Meanings in the Lives of Older Adults: In Their Own Voices

This dissertation attempts to advance our understanding of how we as human beings perceive , judge, and evaluate our lives, particularly in the last stage of the life cycle. The primary goal was to counterbalance the predominantly quantitative approach to meaning in life with a qualitative approach by interviewing 21 older adults as they reflected on and attempted to find purpose and value in the totality of their experiences. This goal was explicit on the point of wanting older adults , those ideally positioned in the lifespan to inform us about meaning in life, to speak without the restriction associated with response options and format. Guided by the approach of phenomenology which emphasizes immediate description, the participants' responses to interview questions were summarized and, to the extent possible, reporting preserved their actual words. Participants articulated in their own voices that (1) "meaning in life" is about helping others; (2) having a sense of purpose; (3) maintaining relationshipsespecially with family; and (4) subjective well-being. They reported that life has been and continues to be meaningful, and they attributed their sense of meaning to three higher order factors: affiliation, agency and control, and self-transcendence. The second goal of this study explored the relationship between meaning in life and quality of life through self-report estimates. Participants indicated on a Likert scale of 1-10 how much meaning in and quality of life they perceive having both in the present, and retrospectively, for life as a whole. Results indicated that quality of life in the present did not diminish relative to the past, meaning in life in the past correlated with quality of life, but estimates of meaning in life in the present did not correlate with quality of life estimates or with meaning in life in the past. This supports the notion that meaning in life is experienced as distinct from quality of life. It also accords with Erikson's psychosocial theory of development which predicts that during late adulthood individuals confront the inevitability of a finite existence. From this perspective, they review the life they've lived, hoping to achieve a sense of integrity. The third goal was to compare a larger sample of older adults with a sample of college students on overall quality of life. Analysis revealed that older adults reported higher quality of life and also differed with regard to the different dimensions that contributed to quality of life. Out of 16 life areas, the following were most closely associated with overall quality of life scores for older adults: relatives, goals and values, home, helping, community and children.

wanting older adults , those ideally positioned in the lifespan to inform us about meaning in life, to speak without the restriction associated with response options and format. Guided by the approach of phenomenology which emphasizes immediate description, the participants' responses to interview questions were summarized and, to the extent possible, reporting preserved their actual words.
Participants articulated in their own voices that (1) "meaning in life" is about helping others; (2) having a sense of purpose; (3) maintaining relationshipsespecially with family; and (4) subjective well-being. They reported that life has been and continues to be meaningful, and they attributed their sense of meaning to three higher order factors: affiliation, agency and control, and self-transcendence. Along the way many individuals have provided inspiration and influence. Early on, a number of professors at the University of Rhode Island made significant contributions. Among them is Susan Brady, with whom I took my first course as a non-matriculated student, and whose words had the power to convince me that I · could aspire to this level of accomplishment; Lisa Harlow who introduced me to meaning-in-life research; Lisa Bowleg who taught me that qualitative study can be scholarly and scientific; and the Department Chair, Dom Valentino, who always received me with kindness, a virtue I much admire. I acknowledge these individuals' influence and express my gratitude. Last, but certainly not least, I acknowledge a debt of gratitude to the men and women who participated in this study. This work is a tribute to them.

Introduction
This dissertation is about the uniquely human impulse to organize experience so that there is coherence and unity to life. With the passage of time, we create and accumulate interpretive stories that document our journey. We maintain a narrative that connects the past, present, and future. We are impelled to create meaning-even when we don't know where the journey will take us and even when unpredictable challenges threaten to disrupt the coherence of our narrative. As Victor Frankl's (1959) story exemplified, meaning is ubiquitous and unconditional; our experiences need to make sense to us. Our stories are precious to us and deeply personal, precisely because they speak to us of ourselves, and because they give us a sense of continuity and consistency.
The primary goal of this study was to receive the narratives of older adults as they reflected on satisfaction, quality, purpose, and meaning in their lives. It was expected that, given their developmental stage and accumulated experience and insight, participants would be in a good position to discern a richer and more abstract understanding of those facets of life that have given it meaning.
Philosophical discourse about meaning in human life dates back to the early Greek philosophers. Aristotle took up this question approximately 2500 years ago. His account is still significant today and is particularly relevant to this study; it is explicit on how one can live a "good" life, it views the person rooted in community life, and it acknowledges that the perspective of older people can be of value.
In Book 1 of the Nicomachean Ethics (Aristotle, trans. 1953), Aristotle introduces the "good" for human life as that for the sake of which human pursuits are aimed, namely, eudaimonia, translated as well-being and flourishing.
Aristotle's eudaimonia is deeply rooted in the biology of how a species of a particular class is designed to flourish (e.g., mature salmon are designed to swim upstream to the place of their birth where they will propagate and subsequently die). How can human beings flourish (i.e., achieve eudaimonia)? Aristotle provides guidance: (1) Individuals should organize their lives around worthy goals, the pursuit of which human beings have been designed; (2) Some ends are better than others, and the best life is one in pursuit of the ultimate good, which involves excellence in reasoning, understanding and logical choices; (3) To flourish, excellence in reasoning needs to be guided by action that accords with moral virtues (i.e., generosity, patience, truthfulness, courage, wisdom, and temperance); ( 4) Love of self (viewed in terms of pursuing understanding and moral activity) should be balanced with concern for others; (5) A balance of family, children and friends is important because" ... man is by nature a social being" (NE, p. 74); and (6) Lower goals such as, pleasure and amusement, are important because they permit a person to pursue higher goals (i.e., ethical activity and reasoning).
How and when is it ascertainable that we have lived well (i.e., achieved eudaimonia)? According to Aristotle, "the quality of a life is determined by its activities . . . " and a happy person is" ... one who is active in accordance with complete virtue, and who is adequately furnished with external goods, and that not for some unspecified period but throughout a complete life" (NE,p.84). In this quotation Aristotle implies that things often associated with quality-of-life (pleasure, amusement, nourishment) are important only because they enable us to pursue the good (i.e., reason and moral virtue). As defined by Aristotle, " ... happiness itself cannot be acquired through sheer good fortune. Though fortune may give us some of the prerequisites and useful tools of a happy life, it does not give us happiness itself, since it takes great individual effort over a long period of time to develop the virtues whose exercise constitutes our ultimate end" (Kraut, 1989, p. 253). He also implies that it is toward the end oflife that we can look back on a "complete life" and judge whether it has been good.
As stated, Aristotle's thinking about how to live a good life has inspired this dissertation. Two of his perspectives are particularly relevant: he emphasizes that the person is embedded in community life, and he posits that older adults are ideally suited to inform us about how to live. In Psychology, Erik Erikson and Victor Frankl have also guided this dissertation. Erikson's (1982) psychosocial theory of development proposes that older adults are especially focused on this question of meaning-in-life; Erikson therefore preserves Aristotle's perspective that older adults can inform us about meaning-in-life. Erikson (1986) also stresses vital communal involvement (p. 44). Frankl emphasizes that it is through involvement in the social world that meaning can be discovered. Frankl's view therefore also accords with Aristotle's emphasis on social activity in describing the "good life." Aristotle, Erikson, and Frankl all provided theoretical foundations for this study.
Erikson's theory of human development emphasizes that development commences at birth and ends in death. The scope and richness of Erikson's theory cannot properly be presented here. However, for the sake of providing a foundation for the reader's understanding of the last two stages, which are most pertinent to this study, a brief overview (Erikson, 1982;1986) follows. The human life cycle is broken down into eight stages. At each stage, the individual strives to understand and organize experience with the social world around a developing sense of personal identity. Each stage involves a dynamic conflict between changes in the individual and change and demands in the social world.
As such, each conflict represents both increased vulnerability and potential for growth. Successful navigation through each stage results in ego strength, but successful navigation does not imply complete resolution of each conflict at its given stage. At each successive stage, earlier conflicts re-emerge and are resolved in relation to the current stage.
During the first year of life, infants need to develop trust in their own senses to convey the world to them, and they also need to trust that caregivers will meet their needs. Needless to say, trust is never complete, and throughout life it is inevitable that issues of mistrust confront the individual. But if foundational trust is secure at this stage, the infant will develop a sense of hope. In early childhood, the child begins to assert autonomy and independence of self, but inevitably encounters situations in which self assertion leads to shame and doubt. The ego strength associated with successful resolution of this conflict is will. The preschooler confronts the conflict of initiative versus guilt. This involves the goal to successfully influence and manipulate the social environment and is associated with the ego strength of purpose. During the school years, children face the conflict of industry versus inferiority as they strive to develop a sense of competence in socially valued skills. In the teen years, adolescents ask the questions "who am I?" as they face the conflict Erikson is most known for, identity versus identity confusion . This stage leads to fidelity, to be understood as loyalty to self-chosen values and commitment to roles. Erikson stressed that identity development begins in infancy and continues throughout adult life, but it is during adolescence that, in preparation for entering the adult world, the individual forges an independent sense of personal identity which integrates past experience, present reality, and future ideals. Young adulthood brings intimacy versus isolation to the fore, as the ability to love is developed and expressed. In middle adulthood, generativity versus stagnation involves a focus on productivity and creativity , leading to a virtue of care for the welfare of others. Finally, old age provides opportunity during the final stage of life, called integrity versus despair, for developing the strength of wisdom. All previous stages culminate in the ability to integrate accrued experiences into discerned wisdom. Erikson (1982) posited that toward the end of life an individual engages in a "life review" and hopes to achieve an overall sense that life has been good. This involves revisiting , reconciling, and reintegrating psychological themes of the previous seven stages of life, and coming to terms with the life one has lived so far. This is a complex stage of psychosocial development. "It is through this last stage that the life cycle weaves back on itself in its entirety , ultimately integrating maturing forms of hope, will, purpose, competence, fidelity, love and care into a comprehensive sense of wisdom" (Erikson, Erikson, & Kivnick, 1986, p. 57).
In Vital Involvement in Old Age (Erikson et al., 1986), Erikson, already in his nineties, conducted in-depth interviews with 29 parents of children ( every third Berkeley birth in 1928 and1929) taking part in the Guidance Study of the Institute of Human Development at the University of California at Berkeley. In this longitudinal study, data were gathered on the children and their parents. The follow-up study conducted by Erikson in 1981 involved interviewing these parents, now octogenarians, for whom life histories had been collected for over 50 years. Based on these interviews which aimed to elicit reflection about the experience of old age, Erikson found strong support for his last stage of integrity versus despair. In brief, this stage involves looking back at choices made along life's journey and working toward a sense of self integrity, while also confronting despair-that everyone experiences to some degree-as they face the inevitable and unavoidable fact of a finite existence. His informants were, for the most part, effectively balancing these tasks. That is, they were balancing efforts to find integrity in life as a whole while also facing the reality of death without despair (i.e., coming to accept death) . Moreover, they were also actively and meaningfully involved in the present. This involvement is vital (thus the title vital involvement) because the awareness that time is running out prompts an examination of how it may be possible to actualize some sense of immortality. This is achieved by caring for and preparing the next generation, a return to the midlife stage of generativity versus stagnation. In Erikson's own words: The capacity for generativity incorporates care for the present with concern for the future-for today's younger generations in their futures, for generations not yet born, and for the survival of the world as a whole.
It contributes to the sense of immortality that becomes so important in the individual's struggle to transcend realistic despair, as the end oflife approaches, inevitably. (Erikson et al., 1986, p. 75) Erikson has made major contributions to the understanding of how older adults experience life during the last stage of the human life cycle. His theory has generated considerable research and, in general, empirical support is found. Older adults do have a tendency toward philosophical reflection about their lives, and they focus on developing a sense of integrity (e.g., Prager, 1998;Reker & Wong, 1988). Erikson's pioneering work is extremely important to this study and will be further discussed in the Results and Discussion sections.
In psychology, the theorist who has most inspired research and analysis of the question of meanings in human life is Victor Frankl. The central assertion of Frankl's theory is that the universal striving to discover meaning-in-life, the "will to meaning," is the primary motivational force in human behavior; it is a uniquely human need and distinct from all others (Frankl, 1959). Moreover, meaning-in-life is unconditional; it can be experienced by all people and under any life circumstances. In his classic book, Man's Search for Meaning, Frankl (1959) contrasts this "will to meaning" with Freud's "will to pleasure" and Adler's "will to power" (all three represent reactions to Nietzsche's view that human beings are free to make decisions and are instinctually driven to express their power to choose values and to determine how they live their lives. Frankl departs from Nietzsche's doctrine that meaning is an outcome of a drive; For Frankl, the will to meaning is about self transcendence, and stresses individual freedom to assume responsibility for fulfilling specific meanings { Gould, 1993} ).
For Frankl, consciousness of latent perceptions involves intentionality and will.
Frankl stresses that to be human is to be conscious and responsible. Thus, human beings are conscious of their responsibility to find meaning. Also stressed is the fact that each individual is free to choose how he/she will live and that this meaning-seeking impulse prompts a phenomenological experience. Of any individual human being, Frankl ( 2000) writes: By virtue of what I call pre-reflective ontological self-understanding, or what is also called "the wisdom of the heart," he knows that being human means being responsible for fulfilling the meaning potential inherent in a given life situation. What is even more important, the man in the street knows that meaning may be found not only in creating a work and doing a deed, not only in encountering someone and experiencing something, but also, if need be, in the way in which he stands up to suffering. (p. 122) Frankl stresses that meaning is to be found "out there" in the world ; a person is more fully human when she/he forgets herself/himself and experiences what Frankl (2000) considers the essence of human existence, self transcendence (p. 138). Self transcendence involves serving a cause or demonstrating care and concern for others. In other words , self transcendence requires that we depart from our customary tendency to be "self' centered and become truly aware of self through its interaction with the social world. Such activity may include: (1) work, achievement or accomplishment; (2) experience of goodness, truth, beauty, and love; and (3) attitude toward unavoidable suffering (Frankl, 1959).
In his last book, Man's Search for Ultimate Meaning, Frankl (2000) argues that there is an ultimate meaning to life, but this spiritual knowledge is unconscious and must therefore be revealed to consciousness. Like the "embedded figure," it is already there; it is not created but discovered. In this work, Frankl emphasizes a personal perceptual experience of this ultimate meaning (i.e., God) through the same "will to meaning" ( p. 113). Frankl, like Freud, acknowledges that the unconscious cannot be known directly. It is revealed to us to the extent that we are able to go beyond ourselves and identify latent meanings manifested in specific life experiences. The intuitive conscience is the spiritual element of life that guides choices, fosters discovery of meaning, and underlies the impulse toward self transcendence. " ... conscience lets us arrive at the unique meaning gestalts dormant in all the unique situations that form a string called a man's life." (p. 114).

