Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Social interactions, wellbeing and health in the oldest old: what can we learn from daily life approaches? Booi, Laura Marie 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2011_fall_booi_laura.pdf [ 223.55kB ]
Metadata
JSON: 24-1.0072122.json
JSON-LD: 24-1.0072122-ld.json
RDF/XML (Pretty): 24-1.0072122-rdf.xml
RDF/JSON: 24-1.0072122-rdf.json
Turtle: 24-1.0072122-turtle.txt
N-Triples: 24-1.0072122-rdf-ntriples.txt
Original Record: 24-1.0072122-source.json
Full Text
24-1.0072122-fulltext.txt
Citation
24-1.0072122.ris

Full Text

Social Interactions, Wellbeing and Health in the Oldest Old: What Can We Learn from Daily Life Approaches?  by  Laura Marie Booi  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF ARTS in The Faculty of Graduate Studies (Psychology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) August, 2011 ©Laura Marie Booi, 2011  ii Abstract Canada, like many Western societies, has an aging population. Past research indicates that social interactions are meaningfully associated with physical and mental health. Unfortunately, very few studies have included individuals older than 85 years. This is important because the Oldest Old are the fastest growing segment of the Canadian population. This study is based on a subsample of the Berlin Aging Study (N = 83; Mean age = 81.1 years) who participated in an intensive, one-week time-sampling module. I examined how older adults regulate their social interactions by looking at two competing theories, the Convoy Model and the Socioemotional Selectivity Theory. I also examined how daily social interactions are associated with concurrent affective experiences and physical symptoms. This study found that even when controlling for cognition and overall health, age was associated with a decrease in social interactions in the present sample of older adults. Results showed that a more limited future time perspective in this sample was associated with a greater amount of time spent alone as well as a greater number of socially unpreferred situations. Furthermore, a greater amount of time spent alone was associated with lower levels of daily wellbeing. Results also showed that time spent alone was a risk factor for mortality. I discuss possible explanations as to why my findings complement past research by showing less favorable associations between social interactions and wellbeing in the Oldest Old.  iii Foreword To answer my hypotheses examining social interactions, wellbeing and health in the Oldest Old I drew on data independently collected in two subprojects of the BASE (Berlin Aging Study). One subproject was on the self in old age (Smith & Baltes, 1999; Smith & Freund, 2002), which collected data from all BASE participants; the other was on everyday competence, and it collected time-use information from a subsample of BASE participants (M. M. Baltes, Maas, Wilms, Borchelt, & Little, 1999; Klumb & Baltes, 1999b). In my study I reorganize and combine these two data sets to answer my unique question to contribute to the invaluable body of literature on health in the Oldest Old.  iv Table of Contents Abstract…………………………………………….……………………………………………………..ii Forward …………………………………………..……………………………………………………..iii Table of Contents……………………………….……………………………………………………….iv List of Tables ………………………………….………………………………………………………..vi Acknowledgements……………………….……………………………………………………………viii Chapter 1: Introduction…………………….……………………………………………………………1 1.1  Our Aging Society …………….………………………………………………...……………….1  1.2  Social Relationships and Health…………………………………………………...……………..1  1.3 Social Relationships and Health in the Young versus Oldest Old ………………....…………….3 1.4 Mechanisms Underlying Social Relationships and Health……………………………………….4 1.5  Limited Access to Social Relationships in the Oldest Old ………...………………………….....4  1.6  Active Relationship Regulation in Old Age ………………………...…………………….……..5  1.7 Daily Social Interactions, Wellbeing and Health ………………………………………………...8 1.8  Daily Social Interactions and Long-Term Outcomes…...……………………………………....10  1.9 Additional Variables that Influence Health and Wellbeing……………………………………..10 1.10 Time Sampling Methods………………………………………………………………………...12 1.11 Summary…………………………………………………………...……………………………12 Chapter 2: Methods…………………………………………………………………………………….14 2.1 Participants……………………………………………………………………………...……….14 2.2  Design ……………………………………………………………………………………...…...14  2.3  Measures……………………………………………………………………………………...…15  Chapter 3: Results …………………………………………………………………………………...…20 3.1  Relevant Descriptive Variables…………………………………………………………………20  3.2 Future Time Perspective, Age and Social Interactions…………………………………….…....21 3.3  Social Interactions, Daily Affect and Physical Symptoms……………………………………...22  3.4 Mortality and Physical Symptoms ……………………………………………………………...24 3.5  Follow up Analyses ………………………………………………………………………...…..26  Chapter 4: Discussion …………………………………...……………………………………………….35  v 4.1  Social Interactions, Age and Future Time Perspective…………………………….…..35  4.2  Benefits of Social Interactions in the Oldest Old………………………………………38  4.3  Limitations and Strengths ……………………………………………………….……..38  4.4  Conclusion and Future Directions ……………………………………………………..39  References……..…………………………………………………………...…………………………….41  vi List of Tables 1  Hypotheses…………………………………………………………………………………………..13  2  Descriptive Variables……………...………………………………………………………………....21  3  Hierarchical Linear Models Predicting Positive Affect and Negative Affect, with Age and Future Time Perspective………………………………………………………………………………….….22  4  Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective from Time Spent Alone………………………………………..23  5  Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective from Time Spent in Socially Unpreferred Interactions………………………………………………………………………………..………….24  6  Survival Analysis Using Cox Regression Examining Time Spent Alone, Controlling for Functional Health, Age, Gender and Cognition ………………………………………………..………………..25  7  Survival Analysis Using Cox Regression Examine Time Spent in Unpreferred Social Situations, Controlling for Functional health, Age, Gender and Cognition ………………………………….....26  8  Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective from Time Spent Alone………………………………………..27  9  Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective and Time Spent in Socially Unpreferred Interactions…………………………………………………………………………………………...28  10 Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective and Overall Health from Time Spent Alone……………………………………………………………………………………….………...29 11 Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective and Overall Health from Time Spent in Socially Unpreferred Interactions…………………………………………………………………………………………...30 12 Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective, Cognition and Time Spent Alone …………………………….31 13 Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective, Cognition and Time Spent in Unpreferred Social Interactions…………………………………………………………………………………………...32  vii  14 Hierarchical Linear Models Predicting Being Alone and Being in Unpreferred Social Interactions with Age, Future Time Perspective and Overall Health……………………………………………..33 15 Hierarchical Linear Models Predicting Being Alone and Being in Unpreferred Social Interactions with Age, Future Time Perspective and Cognition…………………………………………………..33 16 Hierarchical Linear Models Predicting Being Alone and Being in Unpreferred Social Interactions with Age, Future Time Perspective and Social Network……………………………………...……..34  viii Acknowledgments  Particular thanks to my fellow cohort of graduate students at UBC who ensured that we all excelled together. Thank you also to my dear friends Jameson and Daniels for your constant guidance and support. Thank you to my thesis committee members for reading many drafts and gently giving me invaluable advice. Special thanks are owed to my partner who has listened to my presentations, read over countless essays and supported me throughout my many years of education since secondary school. I offer my sincerest gratitude to Dr. Christiane Hoppmann. With her guidance, patience and knowledge my experience as a graduate student and young researcher has been both a pleasure and a learning experience.  Thank you.  1 Chapter 1: Introduction 1.1 Our Aging Society Canada, like many Western societies, has an increasingly aging population (Health Canada, 2008). The proportion of Canadian adults above the age of 65 years is estimated to double within the next ten years, while the number of Canadians above the age of 85 years is going to quadruple within the same time-frame (Statistics Canada, 2007). Empirical research has revealed substantial, average decreases in functioning with aging, but also tremendous variability in aging outcomes, highlighting that some older adults seem to accumulate losses while others age relatively well (Baltes, Lindenberger, & Staudinger, 2006). A proper understanding of the mechanisms involved in successful aging will contribute to the health and wellbeing of a large segment of the Canadian population. Rowe and Kahn (1997) describe successful aging as having three main components: a low likelihood of disease and disease-related disability, active engagement with life and high cognitive and physical functional capacity. This study examined the daily life processes that are involved in successful aging, specifically social engagement and interactions.  