UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Agency in moral action : mapping the moral self Frimer, Jeremy A. 2006

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-ubc_2006-0458.pdf [ 5.63MB ]
JSON: 831-1.0092702.json
JSON-LD: 831-1.0092702-ld.json
RDF/XML (Pretty): 831-1.0092702-rdf.xml
RDF/JSON: 831-1.0092702-rdf.json
Turtle: 831-1.0092702-turtle.txt
N-Triples: 831-1.0092702-rdf-ntriples.txt
Original Record: 831-1.0092702-source.json
Full Text

Full Text

AGENCY IN MORAL ACTION: MAPPING THE MORAL SELF by JEREMY A. FRIMER B.A.Sc, The University of British Columbia, 2002 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Psychology) THE UNIVERSITY OF BRITISH COLUMBIA August 2006 © Jeremy A. Frimer, 2006 Abstract By telling us who we are as well as how to think and act, our self-definitions regulate interpersonal relationships and moral functioning. Aspects of the self-understanding of individuals were examined so as to give a fuller account of moral functioning than those posited by extant rationality-based models. Self-content refers to the seven different domains of life posited by William James (1890): physical, active, social, psychological, agency, continuity, and distinctness. Previous attempts to map self-understanding to moral action have generally produced weak and inconsistent findings. A new model was developed, which attempted to correct some lingering conceptual and methodological issues. Participants were 99 university students, who were recruited through 30 student clubs—a heterogeneous sample with a variety of life orientations. Moral behavior was operationalized as an aggregate of three measures: (a) self-reported altruism, (b) self-reported ecological behaviors, and (c) a behavioral measure of honesty. Participants also responded to an individual self-understanding interview. These interviews were coded for individuals' construal of and emphasis on the seven different aspects of their existence. Associations between these aspects of self-understanding and moral behavior were explored. Results indicate that the new model is predictive of moral action; the present study is the first to demonstrate a significant association between a self-content scheme (namely, agency) and moral action. Individuals that tended to implicate deliberate and volitional efforts of the self in causing some kind of change tended to engage in moral behavior more so than those that took a more deterministic stance. This finding is supportive of Blasi's (1984) theory of the moral self. Discussion focuses on the nature of moral identity and its central role in a comprehensive understanding of moral functioning. Table of Contents Abstract '. • Table of Contents • iii List of Tables v List of Figures vi Acknowledgments vii Chapter I - Introduction 1 Moral Identity as Moral Motivation: Blasi's Self Theory , 2 Previous Attempts to Model Blasi's Theory 3 The Present Attempt to Model Blasi's Theory '. 4 The Subjective Experience of Self 5 Self-Concept ; 6 Self-Understanding 7 Previous Research 8 Hart and Fegley (1995) - 8 Derryberry and Thoma (2005b) 10 The Current Approach • 12 Reconceptualizing Self-Understanding '.. 13 Predictive Validity 16 Summary and Hypotheses 17 Chapter II - Method 18 Participants 18 Sample Size .'. 18 Demographics 18 Procedure 19 Measures 23 Interview 1 23 Reliability 23 Metrics : 24 Questionnaire Measures 25 Prosocial Behavior 25 Ecological Behavior • 26 Ecological Attitudes 26 Materialism 26 Socially Desirable Responding 27 Satisfaction with Life 27 Behavioral Measure 28 Chapter H I - Results and Discussion 30 Moral Behavior ; 30 Reconceptualizing Content 38 Content and Level Confounding 38 Introducing the Scoring Manual for Self-Understanding 2 39 Agency 39 iii Analysis 40 Four Coding Paradigms Compared 42 Agency 45 Limitations 49 Summary and Conclusions 50 Chapter HI - Future Directions 51 Content 51 Values 52 Values and Moral Functioning 54 Values and Self-Content in Concert 56 The Texture of Identity Model 56 Encapsulating Blasi's Theory 56 Predictive Validity 57 References '. 58 Appendix A - Self-Understanding Interview (SUI) 63 Appendix B - Self-Understanding Interview—Transmogrified (SUIT) 64 Appendix C - Email Recruitment Letter 65 Appendix D - Self-Reported Altruism (SRA) 66 Appendix E — General Ecological Behavior-Adapted (GEB-A) 67 Appendix F - New Ecological Paradigm (NEP) 68 Appendix G - Materialism Scale (MS) 69 Appendix H - Balanced Inventory of Desirable Responding (BIDR) 70 Appendix I - Conscientious Behavior 71 Appendix J - The Scoring Manual for Self-Understanding 2 73 Appendix K - Schwartz' Values Typology 82 Appendix L - UBC Research Ethics Board's Certificate of Approval 83 List of Tables Table 1. Centrality-Rating Paradigm, Sheet 1 , for Participant # 75986 21 Table 2. Correlation Table of Measures of Moral Behavior 31 Table 3. Correlations between Self-Understanding (SMSU) and Measures of Moral Behavior 37 Table 4. Change in Percent Variance Explained by Step in the Regression Models 43 Table 5. Correlations between Self-Understanding (SMSU2) and Measures of Moral Behavior 44 Table 6. The 10 Schwartz Values and Corresponding Value Orientations 53 Table 7. Moral Behavior Descriptive Statistics across Value Orientations 55 Table 8. Descriptive Statistics of the Conscientiousness Contributor Variables 72 v List of Figures Figure 1. Centrality-Rating Paradigm, Screen 2, for Participant # 75986 22 Figure 2. Histogram of Honesty Scores 31 Figure 3. Histogram of Self-Theory Levels 34 Figure 4. Scheme Emphasis in SUIT Coding • 36 Figure 5. Scheme Emphases for the Various Coding Systems 41 vi Acknowledgments This research was supported by a Graduate Entrance Scholarship to Jeremy A. Frimer as well as by a Hampton H&SS Research Grant to Lawrence J. Walker. I would like to thank Katie Lightfoot, Melissa Brinkman, Jamie Smith, and Nathalie Oakes, whose contributions as research assistants were instrumental in administering interviews and coding. I am also in gratitude to Dr. Susan A. J. Birch and Dr. Jeremy C. Biesanz for their constructive challenges and helpful suggestions as members of my thesis committee. Finally, I would like to thank my advisor, committee member, mentor, and friend, Dr. Larry Walker, for his on-going encouragement, insight, direction, and thorough involvement in this project. Chapter I - Introduction In an era of genocides, pre-emptive war, and ecological destruction, moral considerations factor strongly into many aspects of our lives. While these blatant forms of immorality are not particularly new to the 21st century, there now exists a more silent, esoteric brand of global immorality that is just as important. With globalization, Western countries are having an increasing impact on less developed nations, which often is expressed as economic and political exploitation and results in dire consequences. The offenses are as often malicious acts as they are failures to act in a humanitarian manner. An example, of the latter would be the West's failure to provide medication and financial aid to African nations suffering with the AIDS pandemic, even while the West continues to benefit from trade with, and the sale of armaments to, these nations. While the citizens, corporations, and governments of the West assume much responsibility for their complicity with these human rights violations, the acts of oppression are often carried out by only a few. In contrast, buying a garment in a store in Canada hardly seems to carry moral implications. But as it turns out, to buy some textiles is to support the exploitation of children and/or those bound to sweatshop labor. Hence, globalization has contributed to making our individual consumer choices into moral ones. In a similar way, our occupational and political decisions are laden with moral implications. This was salient in the 2004 U.S. presidential election, where George W. Bush was re-elected despite his controversial decision to wage war on the independent country of Iraq (and without the multilateral support of the community of nations); a decision that has, to date, directly cost tens of thousands of Iraqi lives and a couple of thousand American lives. In spite of this, in exit polls, American voters were found to report "moral values" as the most important factor influencing their presidential choice. Regardless of how one interprets these statistics, undeniably, moral considerations are of primary concern here. On a more proximal level, interpersonal relationships and intrapsychic well-being similarly depend on mature moral functioning. A propensity for perspective taking and a dedication to integrity strongly influence how we interact with others and resolve interpersonal conflict. Trust and commitment both lie at the core of healthy relationships; these virtues very much rely on morality within each individual. In these ways, we see morality as pervading all aspects of life, both in terms of how we react to difficult situations as well as the lifestyles that we choose. Our understanding of moral failure, maladaptive values, rationalizations, integrity, and so forth are of fundamental conceptual and practical importance to society. Widely accepted among developmental psychologists is the notion that such mature moral functioning comes through deep internalization, rather than from superficial heuristic learning. Inspired by the formalist Kantian tradition of philosophy and by the cognitive-developmental work of Piaget (1932/1977) in developmental psychology, Lawrence Kohlberg (1969, 1981, 1984) established a developmental model of moral reasoning, which emphasized the cognitive aspects of moral functioning. The explicit devotion to pure rationality of Kohlberg's model, however, has come to be recognized as 1 entailing a substantial conceptual skew in our understanding of the moral domain and moral functioning (Walker & Pitts, 1998). Despite decades of development, Kohlberg's most refined version of the model still failed to adequately incorporate other aspects of moral functioning (emotions, values, etc.) and the model explained no more than 10% of the variance in actual moral behavior (Buchanan, 1992). While " behavior was regarded almost as a trivial and inconsequential by-product of moral cognition by Kohlberg, it is a central tenet of this research that its inclusion within the moral domain is of fundamental importance; as was well put by Winston Churchill: "However beautiful the strategy, you should occasionally look at the results." As now is increasingly becoming understood, moral motivation is the missing link in the rationality-action gap. Reflecting this shift in the field, the contributors to the 51st Nebraska Symposium on Motivation (Carlo & Edwards, 2005) set out to review extant theories and future directions on the topic of moral motivation. Some lines of investigation have included: moral emotions (e.g., sympathy, guilt; Eisenberg, 2005), moral sensitivity (Narvaez, 2005), and the role of temperament (Kagan, 2005). But it was in the heyday of Kohlberg's theory that Blasi (1984) pointed to the cognitive-affective-social construct of identity as the core, deep structure behind morality (Walker, 2004). According to Blasi, identity is a primary source of moral motivation and plays a strong causal role in how we go about living our lives and what kind of persons we become. It represents a broader, more encompassing approach to the moral domain—one that seemingly offers greater explanatory power than any other paradigm. Summing up the outlook for moral psychology in the post-Kohlbergian era, Lapsley (2006) concluded that "the desire to understand moral functioning in the context of larger developmental processes, including the self, identity, and personality development, is now the unmistakable next wave in moral psychological research" (p. 61). Moral Identity as Moral Motivation: Blasi's Self Theory While the self has so many facets that its study may be more a collection of loosely-related phenomena than a unified investigation (Baumeister, 1998), it is the single capacity for self-reflection that lies at the core of identity (Leary & Tangney, 2005). While the self has been studied in various different ways—for example, statuses of identity development (Marcia, 1966; Erikson, 1968); self-awareness (Selman, 1980); and self-concept (Hattie, 1992)—only Blasi's theory seems to be configured appropriately to address the question of moral identity. To Blasi (1988), identity has the capacity to integrate otherwise disconnected and even divergent self-observations into a coherent and unified self-theory. But identity is not relegated to the function of serving as a reflective means of reconciling complex information; it can also take on the role of actively representing the self, becoming a part of the agentic centre of deliberate action. It is in this way that identity may provide the motivation for moral action. However, there are vast developmental and 2 individual differences with respect to identity acting as a source of moral motivation (Hardy & Carlo, 2005). For identity to act as a moral motivator, Blasi (1983, 1984, 1993, 1995) points to three major components that must be in place (Walker, 2004), and which are posited to Change through development as well as to vary among individuals, (a) The first component of Blasi's notion of a moral self is a deep desire to be able to honestly compare one's actions and self-perception and see minimal disparity. That is to say that self-consistency (integrity) is a fundamental motive in and of itself. Here, Blasi emphasizes rationality as a key component to moral functioning, (b) Second, the moral prototype's self-perceptions are to be arranged in a specific way according to Blasi: It is moral values and goals that must be deeply embedded in the core of a person's sense of self. In this way, moral people understand themselves in terms of their sense of morality. This phenomenon is illustrated well by Colby and Damon (1992) who, in studying a group of extraordinarily caring people, found a fusion of personal and moral goals: "All these men and women have vigorously pursued their individual and moral goals simultaneously, viewing them in fact as one and the same.... Rather than denying the self, they define it with a moral center.... None saw their moral choices as an exercise in self-sacrifice" (p. 300). (c) And finally, Blasi's model stresses the importance of a sense of personal responsibility for moral action. That is to say that a moral person is not only interested in engaging in moral behavior but in fact feels bound to do so. Essentially, one's sense of self-worth becomes encapsulated in moral considerations. Lapsley (1996) summarized this by juxtaposing such a duty with the kind that comes as a result of forming a moral judgment: "For Kohlberg ... not to act is to betray a principle. For Blasi ... not to act is to betray the self (p. 86). Previous Attempts to Model Blasi's Theory In attempts to bring theory to practice, Blasi (Blasi & Milton, 1991; Glodis & Blasi, 1993) introduced a model intended to capture the first of his key components. In asking respondents to reconcile internal and external observations, the Sense of Self interview assesses integrity. The measure delineated four levels of identity integration: (a) Social Role Identity is, in essence, a pre-identity mode whereby identity is defined by physical appearance and social roles, (b) The second mode, Identity Observed, marks the emergence of potential conflict between action and thought. In this mode, an individual discovers an inner life, which can contrast starkly with external and superficial characteristics, (c) With identity becoming increasingly important to the individual, the third mode is Management of Identity, which entails the appreciation that identity serves not only reflectively but in fact should be constituted by standards and values, and hence must be earned. And finally (d) Identity as Authenticity involves a primary dedication to truth (over simple self-interest). For someone of this advanced stage, a preoccupation to acknowledge inconsistencies, including one's own, results in a concern for all human beings, worldwide. Despite its conceptual strength, an inadequate coding manual has kept Blasi's model from receiving the empirical support for which it has potential. Walker and Pitts (1998) found that 3 integrity figured prominently in naturalistic conceptions of morality, underscoring the need for further research in this domain. . Two independent attempts (Aquino & Reed, 2002; Reimer, Furrow, Baumeister-Peters, & Roth, 2001) have been made to capture the second component of Blasi's model—moral centrality. Both measures developed by these two sets of researchers tap the degree to which one sees him/herself in terms of moral characteristics. The measures ask respondents to rate (using a Likert scale) their level of agreement with a number of self-statements relating to internalization (e.g., centrality) and externalization (behaviors) of a number of moral traits (e.g., caring, fair, honest). For one measure (Reimer et al., 2001), the traits were taken from a corpus of naturalistic conceptions of morality (Walker & Pitts, 1998). The measures have been shown to relate to several aspects of moral functioning including attitudes and behavior (Aquino & Reed, 2002; Reed & Aquino, 2003). The simplicity of this type of measure represents both a strength and a weakness. On one hand, the measure demonstrates (albeit, superficially) the second component of Blasi's model; on the other, the framework is self-limiting, precluding the integration of the other components of Blasi's model. In addition, these measures tap but a segment of Blasi's second tenet. The integration of moral values and goals into one's own identity is a phenomenon that can only be meaningfully measured within the rich context of a person's life story (McAdams, 1993). It is one thing to rate oneself highly in the moral domain; it is an entirely different thing to make moral commitments at the expense of other goals (e.g., monetary wealth). For these reasons, we argue that the trait endorsement approach falls short of fleshing out the full potential of Blasi's second tenet. The third and final component of Blasi's model—responsibility—has been addressed somewhat superficially by Kohlberg and Candee (1984). In the context of moral judgment interviews, people were asked if they would have a responsibility to act on their judgment were they in such a case. They found that those that reported a greater responsibility to act showed an elevated consistency between the decision choice and overt behavior. While tentative, these data support the theory behind Blasi's third tenet. Thus, no comprehensive measure of moral identity exists. The overarching goal of the present study is to advance towards a more encompassing measure of moral functioning. For such a model to succeed, it must be both sufficiently broad—so as to capture the various facets of moral identity—but also sufficiently sensitive to the subtleties of a moral self. The Present Attempt to Model Blasi's Theory The inner self and the outside world interact continuously and bidirectionally in discourse, acting as part of the judgment-action bridge. This open-ended self-understanding is both reflective and directive. In being reflective, self-understanding inherently draws into one's identity contextual information, and thus documents an individual's history, relationships, and subjective experiences. In being directive, self-understanding has the tremendous potential to elucidate a person's motivation. By setting out to capture the unbounded richness of the subjective experience of self within a framework that 4 allows for coding from a variety of different perspectives, we see the self-understanding approach as having the potential to account for all three components of Blasi's model. There is a key assertion here, favoring the phenomenonological approach—that the whole subjective experience is valid, credible, worth studying, and not completely reducible to the sum of its parts. This stands in contrast to more objectifying models of the self (e.g., moral expertise; Narvaez & Lapsley, 2005) where moral personality is reduced to the construct of chronic accessibility. While chronic accessibility of moral schemas has been shown to predict various aspects of moral cognition (Narvaez, Lapsley, Hagele, & Lasky, in press), we contend that chronic accessibility is more so a symptom of moral personality rather than the construct itself. Personality remains a whole phenomenon; one that should most validly be studied in a holistic way. In that a subject's intrapsychic world is not directly accessible to the outside observer, we content that the best way of studying it involves approaching the subject with direct inquiry. The question of which of the two approaches—phenomenological versus symptomological—is more predictive of moral action is an empirical matter, and one for future research. Hence, the model developed here will adhere to the phenomenological approach. Let us next review previous attempts at and existing frameworks for capturing the subjective experience of the self. The Subjective Experience of Self What is meant by "subjective experience of self? Clearly, there are striking differences in the way individuals answer the question, "Who am I?" The Zen Buddhist replies "I am an observer;" the materialistic consumer returns "I am beautiful;" the academic says "I am intelligent;" and the philosopher reflects, "I am my consciousness, traveling through time." We hold that these different self-statements are strongly value-laden and directive. Their existence does not get played out overtly; rather, our self-definition functions implicitly as the lens through which we see and platform from which we interact with the world. Our self-definitions tell us who we are, as well as how to act and think; in this way, self-definition enters the moral domain. Consider the contrast between, and potential real-world implications of, one person believing himself to be friendly, caring, broad-minded, and outgoing and a second person believing herself to be attractive, liked, and important. This point is central to our intuition of the functioning and functionality of the self. Not only does this paradigm offer an inroad to our understanding of the self, it also holds tremendous practical implication. In that self-understanding invites the powerful reifying influence of human language into the core constructs of personality/identity, we see this self-understanding as holding potential for intervention. By using the constructive power of language, it may be possible, to induce "lateral shifts" in self-understanding towards configurations that are associated with optimal developmental outcomes, including moral ones.. What is not yet clear, however, is how the different identities relate to moral functioning. It is the overarching goal of the present study to make a contribution towards the goal of mapping the moral self. 5 The task of eliciting a valid representation of an individual's self-understanding presents unique challenges. Few individuals have sufficient self-awareness and verbal ability to articulate a thorough and accurate account of themselves "off the cuff." In addition, context and culture shape the way we report ourselves. For example, contemporary Western society emphasizes activities and personality as the primary categories of self-report (Derryberry & Thoma, 2005b; Kunda, 1999). Hence, simply asking participants the question "Who are you?" (Spitzer, Couch, & Stratton, 1971) may result in non-representative and impoverished accounts. That is, respondents may rely heavily on culturally-scaffolded responses (which, of course, comprise one part of the self) but neglect to report other important aspects of the self that would rarely enter into common conversation (e.g., regarding one's sense of continuity through time). To address the issue that most people fail to spontaneously report a comprehensive rendition of their self-understanding when prompted in a general way, many researchers have sought a conceptual framework of self-understanding with which to scaffold a person's individual report. It would appear that there is a nearly infinite number of ways an individual can define the self. It would be beneficial if this unwieldy set could somehow be reduced to a more manageable one. With this goal in mind, William James (1890) began by drawing a distinction between the self-as-subject (i.e., the "me-self) and the self-as-object (i.e., the "I-self'). The me-self is "the sum total of all a person can call his [sic]" (1892/1961, p. 44), which includes (Damon & Hart, 1988): (a) material characteristics (possessions, physical characteristics); (b) activities and capabilities (activities, skills); (c) social characteristics (roles, relationships); and (d) psychological traits (personality, thoughts, beliefs, goals). Meanwhile, the I-self emphasizes the subjective aspects of actively experiencing