Literature Review: Meaning-in-life
As mentioned, Frankl' s work achieved distinction for inspiring psychological research on meaning. Since Crumbaugh & Maholick (1964) developed the Purpose in Life Test, which is based on Frankl's theory, it is rare to find any study of meaning-in-life that does not reference the "will to meaning" as philosophical foundation. Furthermore, the Purpose in Life Test was developed specifically to operationalize Frankl's construct of meaning-in-life. The importance of Frankl's theory in generating research in psychology cannot be overstated; the Purpose in Life Test made it possible to objectively study meaning-in-life. Yalom (1980)  More recent research has continued to find significant associations between meaning-in-life and both physical and psychological well-being (Debats, Drost, & Hansen, 1995;Fry, 2001;Harlow, Newcomb, & Bentler, 1986;Harlow, Newcomb, & Bentler, 1987;Reker, Peacock & Wong, 1987;Ryff & Singer, 1998;Takkinen & Ruoppila, 2001;Zika & Chamberlain, 1992). A recent meta-analysis conducted by Pinquart (2002) also corroborates these associations.
Conceptualizing meaning-in-life as roughly synonymous with purpose in life (i.e., "having goals in life and a sense of directedness, feeling that there is meaning to past and present life, holding a belief that gives life purpose ... "), Pinquart synthesized findings from seventy studies focusing on middle and late adulthood.
He concluded that high purpose in life is strongly associated with psychological well-being, while low purpose in life is associated with depression. Additionally, the following were correlated with purpose in life: social integration, relationship quality, better health, higher socio-economic status, being employed, being married, and higher competence. Relationships to spouse, family, and friends were frequently reported as sources of purpose in life, and the quality of relationships in general, and with family in particular, showed the strongest association with purpose in life. Good health and competence were associated with the ability to engage in activities that promote a sense of meaning. Pinquart also found that a sense of being useful and needed is a "core component of purpose in life" (p. 93).
As with the research reviewed by Yalom, Pinquart's analysis was also correlational. A number of more interesting studies have used sophisticated methodologies to demonstrate that meaning-in-life can play a mediating role. For example, Harlow, Newcomb, and Bentler (1986) used structural equation modeling to demonstrate that depression and self-derogation lead to lack of purpose in life among adolescents, and that lack of purpose in life, in tum, leads to substance use and suicidal ideation. The mediational model was validated by demonstrating that depression and self-derogation do not lead to substance abuse and suicidal ideation except through their influence on purpose in life. That is, a depressed person is not likely to abuse drugs unless she/he has low purpose in life, and people with high purpose in life are unlikely to abuse drugs even when depressed. In another study, Newcomb and Harlow (1986) found that purpose in life mediated the relationship between uncontrollable stress and substance use. Harlow and Newcomb (1990) have also used modeling techniques to investigate the components and processes of meaning and satisfaction in life.
Using a sample of 739 young adults, they found support for nine factors associated with meaning-in-life. In their hierarchical model, these nine factors were explained by three second order factors: (1) relationship satisfaction (peer, intimate, and family); (2) perception of purposeful living (purpose in life, meaninglessness, powerlessness, and perceived opportunity); and (3)  As the above review makes clear, since Frankl' s (1959) Man's Search for Meaning, a substantial body of literature has accumulated with regard to the dimensions, structure and correlates of meaning-in-life. In addition to the Purpose in Life Test (Crumbaugh & Maholick, 1964), other measures are also now available: the Life Regard Inventory (Battista & Almond, 1973), and the Sense of Coherence (Antanovski, 1979). While considerable gains have been made, these research contributions still fall short of illuminating how individuals give meaning to lived experience. What is needed is an orientation toward detailed description of how people experience meaning in their lives-in their own words and in their own voices. In other words, we need to examine narratives as individuals focus and reflect on their own lives and articulate experiences with purpose and meaning. Narratives represent the integration of experiences into a unified story that speaks to us of ourselves in the past , present and anticipated future. As Johnson (1989) suggested " ... .life is wrapped in stories , memories, language, and philosophy" (p. xii).
These stories are best captured and analyzed by qualitative approaches and by the interview method in particular (K vale, 1996). The interview can reveal how individuals make sense of their lives and organize subjective experiences into personal narratives. An obvious advantage to this approach is that participants can literally speak for themselves. Surprisingly, few qualitative studies have been conducted by psychologists to investigate meaning-in-life .
However, one must remember that meaning-in-life was not a subject for psychological investigation until it could be objectively measured (i.e., the Purpose in Life Test). Viewed from this perspective, it is not so surprising that qualitative investigations have been limited .
A number of studies have combined qualitative and quantitative methodologies (i.e., written or oral responses to specific questions and some measure of meaning-in-life, respectively). Typically , a qualitative component supplements or complements a quantitative one. For example , Debats , Drost , and Hansen (1995)  Using a similar design, Ebersole and DePoala (1986) administered the Purpose in Life Test (Crumbaugh & Maholick, 1964) to late-life couples and then asked for a written statement about their strongest source of meaning-in-life. The most frequently reported categories of meaning were: family relationships, health, and pleasure. Ebersole and DePoala (1988) also compared older and younger samples, and reported that older adults were found to have higher scores on the Purpose in Life Test but scored significantly lower on an external validation measure, which consisted of a written statement about their strongest meaning-in-life.
Yet another example of a combined qualitative-quantitative research design is provided by Baum and Stewart (1990). The Purpose in Life Test was administered to a sample ranging from age 17 to 96. Respondents also answered questions about their most meaningful life event. Researchers reported the following findings: there were no age or sex-related differences in amount of meaning, the sources of meaning also did not change across age categories, and involvement in love and work were the most often cited sources of meaning.
These four studies were not explicitly designed to reveal how individuals make sense out of their lives through narrative themes. Rather, their focus was more on validating the instrument they used to quantify meaning-in-life. A purely qualitative investigation of meaning-in-life was conducted by O'Connor and Chamberlain (1996). As these authors observed, most of the mixed qualitativequantitative studies such as those discussed above, are inadequate if the aim is to examine the multidimensional structure of this construct. They reasoned that structured interviews would be the best way to approach people for their accounts of life meaning. Fourteen men and twenty-six women at midlife were asked to discuss important sources of meaning in their lives. O'Connor and Chamberlain identified the following as the most frequent sources of meaning: relationships, creativity, personal development, nature, religion/spirituality, and social/political action and beliefs. The importance of relationships was reported by 100% of participants.
O'Connor and Chamberlain's study is noteworthy for its qualitative approach, but unfortunately it does not reveal how meaning-in-life is experienced from the unique and important perspective of older adults. As suggested by Erikson, with advancing age, experiencing life as meaningful takes on greater urgency as individuals face existential questions and integrate their life experiences into a meaningful whole (Battista & Almond, 1973;Baum & Stewart, 1990;Erikson et al., 1986). Frankl's view also suggests that accrued experience with meaning seeking generally leads to greater insights in later life. He uses the analogy of a film to demonstrate this point (Frankl, 2000, p. 143). Each of the thousands of individual pictures constituting a film has meaning in itself, but it is in combination, and not until the last in the sequence of pictures is shown, that the meaning of the total film is discovered. Frankl adds that the whole film cannot be understood unless individual components are first understood. So it is, argues Frankl, with a human life. Unfortunately, much of the research has used convenience samples, predominantly college students and, ironically, there is a paucity of investigation on the life meanings of older adults, those in the best position to describe how life can be lived meaningfully. Kaufman (1986), who utilized interviews to gather life stories focused on exploring the relationship between the experience of aging and identity, agrees that "If we can find the sources of meaning held by the elderly and see how individuals put it all together, we will go a long way toward appreciating the complexity of human aging and the ultimate reality of coming to terms with one's whole life" (p. 188). Kaufman's work will be relevant to the discussion of results.