1.2 Social Relationships and Health Past research indicates that social relationships are closely associated with health (Uchino, Cacioppo, & Kiecolt-Glaser, 1996; Holtzman, Newth, & DeLongis, 2007; Hoppmann & Gerstorf, 2009). For example, isolation and lack of connectedness with others have long been recognized as predictors of morbidity and mortality (Durkheim,  2 1951). It has also been shown that social relationships and social support have positive effects on the health and well-being for adults of all ages (Antonucci & Jackson, 1987). Recent research has shown that strong social relationships have beneficial effects on survival rates in populations of the Oldest Old (Giles, Glonek, Luszcz, & Andrews, 2005). One underlying mechanism may be that seniors with strong social relationships have high levels of social support and experience less stress when they confront a negative experience and that they may also cope with a negative experience more successfully than those with less social support. Social support may thus lower the likelihood of illness (Krohne, 2005) and reduce the risk of all-cause mortality (Berkman et al., 1979). Hence, there is increasing evidence that social interactions help individuals prepare for, cope with, and recover from many of the demands of life that are associated with aging (Lazarus & DeLongis, 1983). However, very few studies have enrolled participants above the age of 80 years and the available evidence is primarily based on cross-sectional data or longitudinal data with relatively long measurement intervals, thus limiting the ability to pinpoint the specific daily life mechanisms that account for differences in aging outcomes (Nesselroade, 1991). In particular, few studies have examined the relationship between social relationships and health in the Oldest Old (Baltes, & Smith, 2003). Researchers have started to call for more in-depth investigations of daily life processes in old age using intense measurement designs. Such approaches may be particularly useful to increase our understanding of the mechanisms that link social relationships, wellbeing, and health among the Oldest Old. This study thus focuses on the daily life approaches linking social interactions, wellbeing and health of the Oldest Old. For the purposes of this study I will  3 examine how social interactions matter to health and wellbeing. Wellbeing has been defined in numerous ways across the literature (Daver, Cummins & Stokes, 2007). The indicators that I will use to define wellbeing in this study are self-rated daily positive affect, daily negative affect and daily physical symptoms.  1.3 Social Relationships and Health in the Young versus Oldest Old There is accumulating evidence for the benefits of social relationships in midlife and early old age (Baltes, & Smith, 2003). However, the Young Old and the Oldest Old are substantially different from one another and it is not always possible to take what we have learned from the Young Old and apply it to the Oldest Old. The Oldest Old are described as those individuals who are 85 years and older (Baltes, & Smith, 2003). Just a century ago very few individuals reached the ages that the population of Canada is reaching today and thus the number of Oldest Old in our society is historically new (Cutler, Poterba, Sheiner, Summers, & Akerlof, 1990). Importantly, the Oldest Old are not only going to represent a growing segment of our population but they will also challenge the health care system. Older adults are better at regulating emotions than young adults. However, if older adults experience high levels of sustained emotional arousal they often have difficulty returning to homeostasis (Charles, 2010). One of the strengths of the proposed study is that it involves a sample of the Oldest Old to further understand the association between their social interactions, wellbeing and health processes.  4 1.4 Mechanisms Underlying Social Relationships and Health There are a number of theories examining how social relationships influence health. First, the Direct Effects hypothesis proposes a direct relationship between social relationships and well-being. It suggests that, in general, social relationships are beneficial during non-stressful as well as stressful times, and that social relationships have direct effects on adaptation (Cohen, & Wills, 1985). Second, the Buffering hypothesis maintains that the benefits of social relationships are mainly evident during periods of high stress (Cohen, & Wills, 1985). The Buffering hypothesis predicts that people who have high levels of social support will have decreased negative reactions to difficult life events (Cohen, & Wills, 1985). Third, the Matching hypothesis by Cohen and McKay (1984) states that to be beneficial, social support must match individual needs. In other words, different kinds of stressors elicit different needs and social support will be most effective if it matches those needs. For example, if someone is emotionally upset, the support they may need is a hug and a listening ear, not help with grocery shopping. Hence, there are multiple mechanisms that might explain previously observed positive associations between social relationships and health.  1.5 Limited Access to Social Relationships in the Oldest Old All of the previously mentioned hypotheses suggest that having limited social relationships is problematic for health, yet few social relationships are the norm in old age and not the exception (Baltes & Smith, 2003). There are multiple reasons why this is the case: First, death plays an important role in reducing social networks in older adults.  5 Older adults’ peers, family members, and other relations may all be well advanced in their years and at a high risk of dying (Fuller-Iglesias, Sellars & Antonucci, 2008). Second, the loss of social roles for older adults is another reason for limited social networks. While in the labor force, individuals have many opportunities to participate in important roles and engage in meaningful social interactions. However, when older adults retire, they lose these important roles as well as the interactions they have accumulated at their place of work. The loss of these social roles eliminates important social contacts in ones life (Pinquart & Schindler, 2007). Third, limitations in functional health may decrease the number of social relationships that older adults may be able to maintain. Older adults are susceptible to many functional health limitations because of their declining health. For example, older adults may not be able to drive, or easily move and therefore spend more time alone at home and not engaging in social interactions (Everard, Lach, Fisher & Baum, 1999). Hence, older adults may be at high risk of having limited access to important social interactions.  1.6 Active Relationship Regulation in Old Age Older adults cannot influence these previously mentioned reasons for declines in their social networks. However, they can actively regulate their social networks in order to maximize the benefits they get from their limited social ties. There are two very prominent models that address why and how older adults regulate their social interactions to receive the maximum amount of benefits in late life. Both of these models describe, in different ways, how older adults actively maintain and  6 regulate their social networks. The two models that this study will target are the Convoy Model (Antonnuci, 1990) and the Socioemotional Selectivity Theory (Carstensen, 1995). This study will compare the predictive effects of these two models. According to the Convoy Model, social networks are dynamic structures that change with the development of the individual (Kahn, & Antonucci, 1980). The Convoy Model predicts continuity and change in social networks as individuals’ age (Antonucci, Akiyama, & Takahashi, 2004). Hence, this model proposes that the composition of convoys changes with age. For example, whereas children’s convoys may include their parents, siblings and perhaps some close peers, middle-aged adults’ convoys typically include their spouse, close relatives, co-workers and friends. Older adults’ convoys have been shown to include their spouse, relatives, neighbors and long-term friends (Antonucci, Akiyama, & Takahashi, 2004). The Convoy Model proposes that chronological age influences relationship regulation and that older adult’s focus on their closest social relationships as represented by the inner circle of their social network. In other words, the Convoy Model proposes that changes in social relationships in old age tend to occur in the peripheral sphere of an individual’s social network because older adults do not have the energy to maintain their peripheral social relationships anymore and thus prioritize their closest social relationships. The decline in social relationships in older adulthood is therefore not arbitrary; it is an active and conscious process. Based on this model I expect that age is associated with a reduced number of distal social relationships but that close social relationships will remain. Second, Socioemotional Selectivity Theory proposes that future time perspective, or time left in life, rather than chronological age, is the important variable affecting  7 relationship regulation in old age (Carstensen, 1999). Specifically, this theory proposes that limitations in perceived time left in life lead to motivational changes that direct attention to emotional goals (Carstensen, 1999). The Socioemotional Selectivity Theory states that with less time left in life older adults prioritize emotionally meaningful social ties. They do not want to spend their limited time with people they do not care aboutinstead they capitalize on the most emotionally meaningful interactions to satisfy their immediate emotional goals. Consequently, with a more limited future time perspective, emotionally meaningful interactions assume unmatched importance (Carstensen, 1999). Appreciation for the fragility