and interacting with the world. It includes (Damon & Hart, 1988): (a) agency (actively forming and interpreting one's experience); (b) distinctness (awareness of the uniqueness of one's own experience); (c) continuity (a sense of connection with one's self of the past and future); and (d) reflection (an awareness of one's own awareness). James' taxonomy of the self has been ubiquitously taken as a comprehensive and valid characterization of the domain. Given this "self-understanding space," the task then turned to drawing self-understanding from the entire geography. Precisely how the theory has been implemented, however, has been the point of considerable divergence. Next a review of these approaches will be provided, followed by a proposed, alternative approach. Self-Concept The self-concept literature makes the assumption that the various dimensions of self (e.g., social, physical, academic) are of equal importance or value to all individuals (e.g., Braken, 1992). Hence, most 6 self-concept scales simply ask participants to rate their subjective aptitude in each dimension in a hierarchical structure. In this way, self-aptitude ratings are given for each of Math, English, etc., in the academic domain, tallied up to form an overall academic score, then pooled with other domains (e.g., social) to establish a global self-concept score. However, even Braken himself acknowledged that "some dimensions are likely to be more important for individual children than for others" (p. 5). To accommodate this, Harter (1986) proposed asking respondents to rate the importance of each of the domains, and then weighting their self-assessment by the domain weighting. Only, this approach has yet to bear fruit. Wainer (1976) reviewed research on self-concept weighting paradigms and concluded that unweighted assessments consistently predict overall self-esteem better than weighted ones, and advocated for a simple tally of self-ratings as the best predictor of life outcomes such as satisfaction with life and academic achievement (Marsh, Trautwein, Ludtke, Koller & Baumert, 2006). Summarizing this failure to bring self-concept theory in line with common intuition, Marsh and Hattie (1996) stated that the "practical implications of Wainer's recommendations have not been seriously threatened by subsequent research" (p. 82). It would seem that this null finding can be explained by one or both of two failures: either the problem is with the notion of weighting itself (which would challenge the phenomenonological view) or the lack of predictive validity could lie with the heavy pre-structuring of the self-concept approach in general, which would suggest that it is not particularly efficacious as a life-outcome model. It is argued here that the latter is the case. Self- Understanding Approaching the task of implementing James' (1890) theory in a different way, Damon and Hart (1988) created a developmental model, which delineates three mutually orthogonal dimensions of self-understanding: (a) four me-self "schemes" (physical, active, social, and psychological) by (b) three I-self schemes (agency, distinctness, and continuity) by (c) four developmental levels1. These developmental levels (called "self-theory") represent the level of sophistication with which people reason about their endorsement of a particular characteristic. The various levels span from simple categorical identifications to competitive motivations to more prosocial interpersonal implications and finally to well-reasoned values or directive notions. In this way, it is somewhat reminiscent of Kohlberg's theory, simply applied to the self rather than moral dilemmas. Advances in level of sophistication are associated with age (Damon & Hart, 1988) and moral functioning (Hart & Fegley, 1995). In order to elicit data to test their model, they developed the Self-Understanding Interview (SUI; see Appendix A). Instead of asking specific questions for each scheme, the SUI asks a series of general questions relating to self-definition, self-evaluation, and so on. Each question was designed to draw responses (or "chunks") nonpreferentially for each of the four me-self schemes. However, considering 1 Damon and Hart (1988) omit the fourth I-self scheme (reflectiveness) delineated by James (1892/1961), citing practical difficulties in measuring this nebulous construct. 7 the rarity with which the average person spontaneously verbally represents themselves in terms of their I-self (e.g., continuity), the SUI includes questions that draw responses specifically for each of those schemes. If the SUI did not prompt directly for responses regarding these schemes, it is unlikely that relevant data would be available. We believe that the nature—in terms of emphasis on (the seven) different schemes—of self-understanding varies dramatically and importantly across individuals. In his popular book, Anthony De Mello (1990) argues that reflectively de-emphasizing the me-self results in an alleviation of suffering. So when you step out of yourself and observe "me," you no longer identify with "me." Suffering exists in "me," so when you identify "I" with "me," suffering begins. Say that you are afraid or desirous or anxious. When "I" does not identify with money, or name, or nationality, or persons, or friends, or any quality, the "I" is never threatened. It can be very active, but it isn't threatened. Think of anything that causes or is causing you pain or worry or anxiety. First, you can pick up the desire very keenly or else you wouldn't be suffering. What is that desire? Second, it isn't simply a desire; there's an identification there. You have somehow said to yourself, "The well-being of T,' almost the existence of T,' is tied up with this desire." All suffering is caused by my identifying myself with something, whether that something is within me or outside of me. (p. 50, emphasis in original) This claim speaks to an interaction between the I-self and the me-self. But unpacking this claim suggests that there should also be a direct relationship between an emphasis on the me-self and suffering, as well as an analogous but inverse relationship between the I-self and suffering. The same general reasoning could be applied not only to subjective well-being but also to moral functioning (as argued by Blasi, 1984). Thus, the first hypothesis of this study—henceforth referred to as the De Mello Hypothesis—is that emphasis on the me-self is negatively associated, and emphasis on the I-self is positively associated, with moral functioning. Previous Research Two previous studies have focused on the relationship between self-understanding and moral behavior. As will be seen, these studies made valuable contributions to our understanding of this complex relationship; however, neither succeeded in cleanly demonstrating the hypothesized relationship. Next, a review of each is presented. Hart and Fegley (1995) This first important study was carried out by one of the original authors of the Self-Understanding model (Daniel Hart). The primary goal was to explore the relationship of a variety of measures of self-understanding (including those derived from the SUI), along with moral reasoning, to moral action. A unique sample of 15 adolescents (M= 15.5 years) were chosen for having engaged in sustained prosocial behavior in a disadvantaged New Jersey community. Fifteen demographically matched teens were recruited as a comparison group. Interestingly, the groups did not differ in terms of moral reasoning; they did, however, differ on a number of measures of self-understanding. 8 First, participants completed the Persons and Their Attributes Questionnaire (PTAQ), a measure distinct from and orthogonal to the SUI. The PTAQ is a complex measure, which allows inferences to be made about the relationship of various facets of the self to each other and to other individuals (the details of which are not discussed here). Hart and Fegley found that caring exemplars tended to focus more so on themes of self-evaluation and self-peer distinction than the comparison group in differentiating various aspects of the self (e.g., actual self, ideal self) from one another and from other individuals (e.g., parent, peer). In contrast, comparison teens relied more on the third dimension—change-over-time—to differentiate between the various selves and others. This can be taken to mean that self-evaluation and self-peer differences are important considerations to the intrapsychic functioning of moral exemplars. In addition, the actual self of the typical caring exemplar incorporated more of his/her ideal self and parentally-related self, and less of his/her friend-related self than did the actual self of the comparison teen. These findings together point to the complex ways in which the moral self is configured and interplays with conceptions of other selves and other individuals. Second, and of more direct application to the present study, Hart and Fegley assessed the self-content and self-theory of the self-understandings of their participants. The SUI was administered and self-theory was assessed by considering the highest level present. It was found that half of the exemplars but only 7% of the comparison adolescents evidenced Level 4 self-theory. Interestingly, self-content was not assessed using the SUI. Instead, participants completed the Free Descriptors of Self (FDS) measure, which asks five questions: (a) What kind of person are you? (b) What kind of goals do you have for yourself? (c) What kind of emotions do you typically experience? (d) What are your typical activities? and (e) What is your personality like? Responses were coded into one of five schemes: the four me-self schemes (physical, active, social, psychological) or the newly-introduced "goal" scheme. What more, schemes were divided into between four and eight sub-schemes for each scheme (e.g., two sub-schemes of the physical scheme are attractiveness and size). There were a total of 28 sub-schemes. Relative to the comparison group, the authors claim that exemplars referenced the following sub-schemes more often: (a) positive, caring, moral personality traits; (b) moral, caring goals; (c) academic goals; and (d) non-moral typical activities. However, they appear to arrive at this result though a series of 28 unprotected t-tests. Performing a Bonferroni correction would reduce a to .0017, a significance level that none of the purported findings reach. These results pertaining to the content and theory of self-understanding suggest that these aspects of identity may be active in moral functioning. However, three further points of concern are noted regarding the methodology used. First, while the metric of self-theory (highest level present) proved successful in distinguishing groups, this method will clearly be ineffective with older populations. In older age groups, Level.4. self-theory is.far.morecommon;.this.will cause a. ceiling effect, and thus render this metric ineffective. Second, it is unclear as to why the SUI was not used to tap content. By introducing the FDS measure, the authors make a questionable move towards fracturing the measurement 9 of the self. In that the SUI already draws narrative on the various schemes, it seems unnecessary to introduce a redundant set of observations. In addition, further fracturing is noted in the introduction of sub-schemes. Their implementation did yield some important findings regarding the complexity of the moral self. For example, exemplars may differ from comparison teens in terms of moral, caring goals but not in terms of occupational goals. However, the generation of the various sub-schemes would appear to be an atheoretical, "scatter gun" approach, rendering the results difficult to interpret. And third, this study had low power to detect differences; a small sample size (JV= 30) is noted. Two improvements from the Hart and Fegley methodology are suggested. First, as noted, it would be beneficial and parsimonious if multiple facets of the self (here, self-theory and self-content) could be measured from the same data set. To achieve this, self-content could be coded from the SUI. And second, a larger sample size is recommended so as to improve statistical power. The present study attempts to achieve both of these suggestions. Derryberry and Thoma (2005b) The goal of this second important study was to understand how self-understanding and moral reasoning act in concert to influence moral behavior. The authors asserted that: (a) the construct of moral action is most appropriately tapped with multiple measures; and (b) moral action is best explained when considering multiple predictors (e.g., moral judgment, self-content) in tandem, as opposed to in isolation from one another. Two differences between the methodologies of the two studies are noted. First, Derryberry and Thoma looked for individual differences within a single sample, whereas Hart and Fegley compared an exemplary group with a comparison one. Whereas the latter relied on group membership as a metric of moral action, the former relied on multiple sensitive, broad-band, and validated questionnaires and behavioral manipulations. The aggregation of multiple observations of behavior increases reliability and validity in assessing personological dispositions (Epstein, 1983). And second, Derryberry and Thoma's participants were undergraduates and were older (M= 20.6 years) than Hart and Fegley's community-based care exemplars (M= 15.5 years). Identically to the Hart and Fegley study, participants completed the same three measures of self-understanding—SUI, FDS, and PTAQ—from which self-theory, self-content, self-concept as semantic space, and self-concept as hierarchy were derived. Moral judgment was assessed using the Defining Issues Test (DIT; Rest, Narvaez, Bebeau, & Thoma, 1999). Finally, moral action was operationalized by three measures: (a) Getz's (1985) Attitudes Towards Human Rights Inventory (ATHRJ), a self-report measure of opinions on a number of issues pertaining to liberty and justice; (b) Clary, Snyder, Ridge, Copeland, Stukas, Haugen, and Miene's (1998) Volunteer Function Inventory (VFI), a self-report measure of an individual's motivations for engaging in altruistic behaviors; and (c) Bersoff s (1999) Honesty measure (HON), a behavioral measure wherein a participant's tendency to refuse an overpayment for participation is assessed. While the authors refer to both the ATHRI and VFI as measures tapping moral action, they are better understood as measures of subjective appraisals 10 (opinions, sentiments) of moral considerations, rather than actual behavior. This problem will be corrected in the present study. Self-theory was assessed using the SUI; the authors note a ceiling effect (and resultant restriction of range). Meanwhile the FDS was used to tap content, where not five but six schemes were coded (a new "moral" scheme was added). Structural Equation Modeling (SEM) was used to assess the relationship between each of the five predictor variables (DIT, self-content, self-theory, semantic space, self-hierarchy) to each of the moral action variables (ATHRI, VFI, HON). Both of the hypotheses were supported. First, the predictor variables related differentially to each of the predicted variables: Moral judgment predicted ATHRI and VFI; semantic space predicted honesty; and hierarchy predicted VFI (all other relationships were not significant). Thus, assessing moral functioning with a single variable would have resulted in misleading findings. Based on this result, the authors advocate for multiple measures of moral functioning in future research. This point will be accommodated in the present study. Second, predictor variables tended to be related to a greater number of predicted variables when multiple predictor variables were considered simultaneously. For example, moral judgment was significantly related to both VFI and ATHRI; but when a second SEM was run with self-understanding omitted, moral judgment no longer predicted VFI. The authors interpret this as suggesting that multiple moral developmental constructs must be functioning simultaneously for appropriate action to result. However, the authors do not go so far as to suggest that singular constructs (e.g., self-understanding) should not be studied in isolation; rather, they advocate for a distal goal being the integration of singular constructs into constellations. Another important consideration is that neither self-content nor self-theory predicted any of the three measures of moral functioning. These null findings may be attributed to the problems with the instruments mentioned previously (restriction of range in self-theory and atheoretical fracturing of self-content). Without sufficient theoretical justification, Hart and Fegley had augmented the list of me-self schemes with a fifth "goals" scheme; similarly in the absence of theoretical justification, Derryberry and Thoma contributed a sixth "moral" scheme. The notion of separating moral content from other facets of life stands in direct contradiction to well-validated models of selfhood (McAdams, 1993). The loss of predictive validity associated with this ad hoc renovation of the me-self poses a serious threat, and perhaps a dead-end for this line of inquiry. While self-theory and self-content (at the sub-scheme level) succeeded in predicting moral action in Hart and Fegley's adolescent moral exemplars, they failed to do so in an older cohort. These young adults should have just entered the developmental period wherein identity functions as a moral motivator (Bergman, 2004). A null finding in this case can be taken to mean that either self-understanding (as a construct) does not strongly relate to moral functioning, or that the methods for tapping this construct are suboptimal for detecting the real phenomenon. In the present study, the latter will be assumed and explored, and the former accepted only if the latter fails. 11 • The Current Approach We persist with the hypothesis that individual differences in emphasis placed on the specific self-schemes are associated with moral action. It is argued that previous weak or null findings are explained by methodological shortcomings. In understanding the precise domains in and mechanisms by which identity causes moral functioning, these result could provide valuable information regarding the way character education interventions would try to induce prosocial outcomes. For example, Hart and Fegley (1995) found that self-understandings emphasizing continuity with the past and future and those emphasizing psychological traits are associated with higher moral functioning. If causality can be established, interventions aimed at inducing "lateral" shifts towards adaptive domains may prove successful. To date, the SUI represents the best compromise between efforts to simultaneously address the issue of lean accounts (as faced by asking "Who are you?") while imposing the least amount of structure possible (the self-concept approach failed on the latter). By prompting for breadth (here, all seven schemes), depth (probing questions designed to tease out unreported motivation and meaning of specific self-statements), and providing an analytic framework for inferring level of sophistication (here, the four developmental levels) of responses, the SUI represents the most systematic and thorough measure of self-understanding to date. However, there are a number of problems with the measure that are keeping it from realizing the inherent potential of self-understanding in capturing moral, identity. The SUI has six significant shortcomings, which are addressed in turn: 1. Bluntness. The questions of the SUI are asked in an abstract, contextually isolated way. After being greeted by the interviewer, a respondent is bluntly asked a series of atypical, personal questions (e.g., What kind of person are you? What does that say about you?). While extracting information of that sort is the goal of the interview, giving respondents a chance to "warm up" by speaking about something more common-place would likely be beneficial for establishing a comfortable rapport, as well as priming for self-reflection. 2. Biasing. The SUI fails to pull for the breadth of responses homogeneously. On the face of it, the question, "What kind of person are you?" does not draw preferentially for any of the four me-self schemes. However, personality traits dominate our reported impressions of others (Park, 1986). This does not necessarily mean that our conceptions of others and ourselves are dominated by traits; it is possible that traits simply offer an efficient means of communicating our impressions. It remains to be seen how much traits dominate our conceptions of ourselves in the absence of contextual pressures (e.g., time). In addition, the natural bias to respond in a socially desirable way, interacting with the SUI's reliance on general questions, skews responding towards culturally acceptable responses and away from a valid representation of the individual. 3. Orthogonality Claim. The Damon and Hart model unambiguously claims that the me-self schemes are 12 orthogonal to the 1-self ones, and the two types are measured in non-parallel ways (general as opposed to specific questions). While their claim may not be unfounded, there are good reasons to place all seven schemes in the same dimension, and allow them to vie for a piece of an individual's self-understanding ipsatively. As stated, De Mello (1990; Gyatso, 1995) makes a spirited argument for not identifying with any of the me-self schemes and instead seeing oneself as "the observer," a purely I-self construct. This is an example of me- and I-self schemes engaged in competition, something that is implicitly precluded in Damon and Hart's model. 4. Equal Weighting. The SUI fails to allow participants to actively construct, organize, and contextualize their responses. It is quite possible that one response given by a respondent is of greater subjective importance than the entire remainder of his or her responses. The SUI, as it stands, does not accommodate this issue. 5. Insensitivity. The SUI is not configured in a way that it will be sensitive to morally relevant self-narrative. Neither schemes nor levels make distinctions between moral self-statements (e.g., "I'm generous") and immoral ones (e.g., "I'm greedy"). Probably noticing this problem, one of the model's own creators (Hart & Fegley, 1995) later introduced the Free Description of Self (FDS) measure (see p. 9), which splits the schemes into sub-schemes. Some of the sub-schemes are moral ones (e.g., "moral, caring personality characteristics"). The FDS conceptually overlaps with the SUI and is intended as an alternate procedure to tap self-content. This introduction represents the beginning of an era of fragmentation in the measurement of self-understanding. It is of practical and conceptual interest to subsume self-understanding measurement under one comprehensive paradigm. 