Literature Review: Quality-of-life
In an effort to answer the question of what constitutes a good life, psychological research has also focused on quality-of-life. Regrettably, qualityof-life research is fraught with theoretical and definitional inconsistencies (Rapley, 2003). A number ofrelated terms are used almost interchangeably: quality-of-life ( e.g., Frisch, 1994), subjective well-being ( e.g., Chamberlain, 1988;Diener, 1984), life satisfaction (e.g., Hamarat, Thompson, Steel, Matheny, & Simons, 2002;Hilleras, Jorm, Herlitz & Winblad, 2001), and happiness ( e.g., Myers, 1992;Ryff, 1989;Seligman, 2002). This precludes a systematic exploration of the construct. Diener, Suh, Lucas, and Smith (1999) made a considerable contribution to the field by demonstrating how these constructs with different labels and theoretical nuances could be integrated into a structural model. They reviewed the literature on subjective well-being and found justification for a subjective well-being factor comprised of independent lower-order components. These components include: (1) pleasant affect (e.g.,joy, contentment, happiness); (2) unpleasant affect (e.g., guilt, sadness, anxiety); (3) life satisfaction, the cognitive component ( e.g., "If I could live my life over, I would change almost nothing"); and (4) satisfaction with specific domains (e.g., home, leisure, health, finances).
Diener et al. thus emphasize affective, cognitive and situational constituents of the construct "subjective well-being." Diener (1984) developed the Satisfaction with Life Scale as a measure of global life satisfaction. It has been subjected to numerous reliability and validity studies (Diener, Emmons, Larsen, & Griffin, 1985;Pavot, Diener, Colvin, & Sandvik, 1991;Pavot, Diener, & Suh, 1998), and has been widely used in psychological investigation. In this scale, a small discrepancy between aspirations and achievement is associated with higher perceived level of satisfaction.
Diener's Satisfaction with Life Scale consists of five items which assess only the cognitive component of subjective well-being (i.e., the extent to which goals, needs and wishes are fulfilled). Frisch (1994) argued that a definition of Inventory (QOLI) based on the articulated theory that life satisfaction is best conceived of as the interaction between cognitive and affective components. The QOLI assesses the degree of satisfaction with 16 life areas ( e.g., health, helping, love, friends, creativity, play, money, community), and for each of these 16 areas, it also asks respondents to indicate how important the area is to her/his happiness.
It is argued that a person's satisfaction with a particular area (with work, for example) contributes minimally, if at all, to life satisfaction, if the area is not important to her/his happiness. Discrepancy between aspiration and achievement in those areas considered important or extremely important to happiness, determines quality-of-life.
As stated, because different operational definitions are used in quality-oflife research, it is difficult to integrate research findings (Rapley, 2003).
However , irrespective of assessment used, studies generally converge on two findings: First, quality-of-life is negatively skewed; most people are satisfied with life ( e.g., Cummins, 1995). Second, older adults are as satisfied-if not more satisfied-with life than younger adults ( e.g., Diener et al., 1999;Hamarat et al., 2002;Pavot et al., 1998;). This is the case in both national and international studies. Diener and Suh (1997) summarized data from five largescale international studies on subjective well-being and report that, even though objective circumstances that generally relate to well-being are lower for older adults (i.e., low income, health concerns), their overall life satisfaction is higher.
This paradox suggests that subjective appraisals of life conditions contribute significantly to quality-of-life judgments. This international finding is robust, but more research is needed to shed light on how and why older adults tend to view life in more positive terms. It is conceivable that at this age in the life cycle, more abstract concerns with meaning-in-life lead to a re-framing of daily life conditions.
The relationship between quality-of-life and meaning-in-life is indeed intriguing. Zika & Chamberlain (1992)  Not surprisingly, they found a significant canonical correlation. The meaning-inlife variate explained about 50% of the variance in the well-being variable set.
They concluded that the two sets of measures do overlap but noted that each set has unique variance, thus supporting the argument that these are distinct constructs. Meaning-in-life is more about purposeful existence while quality-oflife pertains more to satisfaction with life conditions. Zika & Chamberlain speculate on the nature of the obtained correlation by stating that meaning-in-life is often accompanied by feelings of life satisfaction.
Zika and Chamberlain's interpretation is consistent with Frankl's theory.
Frankl's position does not deny the importance of quality-of-life, but it asserts that it is neither necessary nor sufficient for a sense of meaning-in-life. Without a doubt, Frankl would view quality-of-life and meaning-in-life as distinct. He would argue that meaning can be found in any situation, time, or place. In fact, his thesis is that even when severely deprived of physical comforts, even when one is stripped of everything that is important, the one thing that can never be taken away from an individual ( except by death itself), is the ability to think freely. Our freedom to choose how we will interpret and how we will respond to our lived experience is an inalienable and quintessential feature of human existence. It underlies our ability to discover meaning . Frankl is in dramatic opposition to Maslow' s contention that a certain standard of living is essential before an individual can pursue self actualizing needs (i.e. meaning). Since Frankl was fond of the "man on the street," it is worth noting that even lay conceptions of a good life seem to make the distinction between meaning-in-life and quality-of-life (King & Napa, 1998;& Westerhof, Dittmann-Kohli, & Thissen , 2001).
To conclude this literature review, we must recall that investigations into meaning-in-life have predominantly relied on quantitative approaches. As discussed, the Purpose in Life Test was designed so that meaning-in-life could be investigated (i.e., method preceded the larger research questions). To counterbalance these methods, qualitative approaches that are more congenial to the nature of meaning-in-life are needed. Camic, Rhodes, & Yardley (2003) use the analogies of a map versus a video to discuss the differences between quantitative and qualitative methodologies : A map is extremely useful; it conveys with economy and precision the location of a place and its relationship to other places in terms of proximity and direction. However, even the most detailed map is unable to convey an understanding of what it is like to be at that place. In contrast, a video conveys in vivid detail the constantly changing perspectives of the observer. Although this perspective is selective and could not be easily used for navigation, it is able to communicate something of the subjective experience of being there. (p. 10) In the case of meaning-in-life, the investigator wants to be right there as informants reflect on meaning in their lives and articulate their subjective experiences in their own words and in their own voices. Ironically, qualitative approaches to this subject have been limited. Another serious limitation in approaches to this important subject of study is the tendency to use primarily younger samples. As discussed, this neglect of older adults' voices is unfortunate, given their privileged place in the life span.

Study 's Goals and Research Questions
This study will contribute to research on meaning-in-life by using a qualitative strategy of inquiry and by investigating this construct with older adults (i.e., age 65 and older). The primary goal of this study was to capture the thoughts and feelings, largely ignored in the extant literature, of individuals as they reflect on their subjective experience of meaning-in-life. The aim was to investigate how older adults derive meaning from life experiences and how their narratives are articulated in linguistic terms. The following research questions addressed this goal: (1) Do people over the age of 65 experience that their lives in general have been and are currently meaningful? What language do they use to express these meanings?
(2) What are the characteristics ( e.g., events, circumstances, life experiences, insights) associated with the judgment that life has been/is meaningful?
A second goal of this study was to determine how older adults experience and judge their quality-of-life and to examine the relationship between meaningin-life and quality-of-life. This would improve upon Zika and Chamberlain's study by investigating meaning-in-life through the interview method, the benefits of which were previously discussed. It would also improve upon O'Conner and Chamberlain's study by using a sample of older adults. This study also used a quantitative approach, the Quality-of-life Inventory (Frisch, 1994), to assess participants' subjective estimates of quality-of-life. The following research questions pertain to this goal: A final aim of this study is to examine the developmental question of whether older adults' self reports of meaning and quality-of-life differ from estimates of these same constructs as they apply to their retrospective judgments of their earlier lives. Thus, older adults will be asked to report on these constructs in the present and to retrospectively consider how constructs generally apply to the life they've lived. A related developmental issue is whether older and younger adults differ in their subjective self estimates of quality in their lives.
Thus, the following research question will be explored: (6) Are there significant differences between older and younger adults on their self-reported estimates of quality-of-life? And, do these two groups differ in the domains which most contribute to an overall rating of quality-of-life?

The Phenomenological Approach
The strategy of inquiry for the qualitative component of this study was phenomenology. A phenomenological stance is compatible with qualitative methods in general and with interview-focused studies in particular (Georgi, 1985;Georgi & Georgi, 2003;Kvale, 1996). Moreover, studies using a phenomenological perspective with older adults have found it ideally suited for this population. For example , Hedelin & Strandmark (2001) interviewed five elderly women suffering from depression , and Montgomery , Barber , and McKee (2002) interviewed six older adults about their experience of wisdom. Both studies provide descriptions of the respective phenomenon from the perspective of the individuals experiencing it.
The phenomenological approach to scientific investigation offers several guiding principles: First, it emphasizes the description of phenomena as they are experienced by people in everyday life. Husserl called for a return to the things themselves (Moran , 2000 , p. 107) in reference to this re-orientation to science. He argued that science should aim to describe and understand direct experience-as it is experienced-in the world . Secondly, scientists need to describe what is conveyed as it is presented to them, with a conscious effort to avoid preconceptions and biases about the phenomena being studied from distorting what is observed. Husserl referred to this practice as bracketing (Moran, 2000, p . 145). Thirdly , reporting of results should also avoid imposing theoretical interpretations on that which is described . The data should speak for themselves, or better yet, participants should speak for themselves-in their own words, and in their own voices . Husserl used the term reduction to refer to this process whereby what is described is reduced to what is given in consciousness. But as Moran points out, Husserl's followers, and even Husserl himself, subsequently rejected the notion that a complete reduction is possible, and Merleau-ponty in particular, reminds us that this impossibility is important for the scientist to be mindful of (Moran,p. 161 ). While acknowledging the limits associated with any attempt at reduction, it is nevertheless important to describe our experience of phenomena while attempting to minimize the imposition of theoretical interpretations. One must also acknowledge that phenomenal experience essentially grows out of dialogue between the perceiver and that which is perceived. As Johnson (1989) stated "gaining insight about the meaning of another's action and life is a dialogical situation between us who wish to understand and the person who wants to be understood" (p. 110). These principles represent the phenomenal stance adopted in this study; they will guide and inform both attitude and practice at each phase of this study. Frankl (2000) endorsed phenomenology as the method for learning about the wisdom of the "man on the street. He wrote: "As I see it, it is the assignment of phenomenology to translate this wisdom of the heart into scientific terms" (p. 126).

Participants
This study used a sample of younger adults (n = 90, Mage= 22.09, SD= 6.77) and one of older adults (n = 115, Mage= 79.18, SD= 7.04). The sample of older adults consisted of Rhode Islanders aged 65 and over, recruited from Section-Eight housing and senior centers (see appendix A). All 115 participants completed the Quality-of-life Inventory (QOLI). Of those who agreed to participate at each site, two individuals were randomly selected to be interviewed .
At Smithfield, three individuals were interviewed because a third person who did not understand random selection had expected to be interviewed.
The specific recruitment sites were selected in consultation with the Director of Aging 2000 (L. Purcell, personal communication, July 16, 2004). The five Section-Eight housing residences were selected to maximize cultural and income diversity as well as geographic representation in R.I., and the five senior centers were selected based on such diversity, as well as size and activity level.
In order to get a more representative sample, it was necessary to recruit participants from these two types of facilities because, according to the Director of Aging 2000, adults who frequent senior centers are generally healthier, more active, and have greater financial resources than residents of subsidized housing.
This prediction (discussed in Results) was corroborated. It was therefore important to recruit from both Section-Eight housing and senior centers in order to get a wider and more balanced representation of Rhode Island seniors.
The comparison sample of younger adults was randomly selected from the population of day students enrolled in General Psychology at the Flanagan Campus of the Community College of Rhode Island in the spring of 2005. A total of 425 students were enrolled in seventeen daytime sections one week after classes commenced. From these, 110 were randomly selected.