and value of human life increases the value of emotionally meaningful interactions with family and friends, which assume unprecedented importance. Socioemotional Selectivity theory states that with the perception of less time left, individuals reorganize their goals in such a way that goals related to deriving emotional meaning from social interactions are prioritized over goals that maximize longterm payoffs in the future. Therefore, different from the Convoy Model, Socioemotional Selectivity Theory predicts that with limited future time perspective, people seek more meaningful, less conflicting social interactions. Age and future time perspective are negatively correlated in healthy samples, yet this correlation is only moderate (Lang, & Carstensen 2002). Hence, researchers can still expect variability in future time perspective amongst individuals of the same age. For example, Carstensen and Fredrickson (1998) studied social preferences in young gay men, half of whom were Human Immunodeficiency Virus (HIV) positive. They found that, with age held constant, limited future time perspective was associated with an increased focus on emotionally meaningful social relationships. This study indicates that  8 the perception of limited time, as compared to chronological age, is a meaningful variable and influences the regulation of social relationships. Hence, the Convoy Model and the Socioemotional Selectivity Theory answer the question of how older adults actively regulate their social relationships in very different ways. The Convoy Model asserts that the variable that is responsible for the active regulation of social interactions is chronological age. In contrast, the Socioemotional Selective Theory proposes that it is not age, but the variable time left in life, or future time perspective that drives older adults’ active regulation of their social interactions to have fewer situations with social relationships but more meaningful social interactions. This study will examine predictions of both these two models in this sample of older adults.  1.7 Daily Social Interactions, Wellbeing and Health Participation in daily social interactions promotes multiple benefits including better physical and mental health and ultimately survival (Lennartson & Silverstein, 2001). It is not merely that older adults need to have social interactions with anyone; it is the quality of social interactions that is often the important variable. There is a tremendous difference in outcomes if an individual is spending time with high quality social relations or low quality social relationships. People with stronger social ties are shown to report higher levels of well-being (Cohen & Wills, 1985) and live longer; compared to those who are socially isolated or have low-quality social interactions (House, Landis, & Umberson, 1988). Daily social interactions have been associated with both mental and physical health indices (Kamarck, Schwartz, Shiffman, Muldoon, Sutton-Tyrrell, & Janicki, 2005;  9 Nezlek, Richardson, Green, & Schatten-Jones, 2002). In fact, when compared to a variety of other daily life stressors, interpersonal problems have been shown to be most detrimental to daily affective experiences in adult samples (Bolger, DeLongis, Kessler, & Schilling, 1989). Similarly, negative social exchanges have also been shown to be negatively associated with daily emotional experiences in older adults (August, Rook, & Newsom, 2007). Interestingly, recent evidence suggests that older adults show substantial but also lower reactivity to interpersonal tensions than young adults (Birditt, Almeida, & Fingerman, 2005). It is therefore proposed that negative daily social interactions are associated with elevated negative affect and lower positive affect. Positive social interactions in contrast are proposed to be associated with less negative affect and more positive affect. Also, negative social interactions may be more closely tied to negative affect and physical symptoms and not as much with positive affect. I suggest that social interactions will contribute to the maintenance of health and to longevity because these interactions exert their effects on a daily basis and these effects are shown to accumulate over the life course (Seeman et al. 2002; Carstensen et al. 2011). Negative social interactions have also been associated with increases in physical symptoms. For example, Holtzman and DeLongis (2007) found that individuals vary in the extent to which they experience pain throughout the day and that variations in pain catastrophizing are associated with subsequent levels of pain and negative affect. Interestingly, they also found that positive spousal interactions can help attenuate the negative association between pain catastrophizing, pain, and negative affect. Past research also supports the notion that positive responses from others can attenuate the detrimental effects of maladaptive coping (Marin, Holtzman, DeLongis, & Robinson,  10 2007). Therefore, in the present study, I assume that older adults’ daily social interactions are also associated with the experience of physical symptoms. Specifically I expect that older adults who report negative daily social interactions report high levels of physical symptoms.  1.8 Daily Social Interactions and Long-Term Outcomes Social interactions may also impact physical and mental health in the long term (Uchino, Cacioppo, & Kiecolt-Glaser, 1996; Cohen, 2004; Hoppmann, & Gerstorf, 2009). For example, a study by Maier and Klumb (2005) found that time spent with friends was associated with a survival advantage in older adults as compared to other types of activities. Interestingly, research has found that social activities are as powerful as physical activities in lowering the risk of death (Glass, Mendes de Leon, Marottoli, & Berkman, 1999). One of the benefits of examining associations between daily social interactions and health in the context of a long-term longitudinal study is that I can link daily social activities to long-term health outcomes. Specifically, the proposed study will explore the association between daily social interactions and mortality in the Oldest Old.  1.9 Additional Important Variables that Influence Health and Wellbeing This study will also examine a number of other factors that are known to be associated with social interactions, health and wellbeing: functional health, cognition, and number of diagnosed diseases (Baltes & Mayer, 1999). The first variable that will be considered in this study is functional health. Older adults typically accumulate limitations to their functional health. Such limitations play a key role in restricting their opportunities  11 for social interactions. Hence, older adults with low levels of functional health may spend more time at home alone because of these limitations (Everard, Lach, Fisher, & Baum, 1999). This can negatively impact the number of daily social interactions older adults’ engage in. A second variable that must be considered when examining social interactions, wellbeing, and health in older adults is cognition. Cognition influences the health and wellbeing of older adults in many different ways (Seeman, Lusignolo, Albert, & Berkman, 2001). For example, cognitive impairment may limit access to social interactions. There is a growing body of literature that suggests that higher engagement in social interactions alleviates decline in perceptual speed (Ghisletta, Bickel & Lovden, 2006). In order to provide a meaningful interpretation of the association between older adults’ daily social interactions, wellbeing and health I therefore need to account for the influence of cognition in old age. The third variable that will be considered in this study is number of diagnosed diseases. This information was collected at measurement point three in the parent BASE study. Diagnoses were determined in the course of consensus conferences by the research physician and psychiatrist, based on a standardized summary of clinical findings from all diagnostic procedures. Diagnosed moderate and severe illnesses were summed up to form the variable ‘‘number of diagnoses’’. The number of diagnosed diseases will be used as a proxy for overall health status in the following analyses.  12 1.10 Time-Sampling Methods This study will provide insight into the association between daily social interactions, wellbeing and health in older adults. In spite of the labor-intensive and timeconsuming nature of this type of research, it offers many advantages (Holtzman, & DeLongis, 2007). By using time-sampling methods, I are able to get unique insights into older adults’ daily lives, their social interactions and their current health status. Specifically, time-sampling methods allow for valuable assessments of within-person associations in daily processes (Almeida, 2005). In the current study, participants were randomly beeped five times a day. Because these time points were randomly chosen, I can assume that our data portray a representative snapshot of the person’s day. Given reduced cognitive functioning in older adults which may increase retrospective response biases, time-sampling data may also have the advantage of significantly increasing data quality and maximizing ecological validity (Almeida, 2005). For example, the participants in this study do not have to remember what they did two weeks ago, or how they felt the day before because daily social interactions are reported in close proximity to their occurrence. Since older adults tend to see their past in a more positive light (Carstensen, 2003) this may also reduce their tendency to not remember conflictual interactions.  