6. Predictive Validity. Finally (and perhaps most importantly), the SUI has not yet been demonstrated to predict a wide range of moral outcomes. The most convincing associations were found by Hart and Fegley (1995); however, the associations between scheme emphasis (here, certain types of psychological and active schemes) and moral exemplarity were scant. It has been implied that a strong sense of continuity as well as agency are fundamental to one's integrity (Blasi, 1984; Colby & Damon, 1992). Hart and Fegley found this to be the case in yet another parallel measurement of self-understanding (i.e., semantic space analysis) but failed to detect a difference within the SUI itself. Even more discouraging are the results of Derryberry and Thoma's (2005b) study, wherein they failed to detect any associations between either of scheme emphasis or self-theory and moral behavior. Reconceptualizing Self-Understanding In attempting to redress these shortcomings while still allowing for integration of previous findings, in this thesis we reconceptualize the Damon and Hart model into a new and more comprehensive framework, the Self-Understanding Interview—Transmogrified (SUIT; see Appendix B). We address the first shortcoming (bluntness of the interview) by creating a friendlier and hence more comfortable environment; in doing so, we hope to encourage more transparent and elaborated responding. A more comfortable environment is achieved by starting the interview with two questions 13 that prompt the respondent to provide a brief overview of his or her life. This first question is adapted from the Life Story Interview (McAdams, 1995). An additional benefit of beginning with general, familiar questions is that responses would now be contextualized. McAdams (1993) argued that identity is most meaningfully measured within the context of life stories. After re-centering in the present with a question about the current chapter of the respondent's life, the interview proceeds with 14 self-understanding questions. To address the second shortcoming of the SUI model (preferential pull), the SUIT makes a step towards a more rigid structure by asking questions prompting for each of the seven schemes, but then later asks respondents to rate their responses in an ipsative fashion. To accomplish this, participants rate their response stems on a 100-point scale using an active computer spreadsheet. Beside their ratings, a pie chart—representing the entirety of their self-understanding—will automatically display the percentage of each of their responses. In this way, each response will be on a level playing-field with every other one. Note that this methodology simultaneously addresses the third and fourth limitations of the SUI paradigm. . There is a critical assumption here that an individual's well-reasoned views about him or herself are in some way more valid than their unreflective ones. The latter view has dominated past research on self-understanding (Damon & Hart, 1988) and was implicitly supported by Freud, McDougall, and possibly Erikson's views that moral identity produces and influences moral understanding (Blasi, 1984). In contrast, the former has received only modest empirical support (e.g., Thoits, 1992). This critical assumption reveals a fundamental tenet of our model: That contrary to the "unreflective camp," we argue that reasoning should play an important part in the construction and configuration of the self. This notion has been argued consistently by Blasi (e.g., 1984): "the self s very identity is constructed, at least in part, under the influence of moral reasons.... the direction of influence would be from moral understanding to moral identity, rather than the other way around" (p. 138). The SUIT, as it stands, prompts respondents for the entirety of the Jamesian self-understanding space while still being sensitive (and representative, as we have argued) of one's own experience of self. It strikes somewhat of a delicate balance between the rigidity of the self-concept approach and the overly open-ended and fractured Damon and Hart model. Metaphorically, we conceptualize self-understanding as the platform upon which we stand while interacting with the world. The seven schemes form the length of the platform, and the four levels of sophistication the width. The "Platform Model" is intended to provide a valid assessment of Blasi's theory. While several measures have had partial success in accounting for one of the three tenets, it is one of the overarching goals of this study to move towards a broader, more encapsulating framework. It is argued here that the Platform Model partially addresses the first and second tenets of Blasi's model (centrality, responsibility), and has the potential to be extended to address the third (integrity). What the 14 Platform Model loses by as yet capturing the expanse of Blasi's model, it gains in providing an appropriate data set for doing so in future research. That said, let us now consider how the Platform Model fairs in capturing Blasi's model. First, the Plaftorm Model does a partial job of accounting for the centrality of moral goals and beliefs. In eliciting rich, contextualized accounts of self-understanding, the SUIT allows respondents to report whatever it is that is central or important to them. A possible self-understanding is one involving moral goals and beliefs. While the Platform Model does well to bring goals and beliefs into relief (psychological scheme, Level 4) from other aspects of the self, it has yet to be developed to differentiate between moral beliefs and goals, on the one hand, and amoral or immoral ones on the other. That is, as both are coded as Psychological Level 4, the Platform Model stops short of differentiating moral beliefs from amoral and immoral ones. Hence, the SUIT is configured for drawing out the self-narrative in which the centrality of moral facets could be revealed, but does not detect a potentially critical divergence in values (e.g., moral ones). Second, the Platform Model partially and indirectly measures a sense of responsibility. While it does not directly query respondents for their sense of responsibility to act upon their goals, it does tap constructs that are integral to responsibility. Specifically, it is argued that responsibility necessarily entails a valuing of social roles and relationships, a sense of agency, and a sense of continuity—each of which the SUIT is configured to measure. First, Kohlberg and Diessner (1991) suggest a link between personal responsibility and the formation of specific relationships to other people. Experiencing such relationships (it is argued) contributes to such emotions as empathy, guilt, and commitment. Second, a sense of volitional agency—that one's actions are a direct result of one's will—provides a critical connection between one's actions and the evaluation of the self. Only for those actions that result in a choice is it reasonable to implicate the self. And third, responsibility entails a sense of connection to other individuals through time—-a memory of past actions and future implications are integral to a sense of responsibility. Such a sense of continuity is also captured well by the SUIT. Third, the Platform Model does not measure integrity but has the potential to be expanded to do so. Recall that a sense of integrity entails a dedication to the critical assessment (for congruence) of one's actions and self-perception. The SUIT is efficacious in measuring the latter and the present study does measure the former, but the present study does not put respondents "on the spot" to reconcile any disparities between the two. However, there is potential for the methodology used here to be extended to ask respondents to consider potential contradictions inherent in their different responses, in order to tap integrity in a way perhaps.similar to the Sense of Self interview (Blasi & Milton, 1991; Glodis & Blasi, 1993). This is beyond the scope of the present study and a matter for future research. Thus, it is argued that the-Platform Model is more efficacious in encapsulating Blasi's theory than any extant measure. Furthermore, it is argued that most gaps that may exist between the Platform Model and Blasi's theory are probably reconcilable with further research. 15 Predictive Validity Recall that the sixth and final criticism of the SUI was its inability to tap the breadth of the moral. domain. In that the Platform Model better adheres to Blasi's theory, it is argued on a conceptual level that the Platform Model will bridge the judgment-action gap and thus succeed where the SUI fails. But, ultimately, this is an empirical question. By measuring both identity and a wide range of moral behaviors simultaneously, this study will test the predictive validity of the Model. It is a matter for future research to address the directionality of influence in this relationship. At least two issues plague the measurement of moral behavior. First, moral functioning can be expressed within a vast number of domains (e.g., lifestyle, occupational, and political choices; responses to moral dilemmas; ecological behavior; interpersonal relationships; Blasi, 1980; Krebs & Van Hesteren, 1994). It is argued that moral behavior is a result of both the individual's inner constitution as well as r how it interacts with situational influences and context. This makes the inner construct that influences moral behavior slippery when it comes to measurement., Second, the inherent social desirability of the construct also makes the measurement of moral behavior difficult. In essence, nearly everyone believes him or herself to be a "good" person. To eliminate the confound of social desirability, measurement should be opaque, multimodal (Derryberry & Thoma, 2005a, 2005b), and statistically controlled for social desirability. These two issues are here addressed by assessing three different modes of moral behavior while simultaneously controlling for the confound of social desirability. These three modes are all of either self-report or direct behavioral observation, each of which has its benefits and drawbacks. While self-report measures are useful in that they allow for tapping a wide range of behaviors with ease, they are inherently fraught with issues of socially desirable responding. Meanwhile, in presenting an extrinsic reward for "failing" the moral task (e.g., money), behavioral measures reduce the impact of socially desirable responding. The short-coming of behavioral measures is their narrowness—they are resource expensive and hence often relegated to tapping but one behavior. Together, however, self-report measures and a behavioral one make for a robust battery for measuring moral action (see Kunda, 1999). In the present study, moral behavior is tapped broadly with two self-report measures and an actuarial behavioral measure of honesty. First, altruistic behavior (e.g., volunteering) represents the traditional conceptualization of moral action. Second, Kahn (2006) makes the case for the inclusion of ecologically responsible behavior (e.g., recycling) in the moral domain. The basis for his argument is that people have relationships both with other people as well as with animals, plants, and forests, and that our sense of justice, obligations, and rights ought to be applied to both sets of relationships. Third, a measure of actual moral behavior (honesty) will be used wherein participants will be (seemingly accidentally) overpaid for their participation in the study. It is argued that this battery of measures taps the breadth of the moral domain. 16 Summary and Hypotheses Rationality-based models of moral functioning are only weakly predictive of moral behavior. Identity (specifically self-understanding) is theorized to bridge the reason-action gap. Previous attempts at demonstrating the relationship between self-understanding and moral action have failed to demonstrate the effect, perhaps due to conceptual and methodological problems. Attempting to correct these problems, the current study aims to demonstrate the hypothesized effect, and in doing so, move towards a more comprehensive understanding of moral functioning. While the general goal of the present study is to make gains towards the mapping of the moral self, there are three specific hypotheses. First, it is hypothesized that (some of the seven) scheme emphases will be associated with moral behavior. In particular, it is hypothesized that the sum of the I-self schemes will be positively associated with moral behavior—the De Mello Hypothesis. In addition, identifying with certain aspects of the me-self are hypothesized to be associated with moral action (e.g., psychological, positively related; physical, negatively related); whereas the incongruity of these relationships is hypothesized to negate any main effect between the me-self and moral action. Second, it is hypothesized that moral action will be predicted more accurately by scheme emphases derived from the centrality weighting paradigm than those generated using a raw hit count. And third, it is anticipated that self-theory will suffer from a ceiling effect, and thus be unpredictive of moral action. 17 Chapter II - Method A correlational design was used to explore the relationship between self-understanding and moral action. Participants responded to a battery of questionnaires pertaining to moral behavior via an online survey and then participated in an individual interview wherein self-understanding was tapped. The behavioral measure of moral action (honesty) followed the interview in the context of participants' receiving their honorarium for research involvement. Participants To methodologically reduce the influence of confounds such as age and level of education, while at the same time enhancing variability in the primary variables of interest here (e.g., self-understanding, moral functioning), the population of interest was narrowed to young adults attending university in Western Canada. Sample Size • In performing a power analysis, we consider the hypothesis that the reconceptualized model would show a stronger association between scheme emphasis and moral action than the original SUI (Damon & Hart, 1988) did. Hart and Fegley (1995) found a medium effect size (r = .33), albeit for the relationship between emphasizing moral, caring goals and moral exemplarity. Estimating an increase to a large effect size (r — .54) with the new model, power of .80, and p < .05 (one-tailed), a sample size of 100 is estimated (Cohen, 1988). Demographics It is widely held that identity does not act as a source for moral motivation until later in life, when for some people, one's sense of morality and one's sense of self become integrated (Bergman, 2004; Blasi, 1989; Colby & Damon, 1993; Damon, 1984; Hardy & Carlo, 2005). There is some debate over when identity begins to act as a source of moral motivation; however, few disagree that the mechanism is at least partially developed by early adulthood. While it is of paramount importance to better understand the development of identity as moral motivation, the purpose of this study is to explore the relationship between identity and moral functioning once it has been somewhat established. In essence, this study aims to move towards fleshing out the end-point developmental goal of identity as moral motivation. Hence, the sample should be of early adulthood or later. While a sample of convenience (i.e., psychology undergraduates drawn from the department's subject pool) could fulfill this age requirement, in the interest of greater variability (and hence representativeness of a wider population), the sample will be drawn from various clubs of the University of British Columbia student association, the Alma Mater Society. Sixty-one clubs were selected from the publicly posted list of 295 (see http://www.ams.ubc.ca/clubs/) with the intent of sampling from clubs that reflect a variety of worldviews without introducing a confound (e.g., ethnicity). Some of the activities or foci of the clubs include: specific program of studies (e.g., engineering); business, marketing and technology; mentoring children; international political justice; fraternities; religious groups; theatre; 18 music; and athletics. It is important to note that the primary goal of drawing participants from a variety of clubs is to increase variability in the sample. While participants are drawn from specific groups of value orientations, unlike Hart and Fegley (1995), the metric of moral action will not be group membership; instead, it will be derived from a battery of measures of moral functioning. Should the members of the clubs evidence the intuited value orientations (discussed later), it will be noteworthy as a methodological advancement, but not critical to the primary analyses in this study. The website of each club was searched for any and all email contact information. A recruitment letter (see Appendix C) was sent to each contact individually. Of the 61 clubs contacted, responses were received from 30. Out of the total of 316 recruitment emails that were sent out, 111 individuals responded with intent to participate, of which 12 respondents were eliminated for one of three reasons: the respondent dropped out prior to the interview or was truant for the interview (nine cases); the interviewer was previously familiar with the respondent (one case); or experimenter error (two cases). Thus, a total N of 99 remained and was used in all further analyses. The sample was young adult in age (Af= 22.0 years, SD - 2.2 years) and 65% female. Participants identified themselves as European in origin (48%), East and Southeast Asian (28%), South Asian (6%), Latin, Central, and South American (3%), West Asian (3%), Aboriginal (1%), and other (10%). Of the participants, 65% reported having been born in Canada, with the remaining 35% having lived in Canada for on average 6.9 years (SD = 6.0 years). Participants had 3.5 years of post-secondary education (SD = 1.5 years), and belonged to a variety of faculties, the most common being: Arts (45%), Applied Science (17%), Science (15%), and Commerce and Business (6%). Procedure During the course of participants' involvement, the study's narrow focus on morality was occluded; instead, participants were led to believe that the study was interested in many general forms of human flourishing. Participants were primed for the study as follows: "In psychology, much more research attention gets devoted to mental illness than to positive human characteristics such as character, dedication, optimism, skill, and personal control. This research project explores such positive human characteristics by examining the lives of people behind social impact." During initial contact, participants were informed that participation, which would last approximately 1 Vi hours in total, would be rewarded with a $20 honorarium. Participation entailed (a) filling out a 30-minute survey online, and then (b) participating in an individual audio-recorded interview (60 minutes). The interview took place approximately one and a half weeks (Af = 10.0 days, SD = 5.9) after the participant completed the survey. The online survey made a second mention of the $20 honorarium, gave a brief description of the study, and asked participants to provide demographic information as well as the completion of a battery of measures. Upon completing the survey, participants were contacted so as to arrange an individual interview, wherein the SUIT was administered. The interview portion consisted of signing an informed consent 19 form (which included a third description of the $20 honorarium), then responding verbally to a standard set of 16 questions about his/her life and various aspects of his/her self-experience. After prompting the respondent to share an overview of his or her life story (two questions), the self-understanding portion of the interview was initiated with an open-ended scheme-general question ("How would you describe yourself?"). Next, 11 questions—each designed to tap from a specific scheme—were asked along with prompts to elicit explication. The questions were derived from the FDS (Hart & Fegley, 1995), when appropriate, and otherwise newly devised for the present study. Some examples of questions pertaining to specific schemes are: "Which of your activities are most important to you?" (active scheme); "Given that you change from year to year, how do you know that it's still always you?" (continuity scheme). Finally, a question to check if anything important was missed was asked before concluding the interview with a positively valenced question ("What do you like most about yourself?"). The complete SUIT is found in Appendix B. During the course of the interview, the interviewer jotted down a list of "stems" or phrases that summarize each of the answers emphasized by the participant. Upon completing the SUIT question-and-answer period, the audio-recorder was turned off, and the participant was given a 5-minute break while the interviewer typed the stems into a prepared computer spreadsheet. Next, the interviewer placed the computer in front of the participant and asked him/her to indicate the "importance, centrality, or salience" of each of the topics about which he/she had spoken. This was achieved in two steps. In the first step (Sheet 1; see Table 1), participants assigned a number between 0 and 100 to each stem in the rightmost column of Table 1, where a rating of 100 would indicate something of primary importance, 50 would indicate a aspect of moderate importance, and 0 would indicate an insignificant notion to whom he/she is. Before proceeding, the participant was explained that these evaluations should be made independently of one another and should be seen as rough estimates; he/she was made to understand that there would be a second screen forthcoming in which he/she would have a chance to revisit his/her numbers to get them just right. 20 Table 1. Centrality-Rating Paradigm, Sheet 1, for Participant # 75986. Question Response Stem Centrality rating (0-100) How would you describe yourself? social butterfly 60 bad at sustaining long-term and group friendships 70 split between individuality and need for community 70 Do you have a job and/or go to school? international relations student 60 tutoring 30 Which of your activities are most church 70 important to you? my student club2 70 Do you have any habits or unique ways of doing certain things? play with hair 40 Who are the most significant people and/or groups in your life? Roger3 75 family 90 What are the favorite things you have or 60 own? my voice my entire room 40 necklace 50 What's important to you in terms of your physical characteristics? being healthy 60 being small 80 good sense of balance 80 What are your major roles and responsibilities? role as roommate 50 What are the most important psychological aspects of who you are? Christian beliefs 95 strong powers of definition 65 Given that you change from year to year, how do you know it's still always you? still me because of community 70 How did you get to be the kind of person you are now? formed by God and beliefs 90 formed by coming to Vancouver 95 How do you know that you're unique or different from everybody else? unique set of life experiences 80 Once the participant was satisfied with his/her responses, the interviewer pressed a "hot key," which caused the indicated responses to be entered into Sheet 2 (see Figure 1). Sheet 2 was like Sheet 1 with two notable differences. First, the questions were omitted so that participants' focus would not be diverted from the stems. Second, a pie chart representing the individual's responses, the size of each slice corresponding to its weighting, appeared. Participants were instructed that the purpose of this second exercise was to form the most valid picture of who they are, as represented by a pie chart. That is to say that each slice of the pie should occupy a space that corresponds to how big a piece of the participant it feels. They were instructed further that the first task was different than this second one in that they rated each stem independently in the former; in the latter, they were to consider the concept represented by each stem in relation to their whole self. Finally, participants were informed that they were free to change the ratings as much or as little as they wished, and they could do so in one of two ways. First, they could 2 The actual name has been changed to protect the identity of the participant. 3 The actual name has been changed to protect the identity of the participant. 21 simply type a new centrality rating over the original and the pie chart would automatically adjust. Or second, they could make use of three preprogrammed hot keys. The function of the first and second hot keys was to increase and decrease, respectively, the size of a selected stem by 10%. Each time the key was pressed, .a further 10% of stretch/shrink occurred. Thus, participants were able to visually manipulate the size of each slice of their self-understanding pie. The third hot key caused the stems to be arranged in order from highest rating to lowest rating. Reminders of the commands for the three hot keys were visible on the computer screen. Participants were allowed to manipulate their pie chart as much or as little as they wished; the experimenter sat nearby in case any further explication of the procedure was necessary. Importanci' Response Stem Rating Christian beliefs 298 unique set of life experiences 142 split between individuality and need for community 124 formed by coming to Vancouver 115 family 90 church 85 bad at sustaining longterm and group friendships 70 my student club 70 ' social butterfly 60 my voice 60 strong powers of definition 44 Roger 42 being small 41 international relations student 41 play with hair 40 still me because of community 36 good sense of balance 34 necklace 23 role as roomate 23 " being healthy 23 my entire room 23 tutoring 13 formedby God and beliefs 0 being healthj-rck* as roomate-necklace' good sense of balnce-Kil me because of community play with hair international relations student bad at sustaining longterm and group friendships Christian beliefs unique set of life experiences split between individuaEtv and nt for communitj formed by coming to Vancouver Figure 1. Centrality-Rating Paradigm, Screen 2, for Participant # 75986. At the end of the interview, an experimental measure of moral action (honesty) involving slight deception was administered. In concluding the interview session, the interviewer informed the participant that he had to prepare for the next participant; and asked the participant to collect his/her honorarium from a female research assistant in a nearby office. When the participant approached the office, the research assistant appeared to mistakenly award the participant $30 for his or her involvement, rather than the (thrice) promised $20. It was assumed that the three instances wherein the $20 honorarium had been previously mentioned would make the dollar amount unambiguous to the participant. The participant's reaction was coded as a measure of honesty. This measure was developed by Bersoff (1999; and was also used by Derryberry & Thoma, 2005a, 2005b) and is discussed in detail below. Ethical considerations regarding whether or not participants ought to be debriefed on this specific aspect of the study are complex, the discussion of which is provided below in the context of fuller details of this measure. 