Instruments
The Quality-of-life Inventory. The Quality-of-life Inventory (Frisch, 1994) assesses an individual's overall quality-of-life based on how satisfied she/he is with various areas of life weighted by ratings of importance of each of these areas to overall life satisfaction. It consists of 32 items about 16 key areas of life, for example, health, self-esteem, goals and values, love, and community. There are two items for each of the 16 life domains: (1) how important the area is to the respondent's happiness, and (2) how satisfied he/she is with this aspect oflife. In each area, satisfaction scores are multiplied by importance scores to provide an index of weighted satisfaction for each area. The overall Quality-of-life Inventory (QOLI) raw score is the average of the weighted satisfaction ratings. Scores range from -6 to 6, and are classified as high (3.6-6), average (1.6-3.5), low (.9-1.5), and very low (-6 to .8).
The QOLI was normed on a national sample ( N = 798) that closely matched the U.S. 1990 Census with respect to race and ethnicity, and with an age range of 17 to 80 (Frisch, 1994). Psychometrically, the QOLI appears to have reasonable internal consistency; the manual reports coefficient alpha at .79. The test-retest reliability estimate of .73 over a two-week period was also reported in the manual (Frisch, 1994). Convergent validity was demonstrated through significant (p < .001) correlations with other measures oflife satisfaction (e.g., Quality-of-life Index and the Satisfaction with Life Scale). QOLI scores were also significantly correlated with the Marlowe-Crowne Social Desirability Scale (r = .25, p < .001), but only 6.25% of the variance in QOLI scores was explained by the Social Desirability measure. It therefore appears that participants' motivation to present themselves in a socially desirable manner was not a major problem.
The QOLI is appropriate for an adult sample because it is written at the sixth grade level, takes approximately five minutes to complete and, according to its author (Frisch, 1994), scores can be considered a measure of successful aging.
In this study, The QOLI was found to be internally consistent for both the older and younger adult groups. Cronbach's Coefficient Alpha was computed for seniors and college students independently. Alpha values of .88 and .84, respectively, were obtained. These are consistent with and exceed the alpha of .79 reported in the QOLI manual. Additionally, all items were significantly correlated with the overall QOLI score (all at the .01 level of significance).
According to DeVellis (1991), alpha (I-error variance) is the proportion of a scale's total variance that can be attributed to the underlying construct. All items on the QOLI were found to make significant contributions to the overall score.
Thus, the inventory is internally consistent.
Quality-of-life ratings: past and now. The QOLI was augmented by two questions, presented separately, intended to elicit overall ratings of quality-of-life as retrospectively judged for life as a whole, and for life as it is experienced in the present (see Appendix G). These two questions were answered by all younger and older participants as follows: (1) On a scale of 1 to 10 (1 = very low quality, IO = very high quality), please circle the number that best represents how you estimate your overall quality-oflife right now: (2) On a scale of 1 to 10 (1 = very low quality, IO= very high quality), as you look back on your life up to now, please circle the number that best represents how you would rate the overall quality-of-life you've had up until now.
The interview. The interview, according to Kvale (1996), is an intimate conversation between two people on a theme of mutual interest and (hopefully) mutual benefit. It is, states Kvale, "particularly suited for studying people's understanding of the meanings in their lived world, describing their experiences and self-understanding, and clarifying and elaborating their own perspectives of their lived world" (p. 105).
In this study, semi-structured, in-depth interviews were conducted by a professional counselor trained in the practice of empathic listening. An interview guide was used in order to obtain responses on key concepts being explored (see

Procedure
The QOLI: older adults. The researcher contacted administrators at each site to introduce the study and request permission to visit and recruit. As followup to the initial contact, written and detailed information was provided, including a brief description of the study's goals, the informed consent forms, the demographic sheet, the QOLI, and the interview guide. In most cases, this was followed by a meeting where any remaining concerns were addressed with the site director, specific details were discussed, and arrangements were made for future visits.
For the purpose of actually collecting data, the number of visits made to each site varied from one to three, with most requiring two. Visits were scheduled on dates when structured activities were taking place to maximize the participant pool. A flier (see Appendix B) containing the pertinent information for each site was prepared and sent to activities coordinators for posting. Potential participants were asked to arrive an hour prior to the start of the scheduled activity to allow time for participation in the study. With each visit, the researcher introduced herself and briefly spoke about the project's goals. All present were invited to take the QOLI. Anyone who agreed to participate was then given a packet which consisted of the informed consent form (in Appendix C-1), the demographic sheet (in Appendix D), the QOLI, and the quality-of-life ratings for now (QOL-Now) and in the past (QOL-Past), available in Appendix G.
Prior to giving instructions for participants to commence, a lottery selection identified two participants from each site who would be interviewed.
Because the QOLI has the potential to expose gaps between how a person wants to live (i.e., what's important) and how she/he actually does live (i.e., satisfaction), it is possible that for some, an awareness of dissatisfaction in the most valued areas may have lead to negative affect. Mental health referral information (see Appendix H) was provided to all participants as a cautionary measure.
Meaning in life interviews: older adults. Prior to interviewing, the researcher provided participants with a brief overview of how the interview would proceed and answered any questions pertaining to their involvement. The informed consent form for the interview, tape recording, and quoting was then read and signed by interviewees (see Appendix C-2). The fact that participation was voluntary and that participants were free to terminate at any time during the interview was specified in the consent form, and was also stressed in preliminary discussion. The researcher then collected demographic data while stressing to participants that their confidentiality would be protected and that identifying information would not be reported.
Each interview commenced-and the taping began-when it was clear that the participant understood his/her role in the study and was at ease with the decision to be both interviewed and taped. Control over the tape recorder was given to the interviewees; they were explicitly instructed on how to turn it off at any point they wished. These efforts to give as much control as possible to participants were deliberate and were intended to create an environment of mutual respect and trust. It must be stated that none of the interviewees showed any hesitation or discomfort with regard to the taping . None of the interviews were prematurely halted, but two individuals did exercise their control over the taping.
In both cases, they asked if their identity would be protected, and once re-assured, the interview resumed.
Interviews were conducted in a quiet space arranged by the site administrators and conducive to intimate conversation except for two cases where follow-up visits were made to participants' homes. Of those that were randomly selected at all sites, five declined to be interviewed ( one each from Central Falls, East Greenwich, East Providence, Jewish Community Center, Barrington) . For four, the reason given was not having the time. When the suggestion was made to schedule an appointment at their convenience, they responded, "I'd rather not." One person simply said, "No, thank you." In each of these cases, another person was randomly selected to replace the one who declined to be interviewed.
Interview length ranged from fifty minutes to over two hours. Every effort was made to foster a conversation-like experience. Typically, interviewer and interviewee sat across from one another, drinking coffee and sharing a psychological space where trust, understanding, and empathy, permitted connection at a deeper level. Interviewees were treated as valued informants.
They were often reminded that their stories were of great interest. This reminder was both explicit (i.e. verbal) and implicit; the interviewer listened attentively and with undivided attention. Indeed, it became immediately apparent that taking any kind of notes while interviewing was disruptive and counterproductive. Thus, while it was anticipated-and proposed-that notes on major themes would be maintained , such did not occur. Rather, concentration focused fully on the unfolding story. At points of transition in conversation, summaries of major themes expressed in response to each interview question were presented, and participants were immediately given the opportunity to judge the accuracy of these summaries and to clarify or amend them as necessary. Compensation in the form of a $10.00 gift certificate to the Stop & Shop supermarket was made at the conclusion of each interview.
Since interviews may have evoked deeply personal and possibly painful memories in participants, a debriefing was held after each interview.
Interviewees were encouraged to share their experiences relative to the interview .
Without exception, they were thankful for the opportunity to reflect on and speak about their lives. All participants appeared stable in mood and composure.
Nevertheless, as a safety measure, the list of mental health referrals (see Appendix H) was distributed at this time, and they were encouraged to seek assistance if they should need it in the future. Thank you notes were sent to each participant approximately one month after the interview took place (see Appendix I).
QOLJ: the Comparison Sample. As previously described, the comparison sample of younger adults who took the QOLI was randomly selected from General Psychology class rosters at the Community College of Rhode Island in January 2005. For each of 17 sections, the investigator visited each class and (1) introduced the project as it relates to the scientific method in psychology, (2) discussed the survey method of investigation, (3) provided an overview of ethical standards as these apply to research with college students, (4) illustrated how/why random selection had been used in this study, and (5) asked those selected to participate by filling out the QOLI. Out of 110 selected, 15 were not available, 3 declined, and 2 were disqualified due to incompleteness, leaving a final sample size of 90.
It was emphasized to students that their participation was purely voluntary, that their anonymity was guaranteed, and that no extra credit would be awarded. The consent form was distributed, and subsequent to students' review and signature, it was collected (see Appendix C-3). Those students willing to participate were then given the demographic sheet and the QOLI to complete. As with the older sample, the quality-of-life ratings for now (QOL-Now) and in the past (QOL-Past) were supplemental to the QOLI and were obtained separately.
Order of interview and inventory. As originally planned, older adults to be interviewed were to be randomly selected from all those willing to participate at each site, and in order to eliminate influence of the QOLI (i.e., salience of concepts or language) on interview responses, the interview would precede the inventory. However, this was not always practical or feasible. Those selected to be interviewed did not want to sit around and wait while others filled out the QOLI. Suggestions to be interviewed at a future date were not well received.
Generally, people who participated fully expected to contribute on the same day as their peers. It was also discovered that participants were either too tired or were no longer interested in filling out the inventory after having been interviewed.
Consequently, a second visit to each of four sites was necessary for the sole purpose of obtaining the QOLI data from those previously interviewed.
At each of the last five sites, a modification in procedure was made. All adults willing to participate first filled out the inventory. As inventories were collected, the two that were randomly selected were informed and invited to participate in the next phase. Interviews immediately followed the inventories.
Since at half the sites the interviews preceded the inventory, the possibility exists that interviews might have influenced inventories. Conversely, in half the cases, the inventory may have influenced the interview. This issue will be explored in the Results section.