1.11 Summary This study will generate important knowledge on the associations between social interactions, health and wellbeing in older adults. Specifically, this study will examine the effects of daily life processes regarding social interactions and concurrent as well as  13 long-term health outcomes in the Oldest Old. This study will also control for the effects of functional health, cognition and number of diagnosed diseases. Specifically, this study will investigate six main research questions. To answer the following hypotheses examining social interactions, wellbeing and health in the Oldest Old I drew on data independently collected in two subprojects of BASE (Berlin Aging Study). One subproject was on the self in old age (Smith & Baltes, 1999; Smith & Freund, 2002), which collected data from all BASE participants; the other was on everyday competence, and it collected time-use information from a subsample of BASE participants (Klumb & Baltes, 1999b). For this study I reorganized and combined these two data sets to answer my unique hypotheses. TABLE 1. Hypotheses Hypothesis 1  Based on the Convoy Model I expect that, in the present sample, age is associated with a decreased number of daily social interactions but maintained quality of these interactions.  Hypothesis 2  Based on the Socioemotional Selectivity Theory I expect that, in the present sample, limited future time perspective is associated with an increase in meaningful social interactions as indicated by a low frequency of engaging in unpreferred social situations.  Hypothesis 3  In the present sample, daily social interactions are proposed to be associated with an increase in daily affect quality and a decrease in physical symptoms (as compared to non-social activities).  Hypothesis 4  In the present sample, socially un-preferred situations are expected to be associated with lower affect quality and more daily physical symptoms.  Hypothesis 5  In the present sample, few overall social interactions and many unpreferred interactions are expected to be associated with increased mortality.  Hypothesis 6  In the present sample, the proposed associations are expected to go above and beyond other potent variables (functional health, overall health and cognition).  14 Chapter 2: Methods 2.1 Participants This study is based on a subsample of the Berlin Aging Study (BASE; Baltes, & Mayer, 1999). BASE used an age-by-sex heterogeneous (locally representative) sample of 70 to 100+ year-olds. Participants in BASE include 516 older adults who were randomly drawn from the Berlin City Registry. In the course of the third wave of data collection for BASE a sample of 34 women and 49 men (N = 83) volunteered to participate in an additional time-sampling study. Their age ranged from 72 to 97 years, with a mean age of 81.1 years (SD = 5.0). Fortyseven percent of participants in this beeper study were widowed and 53% lived alone. All of these participants were community-dwelling and 34 (42%) lived with their spouses. Selectivity analyses (Klumb, & Baltes, 1995) showed the participants of this study to be a positive selection of the larger BASE sample. The participants in the beeper study were cognitively fitter and healthier then the parent BASE sample. The participants that will be used in this study took part in both the longitudinal aspect of BASE, as well as the one week time-sampling module.  2.2 Design The beeper study by Baltes and Klumb (1998) was a supplementary study added in the center of the large BASE study. The beeper study took place in between Time point 3 and Time point 4 of the longitudinal study. Signal-dependent self-reports were used to assess current activities, contexts, emotional experiences, and physical symptoms. Participants were beeped randomly at five time points across the day; a portable beeper  15 prompted the participants to fill in a sampling form. The shortest time between two successive signals was 15 minutes. The average inter-signal interval was 2.5 hours (SD = 20 minutes). This process was continued on six consecutive days, each with different random patterns. This study combined already collected long-term longitudinal data and time-sampling data.  2.3 Measures Longitudinal Measures Age: This was assessed via date of birth. This information was gathered from the Berlin City Registry (M = 81.60 years, SD= 5.06). Future Time Perspective: This information was assessed at the third measurement point by the following item: “I have the feeling that my time is running out”. Participants rated this item on a 5-point Likert scale with the anchors “definitely does not describe me” (1) and “very good descriptor” (5), (Mean = 2.52, SD= 1.41). This item is also a part of the Future Time Perspective scale (Lang & Carstensen, 2002). Control Variables Gender: For all analyses gender is coded as males= 0 and females= 1. General Intellectual Abilities: BASE included a number of intellectual tests in its cognitive battery. Fourteen cognitive tests were used to assess individual differences in five intellectual abilities: reasoning, memory, and perceptual speed from the mechanic (broad fluid) domain and knowledge and fluency from the pragmatic (crystallized) domain. The T-score of these fourteen tests for the beeper participants was 57.08 (SD=  16 7.53). The measurement of general intellectual abilities in BASE have been shown to have good validity (Lindenberger & Baltes, 1997). Overall Physical Health: To measure overall physical health, the Berlin Aging Study used a number of items including physician diagnoses. Diagnoses were determined from the course of consensus conferences by the research physician and psychiatrist, based on a standardized summary of clinical findings from all diagnostic procedures. Diagnosed moderate and severe illnesses were summed up to form the variable ‘‘number of moderate to severe diagnoses’’. I will utilize the number of medium to severer diagnosed illnesses for participants at time point three (M= 3.08, SD= 2.18). Functional Health: To measure functional health, this study uses assessments of balance and gait, the Romberg Balance task (Tinetti, 1986) from the third measurement occasion, (M = .626, SD= .644). This measure has been shown to have good reliability as indicated by Cronbach’s alpha of 0.89 (Huang, Yang & Lui, 2011) . Mortality: Mortality status information and the date of death for the deceased participants were obtained from the Berlin City Registry. To date there are 69 participants who have died out of a total of 82 (Retrieved August 2010). Loneliness: Loneliness was examined using a slightly modified German version of the eight-item short form of the revised UCLA Loneliness Scale (Russel, Cutrona, Rose & Yurko, 1984). This measure has been shown to have good reliability as indicated by Cronbach’s alpha of 0.84. This scale has been show to have good validity (Russell, Peplau & Cutrona, 1980). This scale reflects two aspects of loneliness: emotional and social. Social loneliness deals with the feeling of not being sufficiently embedded in a social network (Weiss, 1973). Emotional loneliness referred to the statements that  17 describe the feeling of having no close friends and/or being alone in the world (Elbing, 1991, cited after Baltes & Lindenberger, 2006). Emotional loneliness has a mean of 2.26 (SD= 0.85) whereas social loneliness has a mean of 2.11 (SD= 0.71). Network Size: This information was assessed at the third measurement occasion (Kahn, & Antonucci, 1980). It includes information on the number of relatives and friends that were still alive, their relationship status and ages. Participants in this study had a mean of 9.46 (SD= 4.80) in their first three social network circles. In their three closest network circles there was a mean of 2.19 (SD= 1.91) nuclear family members, 4.31 (SD= 3.32) relatives, 1.31 SD (2.38) friends, 1.00 (SD= 1.69) acquaintances and 0.63 (SD=1.29) neighbors. The average amount of medical care aids that participant’s had in their first three social network circles was 0.02 (SD= 0.15).  Daily Measures The beeper module measured many different variables but for the purpose of this study I will just be looking at the following. Of the maximum possible 30 sampling forms, an average of 26.2 (SD = 5.3), i.e., 87% were returned, in varying stages of completeness (Klumb, 2004). Social Activity: At each measurement point participants were asked if they were with someone or if they were alone. A social activity was if they were not alone. Participants answered the item with a “yes”= 1, they were alone or “no”= 0 they were not alone (M= 0.59, SD= 0.49). Social Preference: At each measurement point participants were further asked to report if they were in a socially preferred social situation or not. Social preference was  18 assessed by the following item: “Would you rather have spent the time with someone else?” Participants answered the item with a “yes”= 1 or “no”= 0 response (M = 0.06, SD= 0.23). Current Physical Symptoms: At each measurement point participants were asked to rate their current physical symptoms. Participants rated one item on a 5-point Likert scale with the anchors “not at all” (0) and “very much” (4). Affect Quality: At each measurement point, participants were asked to report their current affect quality. Affect quality was assessed at each time point by asking participants about the extent they were currently feeling: depressed, interested, happy, irritable, lonely, active, bored and relaxed. Eight mood adjectives were to be rated along 5-point scales, with 0 indicating no experience of the particular mood, and 4 indicating that the affect was experienced very intensely. Four items were taken from the Positive-Affect-Negative-Affect Scale, which originally has sixty items (PANAS, Watson, Clark & Tellegen, 1988). The PANAS scale has alpha reliabilities with a median internal consistency estimate of 0.93 (Watson & Clark, 1994). The remaining four items were adopted from the Yesterday Interview (see Klumb & Baltes, 1999b; Positive Affect Mean =2.77, SD= 0.93, α= 0.90, Negative Affect Mean= 0.39, SD= 0.60, α= 0.70). Analysis: This data violates the assumption of ordinary least squares procedure because of the nested data structure (Schwartz, & Stone, 1998). To account for the hierarchically nested structure of this data set (Snijders & Bosker, 1999), I examined my hypotheses  19 using hierarchical linear modeling (HLM; Raudenbush & Bryk, 2002). All occasions were nested within persons. Mortality data was analyzed with Cox regression analysis.  