strain Note. "Resilient" class (i.e., those who remained stable in extraversion) used as reference group. Negative odds of class membership indicate increased odds of class membership in this group. Negative health change excluded as a life event for quasi-complete separation. -12.453% -15.380% 19.842% -0.194 -0.270 -0.391 -0.138 -0.133 -0.167 1.198 0.875 0.846 0.181 0.140 0.149 0.113 0.250 0.285 0.108 0.156 0.107 0.497 0.401 0.570 0.805 244 Table 64. Multilevel logistic regression using actor and partner life events to predict membership in a "non-resilient" class for neuroticism; HRS sample. Odds of class membership (%) Exponen- tiated b LB UB SE p - 0.980 0.790 0.817 - -1.980% -21.022% -18.291% 0.160 -0.169 -0.567 -0.446 1.903 0.129 0.096 0.042 0.445 0.076 0.169 0.125 0.020 0.794 0.163 0.105 b 1.031 -0.020 -0.236 -0.202 0.446 1.562 56.205% 0.293 0.599 0.078 <.001 0.069 1.071 7.144% -0.077 0.215 0.075 0.354 0.090 1.094 9.417% -0.319 0.500 0.209 0.666 0.101 1.106 10.628% -0.274 0.475 0.191 0.598 0.173 1.189 18.887% -0.033 0.379 0.105 0.099 -0.035 0.966 -3.439% -0.252 0.183 0.111 0.755 -0.048 0.953 -4.687% -0.219 0.123 0.087 0.583 -0.038 0.963 -3.729% -0.208 0.131 0.087 0.659 -0.177 0.838 -16.222% -0.342 -0.013 0.084 0.035 -0.005 0.087 0.995 1.091 -0.499% 9.090% -0.168 -0.059 0.158 0.233 0.083 0.074 0.954 0.242 -0.214 0.807 -19.265% -0.364 -0.065 0.076 0.005 0.114 1.121 12.075% -0.206 0.433 0.163 0.486 0.221 -0.028 -0.185 1.247 0.972 0.831 24.732% -2.761% -16.890% -0.086 -0.037 -0.255 0.528 -0.018 -0.115 0.156 0.005 0.036 0.158 <.001 <.001 245 Intercept Moving Birth of child Death of child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent died Partner: Parent died Actor: New job Partner: New job Actor: Retired Partner: Retired Actor: Unemploymen t Partner: Unemploymen t Age Gender -0.473 -0.071 -0.109 Table 64 (cont’d) Person of color Education Wealth Length of marriage Note. "Resilient" class (i.e., those who remained stable in neuroticism) used as reference group. Negative odds of class membership indicate increased odds of class membership in this group. -37.687% -6.854% -10.327% -0.660 -0.094 -0.151 -0.285 -0.047 -0.066 0.623 0.931 0.897 0.096 0.012 0.022 <.001 <.001 <.001 3.666% -0.051 0.416 1.037 0.036 0.124 0.045 246 Table 65. Multilevel logistic regression using actor and partner life events and spousal support and strain to predict membership in a "non-resilient" class for neuroticism; HRS sample. Odds of class membership (%) Exponen- tiated b UB LB SE p b 0.920 -0.032 -0.131 -0.302 - 0.969 0.877 0.739 - -3.149% -12.278% -26.066% -0.321 -0.216 -0.577 -0.601 2.160 0.152 0.316 -0.004 0.633 0.146 0.094 0.736 0.228 0.566 0.152 0.047 Intercept Moving Birth of child Death of child Actor: New chronic illness Partner: New chronic illness Actor: Negative health change Partner: Negative health change Actor: Positive health change Partner: Positive health change Actor: Parent died Partner: Parent died Actor: New job Partner: New job Actor: Retired Partner: Retired Actor: Unemployment Partner: Unemployment Age Gender Person of color Education Wealth 0.252 1.287 28.660% 0.056 0.449 0.100 0.012 0.041 1.042 4.185% -0.147 0.230 0.096 0.667 -0.034 0.967 -3.343% -0.487 0.419 0.231 0.883 0.136 1.146 14.568% -0.293 0.564 0.219 0.535 0.213 1.237 23.738% -0.042 0.468 0.130 0.102 -0.001 0.999 -0.100% -0.260 0.258 0.132 0.994 -0.046 0.955 -4.496% -0.277 0.185 0.118 0.697 -0.020 -0.314 -0.045 0.176 0.980 0.731 0.956 1.192 -1.980% -26.948% -0.251 -0.554 0.210 -0.073 0.118 0.863 0.123 0.011 -4.400% 19.244% -0.268 -0.003 0.178 0.356 0.114 0.694 0.092 0.055 -0.219 0.803 -19.668% -0.416 -0.021 0.101 0.030 -0.032 0.969 -3.149% -0.528 0.465 0.253 0.901 0.368 -0.026 -0.118 -0.534 -0.095 -0.092 1.445 0.974 0.889 0.586 0.909 0.912 44.484% -2.566% -11.130% -41.374% -9.063% -8.789% 247 -0.048 -0.038 -0.210 -0.788 -0.126 -0.152 0.784 -0.013 -0.025 -0.281 -0.063 -0.033 0.212 0.083 0.007 <.001 0.047 0.013 0.129 <.001 0.016 <.001 0.030 0.002 0.099 0.224 1.104 0.920 -0.083 -0.401 -0.026 -0.079 -7.965% 10.407% 0.064 0.122 0.155 0.017 0.153 0.137 0.691 1.256 -0.370 0.228 -0.674 -0.072 -0.065 0.527 -30.927% 25.609% Table 65 (cont’d) Length of marriage Spousal support Spousal strain Actor: New chronic illness*Support Actor: New job*Support Partner: Retired*Suppor t Actor: New chronic illness*Strain Actor: New job*Strain Partner: Retired*Strain Note. "Resilient" class (i.e., those who remained stable in neuroticism) used as reference group. Negative odds of class membership indicate increased odds of class membership in this group. 0.156 0.064 0.172 0.644 0.164 0.529 0.162 0.607 0.187 0.384 0.155 0.051 35.391% 17.704% 33.509% -9.787% -7.596% -0.017 -0.415 -0.203 -0.424 -0.001 -0.103 0.924 1.335 1.354 1.177 0.902 0.596 0.303 0.289 0.163 0.257 0.529 0.218 0.606 0.235 248