Data Analysis
Qualitative analysis: interview data. Georgi & Georgi (2003) outline the specific steps of the phenomenological method once interviews have been fully transcribed to produce the raw data . They emphasize that throughout analysis the researcher must remain within the scientific phenomenological reduction: (1) reading through the entire text for a sense of the whole; (2) establishing meaning units , which involves making meaning discriminations in the text and delineating them; (3) transforming each of these individual meaning units (in the participant's own words) into explicit psychological descriptors with regard to the phenomenon; and (4) determining structure, which involves movement toward psychological essences of the phenomenon by synthesizing across data from all participants. With noted modifications , these steps to analyzing interviews were followed.
Each interview was transcribed from the tape recordings verbatim including repetitions, pauses, sighs, and other expressions . Once this was done , a trained research assistant listened to the tape while simultaneously reading the written text and noted any discrepancies. Discrepancies were subsequently discussed and resolved between the researcher and the research assistant after listening to the tape together. The researcher then read each transcript in full (multiple times as necessary) to form an overall impression of general themes. It was important also to read any relevant notes, such as reflections in the project journal or impressions recorded immediately following the interview session.
The researcher then returned to the beginning of the transcript, and "meaning units" (Georgi & Georgi, 2003) were identified and marked. Each time that a transition in meaning was perceived in the text, it was delineated or tagged (Georgi, 1985). Meaning units tended to correspond to specific questions on the interview guide. Once all the meaning units had been delineated, each was considered with respect to the psychological insight to which it pertained. The coding of these psychological elements was done by two independent readers.
Inter-coder reliability . Inter-coder reliability was addressed in two parts.
First, seven of the twenty one interviews were randomly selected for coding by two independent coders; in addition to the principal investigator, a highly qualified college counselor was enlisted as the second coder. She was first encouraged to assume a phenomenological attitude toward the data. Specifically, she was asked to try to set aside any preconceptions or biases she might have about older adults or about meaning-in-life issues (i.e., bracketing). After receiving guidance on the nature of the phenomenological attitude, she was also instructed to make an effort to describe but not interpret the written text. Copies of the interview transcriptions were provided along with a coding template which guided the coder as she progressed through the text and identified answers to interview questions as well as follow-up queries (see Appendix F). This coding guide was intended to standardize the procedure as much as possible and to increase the ease of coding. The two independent codings were then compared by two psychologists who concurred with the remarkable degree of agreement between the two coders. In fact, out of a total of 223 codes made by each coder on the seven interviews, minor discrepancies were found in only four cases.
This impressive inter-coder reliability ratio is attributed to the ease with which answers to interview questions were identified in the text. In most instances, the coding preserved the exact language used by participants. As noted by the two psychologists, the codings were obvious and self-evident. High intercoder reliability is also explained by the fact that conscious effort was made by both coders not to provide interpretation, clinical or otherwise, but to simply describe and summarize responses to each question on the interview guide.
Because the percentage of exactly replicated codings was so high on these seven interviews (98.2%), coder reliability on the other 14 interviews was demonstrated in a more expedient manner. The principal investigator coded each transcript, and the second coder subsequently read the interview text, considered the codings made, and noted any discrepancy between the existing code and how she would have coded. Once again, inter-coder reliability was exceptional; out of 460 codes, only 10 (2.2%) were slightly discrepant. These discrepancies were resolved through discussion between the two coders and subsequent agreement reached.
Internal consistency of interview guide. In addition to inter-coder reliability, another issue arises with respect to the consistency with which questions were asked of individual participants. In this study the interviewer is viewed as an instrument, and, as such, questions of internal consistency must be addressed. While participants spoke directly and conscious effort was made to represent what they offered in their own voices (i.e., without interpretation), the entire process could be fundamentally compromised by inconsistent modes of questioning. To explore this issue, verbatim interview questions were extracted from each interview and compared across the twenty one interviews.
Examination of results revealed remarkable consistency in the phrasing of interview questions. Table 1 presents a list of the interviewer's actual articulations pertaining to question two. As this table demonstrates, there is minute deviation from the prototypical question found in the interview guide.
While the same interviewer was used for all interviews and asked the same questions of all participants, this does not mean that a standard oral protocol was always followed. In fact, consistent with a phenomenological attitude, the number one priority was to establish an environment where informants could feel that they as unique individuals had much to offer and that their personal and distinct experiences were respected and appreciated. Thus, from the onset, sensitivity, openness, and flexibility were imperatives. Adjustments in the time required to establish rapport, for example, had to be made. Some individuals were eager and ready to begin; others were more reticent. In short, the same questions were asked of all participants, but the pace and sequence varied as appropriate. The internal consistency of the interview guide is important to address for two reasons. First, while it was important to ask the same standard questions of interviewees, the social contexts in which interviews took place were far from standard. In some cases, the individuals were present to a site for the sole purpose of participating in the study, but in others their interest was to engage in some other social activity after their participation in the study. This undeniably-and perhaps automatically and even unconsciously-influenced the pace at which the interview progressed. For example, at the Cranston Senior Center, a holiday dinner was about to take place, and one hour prior to the dinner's announced start time, the center was already buzzing with activity and excitement. The person I interviewed on that day was expected at her reserved table. This circumstance probably exerted some pressure to expedite the interview.
Secondly, participants' responses also varied in their length and complexity. In some cases, they responded with specificity and brevity. In others, they responded to one question and while continuing to elaborate spontaneously began addressing another question that had not yet been asked.
This necessitated adjustment in the sequencing of questions to maintain smooth transitions and flow in the discourse. In other words, intuition and attunement to the interviewee's inferred thought process guided the progression and course of each interview.
The researcher was mindful from the onset that a standard interview protocol would be important to internal consistency and would furthermore facilitate data analysis. However, this focus was tempered by an equally important goal, namely, that of presence, sensitivity, and complete absorption in the intimate human stories that were unfolding. It was important to the validity and reliability of interview data that participants not be intimidated by the formality of the situation. Conscious effort was made to convey genuine interest in their individual thoughts and reflections . It was hoped that this salient interpersonal milieu would evoke spontaneous and authentic descriptions of lived experience.
Some participants initially presented an inability to articulate their thoughts or feelings with respect to certain questions. However, they appeared to have gained insights as the interview progressed. In fact, as they reflected on their experience, answers came rather more spontaneously. This gave the impression that they themselves did not know how they would answer until they started to organize memories and reflections into narratives.
Data management: manual or automated? It had been intended that data management would be handled by QSR NVivo, a software package (Bazeley & Richards, 2000). However, this was re-considered once engagement with the data began. The investigator was intimately involved with the interview transcripts on paper, but they somehow seemed remote when represented on a computer screen.
There was also tension and discomfort with the perception of relinquishing control to a computer. Thus, the decision was made to analyze the data without automated assistance.
The following steps were involved in data analysis once the integrity of the data was demonstrated through inter-coder reliability: 3. The average QOL and MIL ratings for the 10 recruitment sites were computed and compared.
4. The participants from senior centers and Section-Eight housing were compared on their overall QOLI score. They were also compared on demographic characteristics.
5. Those participants who were interviewed prior to taking the QOLI were compared with those who took the inventory first to determine potential influence in either direction. Missing data. The 21 participants that were interviewed had complete data on both inventories and demographics. From the original sample of 215, a total of 10 QOLI were eliminated because more than two items were either left blank or had multiple responses. This is consistent with the QOLI manual (Frisch, 1994) Other demographic items that were left blank were not addressed because this occurred in less than 2% of cases-with one exception, namely age. Five (2.4%) participants, three from the younger and two from the older sample, had missing values. Mean substitutions for each of these groups were used.

Results
Results for quantitative and qualitative analysis will be presented separately. The quantitative section will focus first on the interviewed sample (n = 21) and will examine QOLI scores and the relationship between meaning-inlife and quality-of-life ratings. It will then proceed with analysis of QOLI scores, demographic variables, and quality-of-life ratings for the complete sample of older adults (n = 115). Finally, this sample of older adults will be compared with the younger sample (n = 90) on QOLI scores and quality-of-life ratings. The qualitative analysis will follow the quantitative; it will consist exclusively of describing and summarizing the interviewed sample's responses to interview questions.

Quantitative Analysis
The While many mentioned that with increasing age they have had to deal with health concerns, 16 (76%) reported being in good health and the rest reported fair health.
Given the recruitment sites for this study, it is not surprising that reported yearly income is low: 38% reported under $10,000 and another 52% reported between $10,000-20,000.
QOLI scores. The average QOLI score for interviewed participants was 3.17 (SD= 1.37). According to the QOLI Manual (Frisch, 1994), the range of 1.6-3.5 is categorized as average. The distribution of scores for this small sample closely matches that of the norming sample, and it approaches normality. On the two questions asking participants to rate their quality-of-life in the past and now on a scale of 1-10, the distributions are negatively skewed with a mean rating of 7.8 for past and 8.1 for now. Thus, participants report having high quality-of-life, and of particular interest is the fact that from their perspective, quality-of-life in the present has not diminished relative to the past. It is also worth noting that QOLI scores were significantly correlated with ratings relative to quality-of-life in the past (r = .738, p < .01) and in the present (r = .768, p < .01).  Table 2).

Quality-oflife and meaning-in
This finding is intriguing, but no definitive explanation is available. It is perhaps possible that as interviewees reflected on past lived experience, MIL and qualityof-life were blurred or merged concepts, but that when they reflected on life as they live it right now, MIL was in fact distinct. An alternative and possibly more plausible explanation is that meaning changes radically in later life, but not in any predictable fashion. Neither explanation can be ascertained from the data here presented. .13 Relationships among MIL, QOL and demographic variables. Ratings of MIL (past and present) were also not correlated with most demographic variables: educational attainment, frequency ofreligious practice, self-rating on religiosity, or marital status. In fact, only one of the demographic variables, health rating, was significantly related to MIL in the present (see Table 3). MIL ratings for those reporting good health were higher than for those reporting fair health (M = 8.69, SD = 1.32; and M = 7 .0, SD= 1.32, respectively), and this difference was statistically significant, F (1, 19) = 5.74,p < .05. As Table 3 also shows, QOL estimates cannot be predicted from demographic characteristics. Those reporting good health did not differ from those reporting fair health on QOL ratings. While caution in interpretation is warranted given the limited sample used, other studies also report that demographics have surprisingly little effect on subjective wellbeing (e.g., Diener & Suh, 1997;Diener et al., 1999). Counter-balancing the order of interviews and inventories. As discussed, it was proposed that for participants interviewed, the administration of the inventory would follow the interview. However, as also discussed, alteration to this procedure was deemed necessary and appropriate at half the sites. This resulted in counter-balancing, albeit not in a random fashion. Did interviews influence inventories? There is no evidence of this having happened. That is, there are no significant differences in QOL ratings between those who interviewed prior to and those who interviewed following the inventory . Whether the inventory influenced the interview is more difficult to ascertain, but a comparison of MIL ratings does not suggest that any contamination occurred (see Table 4 for comparisons of QOL and MIL). .17 .29 .08 Comparing sites on QOL and MIL ratings . Ratings of MIL and QOL by site reveal several interesting differences. For example, the Jewish Community Center means are much lower across QOL and MIL ratings , and the lowest MIL Now ratings were reported by interviewees at East Greenwich. However , for each value represented, n = 2, and this precludes any inferential analysis. Overall, and most striking, is the fact that the four mean ratings are, with few exceptions, very similar (see Table 5). Note. Two participants were interviewed at each site.

Comparisons by site type: senior centers and Section-Eight housing.
Participants from senior centers differed from participants in Section-Eight housing in reported yearly income; their income level was slightly higher. The two participants who reported incomes above $20,000 were from senior centers and more participants from Section-Eight housing reported income under $10,000. They also differed on QOL-Past, QOL-Now, and MIL-Past ratings.
Surprisingly, differences are not in the expected direction. That is, while participants from senior centers had higher incomes, it was those from Section-Eight housing that reported significantly higher QOL ratings and MIL-Past rating (see Table 6, and Figures 1 and 2). It is necessary to add that the only rating that is higher for the senior center group (i.e., MIL-Now rating) is not statistically significant. Thus, among the interviewed participants, those from Section-Eight housing reported higher ratings of MIL-Past. However, the two sites did not differ on ratings of MIL-Now.

The Complete Sample of Older Adults
Marital status . The relationship between marital status and quality-of-life was examined for the entire sample of older adults (n = 115). Not surprisingly, the majority of participants (n = 64, 55.7%) were widowed. Additionally, 28 (24.3%) were married, 12 (10.4 %) were single, and 8 (7.0%) were divorced .
Three (2.6%) did not provide this information.
Given the plethora of empirical data supporting a link between marriage and well-being (e.g. Baum & Stewart, 1999;Diener et al., 1999), it was expected that married participants would report higher quality-of-life. Results reveal , first of all, that mean ratings are comparable across marital statuses (i.e., range= 7.6 -8.5). In fact, the scores of widowed , married, and divorced individuals were almost identical.
The most interesting comparison did not involve marital status but, rather, QOL ratings for both now and in the past for single participants. As Figure 3 demonstrates, single participants reported having the highest quality-of-life in the past (8.3) and the lowest quality-of-life in the present (7.6).  (Erikson et al., 1986).

Comparing senior centers and Section-Eight housing. As predicted by
Purcell (personal communication, July 16, 2004), older adults recruited at senior centers were different from those recruited at Section-Eight housing on key demographic variables. As Table 7 shows, compared to Section-Eight housing, senior center participants reported higher educational attainment and higher income levels. On the other hand, Section-Eight housing participants were more likely to report poor health and have less than a high school education. There are other differences not presented in the  In spite of these striking demographic differences, and consistent with the same analysis of the sample interviewed, the senior center group did not report higher quality oflife compared with the Section-Eight Housing residents. The mean QOLI scores were 2.70 and 3.00 respectively. On QOL ratings, the groups also did not differ. This finding suggests that QOL is a subjective judgment that perhaps transcends specific features of life experience. To be more precise, judgments thereof might be more a function of attitude and perspective-and less a matter of circumstances.