20 Chapter 3: Results 3.1 Relevant Descriptive Variables The following section describes the analysis and results for my main hypotheses. Further elaboration of these tables will be discussed in the succeeding section of this document. I first wanted to examine the number of relationships reported. Information about the network size for each participant was assessed at the third measurement occasion (Kahn, & Antonucci, 1980). Participants in this study had a mean of 9.56 (SD= 4.80) relationships in their first three social network circles. In their network there was a mean of 2.19 (SD= 1.91) nuclear family members, 4.31 (SD= 3.32) relatives, 1.31 (SD=2.38) friends, 1.00(SD= 1.69) acquaintances and 0.63 (SD=1.29) neighbors. Hence, there are relationships that older adults can draw on. In the following table (Table 1) descriptive variables for the central study variables are shown. As can be seen the variables age and future time perspective are not correlated (r = .209). This tells me that age and future time perspective capture different pieces of information. Age and future time perspective are associated with positive and negative affect. It can also be seen that the number of time one spends in unpreferred social situations significantly increases with age (r= .366**).  21 TABLE 2 : Descriptive Variables (N= 81) Variables 1. Age  M(SD) 80.82 (5.14)  2 .Future Time Perspective  2.51 (1.14)  3. Negative Affect  0.47 (0.49)  4. Positive Affect  2.73 (0.75)  5. Physical Symptoms  0.74 (0.91)  6. Time Alone  14.15 (7.81)  2  3  4  5  .209  .355**  -.233*  .165  .188  .336**  .241*  -.368**  .119  .076  .287**  -.273*  .258*  .241*  .286**  -.103  .328**  -.152  .153  -.329**  6  7  .137  7. Social Preference Situation 1.43 (2.30) Note: * p <.05, ** p <.01 . 3.2 Future Time Perspective, Age and Social Interactions The first hypothesis, based on the Convoy Model, predicted that the amount of time that participants spend alone will increase as participants’ age due to limited resources. As seen in the following table (Table 2) this hypothesis was not supported by the results. Neither age nor future time perspective predicted social activity. Interestingly, the greater the age of the participant the greater the number of time spent with nonpreferred company (β= 0.069*). This means that the older a participant was the more likely it was that they engaged in unpreferred social situations. Similarly, the more limited a participant’s future time perspective, the more likely they were to report socially unpreferred situations. Hypothesis two, based on the Socioemotional Selectivity Theory,  22 predicted that limited future time perspectives would be associated with a greater number of time spent with people as well as a greater number of time spent in preferred social situations. Remarkably, my data did not support this hypothesis. Instead, it was shown that participants who perceived their future to be limited reported more daily unfavourable social situations (β = 0.318*). Both of these findings do not support my expectations that greater age and a limited future time perspective is associated with a greater number of time spent alone and not as much time spent in unpreferred social situations.  TABLE 3. Hierarchical Linear Models Predicting Alone and Social Preferences, Age and Future Time Perspective (N= 81) Model 1: Alone (1=Alone) β (SE) Fixed Effects Intercepts Age FTP  0.383* (0.383) 0.039 (0.024) 0.181 (0.117)  Model: Model 2: Social Preference (1= not preferred company) β (SE) -2.745* (0.160) 0.069* (0.030) 0.318* (0.114)  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error.  3.3 Social Interactions, Daily Affect and Physical Symptoms The third hypothesis predicted that social interactions would be associated with concurrent higher daily positive affect and lower daily negative affect. In the following table (Table 3) I can see that our data does support our predictions. Being alone was associated with significantly lower daily positive affect (β = -0.015*, SE= 0.017), as well as a significantly higher daily negative affect (β =0.949*, SE= 0.045) and daily physical  23 symptoms (β = 0.735*, SE= 0.034). This model explained 5.1% of the variance in daily physical symptoms, 6.3% in daily negative affect and 11% of daily positive affect as compared to an unconditional model.  TABLE 4. Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective and Time Spent Alone Model 1: Daily Physical Symptoms β (SE) Fixed Effects Intercepts 0.110*(0.181) Alone (1= alone) 0.735*(0.034) Age 0.050 (0.025) FTP -0.032 (0.121) Random Effects Residual 0.523 Intercept Level 1 0.700  Model 2: Daily Negative Affect β (SE)  Model 3: Daily Positive Affect β (SE)  1.259*(0.254)  2.841* (0.080)  0.949*(0.045) 0.091*(0.038) -0.175*(0.165)  -0.184*(0.047) -0.020 (0.016) 0.231* (0.103)  0.195 0.181  0.328 0.464  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error. For hypothesis four, I predicted that socially unpreferred situations would be associated with lower affect quality and greater physical symptoms; my data support this hypothesis (Table 4). Socially unpreferred situations were associated with greater negative affect (β= 0.753*, SE= 0.051) and greater physical symptoms (β= 0.588*, SE= 0.040). I also want to point out that age and future time perspective are both negatively associated with wellbeing across the board (including physical symptoms, negative affect, and positive affect). I will discuss this more in the following section. This model  24 explained 4% of the variance in daily physical symptoms, 12% of daily negative affect and 13% of daily positive affect as compared to an unconditional model. TABLE 5. Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective and Time Spent in Socially Unpreferred Interactions  Fixed Effects Intercept Social Preference (1=no) Age FTP Random Effects Residual Intercept Level 1  Model 1: Daily Physical Symptoms β (SE)  Model 2: Daily Negative Affect β (SE)  Model 3: Daily Positive Affect β (SE)  1.727* (0.190)  2.067*(0.264)  2.735*(0.076)  0.753* (0.051) 0.122* (0.042) -0.188 (0.208)  0.003(0.010) -0.022(0.016) -0.219*(0.071)  0.588*(0.040) 0.074*(0.029) -0.044 (0.151) 0.921 0.669  0.196 0.157  0.336 0.459  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error.  3.4 Mortality In Hypothesis five, I predicted that fewer social interactions and many unpreferred social interactions would be associated with an increased mortality risk. This analysis was based on mortality data from the Berlin City Registry from a total of 67 out of 82 participants. As can be seen in Table 5, spending a lot of time alone is a mortality risk (β =1.05*). It can also be seen in Table 6 that an increase in socially unpreferred situations is not a significant mortality risk (β= -.961). I also included a number of additional variables that are known to be associated with morality. For example, age is a predictor for mortality- I know that the older one is the more likely they are to be closer  25 to death, this information gives me confidence in the data. Similarly, being female also shows reduced mortality risk. What is interesting is that being alone has a similar effect as age. I assume that this week is a representation of a regular week in the participant’s lives and thus I can generalize and say that overall lower numbers of daily social interactions is almost as great as a risk for mortality as age is in this sample of the Oldest Old. A limited future time perspective showed a significant risk for mortality both when being alone and when being in an unpreferred social relationship. This tells me that older adults’ predictions of how long they feel they have left to live to be somewhat accurate. These findings will be discussed in greater detail in the following section. TABLE 6: Survival Analysis using Cox Regression Examining Time Spent Alone, Controlling for Functional Health, Age, Gender and Cognition Alone Controls: FTP Romberg Balance task Age Gender (1= Female) Cognitive Status  Mortality Risk 1.04* 0.801* 0.80 1.07* 0.38* 0.98  Note. ** p < .01, * p < .05.; N = 67 deceased, N = 15 alive  26 TABLE 7: Survival Analysis Using Cox Regression Examine Time Spent in Unpreferred Social Situations, Controlling for Functional Health, Age, Gender and Cognition Socially Unpreferred Situations Controls: FTP Romberg Balance task Age Gender (1= Female) Cognitive Status  Mortality Risk 0.96 0.72* 0.79 1.09** 0.57* 1.00  Note. ** p < .01, * p < .05.; N = 67 deceased, N = 15 alive  3.5 Follow up Analysis In the following table I looked at the predictive value of being alone using future time perspective, age as well as emotional and social loneliness. I have just discussed the main findings and there are still a couple of things that I want to examine so that I may furthertrust my main findings. For example, is it really being alone or loneliness that drives wellbeing in older adults? And if I add cognition and overall health will my effect still hold? I re-ran the previous models in Table 8 and can see that my findings are robust. All of the significant values stay in place when emotional and social loneliness, overall health and cognition are added to the models.  