22 Measures The interview portion of this study tapped the predictor measures (self-understanding), honesty was assessed following the interview (when the honorarium was paid), and the self-report questionnaires tapped the other aspects of moral functioning (altruistic behavior, ecological behaviors). Interview The overarching goal of this study is to investigate how self-understanding relates to moral behavior. Self-understanding is captured by the newly-developed SUIT. From the interview narrative, two measures were extracted: self-content and self-theory. The interview and measures are discussed below. Reliability. The SUIT narrative was coded for two themes: self-content and self-theory. The unit of analysis is called a "chunk," which entails a stem (or summary phrase of a single aspect of an individual's self-narrative) and its associated explication. Reliability is assessed in two steps: stem identification, then self-understanding (self-content or self-theory) coding. Recall that the stems were identified by the interviewer in real time, after which they were rated by the participant for centrality. To assess reliability of the identification of these stems (the first step in the coding process), a second rater later listened to the audio recordings of 25 randomly-selected interviews and generated a list of stems for each interview. These, in turn, were compared with the stems generated in real time, and judged (dichotomously) to be either conceptually similar or conceptually dissimilar. Reliability was found to be excellent, with a reliability coefficient of .90. This figure is similar to the 83% agreement found by Damon and Hart (1988). The second step involves either coding self-content or self-theory. First for self-theory, a given chunk was coded for the highest level that it exemplified, as delineated by the Scoring Manual for Self Understanding (SMSU; Damon, Hart, Pakula, & Shupin, 1988). Level was only coded if the participant elaborated on a stem or was prompted to do so (a rule defined by the SMSU). To assess reliability, a subset of 16 interviews was independently coded by a second rater. Reliability for level was calculated as a reliability coefficient, reflecting the number of agreements while correcting for chance agreements. Self-theory has been summarized previously in three ways. First, Damon and Hart (1988) used a simple mean. Reliability for mean level was found to be moderate, 66% agreement, K„ = .55 (correcting for chance agreement), which was noticeably poorer than that found by Damon and Hart (86% agreement, K n = -79). The difficulty in attaining reliable results in self-theory may be explained by an ambiguity in the operationalization of Level 4 in the SMSU, which will be discussed in detail later. This ambiguity would be less of an issue in the Damon and Hart (1988) study as their sample was of a much lower age group, for whom Level 4 self-statements were rare. A second summary statistic of self-theory is modal level. Structural developmental theories such as that of Damon and Hart (1988) argue that development is discontinuous; an individual remains at a stage for a lengthy period of time before making a quick transition to the next stage. This argument 23 justifies using modal scores as a valid representation of an individual's development. Reliability for modal level, however, was only fair, 53% agreement, K„ = .37. These figures are considerably poorer than those obtained by Hart (1992), who reported 87% agreement; thus analyses with this summary statistic are discontinued. A third summary statistic of self-theory is highest level present. Hart and Fegley (1995) found this metric to differentiate between caring moral adolescent exemplars and a comparison group. However, given a ceiling effect in the current study (every single participant evidenced at least one Level 4 statement), this metric is explored no further. Reliability for self-content for scheme coding was almost perfect with 86% agreement, with KN = .83. This is a marked improvement over the reliability found by Damon and Hart (1988; 71% agreement, K„ = .62). It is likely that this improvement in scheme reliability can be attributed to the methodological differences between the two interviews. Metrics. Data for each participant were then condensed into two sets of eight numbers. The first (unweighted) set assessed the unreflective responses to the interview (i.e. not using the centrality-weighting data). Thus, scheme emphasis for a given scheme (scheme_emphasisurmeighted,i) was its raw "hit count" of stems that received that scheme's code divided by the participant's total number of stems: #hitS: scheme _emphasishledi = ——-# stems For example, if a participant had 20 stems, 4 of which were coded as social, the individual's unweighted social scheme emphasis would be 0.20. This was done for each of the seven schemes. Unweighted Level (Levelunweightect) was calculated as a simple average of all of the level codes that their chunks received: ^ Level Level all mweish,ed # chunks coded The second (weighted) set of eight numbers assessed the reflective reconstruction of the person's self. Here, centrality percent scores ("%self; calculated by dividing the centrality rating of each stem by the sum of all centrality ratings for the participant) were used for weighting. Scheme emphasis for a given scheme (scheme _emphasisweighted,i) was calculated by summing the centrality percent scores of each of the stems that were coded for that scheme: scheme _emphasisweightedl = ^%self i For example, if a participant had three stems coded as active, each of which received centrality percent scores of 1.3%, 8.7%, and 5.0% of the self, respectively, the scheme emphasis for active would be the sum of these, or 15%. Weighted level was calculated by summing (over all chunks coded for level) the value of the level (i.e., 1, 2, 3, or 4) multiplied by the centrality percent score assigned to each stem: Levelled = X (Level x %self) all 24 A potential methodological concern arises here. Weighting self-theory may seem invalid as theory is coded from the explication of each stem, not the stem itself. In that participants ostensively rated the stems and not the explication, one could argue that the weightings do not apply to self-theory. However, in assigning centrality ratings, participants were instructed to consider the stem as well as all the ideas that it represents to him/her. Thus, it is argued that the centrality ratings correspond not to the stem but to the mental schema that it represents to the participant. Hence, weighting level is appropriate and justified. Questionnaire Measures Respondents filled out a battery of questionnaires online prior to the interview, which included: 1. Moral behavior: • Self-Reported Altruism Scale (SRA; Rushton, Chrisjohn, & Fekken, 1981); • an adapted version of the General Ecological Behavior Scale (GEB-A; Kaiser & Wilson, 2000); 2. Moral values: • New Ecological Paradigm Scale (NEP; Dunlap, Van Lier, Mertig, & Jones, 2000); • Materialism Scale (MS; Richins & Dawson, 1992); 3. The tendency to present oneself in a socially desirable way: • Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1991); 4. Contentment with one's own life: • Satisfaction with Life Scales (SWLS; Diener, Emmons, Larsen & Griffin, 1985). The questionnaires were chosen to best measure a range of moral actions, both in terms of interpersonal morality (prosocial behavior) as well as ecologically impactful behavior. It was anticipated that these two self-report measures of moral behavior would converge with the actuarial measure of honesty. For the purpose of cross-validating the behavioral measures, the two measures of moral values were included. The BIDR was also included to assess and be able to partial out (if necessary) the effects of cognitive distortion and exaggerated representation. Finally, the SWLS was included so as to tap another domain of human flourishing, which will be discussed later in tangent to the principal investigation. In order to make the study's focus less obvious, some measures were intermingled online. Prosocial Behavior. Prosocial behavior was operationalized as the Self-Reported Altruism scale (SRA; Rushton et al., 1981; see Appendix D). The questionnaire asks respondents to report the frequency with which they have carried out 20 altruistic acts on a 5-point Likert scale: 1 (never), 2 (rarely), 3 (sometimes), 4 (often), and 5 (very often). The measure, was chosen because of its focus on actual prosocial behavior, rather than subjective experience (e.g., attitudes, emotions). The scale is reported to have good internal consistency (Cronbach as = .78 to .87; Rushton et al., 1981). One item was dropped in the present study to avoid any possibility of priming participants for the behavioral measure of honesty: "I have pointed out a clerk's error (in a bank, at the supermarket, etc.) in undercharging me for an item." Still, the scale was found to have good internal consistency in this study (Cronbach a = .83). 25 Ecological Behavior. While moral psychology has historically focused primarily on interpersonal issues, there are good reasons for including ecological interactions in the moral domain (for a relevant discussion of this argument see Kahn, 2006). The well-being of the planet is of moral concern for both teleological as well as deontological reasons. Teleologically, the planet holds moral value for its ability to sustain beings with inherent moral value (i.e., people); and deontologically, some may argue that the planet holds inherent moral value itself—that nature is important independent of humanity. Either way, it seems that ecological issues are of moral concern. Ecological behavior is notoriously difficult to measure out of historical, geographical, and political context (Kaiser, 1998). Opportunities for, and constraints on, ecologically-relevant behaviors differ significantly from one context to another. For example, recycling increases dramatically when recycling containers become widely available. Nonetheless, Kaiser and Wilson (2000) succeeded in creating a robust, cross-culturally applicable measure of ecological behavior, the General Ecological Behavior (GEB) scale, which assesses 43 ecological and 8 prosocial actions. However, Kaiser and Wilson (2000) found that some behaviors were of differing difficulty depending on geographic location (Switzerland vs. California). In the present study, 17 items relating to ecological behavior were adapted from GEB scale so as to be applicable to an urban Western Canadian population. These items were augmented by four novel items, which were particularly relevant to the population of interest in this study. Hence, the modified GEB scale (GEB-A; see Appendix E) used in this study is comprised of 21 items which have been reworded from the original GEB so as to reflect actual frequency of behaviors rather than yes/no or 5-point agree-to-disagree type endorsements. The GEB-A scale asks respondents to rate the frequency in which they engage in 21 ecological actions using a 5-point Likert scale: 1 {never), 2 (rarely), 3 (sometimes), 4 (often), and 5 (very often). The measure was found to have good internal consistency (Cronbach a = .76) in the present study. Ecological Attitudes. In that the GEB-A is an altered version of the GEB, it will be validated by examining its association to a particularly well-validated and accepted measure in the field of environmental psychology, the New Ecological Paradigm (NEP; Dunlap et al., 2000; see Appendix F). The NEP assess attitudes towards the environment by asking respondents to rate, using a 5-point Likert scale, the extent to which they agree with 15 statements about the relationship between humans and the environment: 1 (strongly disagree), 2 (mildly disagree), 3 (unsure), 4 (mildly agree), and 5 (strongly agree). The measure has good internal consistency (Cronbach a = .83; a = .88 in the present study) and has been shown to relate to a wide range of ecological behaviors and attitudes (Dunlap et al., 2000). Materialism. While prioritizing and valuing material possessions may appear to fall outside of the moral domain, there are compelling arguments to suggest that this is not the case (see Belk, 1983, and Fournier & Richins, 1991). One argument is indirect, and relies on the notion of life motives and outcomes being ipsative: the valuation of material goods usually comes at the expense of valuing other 26 pursuits. However, materialism not only reflects an inferior value but it is also interpersonally and intrapsychically maladaptive and harmful; that is, it is immoral not simply because it precludes more virtuous pursuits but because it is inherently wrong when taken to excess. Materialism is operationalized as the Materialism Scale (MS; Richins & Dawson, 1992; see Appendix G). The measure asks respondents to report, using a 5-point Likert scale, the extent to which they agree with 18 statements concerning the importance material goods play in one's life: 1 (strongly disagree), 2 (mildly disagree), 3 (unsure), 4 (mildly agree), and 5 (strongly agree). The measure has good internal consistency (Cronbach as = .80 to .88; a = .90 in the present study) and does not relate to socially desirable responding (Richins & Dawson, 1992)4. It has also been shown to relate to: value orientation (positively to "financial security" and negatively to "warm relationships with others" and "a sense of accomplishment"), self-centeredness (positively), and satisfaction with life (negatively; Richins & Dawson, 1992). Socially Desirable Responding. With the inherent difficulty in measuring a socially desirable construct such as moral behavior, a control for response bias was included. The Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1991; see Appendix H) asks respondents to rate, on a 7-point Likert scale, the truthfulness of 40 self-statements relating to socially desirable acts that are difficult to perform consistently: 1 (not true) to 7 (very true). In the present study, one item was dropped so as to avoid priming participants directly for the behavioral measure of honesty: "I have received too much change from a salesperson without telling him or her." Interestingly, the inclusion of this item on a measure of socially desirable responding is an indication of how ubiquitous it is in everyday life. The BIDR has two subscales, self-deception and impression management, each comprised of 20 questions. The BIDR has good internal consistency with Cronbach a = .70 and .80 for the self-deception and impression management subscales, respectively, and Cronbach a = .81 for the entire scale (all figures from the present study.) Scoring instructions specify that responses should be converted to a dichotomous scale, with rescoring as follows: socially desirable responses (6 or 7) or honest responses (1 to 5). In this way, scores for the total scale range from 0 to 39 (from 0 to 20 for the impression management subscale and from 0 to 19 for the self-deception subscale). Satisfaction with Life. In that the De Mello (1990) referred primarily to the relationship between the I- and me-selves, on the one hand, and suffering on the other, a measure of subjective wellbeing was included to explore this notion tangentially. The Satisfaction with Life Scales (SWLS; Diener, Emmons, Larsen & Griffin, 1985) asks respondents to rate, on a 7-point Likert scale the degree to which they agree or disagree with five statements that suggest that one's life is a satisfactory one: 1 (strongly disagree), 2 (disagree), 3 (slightly disagree), 4 (neither agree nor disagree), 5 (slightly agree), 6 (agree), and 7 (strongly agree). The SWLS has good internal consistency with Cronbach a = .84 in the present study. 4 Contrary to the Richins and Dawson (1992) study, Materialism was found to be related to the BIDR, r(91) = -.30, p = .002. Thus, the effect of socially desirable responding will be considered when later revisited. 27 The SWLS has been found to be positively related to other measures of subjective wellbeing, but also positively related to socially desirable responding (Diener et al., 1985). Behavioral Measure In order to have a more veridical measure of moral action and one less susceptible to socially desirable responding than self-report accounts, a measure of honest behavior which was developed by Bersoff (1999; and used by Derryberry & Thoma, 2005a, 2005b) was implemented after completion of the SUIT. The procedure involves the (seemingly accidental) overpayment of the respondent. After completing the centrality-rating procedure, participants were informed that their participation was complete and that they were now entitled to their remuneration. The interviewer, however, reported that he must prepare for the next participant. Thus, participants were directed to claim their $20 honorarium from a research assistant in a nearby office. Upon arriving there, participants were greeted by a somewhat distracted (ostensively) assistant but assured that this is indeed where they could claim their remuneration. Only, the research assistant counted out three $10 notes (not the $20 indicated thrice previously), and asked participants to sign a receipt for that amount. If participants accepted the $30 and signed a receipt for that amount, they were thanked for their participation. If, on the other hand, participants corrected the assistant, the assistant acted confused about the correct amount, mentioned something about being an assistant for two studies simultaneously, and thanked the participant for correcting her error. Hence, these participants received $20 for their participation, signed a receipt for that amount, and were thanked for their involvement. Those participants that kept the $30 were contacted by email 1 week later, at which time they were informed that there had been a bookkeeping error and that some participants had been overpaid. Further, they were asked how much their received as payment. Regardless of their answer, they were thanked for their response. Before concluding communication with these participants, a final statement was added so as to strike a balance between competing ethical interests (protections from harm vs. right to be debriefed; discussed later). Participants were informed that given the complex nature of the study, they had not been filled in on all of the details. They were told that if they had any future questions, that they should feel free to contact the principal investigator; they were also provided a contact telephone number. Scoring of honesty was as follows (on a 5-point scale): refusing payment altogether or correcting the research assistant's original "error" = 5; accepting the $30 but then returning it within 1 week without being prompted by the experimenter = 4; accepting the $30 but when contacted 1 week later, admitting the overpayment and offering to return $10 = 3; accepting the $30 but when contacted 1 week later, admitting the overpayment without offering to return $10 = 2; accepting the $30 and then later either claiming to have received only $20 or not replying to the experimenter's inquiry at all = 1. The ethical considerations relating to debriefing participants regarding this part of the study warrant discussion. Situations in which one is apparently overpaid or undercharged are common occurrences. On the other hand, being informed ("debriefed") of a moral failure does have the potential to 28 cause psychological and social distress (guilt or embarrassment). In introducing this procedure, Bersoff (1999) discussed how to minimize or eliminate psychological damage by striking a workable compromise between two opposing ethical considerations. On one hand, participants certainly have the right to be fully informed regarding all experimental procedures; but on the other, participation in the study would have the least potential to cause stress and discomfort if there was no debriefing. Participants who pointed out the overpayment of course would have nothing to feel upset about regardless of whether they were debriefed. More important, participants who kept the overpayment also should not experience much if arty discomfort if there was no debriefing because it is assumed that they will have conceptualized their act in terms with which they feel comfortable. But if those who kept the money were fully debriefed, it would be much harder for them to view their action in a benign manner. They would likely feel that they had, in a sense, flunked.some kind of moral test. (p. 32) Should participants that kept the overpayment happen to encounter another participant that corrected the researcher's error and was then debriefed, the participant who kept the overpayment may inadvertently be debriefed by the one who refused the $30, which may result in unnecessary distress. On the other hand, if a participant should experience distress about having been dishonest (without being prompted by the experimenter), it could be in the best interest of that participant to be fully debriefed so that he/she could see that she was tricked into keeping the extra money. In doing so, the participant could then reason that his/her personal responsibility is dimished. Hence, the ethical right for participants to be fully informed about experimental manipulations is in conflict with the right of participants to be protected from harm. Upon considering these matters carefully, the Behavioral Research Ethics Board required that participants be provided with an option of "self-selecting" themselves for debriefing. For this reason, a clause at the end of the email correspondence was added, wherein participants were encouraged to call back if they had any further questions. Of the 42 participants who accepted $30, none "self-selected" for debriefing. 29 Chapter III - Results and Discussion This chapter focuses on mapping the moral self by exploring the relationship between self-understanding and moral behavior. First, a singular metric of moral action will be established for subsequent analyses. Second, the relationship between each of self-theory and self-content with moral action will be explored. Third, inadequacies in the coding manual will be noted. A solution will be proposed, including a new coding paradigm. Fourth, the effectiveness of this new coding paradigm in predicting moral action will be explored. Fifth, the various models will be compared. And sixth, discussion will focus on the meaning and implications of some of the key findings. Moral Behavior In moving towards exploring the relationship between self-understanding and moral behavior, some work remains so as to achieve a single metric of the latter. After the nature of the distributions of the three measures of moral behavior are examined, a single, aggregate variable of moral behavior will be developed. Epstein (1983) advocates for the aggregation of multiple behavioral measures of the same construct as an effective means of controlling situational influences of singular behaviors, and thus achieving a more valid and reliable assessment of cross-situational and cross-temporal behavioral dispositions. First, the Self-Reported Altruism Scale (SRA; Rushton et al., 1981) is a measure of prosocial behavior in which scores can range from 19 to 95, with higher scores corresponding to more prosocial behavior. The present sample's scores ranged from 33 to 81, with a mean of 60.22 (SD = 9.30), reflecting responses at the midpoint of the scale. This sample reports being more altruistic than undergraduates from the University of Western Ontario (M= 55.6, SD = 10.9), /(708) = 4.47,/? < .001, d = .63. This difference probably reflects the generally agentic and altruistic nature of people active in student clubs, such as those in the present study. Second, the General Ecological Behavior Scale (GEB-A; adapted from Kaiser & Wilson, 2000) is a self-report metric of the extent to which one engages in ecologically sound behavior. The range of possible scores is 21 to 105 with higher scores pertaining to behavior that is more ecologically sound. Ranging from 44 to 97, the present sample's scores had a mean of 79.3 (SD = 9.4), reflecting an endorsement of often engaging in pro-ecological behaviors. The validity of this altered scale is substantiated by its strong correlation with the well-validated NEP, r(97) = .52, p < .001. And third, the honesty measure used in this study (adapted from Bersoff, 1999) is an actual behavioral measure of honesty in the context of overpayment for participation. The range of possible scores ranged from 1 to 5, with higher scores reflecting more honest behavior. As can be seen in Figure 2 below, the modal score in honesty was 5 (or most honest). Indeed, the mean of this sample was 3.7 (SD = 1.5) reflecting highly honest behavior. The distribution of scores was somewhat bimodal, with complete 30 honesty being most common, followed by accepting the overpayment but acknowledging as such later when prompted. 