The sample interviewed compared with the entire sample of older adults .
Since not all randomly selected adults agreed to be interviewed, a concern is raised relative to the representativeness of those interviewed . To address this issue, those interviewed were compared with the entire sample of older adults (who only filled out the QOLI) on demographics and QOL variables. As Table 8 demonstrates, they are remarkably similar. Average QOLI scores for interviewed as compared with the complete sample of older adults (3.16 and 2.82 respectively) were found not to be significantly different. Using Analysis of Variance , F (1, 113) = .741,p > .05. Thus, it appears that those interviewed adequately represent the larger sample of older adults who participated in the study. compares the college students with the older adults on mean QOL ratings (see Table 9). Results show a non-significant main effect for the temporal rating, but a significant effect for the age comparisons, F (1,203) = 6.375,p < .05. That is, older adults had significantly higher QOL ratings than did the college students, but for both groups, QOL-Past and QOL-Now estimates were remarkably similar.
Older adults also had significantly higher mean QOLI scores (

Note. Figures represent means (SD in parenthesis)
The average of Past and Now QOL ratings was significantly correlated with the QOLI scores (r = .684,p < .01), and the older adults also had higher QOLI scores (p < .05). The QOLI allowed for a comparison of older and younger groups, not only on overall scores, but also on the sixteen dimensions comprising it.
As previously discussed, for both older and younger groups, all dimensions of the QOLI were found to contribute significantly to the overall QOLI score. An inspection of mean differences does suggest that some of the dimensions may have contributed more to the differences in overall scores. Table   10 compares the means of the two age groups on all the 16 dimensions. Given the large number of comparisons, only those significant at the .01 level or above are noteworthy: older adults were more satisfied with helping, children, neighborhood, and community.  (Frisch, 1994), "satisfaction in highly valued areas of life is assumed to have a greater influence on evaluations of overall life satisfaction than areas of equal satisfaction that are judged to be less important" (p. 5). The strength of association (i.e., correlation coefficient) between a life area and the overall score thus speaks directly to its contribution to overall life satisfaction.  Note. All coefficients are significant , p < .01.
To enhance the interpretability of the above, operational definitions of these life areas will be presented (Frisch , 1994) as follows: • HOME "is where you live. It is your house or apartment and the yard around it. Think about how nice it looks , how big it is, and your rent or house payment." • HEAL TH "is being physically fit, not sick, and without pain or disability." • RELATIVES "means how you get along with your parents , grandparents , brothers , sisters, aunts, uncles , and in-laws. Think about how you get along when you are doing things together like visiting , talking on the phone , or helping each other out." • COMMUNITY "is the whole city, town, or rural area where you live. Community includes how nice the area looks , the amount of crime, and how well you like the people .. . places to go for fun ... " • NEIGHBORHOOD "is the area around your home. Think about how nice it looks , the amount of crime in the area, and how well you like the people." • HELPING "means helping others in need or helping to make your community a better place to live. Helping can be done on your own or in a group like a church, a neighborhood association , or a political party. Helping can include doing volunteer work at a school or giving money to a good cause. Helping means helping people who are not your friends or relatives. " • CHILDREN "means how you get along with your child ( or children). Think of how you get along as you care for, visit , or play with your child." For both age groups, the cluster ofrelatives, home, and community were among the highest predictors of the overall QOLI score. There were some surprising differences however. For example, it is intriguing that health, neighborhood, and friends were more highly associated with life satisfaction for younger than for older adults. It is not so surprising to find that goals, helping, and children are more important to life satisfaction for older adults. These three life areas will also be discussed as they contribute to a sense that life is meaningful (under Qualitative Results).

Qualitative Analysis
In this section, responses to each question on the interview guide are presented. The principal goal was to summarize themes as they emerged from the data. As discussed in the Method section, inter-coder reliability was extremely high. Analysis thus involved simply summarizing related codes and conceptualizing emerging themes. Thematic connections are presented in figures.
Every effort was made to allow participants to speak directly to the reader. This was achieved by providing illustrative quotations as appropriate.
Conscious effort was made to stay as close as possible to respondents' words in order to avoid misrepresentation . As stated by Seidman (1998), "At the root of in-depth interviewing is an interest in understanding the experience of other people and the meaning they make of that experience" (p. 3). But as Seidman goes on to add, there are limits as to what we can know about another person's lived experience. Thus, the decision was made to stay as close as possible to the actual language used by participants. Each question will now be presented in the sequence in which it was introduced in most interviews.

Do you ever think about meaning-in-life?
Fourteen of the 21 participants responded immediately in the affirmative.
Some stated that meaning-in-life is something "you can't help think about," and they "have more time to think about it now." They reported thinking about meaning-in-life "frequently," "daily," or "all the time." Only one person said an outright "no," three never answered the question, and three reported not thinking about it, but nevertheless went on to define it and to delineate how life has been meaningful for them, both in the past and in the present.

What do you think "meaning-in-life" is about?
Sixteen participants responded with confidence in their own knowledge.
Initially, four hesitated to respond because they didn't know quite how to approach the question but, with a little encouragement, went on to express what this phrase means to them. Only one person appeared to be intimidated by the question. Sam was diffident at the beginning of the interview; he suggested I interview his wife because "she's much smarter." He continued to be selfeffacing and made several references to his low educational attainment.
However, after some reassurance, he submitted a single but parsimonious response: "I guess it's about whether someone thinks their life is worth living and also whether they're happy." Perhaps the most important point to report is that all participants eventually had an answer.
Another important point is that there was not a single item used by all participants to define meaning-in-life. The most frequent semantic categories used to delineate "meaning-in-life" in descending order (frequency in parenthesis) were helping and "doing good" (10), purpose (8)  In Their Own Voices: Conceptions of the Phrase "Meaning-in-life"

Bill
Purpose common to us all; "we're here for a reason." This reason is "to do good ."

Harold
To do the things that make life meaningful-"helping others," and "enough money to support me and my family." Kitty Health, activity, and positive attitude. "There's always someone worse off than you."

Peg
To have purpose; "I have a job to do before I die." Mary "Continuing to do what you think is important," and "getting along with people."

Stanley
Purpose; family; social connection; taking care of yourself Barbara Helping-"going beyond yourself'; family.
Marie "What a person does with life"; family, helping others; "enjoying life one day at a time." Sarah "Realizing that life is the best gift"; positive attitude; health.
Paul "You have a lot of good things in your life": marriage, children that are good to you; children and grandchildren are doing well.
Grace Individual purpose; "being worthy of God's love"; trying to do the right thing. Minnie "Doing good" (benefits the giver and receiver); "tell the truth and look for truth."

Has life up to now been meaningful?
All 21 participants responded that life has been meaningful. However, five added a qualifier. That is, they were not able to respond for life as a whole because they recall two distinct periods of time, one of which was far less meaningful. The experiences of these five will be briefly described.
Bill stated in response to being selected for the interview, "I have had two lives. Which one do you want to know about?" He referred to the fact that he was an alcoholic for 27 years, and during this time life had little meaning; his 1-10 rating for this period in his life was a 4. He has been sober for the past 20 years, and his recovery from alcoholism has been accompanied by a parallel recovery of self worth, control over his life, joy in living, and a deep spiritual awakening. His rating for the past 20 years is a 9.
Harold spoke about the struggle he experienced when he had to raise his two children by himself (implied that the mother of his children abandoned him and them, but was explicit on not wanting to discuss any details) and this was not a meaningful part of his life. However, he goes on to cite raising his children by himself as the biggest source of meaning in the past.
Frances, 86, had a very rich and meaningful life. However, for a period of time after her husband died, she experienced a great loss of meaning. She initially gave herself a 5-7 rating while focusing on this loss, but when asked to reflect on her whole life up to now, she gave a rating of 10.
For Sarah, the early years of marriage were very meaningful, but after her husband had an affair with a friend, they eventually divorced. Those years were very low on meaning, and her sense of loss and meaninglessness lead her to make an attempt at suicide. But the divorce itself also precipitated personal growth and Thus, while all stated that life up to now has been meaningful, the five individuals mentioned above acknowledge that there were distinct periods of low meaning, and these corresponded to life events/circumstances. And for these five, the loss and associated pain and struggle prompted growth and transformation.
This theme will be discussed in the next section .