27 TABLE 8. Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective and Time Spent Alone Model 1: Daily Physical Symptoms β (SE) Fixed Effects Intercept a Alone (1= alone) Age FTP Emotional Loneliness Social Loneliness Random Effects Residual Intercept Level 1  0.660* (0.096) 0.124* (0.052)  Model 2 Daily Negative Affect β (SE) 0.411* (0.051) 0.081* (0.027)  Model 3: Daily Positive Affect β (SE) 2.845* (0.072) -0.192* (0.044)  0.009 (0.020) -0.119 (0.124)  0.019 (0.011) -0.041 (0.053)  -0.007* (0.015) 0.143 (0.110)  0.069 (0.165)  0.130 (0.087)  -0.378* (0.117)  0.158 (0.193)  0.068 (0.101)  0.067* (0.124)  0.532 0.728  0.196 0.176  0.329 0.426  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error. The results in Table 7 show that greater numbers of being alone are associated with more reported daily physical symptoms as well as less positive affect and more negative affect. These results hold even when loneliness is added into the model.  28 TABLE 9. Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective and Time Spent in Socially Unpreferred Interactions Model 1: Daily Physical Symptoms β (SE) Fixed Effects Intercept Social Preference (1=no) Age FTP Emotional Loneliness Social Loneliness Random Effects Residual Intercept Level 1  0.726* (0.094)  Model 2: Daily Negative Affect β (SE)  Model 3: Daily Positive Affect β (SE)  0.429* (0.045)  2.746* (0.073)  0.231* (0.110) 0.014 (0.021) -0.118 (0.122)  0.267*(0.076) 0.016 (0.011) -0.037 (0.053)  -0.258*(0.100) -0.008 (0.015) 0.128 (0.114)  0.051 (0.162)  0.125 (0.068)  -0.360*(0.114)  0.131 (0.195)  0.103 (0.098)  -0.019 (0.108)  0.554 0.721  0.196 0.150  0.336 0.422  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error. The results from Table 8 report that being in unpreferred social situations is associated with more daily physical symptoms as well as more negative affect and less positive affect. These results hold even when loneliness is added into the model.  29 TABLE 10. Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective, Overall Health and Time Spent Alone Model 1: Daily Physical Symptoms β (SE) Fixed Effects Intercept Alone (1= alone) Age FTP Overall Health Random Effects Residual Intercept Level 1  Model 2 Daily Negative Affect β (SE)  Model 3: Daily Positive Affect β (SE)  0.698* (0.105) 0.117* (0.055)  0.422* (0.051) 0.068*(0.026)  2.842* (0.080) -0.185* (0.047)  0.014 (0.021) -0.142 (0.130) 0.042 (0.052)  0.024*(0.010) -0.051 (0.056) 0.031 (0.022)  -0.020 (0.017) 0.225* (0.107) -0.009 (0.040)  0.551 0.769  0.209 0.175  0.567 0.439  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error. The results from Table 9 show that being alone was associated with more reported daily physical symptoms as well as more negative affect and less positive affect. These results hold even when overall health is added into the model.  30 TABLE 11. Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective, Overall Health and Time Spent in Socially Unpreferred Interactions Model 1: Daily Physical Symptoms β (SE) Fixed Effects Intercept Social Preference (1=no) Age FTP Overall Health Random Effects Residual Intercept Level 1  Model 2: Daily Negative Affect β (SE)  Model 3: Daily Positive Affect β (SE)  0.756* (0.101)  0.427* (0.048)  2.747* (0.079)  0.027* (0.053) 0.018 (0.022) -0.137 (0.126) 0.0265 (0.053)  0.326* (0.084) 0.021*(0.010) -0.056 (0.052) 0.030 (0.020)  -0.311(0.112) -0.021 (0.017) 0.227* (0.109) -0.007 (0.039)  0.576 0.759  0.208 0.152  0.326 0.432  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error. The results in Table 10 show that being in unpreferred social situations was associated with more reported daily physical symptoms as well as more negative affect and less positive affect. These results hold even overall health is added into the model.  31 TABLE 12. Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective, Cognition and Time Spent Alone Model 1: Daily Physical Symptoms β (SE) Fixed Effects Intercept Alone (1= alone) Age FTP Cognition Random Effects Residual Intercept Level 1  Model 2 Daily Negative Affect β (SE)  Model 3: Daily Positive Affect β (SE)  0.698* (0.106) 0.116* (0.055)  0.425* (0.050) 0.068* (0.026)  2.841* (0.079) -0.185* (0.047)  0.016 (0.021) -0.154 (0.127) -0.007 (0.015)  0.026* (0.011) -0.039 (0.056) -0.014 (0.007)  -0.020 (0.016) 0.193 (0.105) 0.017 (0.008)  0.551 0.776  0.209 0.170  0.321 0.425  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error. The results In Table 11 show that being alone was associated with more reported daily physical symptoms as well as more negative affect and less positive affect. These results hold even when cognition is added into the model.  32 TABLE 13. Hierarchical Linear Models Predicting Physical Symptoms, Positive Affect and Negative Affect from Age, Future Time Perspective, Cognition and Time Spent in Unpreferred Social Interactions Model 1: Daily Physical Symptoms β (SE) Fixed Effects Intercept Social Preference (1=no) Age FTP Cognition Random Effects Residual Intercept Level 1  Model 2: Daily Negative Affect β (SE)  Model 3: Daily Positive Affect β (SE)  0.756*(0.102)  0.431* (0.047)  2.746* (0.078)  0.270* (0.132) 0.019 (0.022) -0.140 (0.122) -0.006 (0.015)  0.328* (0.085) 0.023* (0.011) -0.043 (0.051) -0.014 (0.008)  -0.313*(0.112) -0.021 (0.015) 0.194 (0.105) 0.017 (0.008)  0.576 0.760  0.208 0.148  0.326 0.418  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error.  The results in Table 12 show that being in unpreferred social situations was associated with more reported daily physical symptoms as well as more negative affect and less positive affect. These results also hold even when cognition is added into the model. My results for hypothesis one (Table 13, 14 and 15), hold even when cognition, overall health and network size are added to the model. It can also be seenthat the results for hypothesis one do not hold for future time perspective when the previous covariates are added to the model, this shows further support for the limited resources hypothesis for the Oldest Old that is discussed in further detail in the following section.  33 TABLE 14. Hierarchical Linear Models Predicting Being Alone and Being in Unpreferred Social Interactions, Age, Future Time Perspective and Overall Health Model 1: Alone (1=Alone) β (SE) Fixed Effects Intercepts Age FTP Overall Health  0.423*(0.136) 0.065*(0.028) -0.080 (0.142) -0.160*(0.064)  Model: Model 2: Social Preference (1= not preferred company) β (SE) -2.725*(0.180) 0.093*(0.030) 0.003 (0.241) 0.007 (0.071)  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error.  The results from Table 13 show that older adults are less likely to engage in social situations with greater age, even when I add overall health. The results also show that FTP no longer predicts socially unpreferred situations when overall health is added to the model. TABLE 15. Hierarchical Linear Models Predicting Being Alone and Being in Unpreferred Social Interactions, Age, Future Time Perspective and Cognition Model 1: Alone (1=Alone) β (SE) Fixed Effects Intercepts Age FTP Cognition  0.418* (0.138) 0.059* (0.027) -0.012 (0.152) 0.022 (0.020)  Model: Model 2: Social Preference (1= not preferred company) β (SE) -2.757* (0.173) 0.094* (0.032) -0.064 (0.259) 0.034 (0.019)  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error.  34 The results from Table 14 show that fewer social interactions occur with greater age, even when I add cognition. The results also show that FTP no longer predicts socially unpreferred situations when cognition is added to the model. TABLE 16 Hierarchical Linear Models Predicting Being Alone and Being in Unpreferred Social Interactions with Age, Future Time Perspective and Social Network Model 1: Alone  Fixed Effects Intercepts Age FTP Social Networks  (1=Alone) β (SE)  Model: Model 2: Social Preference (1= not preferred company) β (SE)  0.416* (0.140) 0.056* (0.025) 0.050 (0.154) -0.013 (0.027)  -2.721* (0.180) 0.089*(0.029) 0.019 (0.240) -0.017 (0.040)  Note. * p<.05; intercepts refer to mean age, daily physical symptoms, daily negative affect and daily positive affect levels. FTP= Future Time Perspective, SE= standard error.  The results from Table 15 show that participants were less likely to engage in social interactions the older they were, even when I add number of individuals in their social networks. The results also show that FTP no longer predicts socially unpreferred situations when number of individuals in their social networks is added to the model.  35 Chapter 4: Discussion The purpose of this study was to examine the associations between social interactions, health and wellbeing in the Oldest Old. Findings indicate that older adults spend more time alone as well as in unpreferred social situations. The results also show that even when controlling for cognition and overall health, age is still associated with a decrease in social interactions in the present sample of older adults. Interestingly, my results showed that a more limited future time perspective in this sample of the Oldest Old was associated with a greater amount of time spent alone as well as a more situations spent in socially unpreferred situations. These results held even after other factors were added such as cognition, overall health and social network size. Findings indicate that greater numbers of daily social interactions are associated with higher wellbeing. The results also showed that time spent alone was a risk for mortality.  