60 Honesty S c o r e Figure 2. Histogram of Honesty Scores As one of the goals of this study is to measure the relationship between self-understanding and moral behavior, it would be advantageous to combine the three measures of moral behavior into a single, aggregated variable. Diminishing the number of outcome variables of a study is beneficial for clarity, statistical power (Cohen, 1990), and validity (Epstein, 1983).. Hence, there are good reasons why aggregating a large number of measures of the same construct into a single outcome variable is beneficial5. Before aggregating these three variables, it is worthwhile to explore how they relate to one another. Given that each was included so as to measure various aspects of the same underlying construct (i.e.,.moral action), it was hypothesized that the three measures would be significantly related to the others. As can be seen in Table 2, SRA is positively related to GEB-A. However, Honesty is related to neither SRA nor GEB-A. It remains unclear as to why this is the case. Table 2: Correlation Table of Measures of Moral Behavior. Self-Reported Altruism General Ecological Behavior Honesty r(97) = -.04 r(97) = .08 (p = Jl) (P = M) Self-Reported Altruism r(97) = .39 Gx.oooi ) Note: a = .017 after performing a Bonferroni correction (n = 3). One possibility is that these three measures are not accessing the same construct. This proposition is supported by domain theory (e.g., Turiel, 1983; Nucci, 2001) in which convention and 5 During the course of administering interviews, it became apparent that the conscientiousness of the behavior of participants varied substantially. An ad-hoc measure of conscientious behavior was explored (see Appendix I) but later discarded. 31 morality are considered to be mutually independent constructs. It is possible that the honesty metric taps a construct that lies within the moral domain whereas the altruism and ecological behavior measures tap a construct outside of the moral domain (or vice versa). This seems to be an untenable proposition as, for one, it is unclear as to which (of honesty on the one hand or SRA and GEB-A on the other) lies on which side of the boundary of the moral domain. A second possibility is that the different modes of measurement (i.e., self-report vs. singular behavioral observation) leave the former prone to the confound of social desirability. According to this notion, when the effect of social desirability is removed, the relationships between GEB-A and Honesty as well as SRA and Honesty should strengthen. These hypotheses were tested by regressing GEB-A on Honesty. When partialling out the effect of socially-desirable responding, the relationship between GEB-A and Honesty remains non-significant, ^ = .03. The relationship between SRA and Honesty remains similarly weak when performing an analogous regression (8 = -.06). Therefore, social desirability does not account for the noted, puzzling independence. A third possibility is that the honesty measure taps an aspect of moral functioning that the SRA and the GEB-A do not. While this hypothesis cannot be tested with the present data, it will be taken as a premise. While it is possible that the three measures of moral behavior do not tap the same construct, they are aggregated nonetheless in the present study. To do so, each metric is converted to a z-score, and the three are summed: MoralBeh = zSRA + zGEB + zH0NESTY As a validation check, the relationship between MoralBeh and two moral values (materialism and ecological attitudes) are explored. Attitudes and values are subjectively-held stances that pervade all aspects of life, directing the individual towards particular ends. For this reason, self-report measures of morally-relevant attitudes (ecological) and values (materialism) ought to be predictive of actual moral behavior. And indeed, MoralBeh relates positively to ecological attitudes as tapped by the NEP, r(97) = 34, p = .001; and negatively to the immoral values of MS, r(97) = -.32, p = .001. Self-Understanding The audio-recorded segment of the interview lasted on average 32 min (SD = 9 min), during which time an average of 24.0 stems (SD = 4.2 stems) were generated by the interviewer in the course of the standard 14 questions that elicited self-understanding6. After completion of the audio-recorded interview, participants were given an opportunity to adjust the size of the slices of their "self-understanding pie." To do so, a participant should have reflected upon which aspects of him- or herself are more important and which are less important. As this is a new methodology, it is worth briefly considering whether or not this may have actually happened. If participants had weighted all of their pieces evenly, they would have each represented 4.2% of the pie. 6 The first two of 16 questions elicited background information about the individual's life, and were thus not coded. 32 However, the average largest piece occupied 11.7% (SD = 5.6%) of the pie. In that participants almost tripled the weighting of their most important chunk, reflection of some sort seems reasonably implicated. It remains and open question, however, as to whether or not this reflection is in some way more valid than unreflective views. In this study, this will be assessed by exploring the predictive validity of the different paradigms. SUIT Coding Each interview was coded for both self-content and self-theory. This was done by coding the interview spreadsheet file while listening to the audio recording. Each stem was coded simultaneously for content and theory. When there was insufficient information upon which to assign a self-theory code, a code was not assigned. However, a decision rule was created whereby a self-content (scheme) code was necessarily assigned for each chunk. This was necessary for direct and fair between-subject comparisons in analysis. Two ambiguities in the manual were noted in the course of coding. Each of these potentially threatens the validity of the paradigm. First, "social characteristics" are coded in the social scheme (Level 3) whereas "psychological attributes that influence one's social interactions" (Damon et al., 1988, p. 33) are coded in the psychological scheme (Level 3). While this may sound like an intuitive distinction, it is not a practical one. Socio-personality characteristics are psychological attributes that get played out in social settings. For example, being "friendly" is both a social as well as a psychological characteristic. It is meaningless to have these sorts of characteristics outside of social contexts; but it is equally outlandish to consider friendliness outside of the intrapsychic realm. The SUI manual rules force the coder to choose between assigning these two schemes when the authentic desire is to assign both. This issue will be revisited later; but, in the meantime, a tentative decision-rule was adopted that socio-personality characteristics are coded in the psychological and not the social scheme. Second, the distinctions between the different levels of self-theory were unclear. Recall that, through development, individuals come to reason in increasingly sophisticated ways about the self. For Damon and Hart's model, the four levels represent a sequence of progressively more sophisticated justifications for one's self-statements. At Level 1, categorical identifications have no further meaning than the label itself. A Level 2 (first present at about age 9) response entails defining oneself in comparison or competition with other traits, people, or norms. In Level 3 (age 11), the self is understood in terms of interpersonal implications where the emphasis shifts to belonging. And, in Level 4 (first appearing about age 13), the adolescent uses systematic beliefs and life plans to justify his or her self-statements. Here, the self is understood relative to personal or moral evaluations. Damon and Hart (1988) state explicitly that these are not mutually exclusive categories; rather, higher levels subsume the levels below them. However, an ambiguity exists pertaining to the Level 4 (systematic beliefs and plans). In particular, the degree to which a self-statement about an activity (Level 2) must be systematic or planful (Level 4) remains unclear and highly subjective. For example, if an individual speaks of plans to see a 33 movie next week and go to graduate school next year, how much more he/she must say to have the self-statement qualify as Level 4 remained a difficult and subjective decision. Reflecting this ambiguity, in the subsample of interviews coded for interrater reliability, the number of instances where one rater coded Level 4 and the other coded Level 2 (26 cases) was virtually the same as when they both coded Level 2 (27 cases), and about one quarter as many 4-4 agreements (96). It was similarly difficult to distinguish between social relationships that were seen in terms of interpersonal implications (Level 3) and systematic and planful terms (Level 4). There were a substantial number of 3-4 and 4-3 disagreements between raters (53 cases) compared to the number of 3-3 agreements (91 cases) and 4-4 agreements (96). Self-Theory Results Forty-seven interviews were randomly selected for coding. Tallies for each level across all participants are shown in Figure 3. A whole 49% of all level codings were for Level 4. A ceiling effect is noted: the mean level across participants before weighting was 3.25 (SD = .31) and 3.36 (SD = .31) after. Thus, most self-statements qualified as either Level 3 or Level 4. Note the restriction of range, as evidenced by standard deviations of just one-third of a level. These findings are generally consistent with those of Hart (1992), who found that the mean level for men between the ages of 20 and 22 was 3.0 (SD = .63) and 3.4 (SD =.51) for men ages 24-26. It is interesting to note that the weighted level is greater than the unweighted level, t(46) = 4.93, p < .001, d = .35. That is to say that participants tended to weight chunks that they had justified with more sophisticated self-statements as more important to who they are than those that were justified with less sophisticated self-statements. Mean level was not association with age for either of the weighted or unweighted paradigms, r(45) = .19 and .21,/? = .20 and .15, respectively. Hart (1992) found somewhat puzzling developmental trends in this age range. Between the ages of 10 and 36, there was a strong positive relationship between mean level and age, r = .52. However, a negative trend was found for the older men from the same sample (ages 24-36), r = -.34. This trend suggests that the relationship between level and age cannot be adequately described with a linear model. 600 500 400 t 300 200 100 0 Level 1 Level 2 Level 3 Level 4 Figure 3. Histogram of Self-Theory Levels 34 Issues pertaining to Damon and Hart's self-theory model have been noted previously. Of particular concern is the conceptualization of the highest stage of development, Level 4. What lies beyond socially-bounded justifications of self-statements, however, remains a topic of debate. D. K. Lapsley (personal correspondence, April 12, 2006) elucidates the vision for Level 4: At the highest stages of self-understanding we [ought to] see movement in two directions: externally—a systems perspective that allows us to see how our action affects the widest possible polity; and internally, in depth, as we come to discern and understand the motives behind our actions. Both of these Level 4 accomplishments—the systems and in-depth perspective—clearly have moral implications. We see how our actions affect others (society), which requires a systems perspective; and we see how our actions spring from certain (vile, praiseworthy) motivations, which requires the in-depth perspective. Our actions have implications for the sort of self we are... I think this is what Damon and Hart was [sic] after. I argue that while Damon and Hart's vision for Level 4 was likely the concert between the systems and in-depth perspectives, they failed to operationalize this phenomenon in their coding manual. To qualify for Level 4, a self-statement must merely be systematic or planful (e.g., reference to the type of lifestyle an individual hopes to achieve). Such a statement satisfies but one half of the intended concert—the systems perspective. What a statement is not required to state to be classified as Level 4 is reference the in-depth perspective; that is, a statement need not delve into the root motivation or intrapsychic action behind the identification. Thus, as operationalized, Level 4 accepts many statements that are perhaps more superficial than those that would satisfy the systems-in-depth concert. It remains a matter for future investigation to tighten the criteria for Level 4. Despite the flagged issues with self-theory, it is still worth investigating the relationship between mean level and moral behavior. Recall the Hart and Fegley (1995) finding that half of the adolescent caring exemplars (mean age 15.5 years) evidenced some Level 4 while only 7% of the comparison teens did. Recall also that Derryberry and Thoma (2005b) failed to find any associations between modal level of self-theory and moral behavior in a young adult population (mean age 20.6 years). Similar to the latter (null) finding, no evidence for an association between level and moral behavior was found in the present study: r(45) = .13, p = .46 for the unweighted values; r(45) = .08, p = .59 using weighted values. It may be the case that level is only predictive of moral action for adolescents. Alternatively, recent findings may call into question whether Hart and Fegley's (1995) finding—which potentially rests on as few as seven judgments—-was, in fact, spurious. In sum, self-theory failed to predict moral behavior; its predictive validity is called in question on both conceptual as well as empirical grounds. Self-Content Results Recall that the nature of the SUI precludes direct comparison of scheme emphasis among any of the I-self schemes (agency, continuity, distinctness). With the SUIT paradigm, it is now possible to make such comparisons. Figure 4 (black bars) presents the mean percent of self-statements7 that correspond to each of the seven schemes, relative to the individual's total number of self-statements. Thus, if a 7 Derived from a subset of 47 randomly-selected participants. 35 participant had 20 chunks, and 4 of which were coded for the social scheme, that would constitute a score of 20% in that scheme for that individual. Occupying 28% of the self, the psychological scheme dominates, followed closely by active. The social and physical schemes each comprise 15% of the self. And finally, the three I-self schemes comprise 5%-6% each. The crossed bars of Figure 4 show the weighted averages (from the centrality-rating paradigm) for each scheme. Through the course of the weighting paradigm, participants de-emphasized physical, t(46) = 7.14,/? < .0001, d= .99; continuity, ?(46) = 3.86,p = .0004, d= .54; and distinctness, t{46) = 3.68,/; = .0006, d= .43, self-statements. Meanwhile, participants tended to place significantly more weight on psychological ones, /(46) = 4.86,/? <.00018, d= .62. 40% , 1 s c h e m e Figure 4. Scheme Emphasis in SUIT Coding Damon and Hart (1988) report scheme emphasis data for the four me-self schemes. These data were collected from the oldest cohort in their study, namely 16-year-olds. It is important to keep in mind the 6-year age difference between that sample and the current one. Nonetheless, the same general pattern of scheme emphasis is found, albeit with some minor differences. Most notably, the rank order of social and active schemes are reversed, with the active scheme being emphasized more in the present study than in Damon and Hart (1988). This difference may be attributable to the ambiguity in coding socio-personality self-statements (as noted above). While it is of general interest to note the normative trends in emphasizing the various schemes, the predominant interest of this study is in explaining individual differences. It was hypothesized that 8 As determined relative to a Bonferroni-corrected a = .007. 36 individual differences in scheme emphasis will relate to moral behavior. In particular, the De Mello Hypothesis states that emphasis on the I-self will be positively, and emphasis on the me-self will be negatively, related to moral behavior. In that the I-self and me-self scores are ipsative (necessarily summing to 1), this question can be examined by simply examining the correlation between the I-self schemes with MoralBeh. Contrary to the De Mello Hypothesis, emphasis on the (unweighted) I-self is unrelated to MoralBeh, r(45) = .10,/? = .53. There are at least two reasonable explanations for these null findings. First, it is possible that methodological weaknesses prevented detecting a real trend. And second, it is possible that the De Mello Hypothesis is false. The former of these explanations will be closely examined next, and if further methodological inquiries yield similarly null results, the latter conclusion will be tentatively accepted. But first, it is worth pausing to investigate the individual relationships between each of the schemes and moral behavior. It is quite possible that the De Mello Hypothesis is correct in general—that scheme emphasis relates to moral behavior—but not in specific. An exploratory analysis suggests that this is also not the case. Zero-order correlations between each of the seven schemes and moral behavior were uniformly weak, regardless whether using the unweighted paradigm (-.13 < r(A5) < .15,/?s > .30) or the weighted one (-.13 < r(45) < .15,/?s > .31). Thus, the SUIT paradigm as presented failed to reveal any relationships between moral behavior and scheme emphasis. Finally, consider Derryberry and Thoma's (2005b) finding that various personological variables predicted moral functioning differentially. It is quite possible that self-content and self-theory predict each of the measures of moral action differently. Thus, an exploratory analysis is conducted where the zero-order correlations between each of the ten metrics of self-understanding from the weighted paradigm—seven schemes, me-self, I-self, and mean level—and each of the three measures of moral action are examined. As can be seen in Table 3, no relationship approached significance (all ps > .06 with a Bonferroni-corrected a = .002), further pointing to the lack of profitability in using this paradigm of self-understanding as a predictor of moral action. Table 3. Correlations between Self-Understanding (SMSU) and Measures of Moral Behavior Physical Active Social Psychological Agency Cont. Distinct. Me- • self I-self Mean Level H O N 0.08 -0.06 0.02 -0.04 0.10 0.03 -0.07 -0.07 0.05 0.02 SRA 0.01 0.07 -0,02 -0.11 -0.13 0.26 0.15 -0.09 0.11 -0.07 G E B -0.13 0.28 -0.25 -0.04 0.20 -0.03 -0.17 -0.05 0.04 0.20 However, an important challenge to the SUI coding rules was noted earlier. Before abandoning self-understanding content as a predictor of moral behavior, let us stop to explore this challenge in detail. 37 Reconceptualizing Content When attempting to detect a trend within the context of the vast complexity of real-life narratives, it is of the utmost importance that the methodology be properly tuned to the question of interest, and that the instrument be as sensitive as possible. A null result obtained by an untuned and insensitive instrument does little to inform the question of interest. The devil of this discussion is in the details, the topic to which we now turn. Content and Level Confounding As it exists, the SMSU (Damon et al., 1988) requires a coder to assign to a given chunk one and only one scheme code. For example, the as-mentioned socio-personality characteristics are to be coded as either social or psychological, but not both. It is debatable whether forcing a given chunk into one particular scheme is an authentic way of representing the content of that self-statement. Further, in the course of explaining and contextualizing self-identifications, people inevitably speak of content from many different schemes. Consider the following example (participant # 30078; the stem for this example was "competing and training - triathlon"): WHICH OF YOUR ACTIVITIES ARE MOST IMPORTANT TO YOU? It would be a combination of competing and training with my friends from the triathlon community as well as going out and having a good time with friends and family, be it dinner or a party or whatever. Those two things are kind of the most important. WHY IS COMPETING AND TRAINING FOR TRIATHLON SO IMPORTANT TO YOU? I think it's kind of habitual. Since my first race, I so thoroughly enjoyed it that I wanted to keep going. After doing I guess my fourth year now, I've kind of made the jump into the elite level. It's still really enjoyable but also kind of seems almost like a job. I've had to do a lot of training. But it's important because it helps me achieve goals in my life—set out and achieve things. In that sense it gives me a bit more structure to my life, and sets me on another , path of being positive... In this example, the respondent refers to a variety of asocial activities (active scheme) as well as social activities (social scheme). The participant also mentions his/her sentiments towards goals relating to the activities (psychological scheme) as well as a reflection on how these activities play into the formation of the self (agency scheme). In the existing SUI manual, this chunk is scored simply as the active scheme and as Level 4. I contend that these codes do violence to the rich content present in this self-statement. In addition, I argiie that self-content and self-theory are conflated. Level 4 theory implicates systematic beliefs and plans (which is conflated with the psychological scheme); Level 2 and 3 each make reference to the self in relation to others (social scheme). While Damon and Hart (1988) find no empirical evidence for such a conflation, this is not the case in the present study. To explore the relationship between scheme and level, we begin by noticing that some of the simple zero-order correlations between mean weighted level and weighted scheme emphasis as examined from the SUIT coding data are significant: Physical-Level, r(45) = .00, p = .99; Active-Level, r(45) = -.09, p = .56; but Social-Level, r(45) = -.45, p = .002, and Psychological-Level, r(45) = .44, p = .002. After making a Bonferroni correction, we see that the social scheme is negatively and the psychological scheme is positively related to level. Given that the vast majority (81 %) of stems were coded as either 38 Level 3 or 4—levels that are allegedly conflated with the social and psychological schemes—it is understandable that the relationships between level and scheme would only be significant for these two schemes. Note also the direction of the relationships. Social is negatively related to level while psychological is positively related to level. This makes sense and supports the asserted conflation. Damon and Hart (1988) acknowledge that it would appear level and scheme appear to be conceptually conflated but fail to find empirical support for this assertion with their young sample. The data presented here appear to offer empirical support for such a conflation. Introducing the Scoring Manual for Self-Understanding 2 To address these issues with the SMSU, I propose a new coding paradigm—one which sacks the rule that chunks are coded for one and only one scheme. The new coding paradigm—the Scoring Manual for Self-Understanding 2 (SMSU2; see Appendix J)—is grounded on the premise that content of self-narrative is rich and complex, and that the most authentic way of representing it is by allowing for that complexity to show itself. In doing so, the SMSU2 coding focuses solely on self-content. In the SMSU2, coding for each scheme is performed independently of coding in each of the other schemes. Thus, a given chunk can be coded as a "hit" for anywhere between one and seven schemes. Thus, a total of 16,842 independent hit/miss judgments were made across all 99 participants and all seven schemes in the present study. An average of 2.8 schemes were coded as hits for each stem. The SMSU2 gives a general description of each scheme, provides a list of examples of self-statements, and lists off a number of notes and distinctions. Importantly, the manual gives explicit instructions on some of the more ambiguous self-statements. For example, it states that socio-personality characteristics are coded as hits for both the social and psychological schemes. Agency. A critical conceptual point of divergence between the SMSU2 from the previous paradigm is in the definition of agency. To Damon and Hart (1988), agency in the context of self-understanding referenced the formation of the self. The entity responsible for this formation is an open matter; that is, any of one's upbringing, family, surrounding, activities, will, and so forth can be implicated as responsible for the change. In describing the moral personality, Blasi (1983) emphasized a sense of personal responsibility for moral outcomes. This is a crucial aspect of the moral personality— the individual's own volitional choices. For this reason, the SMSU2 definition of agency turns Damon and Hart's (1988) definition "on its head." Instead of implicating something in changing the self, the SMSU2's definition of agency holds the self as the alterer of something. The SMSU2's definition of agency is stated as follows: Implicitly or explicitly seeing oneself as an agent: Volitional action of the self that affects actual change of any kind, be it change of the physical, active, social, or intrapsychic nature. This definition is more expansive than and diverges from the notion on agency concerning the formation or existence of the self. That is, this definition does not require change to affect oneself but it does require the I-self to be the source of the change. There must be explicit (or strongly implicit) indication that the volitional action of the self causes the change. 