What are the different things that up to now, have contributed to meaning-in-life?
Responses to this question were coded and summarized in Table 13. An attempt was then made to represent schematically the nine major themes outlined in this table in terms of two over-riding super-ordinate themes (see Figure 4). It must be stated up front that these are not "essences" but, rather, one way in which informants' responses could be semantically organized. Affiliation and self transcendence emerged as distinct super-ordinate categories. Work and activities are two other independent categories.
As participants reflected on the sources of meaning in their past, the primacy of social motives was consistent. They emphasized emotional attachments, mutual support, and shared goals with their spouse, children, family, and friends. Collectively these codes are subsumed by the super-ordinate category affiliation. Marriage was mentioned by 16 interviewees. Since only one person has been single his entire life, this means that 80% (16/20) of those who are married, divorced, or widowed think that marriage contributed meaning to their lives in the past. It was reported by 100% of those currently married (n = 4), by 9 out of 11 (82%) of those that are widowed, and by three out of five (60%) of those that are divorced. As Table 13 indicates, while a total of eleven mentioned their marriage, five spoke specifically about their spouse. Mary, 77, remarked about the intimacy she shared with her late husband: "We were very close. It was easy to be together. I knew that he loved me more than he loved himself." For Mary and other widows, marriage was a focal point, and the memories thereof continue to be deeply cherished.
Children and grandchildren were also cited by 16 participants. Three reported both children and grandchildren , seven mentioned just children , and six emphasized that raising their children was meaningful. Of the last group, two women spoke about how meaningful it was to stay home and raise their children , and Harold spoke with pride about raising his sons as a single parent. Family was also cited. Five of the eight codes refer specifically to family of origin-good parents and a happy childhood, whereas three simply stated family, the precise meaning of which is unclear . For six participants , friends also contributed to meaning-in-life.
Collectively then, marriage , children, family, and friends speak to the fact that meaning in life involves intimate human connections. The importance of affiliation was stressed by Erikson, Frankl, and Aristotle . According to Erikson (1986), growth during adulthood involves the virtues of love and care, Frankl (2000) emphasized that meaning is found through human connection , and through love, and Aristotle emphasized the importance of social life and in particular the role of parents , spouse, children, and friends (Kraut, 1989).  Another super-ordinate category presented in Figure 4 is called selftranscendence, a uniquely human capacity (Frankl , 1959). For Frankl, human existence is optimally always directed toward something or someone other than itself . Human beings strive to find and fulfill meaning and purpose, and life's meaning is unconditional-it is even found in unavoidable suffering. But the most important point is that meaning is found through activity in the world, not within self. Frankl proposes that it is when we forget ourselves that we experience our essential humanity. This also applies to knowledge of "ultimate meaning" (i.e., knowledge of God), a profoundly personal and spiritual phenomenon.
Self transcendence as it pertains to MIL-Past subsumes benevolence, finding meaning from adversity, and faith. It is significant that 62% of participants mentioned some aspect of benevolence, a label which includes helping, volunteering, and showing kindness . This theme was widely articulated in response to this question and reiterated throughout the interviews. Its centrality is exemplified by Eneka's statement that "to be alive is to do for others." For some, benevolence was associated with religious prescriptions for a good life. As stated above, the centrality of helping and its relation to meaning-inlife is reiterated throughout the interviews of most participants. It will therefore be discussed later with regard to meaning-in-life in the present, future goals, and life lessons.
Religious concepts such as faith, church, and prayer were mentioned by only seven participants (33%). This is surprising, particularly for this cohort of predominantly Catholic individuals. I must confess that in my initial striving to practice bracketing , I made a conscious effort not to ask about religion. This exhortation, however, was short lived. Impulsively, I started to lead participants into a consideration of the role of religion. I did so because I was so surprised at their exclusion of religion in discussing what for them has contributed to meaning. The independent coder in this study was quick to identify this trend.
She also astutely observed that, when prompted to speak about religion, most discussed it but in a superficial manner. In this analysis, if I along with the second coder agreed that religion was only addressed because of a leading question and then only superficially, its coding was excluded.
Most of the seven that did speak about the importance of religion without being prompted by me, expressed strong beliefs about the role of spirituality or religious practice and its essential nature to meaning-in-life. Peg stated, "There was a definite purpose in my life" and recalls that she "always had a feeling the lives. They had vivid recollections of these events and circumstances but ironically framed these experiences as growth enhancing. Thus, these periods of life were painful but meaningful. According to Victor Frankl (1975), " ... despair is suffering without meaning" (p.137). It appears that for these respondents, suffering did indeed lead to meaning, a phenomenon that has been empirically validated ( e.g., De bats, Drost, & Hansen, 1995). They described how such experiences profoundly shaped them or in some cases transformed them for the better. As some of their stories are now re-told, the reader may recall that some of these are the same individuals who found it impossible to respond to whether life as a whole had been meaningful. For them, the past had been demarcated by these self-transforming events or circumstances.
For some, not only was the adversity or struggle associated with MIL-Past, but the growth and resolve that it engendered is considered that which most contributed to MIL-Past. For example, Bill was transformed by his recovery from alcoholism. It was such a powerful experience that he now does what he can to help others with the same recovery process. He considers this to be the most meaningful part of his life. Another example is provided by Harold who was faced with raising two young children by himself. Initially, he states that life in his "younger days" was not meaningful. He continues, "I just went along and did whatever I had to do." As the interview progressed and he was asked about what most contributed to MIL-Past, he later acknowledged that raising his children the "right way" has indeed been the most meaningful part of his life.
Sarah's story also involves having to bring up her children as a single parent. She found herself without any child support following a painful divorce.
She experienced such deep loss that she attempted suicide. But she now recognizes that the divorce actually precipitated vital growth because by working through the pain she learned to like herself-and her life. She had lost her reason for living but now values life above all else. She had felt overwhelmed and paralyzed by fear, but now she is clear and focused on her mission which is to minister to others in need. She felt alone, but through this she has now found the Lord and is alone no longer. Finding the Lord has contributed the most to MIL-Past. She concludes, "It changed my life completely and I'm happy with that." Kitty spoke most poignantly about her childhood struggle with poverty, and also associates this struggle with the most meaningful part of her life. She said, "I think I always feel as though I'm overweight because we never had anything to eat when we were kids. How many days we didn't have any food, and ha .. .I went to a Catholic school, and the nuns would say ' ... and what did you have for breakfast?' Well, whatever the kid in front of me had, I would say the same thing, because I hadn't had any breakfast-and that stands out more than anything." Kitty explained that the poverty stemmed from her father often not working, and when he did work, he didn't come home with the money. She articulated how she went looking for him: "I would have to wait and meet him downtown and catch up with him ... I had to go looking for him to ask him for money to eat, so my mother could have it." Kitty's father was not a good provider in terms of meeting the family's basic needs. Neither did he provide for his family's emotional and psychological needs. Kitty recalled: " ... and he called us all chicken because he didn't know our names-that's the way I felt." Kitty's story evoked both empathy and sympathy. However, she made it clear that she is strong today as a direct consequence of these early experiences. She is indeed resilient.
Mary recalls that as one of twelve children she had a "hard life." She never finished high school because she was "third in line" and had to work to help the family. She also recalls that she loved to sing and dance but was not able to pursue either of these talents because the family's basic needs were foremost. For Mary , childhood was brief and punctuated by hardship. For example, she told me that she and her sister shared the same pair of shoes and sometimes the same dress, and so they had to attend church at different times. They often fought.
Despite this hardship, Mary remembers a poor but happy childhood. Therein lies the most important lesson, a lesson that would influence her entire life. She and her husband have lived a very modest life. They have raised three boys but have never owned a car or a home. Even today, they manage to have what they need and seem not to ask for more.
Peg also recalled financial hardship during childhood. She is particularly explicit on lessons learned: ( 1) "You don't need money to be happy", (2) "You can get through the worst troubles that face you if you work at it," and (3) "I know that you can make fun out of most anything." For eight participants, work and activities also contributed to MIL-Past.
Work contributed to meaning-in-life in a variety of ways. Stanley talked most about the importance of work. For him, work brings you into contact with people, engages the mind, and keeps you physically fit. At the age of 85, Stanley's goal is to win the lottery so that he could help his family because "they don't have much." But if this were to happen, he would stipulate that everyone would have to continue to work. In his own words, "There is nothing without work-not only because you get some money. It's what it does for you. It keeps your mind busy, you forget all your problems while you're working ... but the most important thing, it keeps your body going and it keeps your mind wide awake." For Sam, a 79-year-old with only eight years of education, it has been through his work that he met the people who would encourage him to strive for a better life. He recalls that "spending time with educated people kind of gave me an education ... they respected me. They taught me a lot and they encouraged me to make something of my life." Dannie, who is 80, recalls that the people she worked with were her best friends. In addition, work contributed to self-efficacy, was enjoyable, and enabled her to contribute to the household. She opened the original Gregg's on the East Side of Providence and later also worked at Cohoe's.
As she thought about these work experiences, she spoke with both pride and nostalgia.
Staying busy and engaging in other rewarding activities was also often articulated. Stanley talked about how he and his wife left home at 8:00 am and went swimming at the YMCA. Then they went to the senior center for lunch. But no one spoke with greater fervor about the importance of staying busy than Minnie who is 95. Even now, she is active with the senior center and with the church, she makes crafts and donates them, keeps her apartment immaculate, makes time to cook and bake, sews for anyone in the building who seeks her help, visits with her children and grandchildren, and organizes trips from the senior center to Foxwoods Resort Casino. She proudly displayed a list of people that have already paid for the next excursion and informed me that the "Indians" let her go for free, throw in a meal and give her $40.00 for trying her "lady luck." She stresses that she has always been active.
Four participants responded that they have no regrets. Of the other seventeen who did, the most frequently expressed regret was not having achieved more education. The common theme expressed by these six individuals is that education is equated with greater opportunity, particularly in the workplace, and that better jobs might have meant better pay. Their six explanations are as follows. Joseph regrets that he was a slow learner and could not pursue a career in law. Mary regrets dropping out of school in the eighth grade and believes she might have had a better life had she continued. She also regrets not having pursued dance and music-two things she has been passionate about all her life.
Sam is convinced he would have had greater opportunities. Paul quit in the seventh grade. With more education, he might have had better work and better pay. Barbara received an associate's degree. She worked as secretary and as a CCD teacher. She loved teaching. Had she continued with school, she could have been a certified teacher. Kitty would have liked to go to college but lacked the financial means. She said with confidence, "I could have been somebody you know." She particularly regrets that she wasn't financially independent: "I don't think I ever would have got married ifl'd had my own money." Kitty stayed married with an unfaithful man for 34 years.
Marriage or, more to the point, the loss and disappointment associated with divorce, was the biggest regret for Sarah and Anne. Sarah was happily married until her husband had an affair with a friend of hers. The divorce was painful-particularly because of the negative impact it had on her sons. Anne also regrets that her husband was unfaithful. He had an affair and later married his secretary, who was seventeen years younger than he.
Having had no children was regretful for two women. Barbara was married and had wanted to have children, but in her own words, "It just wasn't meant to be." Mary and her husband also tried to have children; they had one but it was stillborn.
Two women regret that they moved away from their family of origin to get married. Eneka moved to the U.S. from Holland and Marie moved to RI from Massachusetts. Both women missed their families. Eneka was particularly nostalgic in responding to this question. She starts off her answer in this way: "I can think of my biggest regret. .. was not having my mother at the birth of my children ... that always has stayed with me ." Other regrets, each expressed by a single individual, included losing a spouse, addiction, not having been a better parent, having a son with a mental illness, not having saved enough for retirement, and not having had the social support needed at the loss of a spouse.

What people in your life have had the greatest influence on you?
Eight people acknowledged that their parents had the greatest influence.
Of these, six spoke of their mothers in particular. Anne, Kitty, and Mary articulated best how their mothers' influence had a profound impact. Anne's mother "was a good person, and lived a good life .. . She was good to everyonealways ready to help others. Parenthetically, Anne defined meaning-in-life as doing "good for other people." Kitty returned to her difficult childhood . She recalled how her mother coped with the challenges of poverty and never complained. "I never heard an argument. I never heard a harsh word." When asked what she learned from her mother, she replied, "I'm strong. Dannie, whose biggest regret was losing her husband, explains how he has influenced her: "He was the kindest man I have ever known. He never criticized me or found fault with me." Paul also stated that his wife influences him by not criticizing him. He spoke about his life partner with great tenderness and asserts that she has contributed most to MIL-Past. As a tear gently rolled down his cheek and his chin and lips quivered, he said, "I still kiss her once in a while, and I tell her I love her." He goes on to say that she continues to give the most meaning to his life.
Six individuals spoke broadly about family influence and one about her grandmother . These and other responses to this question were far less elaborate than the above and included friends (4), work/career coaches (2), therapist, priest, AA sponsor (2), people in general (2), the Lord (1), and no one (2).

Is life meaningful for you right now?
As previously discussed in the quantitative analysis section, there appears to be a relationship between health and MIL.

Describe the different things that now make life meaningful
The different things that provide MIL-Now seem to cluster into the three structural representations depicted in Figure 5. Recurrent codes were readily categorized into one of the three themes, namely, affiliation, agency and control, and self-transcendence. As with Figure 4, these three super-ordinate categories are not presented as essences of the phenomenon of meaning-in-life. It is acknowledged that other configurations are possible. These are however consistent with the theories and research discussed.
Living a longer life impacts the types of affiliation featured in Figure 5, and this impact is only now beginning to be understood (Bengtson, 2001).
Married couples are spending more post-retirement years together. In addition, older adults are living long enough to celebrate their generativity as they observe the lives not only of their children but also of their grandchildren and greatgrandchildren unfold before them . Finally, in the post-retirement years , time previously devoted to children and work is re-invested in social life. Carstensen (1995) argues that social contact at this stage reflects a growing need for emotional support that is met by family and close friends. There is also empirical evidence for older people's preference for strengthening existing relationships, particularly with family members (Carstensen , Isaacowitz , & Charles, 1999).
It was expected that affiliation would be important to MIL-Now , but it must be noted that family was by far the most frequently mentioned. Actual codes integrated into family include family (n = 13), children and grandchildren (n = 7), and sisters (n = 2). All currently married participants expressed the idea that married life (n = 1) or specifically their spouse (n = 3) contributes to meaning-in-life. For many, family was clearly associated with generativity issues (Erikson et al., 1986). According to Erikson, generativity  Approximately half of respondents referred to friends. It appears that friends are more important to MIL-Now than they were to MIL-Past; the frequency of its reporting increased from six to ten respectively. This change may reflect a positive correlation between age and degree of common interests and perspectives shared by one's social convoy (Antonucci, 2001 While the importance of friends to MIL-Now seems to have increased, the percentage of respondents (i.e., 48%) who consider it important is still surprisingly low, given the multiplicity of research findings linking friends with good adjustment and well being, particularly among older adults ( e.g., Vaillant, 2002). This is also surprising when we consider that for the entire sample of older adults, the friends dimension of the QOLI had the highest mean score (M= 3.55, SD= 2.93; see Table 10). Extrapolating from these data, it is a plausible hypothesis and, one worth testing, that friends are very important to quality-of-life but only moderately important to meaning-in-life. On the other hand, satisfaction with familial relationships is important to both quality-of-life and meaning-in-life.
Along this same vein, a meta-analysis conducted by Pinquart (2002) that investigated purpose in life in old age found that frequency of contact with family members was more strongly associated with purpose in life than frequency of contact with friends. Discussion on relationships between quality-of-life and meaning-in-life will follow this section.