4.1 Social Interactions, Age and Future Time Perspective The results I obtained for my first hypothesis show that both age and future time perspective were associated with less time spent in socially preferred situations and more time spent alone. These results were not consistent with previous research examining older adults and social interactions. As proposed by Socioemotional Selectivity Theory older adults are proposed to spend more time alone but they also spend more time in socially preferred situations to maximize the positive effects of being with those that they would most rather be with (Carstensen, Fung & Charles, 2003). There are a number of reasons for the discrepancy of my results in comparison to previous literature. One of the main reasons that my results show a bleaker future for the number and quality of social  36 interactions in older adulthood is because the sample was composed of the Oldest Old, the mean age of these participants was 81.60 years, whereas most studies looking at older adults have a sample with a mean age in the early to mid-70s (Baltes & Smith, 2003). The sample used in this study is much older than those who are in their mid-60s or 70s and thus they encounter different social and health issues. There is a great deal of information on older adults around the age of 65 but there is very limited information on the Oldest Old. In these analyses I wanted to replicate previous findings showing that older adults actively regulate their social interactions, however this was not found in the sample of the Oldest Old. I speculate that these results are due to the Oldest Old lacking the energy and resources to actively regulate their social relationships, thus spending more time alone and in the presence of unpreferred social interactions (Charles, 2010). Older adults may know exactly who they want to spend time with but the problem lies in their inability to act upon these goals. The Socioemotional Selectivity Theory states that with a limited future time perspective, emotionally meaningful social interactions attain unmatched importance. I suspect that individuals lack the ability and resources to be able to attain these emotionally meaningful social interactions. This draws from recent literature looking at the 4th age (Oldest Old) and how this age is significantly different than the 3rd age (Baltes, & Smith, 2003). It is thus possible that older adults know who they would benefit the most from spending time with but because of limited resources they are unable to act on this. Resource constraints may limit well-being in old age for a number of different reasons including but not limited to: biological or functional, cognitive, social and financial constraints.  37 Results showed that loneliness is associated with low wellbeing. In my analyses, results showed that it is the actual context of being alone that matters more than the feeling of social or emotional loneliness. I speculate that these results may have occurred because older adults need the stimulation of actually being involved in a social interaction to gain the benefits. When examining the association between future time perspective and unpreferred social situations I found that the negative association between future time perspective and unpreferred social situations disappeared when the variable overall health is added to the model. This suggests that perhaps older adults with poorer overall health are spending time in social interactions with people such as nurses or registered care aids and are thus not reaping any benefits from this kind of social interaction. In a perfect world I would see the Oldest Old gearing down and spending more time alone, maximizing the quality of their limited social interactions by being more selective about who they spend their time with. Due to the fact that the sample in this study was of the Oldest Old, I speculate that this may be mortality related processes which foreshadow the onset of death. In the terminal decline literature they describe a steep decline in wellbeing near the very end of life (Gerstorf et al, 2010). Perhaps what these results show is that the Oldest Old have limited wellbeing and preferred social interactions because they may be on the downward trajectory of limited wellbeing and preferred social interactions preceding death.  38 4.2 Benefits of Social Interactions in the Oldest Old Although the previous section painted a dark picture, I did find positive results regarding the association between social interactions and wellbeing in the Oldest Old. Even though the Oldest Old may not be spending as much time as they would like in the presence of their preferred company, they do still receive positive health benefits from engaging in daily social interactions. These results show that a greater numbers of time spent with others is associated with more daily positive affect, less daily negative affect and fewer reported daily physical symptoms - showing us the health benefits of daily life social interactions. Since this data set is linked to the parent Berlin Aging Study, I have the unique opportunity to link daily life processes with long term outcomes. Carstensen et al. (2000) found that daily beeper patterns in adults are generally stable over the years, because of this I hope that the week of data that I am using is representative of a regular week in participants’ lives. The results also show that being alone predicts mortality almost as well as age does. What this tells us is that being alone is almost as great a risk for death as age is even when controlling for other factors such as cognition and overall health. These findings show us that my results are not spurious and that daily life processes and social interactions go beyond moment to moment effects and that they impact long-term outcomes.  4.3 Limitations and Strengths Although this is a unique study that utilizes two prestigious data sets examining health and wellbeing in the Oldest Old, there are also limitations that must be noted. Since I analyzed this data using cross-sectional analyses I do not have information on temporal  39 order or causality. Although the sample size is 82, it is still a relatively small sample size and future research should be conducted with a larger sample of the Oldest Old. Hopefully in the future with a larger sample size these effect sizes will also increase. Another limitation of this study was that wellbeing was completely self-reported. Participants rated their current positive affect, negative affect and physical symptoms multiple times each day for a week. Although this gives me agreat breadth of knowledge on daily processes I still must consider biases that participants may have in reporting on their own wellbeing.  4.4 Conclusion and Future Directions The following study examined social interactions in the Oldest Old and how social interactions relate to health and wellbeing. I found that as age increases the number and quality of social interactions decreases. The results also showed that as future time perspective becomes more limited, older adults spend more time alone and with unpreferred company. The results of this study add to the conception that the association between daily social interactions and health may be a reciprocal one (Maier & Klumb, 2005). On one hand, social activity appears to be beneficial for health outcomes, but on the other hand, it is clear that good health in turn promotes participation in social interactions. This study makes the case that older adults, especially the Oldest Old, receive health and wellbeing benefits from daily social interactions. Future interventions could be implemented to ensure that older adults are engaging in positive daily social interactions with preferred company. This type of intervention would be easier to implement and  40 perhaps more appealing to older adults than an exercise intervention also trying to benefit older adults’ health. Future research should go beyond examining daily life processes only and also use information from long-term outcomes to better understand the complex relationship between health and social interactions in the Oldest Old.  41 References Almeida, D. M. (2005). Resilience and vulnerability to daily stressors assessed via daily diary methods. Current Directions in Psychological Science, 14(2), 64-68. Antonucci, T. C. (1990). Social supports and social relationships. In R. H. Binstock, & L. K. George (Eds.), Handbook of aging and the social sciences (3rd ed., pp. 205227). San Diego, CA: Academic Press. Antonucci, T. C., Akiyama, H., & Takahashi, K. (2004). Attachment and close relationships across the life span. Attachment & Human Development, 6 (4), 353370. Antonucci, T. C., & Jackson, J. S. (1987). Social support, interpersonal efficacy, and health: A life course perspective. In L. L. Carstensen & B. A. Edelstein (Eds.), Handbook of clinical gerontology (pp. 291–311). New York: Pergamon Press. August, K .J., Rook, K. S. & Newsom, J. T. (2007).The joint effects of life stress and negative social exchanges on emotional distress. The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences. 62B(5), 304-S314. Baltes, P. B., Lindenberger, U. & Staudinger, U. M. (2006). Life span theory in developmental psychology. In R. M. Lerner & W. Damon (Eds.), Handbook of child psychology (6 ed., Vol. 1, pp. 569-664). Hoboken, NJ: John Wiley & Sons, Inc. Baltes, P. B. & Mayer, K. U. (1999). The Berlin Aging Study: Aging from 70 to 100. Cambridge University Press. Baltes, P. B. & Smith, J.(2003) New frontiers in the future of aging: from successful aging of the young old to the dilemmas of the fourth age. Gerontology: Behavioral Science Section/Review, 49, 123–135. Berkman L. F. & Syme S. L.