39 The definition is accompanied by a 5-point checklist for coding. A self-statement must meet all five criteria to qualify as a hit for agency: o Is there volition for change? o Is it explicit or strongly implicit? o Does it come from the self? o Does it necessarily cause change (and not the other way around)? o Is the change actual as not only desired? Conceptual and empirical considerations relating to the multitude of definitions of agency are discussed later. Analysis. Interrater reliability was assessed by a second rater independently coding 25 randomly-selected interviews. Agreement was almost perfect with percent agreement for each of the seven schemes in excess of 92%, K > .83. In considering the sum (weighted or unweighted) totals for each scheme, it is apparent that direct comparisons cannot be made with the SMSU data without a correction factor. While the SMSU requires that a single scheme be coded per chunk, the SMSU2 allows for multiple schemes to be coded per chunk (in this sample, there were 2.83, SD = .33). Hence, the sum total of the seven SMSU schemes add up to 100% (SD = 0%) whereas the sum total of the SMSU2 schemes add up to 283% (SD = 33%). To correct for the multiple codes, as well as to correct for the variability in summed scheme emphasis in the SMSU2 data, the weight of each scheme is corrected by dividing by the average number of schemes coded per chunk for that participant. In doing so, the sum total of the SMSU2 schemes will now add up to 100% (SD = 0%), identical to the SMSU data. In making this correction, the SMSU and SMSU2 data become directly comparable. The percent of the self comprising each scheme is presented in Figure 5 below. For reference, the emphasis derived from the SMSU paradigm are presented as black (unweighted) and hatched (weighted) bars'. The SMSU2 data are presented as grey bars (unweighted) and dotted bars (weighted). Visual inspection suggest that the overall pattern of scheme emphasis is similar between the two scoring paradigms. To test this observation more rigorously, a 2 (scoring paradigm) x 2 (weighting) MANOVA was performed with the seven scheme weights as dependent variables. With the use of Wilks' A criterion, the seven dependent variables were significantly affected by paradigm, F(7,282) = 64.6, p < .001, n2 = .62; weighting, F(7,282) = 13.1,/? < .001, t]2 = .25; and the interaction between the two, F(7,282) = 3.37,p = .002, ij2 = .08. The point here is to understand that scheme emphasis differed across the four different modes of analysis; the exact locus or loci of these differences are not of particular interest (thus no follow-up analyses are explored). Generally speaking, scoring paradigm accounted for the largest portion of the variance, while weighting accounted for considerably less, and the interaction 9 These are identical data to those presented in Figure 4. 40 between the two accounted for even less again. That is to say that, in the end, the SMSU and SMSU2 produced significantly different scheme emphases. x While differences exist between the SMSU and SMSU2 data, a striking similarity is that the self is vastly dominated by the more tactile and concrete me-self constructs (upwards of 85%) and less so characterized by the more transcending, albeit nebulous, I-self ones. Recall that De Mello (1990) asserts that suffering is alleviated by disengaging from one's me-self. The data presented above suggest that people identify very strongly with their me-selves. One half of the De Mello Hypothesis states that emphasis on the I-self relates positively to moral behavior. Recall that addressing this question with the SMSU data, no support was found, r(45) = .10, p = .51. Thus, this question is revisited with the SMSU2 data. Let us begin by revisiting the De Mello Hypothesis. Does identifying with the I-self relate positively and identification with the me-self related negatively with moral behavior? Recall that the answers to these questions are mutually determined due to the ipsative nature of the data generated by the scoring paradigm. Thus, we only investigate one of these questions. The I-self, determined by summing the emphases of each of the three I-self schemes, using the SMSU2 scoring paradigm, is positively associated with moral behavior, r(97) = .30,/) = .003. Statistically controlling for the effect of socially-desirable responding, age, gender, and ethnicity10 does not diminish the I-self s explanatory power on • S M S U UrrAeighted 0SMSU\Afeighted m SMSU2 Unweighted • SMSU2\Afeighted 4* Figure 5. Scheme Emphases for the Various Coding Systems Ethnicity is operationalized dichotomously as Canadian-born or immigrant. 41 moral behavior, p = .33. This is the first support we see for the De Mello Hypothesis". Those individuals that identify with the abstract notions of self-as-knower as opposed to the more concrete self-as-known are more likely to engage in moral behavior. Next, let us consider the locus of this interaction. Which of the schemes contribute to this association? It could be the case that several or all of the schemes contribute equally to this effect or it could be the case that one or two schemes are by-and-large responsible. The latter turns out the be the case, with agency being positively related to moral behavior, r(97) = .31,/? = .002. None of the other scheme emphases were related to moral behavior, |r|s(97) < .17,/?s > .0912. Thus, only one of the seven schemes—agency—is significantly related to moral behavior. Recall that it was hypothesized that the psychological scheme would relate positively and the physical scheme would relate negatively to moral action. These hypotheses were not supported, r(97) = -.05 and -.03,/? = .62 and .77, respectively. Regardless, the relationship between agency and moral action is the first demonstration of an association between the content of self-undertanding and moral behavior. Before entering into discussion on the meaning and implication of this finding, it is worth pausing to compare the predictive validity of the four content paradigms. Four Coding Paradigms Compared. Recall that content was coded in two ways: with the SMSU, where a given chunk was assigned identically one scheme and agency was defined in terms of formation of the self; and with the SMSU2, where a given chunk's content coding was distributed amongst multiple schemes and agency was defined in terms of volitional efforts of the self. For each coding paradigm, we have considered both the unweighted scores, reflecting the simple tally of scheme codings as well as the weighted scores, reflecting the participant's active evaluation and weighting of his/her various responses. Hence, we have four different paradigms. To assess the predictive validity of each, and relative to one another, it is worth considering the predictive power of each of the seven schemes in conjunction. To do so, a regression analysis is performed, wherein demographic variables (age, gender, and ethnicity) are entered in the first block, the tendency to respond in a socially desirable way (BIDR) is controlled in the second block, and the seven schemes are entered in the third. The four different paradigms are evaluated by considering the change in variance explained in the third step (see Table 4). 1 1 De Mello's (1990) original reference was to suffering as opposed to moral behavior. This notion is explored, by operationalizing suffering as Satisfaction with Life Scales (Diener et al., 1985). Contrary to De Mello's assertion, emphasis on I-self does not relate to SWLS, r(97) = .1 \,p = .21. Further investigation revealed than no single scheme emphasis (using the SMSU2 weighted paradigm) relates to SWLS (-.15 < r{91) < .13, ps > .15) and content did not explain a significant portion of the variance (R2 = 5.1%, p = .57) in SWLS in a regression analysis analogous to the ones used for moral behavior (discussed below). Therefore, the data do not support De Mello's (1990) claim that content emphasis relates to subjective well-being. 1 2 Social scheme emphasis was the strongest correlate of moral behavior. The />-value of .09 should be considered relative to the Bonferroni-corrected a = .007. 42 Table 4. Change in Percent Variance Explained by Step in the Regression Models Model SMSU SMSU2 Unweighted Weighted Unweighted Weighted 1. Demographics 16.0 12.2 2. BIDR 11.2 7.3 3. Self-content 15.9 14.0 12.0 18.5 Demographics play a notable role in explaining variance in moral behavior (12% to 16% for the various models). Closer inspection of the three demographic variables reveals that ethnicity significantly predicts moral behavior, r(97) = 30, p = .003. Those born in Canada tended to engage in moral action more so than those that were not. Neither age nor gender13 are significantly related to moral behavior, r(97)s = .10 and-A7,ps = .34 and .09, respectively. However, there is a weak trend for gender, where females tend to engage in moral acts more so than males. The relationship between demographics as a whole and moral behavior is marginally significant for the SMSU, r(45) = .40, p = .056, and significant for the SMSU2, r(97) = .35, p = .006. The difference in significance values here should be interpreted in light of the different number of cases in each (47 for the SMSU and 99 for the SMSU2); effect sizes should be taken as a more appropriate direct comparison. Under this metric, demographics play a similar role in predicting variance in moral behavior in the two models, with demographics predicting slightly more in the SMSU. Also notable is a significant effect of socially desirable responding. For some or all of the measures of moral behavior, a normative desire to present oneself in a favorable way factors in. This could be seen as a confound that challenges the validity of the operationalization of moral behavior. However, re-running the regression analyses with the second step (BIDR) omitted does little to change the variance explained by content (R2 = 15.2%, 15.7%, 12.3%, and 19.2% for the four paradigms in the order presented above, respectively). Thus, the variance in moral behavior explained by socially desirable responding does not overlap with the variance explained by content. This suggests that the desire to be seen favorably by others is a source of moral motivation independent of scheme emphasis. Discussion of the virtuosity of this reason for doing good aside, for the present discussion, it is safe to assume that socially desirable responding is not a confound in terms of predicting moral behavior from scheme emphasis. Finally, we consider the predictive validity of the four models of self-understanding. In that the SMSU data are based on approximately half of the sample while the SMSU2 data are based on the full sample, a fair metric for comparison is effect size, rather than significance level. While it was anticipated that the models would predict variance with differing levels of success, we see that they indeed are 1 3 Gender dummy coded with female = 0. 43 relatively similar. They each predict between 12.0% and 18.5% of the variance, with the most developed model—the SMSU2 weighted paradigm—being the strongest. Recall Wainer's (1976) recommendation that self-concept data be simply tallied rather than weighted; Marsh and Hattie (1996) argued that in the 20 years of subsequent research, no serious threat to his recommendation was produced. The current findings can be seen as a challenge to Wainer's recommendation. The SMSU2's predictive validity increased by 54% in the course of weighting. Nonetheless, the development of the SMSU2 paradigm resulted in only a minor gain in predictive validity over the SMSU. Recall that the impetus for this research is to explain some of the 90% of variance in moral behavior left after controlling for moral reasoning. In that just one aspect of identity—self-content—explains more variance (18.5%) than reasoning (10%) suggests that further investigation into the moral identity is warranted. But before investigating future directions, let us first summarize the findings of the current study in relation to the stated hypotheses, and then revisit the key finding from the SMSU2 coding. Recall that Derryberry and Thoma (2005b) found that different personological measures predicted moral functioning differentially. Thus, it is worth exploring the relationships between the nine content categories from the SMSU2 (seven schemes, me-self, and I-self) against each of the three measures of moral behavior. As can be seen in Table 5, the correlations are weak in general, |r5(98)| < .23, ps > .02. These figures should be interpreted in light of Bonferroni-corrected a = .002. Thus, no single scheme significantly predicts any single measure of moral behavior; whereas recall that agency does predict the aggregated measure of moral action, r(97) = .31,/? = .002. This finding serves to support Epstein's (1983) assertion that aggregating multiple measures that tap the same construct can serve to increase the possibility of detecting effects. Table 5. Correlations between Self-Understanding (SMSU2) and Measures of Moral Behavior Physical Active Social Psychological Agency Cont. Distinct. Me-self I-self H O N -0.13 -0.10 0.04 0.01 0.23 0.10 -0.09 -0.18 0.22 SRA 0.03 0.00 -0.19 -0.05 0.17 0.01 0.12 -0.19 0.20 G E B 0.04 0.06 -0.19 -0.06 0.22 0.01 -0.16 -0.13 0.16 Next, let us revisit the main hypotheses of this study. The first hypothesis dealt with the relationship between scheme emphasis and moral action, and was twofold. One, the first hypothesis contended that (some of the seven) scheme emphases would be associated with moral behavior. This hypothesis was supported with the SMSU2 paradigm, with agency being the one and only significant predictor, r(91) = .31,/? = .002; whereas the hypothesis was not supported with the SMSU paradigm. In neither paradigm was there support for a relationship between either of the physical or psychological schemes, on the one hand, and moral behavior on the other. And two, ari anecdote provided by De Mello (1990) theorized that emphasizing one's identification with the I-self leads to an alleviation of suffering. The De Mello Hypothesis extended this to the moral domain. Again, this hypothesis was supported for 44 the SMSU2, r(97) = .30,/? = .002, but not for the SMSU, r(45) = .10,/? = .51. The second hypothesis addressed the issue of centrality-weighting, stating that scheme emphasis weighted by subjective appraisals of centrality would better predict moral action than pre-weighted tallies. Indeed this was the case for the SMSU2 (54% increase) but not for the SMSU, where a slight (12%) decrease in predictive validity was observed. The third and final hypothesis had to do with self-theory. It was hypothesized that a ceiling effect would be present. This hypothesis was supported, as 49% of all self-statements were coded as Level 4, and variability in level was restricted (SD = .31). Resultingly, level was unpredictive of moral action. Thus, there was some support for each of the three hypotheses. Agency Recall that the key finding from the analysis of self-theory was that agency predicted moral action. Agency was defined in terms of deliberate efforts of the self that effected actual change of some kind. Intentional action plays a key role in Blasi's (e.g., 1984) moral self. An agent cannot reasonably be held responsible for outcomes beyond his/her control. However, when an outcome is brought upon by a choice made freely, responsibility—be it praise or blame—is often assigned. In sum, choice entails responsibility. Recall that a sense of responsibility for one's actions is the third of the components of Blasi's model. It is thus that a sense of intentional action is a necessary criterion for bridging the judgment-action gap. There exist, however, a number of different conceptualizations of agency in the personality literature. A critical review of the most prominent of them is presented, with reference to their efficacy in capturing Blasi's vision for the construct. Blasi (1988) saw intentional action as entailing three orthogonal components. First, and most importantly, a positively-orientated (or approach) stance must exist within the actor towards the goal. Only when desire, hope, or control are present can we say that an action was carried out intentionally. Second, operators—senses as well as cognitive structures and processes—must be available and functional. And third, some goal or product of the action must have existed and must come to be as a result of the action. In sum, for Blasi, intentional action requires that an agent must have held some affinity towards achieving a particular end, and must have possessed the requisite physical and mental tools for carrying it out. With Blasi's vision in hand, let us examine several historical and contemporary views on human agency. Wiggins (1991) reviewed various different views on human agency through history. A notable point of contention between conceptions is the morality of agency. The majority of theorists (e.g. Freud, Erikson) probably saw agency as an amoral human capacity, which led an individual to ends outside of the moral domain such as work and autonomy. This morally-neutral view on agency has filtered down into contemporary models of agency. While unpopular, it is worth noting that some theorists (e.g., Horney, 1937) saw agency as a way of moving away from others, thus pitting it in direct 45 competition with prosocial motives. The way in which agency has come to be seen as being an amoral human quality is best explored through examination of five models that presently hold currency. First, two models dominate the study of personality traits: the Five-Factor Model and the Interpersonal Circumplex (Wiggins, 1995). The Five-Factor Model taps five ubiquitous and independent dimensions of personality: (a) extraversion, (b) agreeableness, (c) conscientiousness, (d) emotional stability, and (e) openness to experience. The proponents of this model contend that this set comprises a necessary and sufficient mapping of the entire personality domain. The Interpersonal Circumplex focuses on the orthogonal dimensions of dominance and nurturance as the primary dimensions underlying interpersonal relationships (involving the granting or withholding of power/status and nurturance/affection, respectively). Dominance corresponds to the extraversion factor of the Five-Factor Model of personality whereas the nurturance dimension corresponds to agreeableness (Wiggins, 1995). This first dimension, dominance/extraversion, encapsulates strivings for the achievement of goals, and is thus most relevant to the current discussion on the topic of agency. To Wiggins (1991), the agency captured by the dominance dimension refers to "the condition of being a differentiated individual, and it is manifest in strivings for mastery and power which enhance and protect that differentiation" (p. 89). Note that Wiggins' agency closer resembles Damon and Hart's (1988) distinctness rather than their agency. To Wiggins and other personality-trait theorists, a sense of differentiation of the self from others is the primary feature of agency; desire or stance would seem to be a product, rather than a core feature in itself. While Wiggins sees a "blend" of dominance and nurturance as the most desirable position on the circumplex, his theory implicates nurturance/agreeableness as a more evaluative (i.e., morally relevant) dimension whereas dominance/extraversion is less evaluative. In contrast, Blasi's vision of intentional action is as a tool for prosocial behavior. Second, Emmons (1999) taps personality in a more contextualized way by asking about everyday personal concerns and eliciting aspects of goal motivation. Self-statements are coded for a variety of motivational themes, some of which include: (a) power—concern with influencing others, having control, or seeking status; (b) self-sufficiency/independence—concern with being autonomous and asserting oneself; (c) identity—concern with having greater self-choice with respect to one's internal processes and greater freedom with respect to others; (d) personal growth—concern for personal well-being and for improving aspects of the self; (e) achievement—concern with success and performing well against an external or internal standard. Each of these themes entails facets of intentional action. Congruent with Horney's (1937) antisocial notion of agency, Emmons' themes of power, self-sufficiency/independence, and identity entail effects in the interpersonal domain that generally point towards the distinguishing or distancing of oneself from others. Meanwhile," personal-growth and achievement both entail deliberate efforts aimed at asocial goals. Emmons' personal striving themes do well to capture agency in the specific contexts in which they occur. However, in their foci on motivation and reticence on actual effects, they fall short of epitomizing Blasi's vision. 