Marriage n=4
Family n=22 Enjoying Life n = 2; Sufficient funds n = I

Agency and Control
Freedom and Independence n=7 Health and taking care of yourself n=9

Self Transcendence
Helping n = 13 The agency and control factor subsumed five different code categories.
These converged on a central theme-articulated by most participants-that speaks to the judgment that they are able to do what is important to them.
Generally, the pattern of responses depicts older adults as active agents who select and wish to optimize engagement in their social world ( Carstensen et al., 1999).
Active engagement in activities was an important aspect of living a meaningful life for eleven participants. But as Kaufman's (1986) sample also reported, staying busy is not sufficient. Seven participants clearly articulated that being able to choose activities was the key to meaning. Even the importance of maintaining good health was discussed by nine participants in terms of how health directly impacts their ability to engage in valued activities, express their freedom and independence, and maintain their home. It also appears that when physical decline impedes ability to engage in valued activities, participants experience a loss of meaning. This was most poignantly expressed by Frances who, due to severe arthritis, is no longer able to practice her arts and crafts which have most contributed to meaning throughout her life. She is also not able to take care of her apartment or move without pain. Frances reported the lowest MIL-Now rating, a five. She attributes this low rating to her lack of agency and control. She is determined to at least still try to keep her mind active. She reads avidly.
In order to maintain fidelity to description , this factor is substantiated with excerpted exemplifications in respondents' own voices. Some of the participants will speak for themselves (see Table 14). 138) The self transcendence factor comprises two code categories: helping and faith. It is worth noting that all but two respondents discussed at least one of these, and many spoke of both. According to Frankl, "the man on the street" prereflexively understands what being human is all about, and that is, to discover his/her own meaning "out there" in the world. The finding that 13 respondents seem to be oriented intuitively toward helping or serving others seems to support Frankl's claim, at least for older adults.
For Frankl, meaning can be found in work. Barbara said early in the interview that "to have meaning you have to extend yourself and help others." She has devoted her entire life to work that brings about improvement in others' lives but is now feeling very guilty because since her retirement several years ago she has taken time to enjoy herself and has done little to help others. In response to the question "Is life meaningful right now?" she responded, "It could be more meaningful. I'm enjoying myself now, but something inside me is incomplete." She describes her experience of guilt: "Guilt isn't all bad. Guilt is that part of us that knows better, reminding us of what we have conveniently forgotten ... so I know I need to make more effort .... " Frankl also said that meaning is found in love. Sam spoke more broadly about his love and concern for others: "I love people and I try to help them. That is meaningful to me." He goes on to describe how at age 78 he runs errands for many people in his building and concludes: "I like to see people especially at this age, I like to see them happy. I do anything to make them happy." Sam continues: "I think too many people are stuck in their own routines. They don't stop to think about how insignificant they really are. As I see it, it's when I help others that I am worthy of living. It's the greatest reward to know that I have privilege of adding to someone's happiness." Sam implies that adding to others' happiness is sought for its own sake, a sentiment expressed by others. For Marie, "It feels good to do good" and for Minnie, "Happiness comes from making others happy." Bill was more philosophically encompassing when he stated that the human purpose common to us all is to do good.
For Frankl (1975), searching for meaning also involves discovering one's own spiritual connection with God, a striving which is deeply rooted in the unconscious. For 12 respondents, meaning and purpose are conceptualized in terms of faith. Peg, for example, said that life is meaningful because "I have a job to do before I die and I haven't done it yet." Mary asks, "What is life without The responses to this question fall into the following categories: helping (n =6), faith (n = 6), family (n = 4), health ( n = 3), friends (n = 3), children-that they "turned out good" (n = 2), spouse (n = 2), and son (n = 1 ).
Some respondents gave multiple answers, and two did not answer.
What is good about being your age?
This question was not originally part of the interview guide , but once it was asked spontaneously of the second participant, it was added on. Responses include: • Freedom to do as he/she pleases (n = 6) • Enjoying life (n =3) • Wisdom; have had enough time to "get it straight " (n = 2) • Looking back, and concluding it's been a good life (n =2) • Sufficient funds (n = 2) • Health (n = 2) • Friends (n = 2) • Nothing-due to poor health (n = 2) The above accounts (i.e., lack of future goals) may strike some as supportive of a deficit model of aging (i.e., loss and decline). This reflects the power of ageist stereotypes, and unfortunately, this tendency toward stereotypic thinking is spontaneous, automatic, and largely unconscious. As pointed out by Carstensen et al. (1999), a shift in focus from the future to the present and the inclination to set short-term goals is highly adaptive for this stage of life.
"Something in the finite nature ohime appears to make life precious, especially as the end nears. Within this framework, age changes may reflect increased appreciation of life rather than despair about loss" (p. 175). Implicit in Cartensen's theory is the notion that, as a result of changing perceptions of the future/past ratio, personal narratives take on different dimensions, as individuals reflect on major themes and attempt to integrate these into a meaningful whole.
What is the single most important lesson life has taught you?
Responses to this question are reported verbatim in Table 15.

Sources of Meaning: Past and Now
A comparison of the sources of meaning in the past and now lead to the following conclusions: • Marriage was and continues to be important. Sixteen reported it as important to MIL-Past, and 4 reported it as important to MIL-Now. This 4 represents 100% of currently married participants. Therefore, it continues to be important. In summary , relationships are strongly associated with meaning-in-life; marriage and family are equally important to past and current meaning, but friends are more important to meaning now. Raising children was important to meaning in the past and evaluation of how children were raised is important to meaning now. Agency , self-sufficiency and self control are more relevant to meaning-in -life now. In particular , freedom and control over choices made , health, and activities undertaken are more important to meaning now. Helping is as important now as it always has been , and faith is more important to meaning-in-life now. Finally , work and meaning from adversity are associated with meaning in the past but not in the present.

Summary of Findings
The phrase "meaning-in-life" as used here refers to actual experience and hence had relevance for the group of older adults used in this study. Characteristics associated with the judgment that life is meaningful in the present differ to some extent from the characteristics associated with the same judgment of life as a whole. Friends are more important now than in the past.
Consistent with Erikson's theory, issues of generativity such as raising children and work were more prevalent in the past, and agency and control are more important to meaning now. Finally, faith was more strongly cited with reference to meaning in the present, and deriving meaning from adversity was only associated with meaning in the past.
Kindness and generosity in service to others was important to meaning both in the past and in the present. This source of meaning was articulated with conviction and pride, which prompted some exploration of this theme. In general, individuals derive a deep sense of fulfillment not just from the knowledge that others benefit from acts of kindness but also perhaps at an unconscious level from the sense that one's actions matter and make a difference in the lives of others.
The former focuses on the beneficiary while the latter focuses on the benefactor.
This notion of action that benefits both the receiver and giver coheres with Aristotle ' s views that (1)  Some participants did indeed articulate the perspective that in evaluating their own life conditions, they take into account others for whom the state of affairs is much worse . Kitty, for example, has had two strokes and recently had to give up driving due to serious visual impairments, but she reported never getting depressed because "There's always someone worse off than you." Extending this line of reasoning, if quality-of-life is a matter of comparative analysis that situates the person in social and temporal contexts, what then can we speculate relative to meaning-in-life? Is it not also a matter of perception and personal narrative? Undeniably it is. But what makes meaningin-life distinct, is perhaps that it involves narrative at a higher level of abstraction and involves an effort to integrate lived experience into an autobiographical life story that is personal and unique, while also encompassing themes and goals that speak teleologically to how humans have been designed to live.
The complete sample of older adults (n = 115) was compared with a younger sample (n = 90) on quality-of-life. The older adults reported significantly higher quality estimates for both life in the past and in the present.
Older adults also had significantly higher QOLI scores. Of the sixteen QOLI dimensions, those that had the greatest power to predict the overall score for older adults were: relatives, goals and values , home, helping, community, and children.
For the younger sample, the following dimensions were most highly associated with the overall score: home , health, relatives, community, neighborhood, and friends. Thus it appears that goals and values, helping, and children were more important to older than to younger participants. As discussed above, these dimensions were also important to meaning-in-life. Unexpectedly, the correlation between the health dimension and the overall score was higher for the younger than for the older participants (r = .698 and .553, respectively).

Study 's Strengths and Limitations
This study has several strengths. It successfully combined qualitative and quantitative methods to investigate two phenomena that are of universal importance to human life. It also allowed participants to speak in their own words and in their own voices about their experience of meaning and paid careful attention to data integrity. It was inspired by three time-honored theories, and it was guided by the phenomenological approach which is ideally suited for qualitative studies.
There are, of course, also limitations. In any psychological investigation, the issue of construct validity can never be put to rest. This study's principal goal was, to the extent possible, to investigate both quality-of-life and meaning-in-life from the vantage point of participants themselves. Meaning-in-life was summarized and presented while staying as close as possible to participants' own words. As discussed, the QOLI quantified level of subjective satisfaction in each of 16 domains weighted by each domain's perceived importance to participants.
Each dimension allowed for a wide range of responses because participants took into account first, how important each dimension was to them and secondly, how satisfied they were. Scores were therefore more individualized. Thus, while it cannot be asserted that this study has investigated these constructs as they exist in real life, or that participants accurately represented their experience, at least it can be claimed that every effort was made to study and report from the point of view of participants.
A more serious limitation relates to generalizability of findings given the sample constraints. The critical reader has undoubtedly already raised concerns about whether the small sample of 21 individuals interviewed spoke for other older adults, whether the community college students represent young adults of the same age, and whether older adults of middle and higher income levels might hold very different conceptions of the constructs investigated. These are valid concerns warranting explicit treatment. Further discussion on this point will focus on the older sample, while acknowledging that the younger sample also affords limited generalizability.
Residents of Section-Eight housing and individuals who frequent senior centers in R.I. do not, indeed cannot, speak for older adults in general. Their most obvious and distinguishing characteristic is their low income level; low income is a qualifying criterion for residence in Section-Eight housing. Another obvious implication is that a confluence of factors associated with their low income status may invariably set constraints on opportunities for self determination. Some, for example, wish that they could take more vacations, others wish they could provide more financial assistance to their children, and still others articulated fears about the necessity of nursing home placement if their ability to take care of themselves and their homes is compromised. It is reasonable to assume that these issues are not as salient for individuals with greater financial security. Those living in retirement communities or in assisted living communities, for example, might look upon issues of quality-of-life and meaning-in-life from very different life contexts and perspectives.
Higher income in late adulthood is also associated with certain prior life experiences. For example, it is more likely that older adults with higher socioeconomic status have attained a college education and have chosen a vocation that was and perhaps continues to be important to meaning-in-life. It is important to recall that work was not frequently mentioned in the context of describing meaning-in-life or quality-of-life. A plausible explanation for this is the fact that for most participants work was not chosen for any intrinsic value but was rather pursued as the means to an end, and constrained by opportunity. It is therefore not surprising that work is something from which most had distanced themselves.
Another obvious problem with this older sample is that it was primarily white. With direction from Aging 2000, sites were selected to maximize diversity. Regrettably, however, the sample was 94% white. Participants were also predominantly Catholic (i.e., 75%), and they held conventional values relative to family structure and affiliation. In short, these participants' voices are important, and they speak for many older adults, but certainly not for all.

Implications for Future Research
As information for mental health services will be provided to you, in case you decide you need help.

Confidentiality
Your confidentiality is protected. None of the information you share will be connected to your name or identity. You may be quoted in the study, but your actual name will never be used. You will be able to assign yourself a pseudo name for the purposes of the study. The audiotape of your interview will be secured by the researcher, and will be used for research purposes only. The tape will be retained for three years following completion of the study, and will be destroyed after that.

Decision to terminate at any time
The decision to take part in this study is up to you. If you decide to be interviewed, you may terminate the interview whenever you choose. If you wish to quit, simply inform the interviewer.

Rights and Complaints
If you are not satisfied with the way this study is conducted, you may discuss your complaints with Isabel Trombetti at (401)  You have been asked to participate in a research project that is investigating meaning-in-life and quality-of-life in people over the age of 65. All the details you may need as you consider participating in this study are outlined below. Please read carefully. If you have any questions or concerns, please don't hesitate to bring them up. If you think of questions later on, feel free to bring them to the attention of Isabel Trombetti, the person responsible for this study.
What will be asked of you: If you decide to take part in this study, first you will fill out a brief demographic information sheet. Then you will complete a survey called the Quality-of-life Inventory. It should take you no more than 10 minutes to complete it, but you may take as long as is necessary.

Possible benefits of participating
There are no benefits to participating. However, you may enjoy taking the time to reflect on your life.

Risks or discomfort
There is no anticipated risk or discomfort associated with your participation in this study. Referral information for mental health services will be provided to you, in the event that you do feel discomfort after completing the survey.
Your confidentiality is protected. None of the information you share will be connected to your name or identity. Your actual name will never be used in this study. You will be assigned a participant number.

Decision to terminate at any time
The decision to take part in this study is up to you. If you wish to quit while completing the survey, simply inform the investigator. Whatever you decide will in no way affect your grade or status as a student. You will not be penalized in any way.

Rights and Complaints
If you are not satisfied with the way this study is conducted, you may discuss your complaints with Isabel Trombetti at (401)