(1979). Social networks, host resistance and mortality: a nine-year follow-up study of Alameda County residents. American Journal of Epidemiology, 109, 186 –204. Birditt, K. S., Almedia, D. M. & Fingerman, K. L. (2005). Age differences in exposure and reactions to interpersonal tensions: A daily diary study. Psychology and Aging, 20 (2), 330-340. Bolger, N., DeLongis, A., Kessler, R. C. & Schilling, E. A.(1989). Effects of daily stress on negative mood. Journal of Personality and Social Psychology, 57, 808-818.  42 Carstensen, L. L. (1995). Evidence for a life-span theory of socioemotional selectivity. Current Directions in Psychological Science, 4, 151-156. Carstensen, L. L., & Fredrickson, B. L. (1998). Influence of HIV Status and Age on Cognitive Representations of Others. Health Psychology, 17, 494-503. Carstensen, L. L., Fung, H., & Charles, S. (2003). Socioemotional selectivity theory and the regulation of emotion in the second Half of life. Motivation and Emotion, 27(2), 103-123. Carstensen, L. L., Isaacowitz, D. M., & Charles, S. T. (1999). Taking time seriously: A theory of socioemotional selectivity. American Psychologist, 54, 165–181. Carstensen L.L, Pasupathi, M, Mayr, U. & Nesselroade, J.R. (2000). Emotional Experience in Everyday Life Across the Adult Life Span. Journal of Personality and Social Psychology. 79(4):644–55. Carstensen, L. L., Turan, B., Scheibe, S., Ram, N., Ersner-Hershfield, H., SamanezLarkin, G. R., Brooks, K. P. & Nesselroade, J. R.(2011). Emotional experience improves with age: evidence based on over 10 years of experience sampling. Psychology and Aging. DOI: 10.1037/a0021285. Charles, S.T. (2010). Strength and vulnerability integration: A model of emotional well-being across adulthood. Psychological Bulletin 136(6), 1068–1091. Cohen, S. & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98, 310–357. Cohen, S. & McKay, G. (1984). Social support, stress and the buffering hypothesis. A theoretical analysis. Handbook of Psychology and Health.253-267 Crawford, J.R. & Henry, J. (2004). The positive and negative affect-schedule (PANAS): Construct validity, measurement propoerties and normative data in a large nonclinical samples. British Journal of Clinical Psychology. 43(3), 245-265. Cutler, D. M., Poterba, J. M., Sheiner, L. M., Summers, L. H. & Akerlof, G. A.(1990). An aging society: opportunity or challenge? Brookings Papers on Economic Activity, 1, 1-73. Davern, M. T., Cummins, R. & Stokes, M. (2007). Subjective wellbeing as an affective-cognitive construct. Journal of Happiness Studies, 8, 429– 449. doi:10.1007/s10902-007-9066-1 Durkheim, E. Suicide. New York: Free Press, 1951.  43 Everard, K. M., .Lach, H. W., Fisher, H. B. & Baum, M. C.(1999). Relationship of activity and social support to the functional health of older adults. The Journals of Gerontology: Series B, 55(4), 208-212. Elbing, E.(1991). Eisamkeit. GÖttingen: Hogrefe. Fuller-Iglesias, H., Sellars, B. & Antonucci, T. C. (2008). Resilience in old age: Social relations as a protective factor. Research in Human Development. 5(3), 181-193. Ghisletta, P., Bickel J., & Lovden M. (2006). Does activity engagement protect against cognitive decline in old age? Methodological and analytical considerations. Journal of Gerontology and Psychological Science. 61B(5), 253–261. Giles, L. C., Glonek, G. F. V., Luszcz,, M. A. & Andrews, G. R.(2005). Effect of social networks on 10 year survival in very old Australians: the Australian longitudinal study of aging. Journal of Epidemiology Community Health 2005, 59(7), 574– 579. Glass, T. A., Mendes de Leon, C., Marottoli, R. A. & Berkman L.F.(1999). Population based study of social and productive activities as predictors of survival among elderly Americans. British Medical Journal. 319, 478-483. Health Canada. (2002). Canada's aging population. Retrieved November 22, 2008, from http://www.hc-sc.gc.ca/seniors-aines Holtzman, S., & DeLongis, A. (2007) .One day at a time: The impact of daily satisfaction with spouse responses on pain, negative affect and catastrophizing among individuals. Pain, 1633-1655. Hoppmann, C. & Gerstorf, D. (2009). Spousal interrelations in old age- A mini review. Gerontology, 55, 449-459. House, J. S., Landis, K. R. & Umberson, D. (1988). Social relationships and health. Science, 241, 540–545 Huang, T. Yang, L. & Lui, C. (2011). Reducing the fear of falling among communitydwelling elderly adults through cognitive-behavioural strategies and intense Tai Chi exercise: a randomized controlled trial. Journal of Advanced Nursing. 67(5), 961-971. Kahn, R. L., & Antonucci, T. C. (1980). Convoys over the life course: Attachment, roles, and social support. In P. B. Baltes, & O. Brim (Eds), Life-Span Development and Behavior (3). New York: Academic Press. Kamarck, T. W.; Schwartz, J. E., Shiffman, S., Muldoon, M. F., Sutton-Tyrrell, K., & Janicki, D. L. (2005). Psychosocial stress and cardiovascular risk: What is the role of daily experience? Journal of Personality, 73(6), 1749-1774.  44 Klumb, P.L. (2004). Benefits from productive and consumptive activities: results from the Berlin Aging Study. Social Indicators Research 67: 107–127. Klumb, P. L. & Baltes, M.M. (1999b).Validity of Retrospective Time-Use Reports in Old Age. Applied Cognitive Psychology, 13, 527-539. Klumb, P. L. & Baltes, M.M. (1995), ‘Berlin Aging Study: Selective Participation and Attrition in a Study On Everyday Competence in Old Age Based on the Experience-Sampling Method’, Paper presented at the EURODEP Conference, Dublin, Ireland, November 17–19. Lang, F. R., & Carstensen, L. L. (2002). Time counts: Future time perspective, goals, and social relationships. Psychology and Aging. 17(1). 125-139. Lang, F. R., & Carstensen, L. L. (1994). Close emotional relationships in late life: Further support for proactive aging in the social domain. Psychology and Aging, 9, 31524. Lazarus , R. S., & DeLongis, A. (1983). Psychological stress and coping in aging. American Psychologist. 38(3), 245-254. Lennartsson C. & Silverstein M. (2001). Does engagement with life enhance survival of elderly people in Sweden? The role of social and leisure activities. Journal of Gertontological Social Science. 25:S335–S342. Lindenberger, U., & Baltes, P.B (1997). Intellectual functioning in old age. Psychology and Aging, 8, 207-220. Maier, H., & Klumb, P.L. (2005). Social participation and survival at older ages: Is the effect driven by activity content or context? European Journal of Aging. 2, 3139. Marin, T., Holtzman, S., DeLongis, A. & Robinson, L.(2007). Coping and the response of others. Journal of Social and Personal Relationships, 24, 951-969. Marsiske, M., Delius, J., Maas, I., Lindenberger, U., Scherer, H. & Tesch- Romer, C. (1999). Sensory systems in old age. In P. B. Baltes & K. U. Mayer (Eds.), The Berlin Aging Study: Aging from 70 to 100 (pp. 360-383).New York: Cambridge University Press. Nesselroade, J. R. (1991). The warp and woof of the developmental fabric. In R. Downs, L. Liben & D. Palermo (Eds.), Visions of development, the environment, and aesthetics: The legacy of Joachim F. Wohlwill (pp. 213-240). Hillsdale, NJ: Erlbaum.  45 Nezlek, J. B., Richardson, D. S., Green, L. R. & Schatten-Jones, E. C. (2002). Psychological well-being and day-to-day social interaction among older adults. Personal Relationships. 9(1), 57-71. Pinquart, M. & Scnindler I. (2007). Changes of life satisfaction in the transition to retirement: A latent-class approach. Psychology and Aging. 22(3), 442–455. Raudenbush, S. W. & Bryk, A. S. (2002). Hierarchical linear models. Rowe, J. W. & Kahn, R. L. (1997). Successful aging. The Gerontologist, 37(4), 433440. Russell, D., Peplau, L.A. & Cutrona, C.E.(1980). The revised UCLA loneliness scale: Concurrent and discriminent validity evidence, Journal of Personality and Social Psychology, 39, 472-480. Schwartz , J. E. & Stone, A. A.(1998).Strategies for analyzing ecological momentary assessment data. Health Psychology, 17(1), 6-16. Seeman, T.E., Singer, B.H., Ryff, C.D., Dienberg, Love, G. & Levy-Storms L. (2002). Social relationships, gender, and allostatic load across two age cohorts. Psychosocial Medicine. 64(3):395–406. Seeman, T. E., Lusignolo, T. M., Albert, M. & Berkman, L. (2001). Social relationships, social support, and patterns of cognitive aging in healthy, high-functioning older adults: MacArthur studies of successful aging. Health Psychology, 20(4), 243255. Smith, J. & Freund, A. M. (2002). The Dynamics of Possible Selves in Old Age." Journals of Gerontology: Series B: Psychological Sciences & Social Sciences. 57B (6), 492-500. Statistics Canada (2007). Portrait of the Canadian population in 2006, by age and sex, 2006 Census. Ottawa. Snijders, T. & Bosker, R. (1999). Multilevel analysis. London: Sage Publications. Tinetti, M. E. (1986). A performance-orientated assessment of mobility problems in elderly patients. Journal of the American Geriatrics Society, 34, 119-126. Watson, D., & Clark, A. (1994). Manual for the Positive and Negative Affect Schedule – Expanded Form. The University of Iow. Watson, D., Clark L.A. & Tellegen, A. (1988). A Development and validation of brief measures of positive and negative affect: The PANAS Scales’, Journal of Personality and Social Psychology 54, 1063 – 1070.  46  Weiss, R. (1973). Loneliness: The experience of emotional and social isolation. Cambridge, MA: MIT Press. Uchino, B. N., Cacioppo, J. T. & Kiecolt-Glaser, J.,K. (1996) The relationship between social support and physiological processes: A review with emphasis on underlying mechanisms and implications for health. Psychological Bulletin. 119(3), 488-531.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0072122/manifest

Comment

Related Items