46 Third, McAdams, Hoffman, Mansfield, and Day (1996) capture agency within the context of integrative narratives of the self, which give a sense of the coherence, meaning, and purpose of an individual's life. McAdams et al. identify four types of agency: (a) self-mastery—efforts to "master, control, enlarge, or perfect a self that has already attained some measure of autonomy" (pp. 346-347); (b) status—striving to achieve acknowledgement from others or advance in a social hierarchy; (c) achievement/ responsibility—efforts to perform at a high standard, succeed at a task, and progress positively in accomplishing sought objectives; and (d) empowerment—the improvement of the self arising from interactions with stronger, wiser, or more powerful others. Some congruence with themes of personal strivings becomes apparent here. Like Emmons' themes, McAdams' agency types are each highly contextualized and specific, and fall short of making requirements regarding the actual accomplishment of the motivated theme. But like some of Emmons' immoral themes, McAdams' theme of status entails a conceptual tension with notions of belonging, preventing it from being applied to moral ends. However, like the more amoral of Emmons' themes (achievement, personal growth), McAdams' themes of self-mastery and achievement/responsibility leave open the possibility that these motives can be carried out towards moral ends. Also noteworthy is the inclusion of empowerment, which falls short of making mention of the core principle of Blasi's intentional action—stance. Empowerment entails effects made upon the self, not by the self. In this way, it diverges from most conceptualizations of agency. Fourth, Damon and Hart (1988) evaluated agency in the context of self-understanding interviews (as discussed above). Their agency referenced the existence or formation of the self, and in this way is similar to McAdams' empowerment. It is somewhat puzzling as to why Damon and Hart include their "self-agency" within the I-self, as the self is clearly the object of change, rather than the source of it. Fifth, Kohlberg and Candee (1984) emphasize the necessary role of judgments that one is responsible to act based on a moral judgment. In drawing attention to the necessity of making such a judgment of responsibility, Kohlberg and Candee implicitly point to the individual's choice to follow those judgments or not. But Kohlberg was a strict Kantian, and placed little interest in the types of virtues that would bring such judgments to life. For this reason, Kohlberg and Candee's treatment of judgments of responsibility lack any assessment of volition or character. Thus, agency has most often been seen as an amoral but positive human characteristic. Some have gone further, giving agency a competitive or antisocial bent. Those that have seen agency as morally neutral have at least allowed for the possibility that agency be applied towards prosocial ends. These conceptualizations, however, tend to focus on the motivational aspect of intentional action, stopping short of requiring the agent to believe that they indeed are successful. This marks the major point of divergence with the conceptualization of agency developed in this study (henceforth referred to as agency-in-action). Agency-in-action is perhaps most reminiscent of McAdams' and Emmons' achievement/responsibility, but entails an important addition—the striving must not only be geared 47 towards a specific success, it must actually succeed in effecting some change. In making this distinction, agency-in-action explicitly requires the incorporation of product into a self-statement; agency-in-action thus epitomizes Blasi's vision of intentional action better than any of the existing models. On the face of it, Blasi's assertion that volitional agency helps bridge the judgment-action gap could be misinterpreted to suggest that agency alone motivates moral action. One may reason that volitional choices causing action quite naturally point one towards moral ends. But this is not Blasi's claim. Instead, Blasi (2005) sees agency as a necessary but not sufficient facet of the moral self. He notes that "like memory or intelligence, willpower can be used for moral purposes, but it can also be used to pursue pragmatic, or even immoral goals" (pp. 75-76). This was indeed the case in the present study. For example, the following amoral statement of participant # 26550 was scored as having agency-in-action present: "I like to consider myself to be fairly motivated, ambitious... even more than ambitious. I'd say hungry in terms of setting goals and getting them. I feel like I will get what I want no matter what and*nothing will get in my way." Similarly, the following immoral statement of participant # 97666 was also scored as evidencing agency-in-action: "I cheated on my girlfriend. I knew I was making a mistake at that time but I did it anyway. I felt like I needed to make that mistake, I needed to see what it was like. It seemed like something that had to be done. If I was going to see if this was going to be a serious relationship, I had to see what it was like to be with somebody else." While agency would seem to be an amoral construct in light of these examples, Blasi (2005) asks an interesting follow-up question: "Could it be that people who identify themselves with moral concerns also tend to develop a higher ability to control their desires? In general, because the ability to control one's desires has widespread effects in people's functioning, we could expect that willpower will also affect the capacity to translate good intentions into action. Unfortunately, not much can be said beyond this general statement" (p. 76). The findings of the present study speak directly to this question, and present new support for this intuition. It is important to bear in mind that this positive trend between agency-in-action and moral behavior existed in spite of many amoral and immoral statements of participants such as # 26550 and # 97666. That is, a general pull for applying agency to prosocial ends exists. This evidence poses a challenge to the antisocial bent of some of the conceptualizations of agency presented above. Blasi (2005) rightly noted that agency can lead intuitively to immoral or amoral goals. Why then does seeing oneself as willful and efficacious go along with leading a moral life? Three possibilities, each with different implications for human nature, are discussed in brief. Consider that the relationship between agency-in-action and moral behavior is correlational. First, consider that a third variable may account for this relationship. It may be the case that parenting or quality of education (for example) causes individuals to both believe in their ability to create change as well as to follow a virtuous path. This notion implicitly places much emphasis on situational influences, and their ability to influence individuals ubiquitously. In this view, positive social change should come through the altering of social 48 structures and institutions. Second, consider that moral behavior may cause one to see oneself as an efficacious agent. A viable explanation for this goes as follows: acts are followed by reflection where the individual credits him or herself for the act if it was favorable but excuses the self if the act was unfavorable. If a general (non-agentically-generated) disposition to engage in acts exists, those that have a disposition for moral acts will have enhanced disposition to implicate the self in terms of outcomes over those that have dispositions for amoral or immoral acts. While also viable, this implicates a highly deterministic and cynical view of human nature where we not only lack free will but also become unrealistically egotistical and self-crediting about that which we have passively inherited. Similar to the first possibility, the second implicates structural changes of society so as to induce behavioral changes. The will and agency-in-action is at best a symptom of moral action in these two views. And third, it is possible that agency-in-action causes moral behavior. Perhaps it is the case that upon realizing that one's inner states and decisions have real and meaningful impacts on the outside world, individuals are left with the unavoidable conclusion that they are responsible for outcomes that result from their intrapsychic action. These results, in turn, reflect on the kind and quality of persons that we are. It seems reasonable to presuppose that there exists a normative desire to see oneself as a good person overall. With such a desire salient, the agentic person sees the connection between one's will, one's actions, and who one is, which causes him/her to then deliberate upon, choose, and alter the willful self that he/she is. It is in this way that the agentic individual would come to lead a moral life. This final possibility is the most optimistic of human nature. Assuming this mechanism, societal changes cannot afford to focus solely on inducing behavioral changes; the self behind the behavior is both critical and non-redundant. The causal nature of the agency-behavior relationship remains an open question, and one for future research. Limitations Several methodological, empirical, and conceptual limitations of the present study are noteworthy. Methodologically, it would have been of benefit if all of the interviews were coded under the SMSU paradigm. Without the complete data set, it was difficult to make direct comparisons between the two coding paradigms. Also, it would have been of benefit had resources been available to hire a primary coder that was blind to the research hypotheses. Several weaknesses of the SMSU2 coding paradigm will be noted in future directions. Empirically, one of the quandaries of the present study is the lack of convergence between the three measures of moral behavior, namely between honesty and the other two measures (ecological behavior and altruistic behavior). While agency predicted the three measures of moral action similarly, it remains unclear as to why there were null inter-relationships between the three measures of moral action. One possibility is that the honesty measure suffered from a restriction of range, where approximately half of the participants scored at the highest level. 49 Finally, an important conceptual limitation is noted. A lack of developmental data prevents deriving insights regarding the causality of some of the relationships, and hinders efforts to uncover the mechanisms that underlie these phenomena. For Blasi's theory to be realize the potential which it appears to have, the key claim that identity causes moral action must one day be put to the test. Summary and Conclusions This research was premised on the notion that extant rationality-based models have failed to adequately capture the complexity of moral functioning. Moral action was operationalized as two self-report measures and a more veridical measure of honesty. An unanticipated finding was that honesty did not relate to either of the other two measures. It was argued that this finding points to the complexity of the moral domain. A novel model was presented here, which succeeded in making the first demonstration of an association between a content scheme of self-understanding and moral action. Agency was found to lie at the core of this effect. It should also be noted that more than just one significant association was hypothesized. The scarcity of positive findings here probably points to the inherent difficulty in adequately capturing a valid representation of the self, and in devising a framework that meaningfully captures morally significant themes. Rather than acting to discourage this enterprise, this should point to the need for further creativity and dedicated efforts to reveal a most nebulous but vital phenomenon. 50 Chapter III - Future Directions In this study, conceptual and empirical gains towards mapping the moral self have been realized. Nonetheless, much work remains before that goal is achieved. In this chapter, two such avenues are explored. Content In this thesis, the SMSU has been criticized on conceptual and empirical grounds as being an inadequate tool for capturing the moral self. Still based on James' taxonomy of the self, the SMSU2 was developed, which showed improved predictive efficacy. However, further improvements are foreseeable; a discussion of three is presented here. First, recall that the Self-Understanding Model (Damon & Hart, 1988) is based on the Jamesian taxonomy of the self. While the taxonomy has been popular amongst researchers, it remains to be convincingly demonstrated that it is useful for capturing the moral self. The present study demonstrates the first relationship between a scheme and moral functioning, pointing to the efficacy of the model. However, some reconceptualizations of some schemes may be justified, so as to streamline the taxonomy into the moral domain. Hart and Fegley's (1995) findings may be informative of how schemes ought to be reconfigured. Recall that, relative to the comparison group, caring exemplars referenced the following sub-schemes more often: (a) positive, caring, moral personality traits; (b) moral, caring goals; (c) academic goals; and (d) non-moral typical activities. The first three differences point to the functional heterogeneity of the psychological scheme. One suggestion is therefore that the psychological scheme be divided into agentic (goals, beliefs) and non-agentic (personality, emotions) components. It is also recommended that the agentic component be relocated in the I-self. A second issue within the me-self is the physical scheme. On the one hand, materialism, an immoral value, entails emphasizing one's possessions, body, and appearance. These are all captured within the physical scheme. On the other hand, also in the physical scheme is one's physical environment. In that valuing one's global environment is a critical component of an ecologically-sound worldview, it would seem that the lumping together of possessions, body, appearance, and physical environment effectively cancels out the physical scheme's moral weights. Thus, it is suggested that these two sets of physical entities be somehow differentiated within content coding. And a third and final issue to be discussed here is the operationalization of the continuity and distinctness schemes. Recall that the reconceptualization of agency proved to be fruitful in better capturing Blasi's notion of responsibility. However, the notion of personal identity—the sense of being oneself and not anyone else—has also been implicated as foundational to human flourishing (Chandler, Lalonde, Sokol, & Hallett, 2003). Nozick (1981) argues that personal identity can be reduced to two sentiments: the sense that one is the same person through time; and the sense that one is oneself and not anyone else. Thus, continuity and distinctness are said to comprise personal identity. Continuity, as 1 51 described in the SMSU and SMSU2, references a defense of knowing the continuing existence of the self through time. However, it is argued here that a sense of continuity can be evidenced without defending the self per se. In that the core principle of the Self-Understanding Model is that individuals define the self differently, a sense of continuity can also be defended in a multitude of ways. For example, when asked about why his or her parents are important to him/her, the participant often emphasized the fact that his/her parents had always been a part of his/her life. Thus, a sense of continuity was expressed not in terms of the self, but rather in interpersonal terms. It is reasoned here that such a statement functions as a connector through time. Therefore, it is recommended that the scheme definition of continuity be expanded to involve more than just the continuity of the self, and that include continuity of other facets of the individual's life. Similarly, the definition of distinctness would appear to be unnecessarily restrictive. Distinctness is defined in terms of knowing that one is unique or different from everyone else in the world. Often, participants emphasized how they were different from certain specific people or groups. It is argued here that the psychology of differentiating between the self and every other person, on the one hand, and differentiating between the self and some sub-group of persons, on the other, are functionally equivalent. Thus, it is recommended that the definition of the distinctness scheme be broadened to accommodate this. Values To this point, we have thoroughly examined the Platform Model of self-understanding, with elaboration on the topics of self-theory (level) and self-content (scheme). As mentioned, self-theory failed to predict moral behavior whatsoever; limitations of the construct were noted. Self-content proved to be more predictive of moral behavior with a new brand of agency alone accounting for a significant portion of the variance in moral behavior. As it stands, however, the Platform Model falls prey to the fifth criticism of the SUI discussed in the introduction—-that morally relevant information remains untapped. For example, the moral exemplars of both Hart and Fegley's (1995) as well as Colby and Damon's (1992) studies emphasized not just any goals (psychological scheme) as core to the self-understanding but moral, caring goals. We argue that part of such orientation or "texture" of self-understanding can be encapsulated within extant value theory. It is to this topic that we now turn. Values represent an important teleological component behind specific motivation, including moral motivation, that has perhaps not received due attention in dominant theories of moral development that have appropriated deontological frameworks for the moral domain (e.g., Kohlberg). Values are seen as situation-transcending concepts or beliefs that pertain to desirable end states. Schwartz (1992) has developed a comprehensive framework, represented by a set of 10 universal values: achievement, power, hedonism, stimulation, self-direction, benevolence, universalism, tradition, conformity, and security. The Schwartz Values Survey (SVS; Schwartz, 1992; see Appendix K for Schwartz's typology of values) prompts respondents to rate "as a guiding principle in [his or her] life" 56 value-laden stems (e.g., 52 PLEASURE-gratification of desires) on a 9-point Likert scale: 7 (supreme importance) to 0 (not important) and -1 (opposed to my values). Thus, ten value scores are derived; they can be combined further into four types of worldviews or "value orientations" (see Table 6). Table 6. The 10 Schwartz Values and corresponding Value Orientations. Value . Value Orientation Achievement Power Self-Enhancement Hedonism Openness to Change Stimulation Self-Direction Benevolence Universalism Self-Transcendence Tradition Conformity Security Conservation Importantly, Schwartz sees values as being ordered in terms of relative importance. In fact, he places the 10 values on a circumplex where any values opposite one another entail a negative conceptual and empirical association. In this way, Conservation opposes Openness to Change while Self-Enhancement opposes Self-Transcendence. On the surface, it would seem redundant and inefficient to tap implicit value statements that are embedded within self-narrative when a quicker, more direct pencil-and-paper measure (i.e., the SVS) already exists. However, two reasons justify the future work, the first is empirical and the second, conceptual. First, the SVS has had mixed results in predicting behavior. Bardi and Schwartz (2003) found that universalism related only moderately to universalistic behaviors (r = .27) whereas benevolence related only marginally to its associated behaviors (r = .13) when behaviors were determined by peer report. Clearly, there is much variance in moral behavior that the SVS is failing to capture. This suggests that either (a) moral values are only weakly predictive of moral behavior and that situational influences account for the remainder of the variance in behavior, or (b) something about the nature of the SVS (e.g., its transparent and self-report nature) prevents it from capturing an existing relationship. By tapping values in a more life-like manner, it is hypothesized that the SUIT provides a more authentic set of observations of an individual's values. If this hypothesis is correct and these values can be validly teased out of self-narrative, an increase in predictive validity will result. 53 The second reason for tapping values from self-narrative is conceptual. Amalgamating theories and findings in moral personality—specifically identity, self-understanding, and values—has been identified as a promising and important direction for future research (Hardy & Carlo, 2005; Schwartz, 2001). The notion held within the Schwartz Value Typology and Circumplex of seeing values in tension dovetails well with the centrality-rating paradigm of the SUIT, wherein respondents are faced with the task of rating the elements of their self-understanding, not in isolation, but in direct competition with one another (i.e., ipsatively). While such a measure of implicit values has yet to be developed, a preliminary analysis is possible with the current data set. As mentioned earlier, participants were drawn from a variety of clubs chosen so as to sample a variety of value orientations. What this means exactly should now be clearer. It remains to be seen whether or not this goal was achieved. Clubs were coded as being of one of the four value orientations (Conservation, Self-Enhancement, Openness to Change, Self-Transcendence) by giving consideration both to the focal activities of the club as well as to the club objectives (as advertised on club websites, when available). This is a rather "quick-and-dirty" method of assessing value orientation as it ignores all of the overlap and multiplicity of values naturally present. For example, a club whose main focus was of camaraderie and belonging (hence classified as exemplifying the Conservation value) emphasized community involvement and philanthropy in its mission statement. Despite the crassness of this measure, reliability between raters in the coding of the predominant value orientation of the 30 clubs represented in the sample was substantial, with 83% agreement and KN = .78. Classifying participants by the value orientation of the club through which they were recruited, we see that the sample was indeed well balanced. Between 21 and 28 participants were classified to each of the four value orientations in this way. Recall that the intent was to sample from a variety of value orientations without introducing demographic confounds such as ethnicity. As intended, the groups did not differ in terms of whether or not they were born in Canada, %2(3, N= 99) = .88, p = .83, cj> = .09. Similarly, the groups did not differ in age, F(3,95) = A3,p = .73, n2 = .01. However, the groups did differ in terms of gender, x2(3, N= 99) = 13.1,p = .003, <) = .37. The Self-Transcendence and Conservation groups appear to be constituted of a higher proportion of females (24 of 28, and 20 of 27, respectively) than either of the Self-Enhancement (11 of 23) or Openness to Change (9 of 21) groups. This trend should be noted, and perhaps corrected, in future investigations using this methodology of recruitment. Values and Moral Functioning It is contended that certain implicit value statements embedded in self-understanding narrative will be associated with moral functioning and moral development. For example, the values of benevolence (preservation and enhancement of the welfare of people with whom one is in frequent personal contact) and universalism (protections for the welfare of all people) carry with them explicit moral themes. Thus, it is hypothesized that self-understandings lined with these themes will be associated with moral behavior. A key feature of Schwartz's (1992) model is that it pits values in conceptual and 54 empirical opposition to one another. In opposition to benevolence and universalism is power and achievement. A second hypothesis is therefore that self-understandings high in power and achievement will be negatively associated with moral behavior. While on the surface, some of the other values (e.g., security, self-direction) seem to be value-neutral, Colby and Damon (1992) and Hart and Fegley (1995) found that moral exemplars stood out in terms of both their moral goals as well as value-neutral ones. Congruently, Blasi (1999) emphasized that in establishing an identity, we "agentically structure our motives and desires" (p. 11). All of this points to self-direction; hence, it is also hypothesized that self-narrative textured with self-direction will be associated positively with moral action. With the available data, we can examine the mean score of the aggregated index of moral behavior of the participants of each of the four value orientations (see Table 7). An ANOVA revealed that indeed moral behavior did differ by value orientation, F(3,95) = 4.63,p = .005, n2 = .13. The locus of this effect was examined. First, it was hypothesized that the Self-Transcendence group would be above average in terms of moral behavior; thus a planned comparison was made between the Self-Transcendence group against the other three and was found to be significant, t(95) = 3.18, p = .002, d = .73. Second, to test Schwartz' (1992) assertion that power and achievement act in competition with moral goals, a planned comparison was made between the Self-Enhancement group against the other three. While it was expected that the Self-Enhancement group would score below the mean on moral behavior, this hypothesis was not supported, /(95) = .71,/? = .48, d= .18. Third, the data were explored post hoc using a Scheffe correction and found that the Self-Transcendence group scored higher on moral behavior (MoralBeh = .99) than the Conservation group (MoralBeh = -.82). While Schwartz posits that Self-Transcendence stands in empirical and conceptual opposition to Self-Enhancement, the former was found instead to stand in empirical opposition to Conservation. This is an interesting finding that points to the need for future research. Table 7. Moral Behavior Descriptive Statistics across Value Orientations MoralBeh M SD Self-Enhancement -.22 1.63 Openness to Change .11 1.90 Self-Transcendence .99 1.80 Conservation -.82 2.01. It is important to bear in mind the crassness of this approach of assigning value orientation. This label is akin to the judgment of a book "by its cover." Nothing about the individual is considered apart from the mission statement of the club through which he/she was recruited. Instances were common where an individual was recruited through a club that typified one particular orientation but the individual 55 happened to be on the executive of a club that exemplified a different value orientation. It is somewhat surprising that the hypothesized effect was observed despite this insensitivity. Values and Self-Content in Concert. The question remains, however, as to whether this effect is merely redundant with that the effects of content (scheme emphasis) of self-understanding in predicting moral action. For example, agency may differ by value orientation, and thus explain the between-group differences. However, agency (SMSU2 weighted) did not differ by value orientation, F(3,95) = 1.45, p = .23, n1 = .04. Next, it is worth asking whether or not value orientation adds to the prediction of moral .behavior. To assess this, an ANCOVA was conducted in which value orientation was used to predict the aggregated index of moral behavior, using the seven scheme emphases (as per the SMSU2 weighted paradigm) as covariates. Even with the effect of self-content removed, value orientation still predicts moral behavior, F(3,88) = 3.46,p = .02, n2 = .09. Value orientation and scheme emphasis when combined accounted for 27.8% of the variance, a marked improvement over the 10% typically explained by moral reasoning alone. What remains to be constructed, however, is a sensitive and valid method of tapping the values of the individual within the rich context of self-understanding. The Texture of Identity Model Recall the conceptual reason that would justify the effort of teasing implicit value statements out of self-narrative. Not only does the integration of values into a self-understanding rubric have functional (predictive) utility, we also argue that it is theoretically justified. Values and self-understanding are both person or personality constructs that influence each other reciprocally. It is through the lens of values that we strive for goals and interpret the world, including the world within. We strive to understand ourselves in terms of our values. Hence, a comprehensive model of self-understanding requires some treatment of values. Metaphorically, self-understanding could then be conceptualized as the platform upon which we stand while interacting with the world. The seven schemes form the length of the platform, and the four levels of sophistication the width, making up the Platform Model. We then introduce themes of values inherent in self-understandings to represent the surface characteristic of the platform—the texture of identity. Hence, another goal of future research would be to contribute to an increasingly comprehensive understanding of the relationship of the self and moral functioning by working towards the development of a unified framework that will integrate extant findings as well as provide a more thorough, systematic paradigm for examining the content of self-understanding. Encapsulating Blasi's Theory. The Texture of Identity (TI) Model is intended to provide a valid assessment of Blasi's theory. While several measures have had partial success in accounting for one of the three tenets, it would be one of the overarching goals of future work to move towards a broader, more encapsulating framework. It is argued that the TI Model makes significant gains over the Platform Model in accounting for the second tenet of Blasi's model (centrality). The TI Model would appear to do an excellent job of accounting for the centrality of moral goals and beliefs. In eliciting rich, contextualized 56 accounts of self-understanding, the SUIT is configured to allow respondents to report whatever it is that is central or important to them. One possible self-understanding is one involving moral goals, beliefs, and acts—all facets captured well by the TI Model. Hence, the SUIT is configured for drawing out the centrality of moral facets of self-understanding while coding for texture would aptly detect these individual differences. Thus, it is argued that the Texture of Identity Model is highly efficacious in encapsulating more of Blasi's theory than any extant measure. Furthermore, it has been argued that most gaps that may exist between the TI Model and Blasi's theory are probably reconcilable with further research. Predictive Validity. Recall that the sixth and final criticism of the SUI was its inability to tap the breadth of moral domain. In that the TI Model better adheres to Blasi's theory, it is argued on a conceptual level that the TI model could better bridge the judgment-action gap and thus succeed where previous models have failed. But, ultimately, this is an empirical question. Recall that the Platform Model explained 18.5% of the variance in moral behavior (beyond demographics and socially desirable responding), while adding a crass measure of texture increased the predictive validity further by 7%. It is hypothesized that with the TI Model, further gains in predictive validity could be made. 57 References Aquino, K. F., & Reed, A., II. (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83, 1423-1440. Bardi, A. & Schwartz, S. H. (2003). Values and behavior: Strength and structure of relations. Personality and Social Psychology Bulletin, 29, 1207-1220. Baumeister, R. F. (1998). The self. In D. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (pp. 680-740). New York: Oxford University Press. . Bergman, R. (2004). Identity as motivation: Toward a theory of the moral self. In D. K. Lapsley & D. Narvaez (Eds.), Moral development, self, and identity (pp. 21-46). Mahwah, NJ: Erlbaum. Bersoff, D. M. (1999). Why good people sometimes do bad things: Motivated reasoning and unethical behavior. Personality and Social Psychology Bulletin, 25, 28-39. Belk, R.W. (1983). Worldly possessions: Issues and criticisms. Advances in Consumer Research, 10, 514-519. Blasi, A. (1980). Bridging moral cognition and moral action: A critical review of the literature. Psychological Bulletin, 88, 593-637'. Blasi, A. (1983). Moral cognition and moral action: A theoretical perspective. Developmental Review, 3, 178-210. Blasi, A. (1984). Moral identity: Its role in moral functioning. In W. M. Kurtines & J. L. Gerwirtz (Eds.), Morality, moral behavior, and moral development (pp. 128-139). New York: Wiley. Blasi, A. (1988). Identity and the development of the self. In D. K. Lapsley & F. C. Power (Eds.), Self, ego, and identity: Integrative approaches (pp. 226-242). New York: Springer-Verlag. Blasi, A. (1989). The integration of morality in personality. In I. E. Bilbao (Ed.), Perspectivas acerca de cambio moralposibles intervenciones educativas (pp. 121-131). San Sebastian, Spain: Servicio Editorial Universidad Del Pais Vasco. Blasi, A. (1993). The development of identity: Some implications for moral functioning. In G. G. Noam & T. E. Wren (Eds.), The moral self '(pp. 99-122). Cambridge, MA: MIT Press. Blasi, A. (1995). Moral understanding and the moral personality: The process of moral integration. In W. M. Kurtines & J. L. Gewirtz (Eds.), Moral development: An introduction (pp. 229-253). Boston: Allyn & Bacon. Blasi, A. (1999). Emotions and moral motivation. Journal for the Theory of Social Behavior, 29, 1-19. Blasi, A. (2005). Moral character: A psychological approach. In D. K. Lapsley & F.C. Power (Eds.), Character Psychology and Character Education (pp. 67-100). Notre Dame, Indiana: Notre Dame Press. Blasi, A., & Milton, K. (1991). The development of the sense of self in adolescence. Journal of Personality, 59, 217-242. Bracken, B. A. (1992). Multidimensional self-concept scale. Austin, TX: Pro-Ed. Buchanan, T. (1992, November). Why is the literature examining the moral cognition-moral action relationship inconsistent? A meta-analytic investigation of five moderating variables. Paper presented at the meeting of the Association for Moral Education, Toronto. Carlo, G., & Edwards, C. P. (Eds.). (2005). Nebraska Symposium on Motivation: Vol. 51. Moral motivation through the life span. Lincoln: University of Nebraska Press. Chandler, M. J., Lalonde, C. E., Sokol, B. W., & Hallett, D. (2003). Personal persistence, identity development, and suicide. Monographs of the Society for Research in Child Development, 68(2, Serial No. 273). 58 Clary, E. G., Snyder, M., Ridge, R. D., Copeland, J., Stukas, A. A., Haugen, J., & Miene, P. (1998). Understanding and assessing the motivations of volunteers: A functional approach. Journal of Personality and Social Psychology, 74, 1516-1530. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Cohen, J. (1990). Things I have learned (so far). American Psychologist, 45, 1304-1312. Colby, A., & Damon, W. (1992). Some do care: Contemporary lives of moral commitment. New York: Free Press. Colby, A., & Damon, W. (1993). The uniting of self and morality in the development of extraordinary moral commitment. In G. G. Noam & T. E. Wren (Eds.), The moral self (pp. 149-174). Cambridge, MA: MIT Press. Damon, W. (1984). Self-understanding and moral development from childhood to adolescence. In W. M. Kurtines & J. L. Gewirtz (Eds.), Morality, moral behavior, and moral development (pp. 109-127). New York: Wiley. Damon, W., & Hart, D. (1988). Self-understanding in childhood and adolescence. New York: Cambridge University Press. Damon, W., Hart, D., Pakula, K., & Shupin, J. (1988). Scoring manual for self-understanding. Unpublished manuscript, Clark University,Worcester, MA. De Mello, A. (1990). Awareness. New York: Doubleday. Derryberry, W. P., & Thoma, S. J. (2005a). Functional differences: Comparing moral judgement developmental phases of consolidation and transition. Journal of Moral Education, 34, 89-106. Derryberry, W. P., & Thoma, S. J. (2005b). Moral judgment, self-understanding, and moral actions: The role of multiple constructs. Merrill-Palmer Quarterly, 51, 67-92. Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The Satisfaction with Life Scale. Journal of Personality Assessment, 49,11-15. Dunlap, R. E., Van Lier, K. D., Mertig, A. G., & Jones, R. E. (2000). Measuring endorsement of the New Ecological Paradigm: A revised NEP scale. Journal of Social Issues, 56, 425-442. Eisenberg, N. (2005). The development of empathy-related responding. In G. Carlo & C. P. Edwards (Eds.), Nebraska Symposium on Motivation: Vol. 51. Moral motivation through the life span (pp. 73-117). Lincoln: University of Nebraska Press. Emmons, R. A. (1999). The psychology of ultimate concerns: Motivation and spirituality in personality. New York: Guilford. Erikson, E. H. (1968). Identity, youth, and crisis. New York: Norton. Epstein, S. (1983). Aggregation and beyond: Some basic issues on the prediction of behavior. Journal of Personality, 51, 360-392. Fournier, S., & Richins, M. L. (1991). Some theoretical and popular notions concerning materialism. Journal of Social Behavior and Personality, 6, 403-414. Getz, I. (1985). The relation of moral and religious ideology to human rights. Unpublished doctoral dissertation, University of Minnesota. Glodis, K. A., & Blasi, A. (1993). The sense of self and identity among adolescents and adults. Journal of Adolescent Research, 8, 356-380: Gyatso, T. (1995). The world of Tibetan Buddhism: An overview of its philosophy and practice. Somerville, MA: Wisdom Publications. 59 Hardy, S., & Carlo, G. (2005). Identity as a source of moral motivation. Human Development, 48, 232-256. Hart, D. A. (1992). Becoming men. The development of aspirations, values, and adaptational styles. New York: Plenum Press. Hart, D., & Fegley, S. (1995). Prosocial behavior and caring in adolescence: Relations to self-understanding and social judgment. Child Development. 66, 1346-1359. Harter, S. (1986). Processes underlying the construction, maintenance and enhancement of self-concept in children. In J. Suls & A. Greenwald (Eds.), Psychological perspectives on the self'(Vol. 3, pp. 136-182). Hillside, NJ: Erlbaum. Hattie, J. (1992). Self-concept. Hillsdale, NJ: Erlbaum. Horney, K. (1937). The neurotic personality of our time. New York: Norton. James, W. (1890). The principles of psychology. New York: Holt. James, W. (1961). Psychology: The briefer course. New York: Harper & Row. (Original work published 1892) Kagan, J. (2005). Human morality and temperament. In G. Carlo & C. P. Edwards (Eds.), Nebraska Symposium on Motivation: Vol. 51. Moral motivation through the life span (pp. 1-32). Lincoln: University of Nebraska Press. Kahn, P. H., Jr. (2006). Nature and moral development. In M. Killen & J. G. Smetana (Eds.), Handbook of moral development (pp. 461-480). Mahwah, NJ: Erlbaum. Kaiser, F. G. (1998). A general measure of ecological behavior. Journal of Applied Social Psychology, 28, 395r422. Kaiser, F. G., & Wilson, M. (2000). Assessing people's general ecological behavior: A cross-cultural measure. Journal of Applied Social Psychology, 30, 952-978. Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental approach to socialization. In D. A. Goslin (Ed.), Handbook of socialization theory and research (pp. 347-480). Chicago: Rand McNally. Kohlberg, L. (1981). Essays on moral development: Vol.1. The philosophy of moral development. San Francisco: Harper & Row. Kohlberg, L. (1984). Essays on moral development: Vol.2. The psychology of moral development. San Francisco: Harper & Row. Kohlberg, L., & Candee, D. (1984). The relationship of moral judgment to moral action. In L. Kohlberg, Essays on moral development: Vol. 2. The psychology of moral development (pp. 498-581). San Francisco: Harper & Row. Kohlberg, L., & Diessner, R. (1991). A cognitive-developmental approach to moral attachment. In J. L. Gewirtz & W. M. Kurtines (Eds.), Intersections with attachment (pp. 229-246). Hillsdale, NJ: Erlbaum. Krebs, D. L. & Van Hesteren, F. (1994). The development of altruism: Towards an integrative model. Developmental Review, 14, 103-158. Kunda, Z. (1999). Social cognition: Making sense of people. Knowledge about other's attitudes, behavior, and personality (pp. 395-450). Cambridge, MA: MIT press. Lapsley, D. K. (1996). Moral psychology. Boulder, CO: Westview. Lapsley, D. K. (2006). Moral stage theory. In M. Killen & J. G. Smetana (Eds.), Handbook of moral development (pp. 37-66). Mahwah, NJ: Erlbaum. 60 Leary, M. R., & Tangney, J. P. (2005). The self as an organizing construct in the behavioral and social sciences. In M. R. Leary & J. P. Tangney (Eds.), Handbook of self and identity (pp. 3-14). New York: Guilford. Marcia, J. E. (1966). Development and validation of ego identity status. Journal of Personality and Social Psychology, 3,551-558. Marsh, H. W., & Hattie, J. (1996). Theoretical perspectives on the structure of self-concept. In B. A. Bracken (Ed.), Handbook of self-concept (pp. 38-90). New York: Wiley. Marsh, H. W., Trautwein, U., Liidtke, O., Koller, O., & Baumert, J. (2006). Multidimensional self-concept and core personality constructs: Construct validation and relations to well-being and achievement. Journal of Personality, 74, 403-456. McAdams, D. P. (1993). The stories we live by: Personal myths and the making of the self New York: Guilford. McAdams, D. P. (1995). The life story interview (Rev.). Unpublished manuscript, Northwestern University, Evanston, IL. McAdams, D. P., Hoffman, B. J., Mansfield, E. D., & Day, R. (1996). Themes of agency and communion in significant autobiographical scenes. Journal of Personality, 64, 339-377. Narvaez, D. (2005). The neo-Kohlbergian tradition and beyond: Schemas, expertise, and character. In G. Carlo & C. P. Edwards (Eds.), Nebraska Symposium on Motivation: Vol. 51. Moral motivation through the life span (pp. 119-163). Lincoln: University of Nebraska Press. Narvaez, D., & Lapsley, D. K. (2005). The psychological foundations of everyday morality and moral expertise. In D. K. Lapsley & F. C. Power (Eds.), Character psychology and character education (pp. 140-165). Notre Dame, IN: University of Notre Dame Press. Narvaez, D., Lapsley, D. K., Hagele, S., & Lasky, B. (in press). Moral chronicity and social information processing: Tests of a social cognitive approach to the moral personality. Journal of Research in Personality. Nozick, R. (1981). Philosophical explanations. Cambridge, MA: Harvard University Press. Nucci, L. P. (2001). Education in the moral domain. Cambridge, UK: Cambridge University Press Park, B. (1986). A method for studying the development of impressions of real people. Journal of Personality and Social Psychology, 51, 907-917. Paulhus, D. L. (1991). Measurement and control of response bias. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures ofpersonality and social psychological attitudes (pp. 17-59). New York: Academic. Piaget, J. (1977). The moral judgment of the child (M. Gabain, Trans.). Harmondsworth, England: Penguin. (Original work published 1932) Reed, A., II., & Aquino, K. F. (2003). Moral identity and the expanding circle of moral regard toward out-groups. Journal of Personality and Social Psychology, 84, 1270-1286. Reimer, K., Furrow, J. L., Baumeister-Peters, B., & Roth, N. (2001, April). Naturalistic conceptions of moral, religious, and spiritual development in urban youth. Paper presented at the meeting of the Society for Research in Child Development, Minneapolis. Rest, J., Narvaez, D., Bebeau, M. J., & Thoma, S. J. (1999). Postconventional moral thinking: A neo-Kohlbergian approach. Hillsdale, NJ: Erlbaum. Richins, M. L., & Dawson, S. (1992). A consumer values orientation for materialism and its measurement: Scale development and validation. Journal of Consumer Research, 19, 303-316. Rushton, J. P., Chrisjohn, R. D., & Fekken, G. C. (1981). The altruistic personality and the self-report altruism scale. Personality and Individual Differences, 1, 292-302. 61 Schwartz, S. H. (1992). Universals in the content and structure of yalues: Theoretical advances and empirical tests in 20 countries. Advances in Experimental Social Psychology, 25, 1 -65. Schwartz, S. (2001). The evolution of Eriksonian and neo-Eriksonian identity theory and research: A review and integration. Identity: An International Journal of Theory and Research, 7, 7-58. Selman, R. L. (1980). Growth of interpersonal understanding: Developmental and clinical analysis. New York: Academic. Spitzer, S., Couch, C , & Stratton, J. (1971). The assessment of the self. Iowa City, IA: Escort. Thoits, P. A. (1992). Identity structures and psychological well-being: Gender and marital status comparisons. Social Psychology, 55, 236-256. Turiel, E. (1983). The development of social knowledge: Morality and convention. Cambridge, UK: Cambridge University Press. Wainer, H. (1976). Estimating coefficients in linear models: It don't make no nevermind. Psychological Bulletin, 83, 312-317. Walker, L. J. (2004). Gus in the gap: Bridging the judgment-action gap in moral functioning. In D. K. Lapsley & D. Narvaez (Eds.) Moral development, self, and identity (pp. 1-20). Mahwah, NJ: Erlbaum. Walker, L. J., & Pitts, R. C. (1998). Naturalistic conceptions of moral maturity. Developmental Psychology, 34, 403-419. Wiggins, J. S. (1991). Agency and communion as conceptual coordinates for the understanding and measurement of interpersonal behavior. In W. M. Grove & D. Cicchetti (Eds.), Thinking clearly about psychology. Volume 2: Personality andpsychopathology (pp. 89-113). Minneapolis: University of Minneapolis Press. Wiggins, J. S. (1995). Interpersonal Adjectives Scales: Professional manual. Odessa, FL: Psychological Assessment Resources. 62 Appendix A - Self-Understanding Interview (SUI) (Damon & Hart, 1988) 1. Self-definition. What are you like? What kind of person are you? What are you not like? Probes: What does that say about you? Why is that important? What differences does that (characteristic) make? What would be different if you (were not) like that? 2. Self-evaluation. What are you especially proud of about yourself? What do you like most about yourself? What are you not proud of? What do you like the least about yourself? Probes: What does that say about you? Why is that important? What differences does that make? 3. Self in the past and future. Do you think you'll be the same or different five years from now? How about when you're an adult? How a about five years ago? How about when you were a baby? Probes: What will be the same? What will be different? Why is that important? 4. Self-interest. What do you want to be like? What kind of person do you want to be? What do you hope for in your life? If you could have three wishes, what would they be? What do you think is good for you? Probes: Why do you want to (be that way, have those things you hope for, have those wishes)? What else do you (Hope for, wish for, believes is good for you)? Why is that good for you? 5. Continuity. Do you change at all from year to year? How (how not)? If you do change from year to year, how do you it's still always you? Probes: In what ways do you stay the same? Is that an important thing to say about you? Why? 6. Agency. How did you get to be the way you are? How did that make you the kind of person you are? How could you become different? Probes: What difference did that make? Is that the only reason you turned out like you did? Is that the only reason? What else could make you difference? How would that work? 7. Distinctness. Do you think there is anyone who is exactly like you? What makes you different from anyone you know? Probes: Why is that important? What differences does that make? In what ways are you different? Ware you completely different or just partly different? How do you know? Ware you different from everybody or just from some people? How can you be sure that you're different from everybody else when there are many people in the world that you don't know? 63 Appendix B - Self-Understanding Interview—Transmogrified (SUIT) 1. This is an interview about you and your life. I am asking you to play the role of narrator. People's lives vary tremendously, and people make sense of their own lives in a variety of ways. In telling me about your life, you do not need to tell me everything that has ever happened to you. The idea is to focus on a few key aspects or themes that are hopefully relevant to your life and who you are. The interview is divided into a number of sections. The interview starts with general things and moves to the particular. I think you will enjoy the interview. Most people do. I would like you to begin by thinking about your life as a story. A long story may even have chapters. Think about your life story as having at least a few chapters. I would like you to describe for me the main chapters of your life story. If you can, identify each chapter and summarize what each is about. This part of the interview can expand forever, so I would like you to keep it relatively brief, say, within 10-15 minutes. So how does your life story begin? 2. What's the current chapter all about? As I mentioned, this interview gradually progresses from the general to the specific. Now switching gears from where you are coming from to who you are, 3. How would you describe yourself? • Why is it important for you to be ? 4. Do you have a job? Do you go to school? • Why is being a important to you? 5. Which of your activities are most important to you? • Why? 6. Do you have any habits or unique ways of doing certain things? • Why do you ? 7. Who are the most significant people or groups in your life? • Why is important to you? 8. What are the favorite things you have or own, your possessions? • Why is your important? 9. What's important to you in terms of your physical characteristics? • Why is your important? 10. What are your major roles and responsibilities? • What aspects of those are important to you? • Why? 11. What are the most important psychological aspects of who you are? Among other things, this may include beliefs, values, goals, personality, intelligence, or emotions. Again, what are the most important psychology aspects? • Why is that important? I'm now going to ask you some questions that are perhaps less typical in everyday conversation. 12. Everyone changes over time. Given that you change from year to year, how do you know it's still always you? ; 13. How did you get to be the kind of person you are now? 14. Do you think that there is anyone who is exactly like you? • How do you know that you're unique or different from everybody else even when there are many people in the world that you don't know? 15. Is there anything else that defines you or is important to who you are that we have not yet spoken about today? 16. What do you like most about yourself? • Why? 64 Appendix D - Self-Reported Altruism (SRA) (Rushton et al., 1981) 1 2 3 4 5 Never Rarely Sometimes Often Very Often 1. I have helped push a stranger's car (out of the snow, ditch, etc.) 2. I have given directions to a stranger. 3. I have made change for a stranger. 4. I have given money to charity. 5. I have given money (or food, etc.) to a stranger who needed it (or asked me for it). 6. I have donated goods or clothes to a charity. 7. I have done volunteer work for a charity. 8. I have donated blood. 9. I have helped carry a stranger's belongings (books, parcels, etc.) 10. I have delayed an elevator and held the door open for a stranger. 11. I have allowed someone to go ahead of me in a lineup (at Xerox machine, in the supermarket, etc.) 12. I have given a stranger a lift in my car. O 13. I have pointed out a clerk's error (in a bank, at the supermarket, etc.) in undercharging me. 14. I have let someone whom I didn't know too well borrow an item I owned of some value to me. 15. I have bought something (Christmas card, etc.) from a charity deliberately because I knew it was a good cause. 16. I have helped a classmate who I did not know that well with a homework assignment when my knowledge was greater than his or hers. 17. I have before being asked, voluntarily looking after someone's pets or children without being paid for it. 18. I have offered to help a handicapped or elderly stranger (e.g., across the street) 19. I have offered my seat on a bus, train, or in a waiting room to a stranger who was standing. 20. I have helped an acquaintance to move households. 21. I have helped push a stranger's car (out of the snow, ditch, etc.) O - omitted in the present study. Appendix E - General Ecological Behavior-Adapted (GEB-A) (adapted from the General Ecological Behavior scale; Kaiser & Wilson, 2000) 1 2 3 4 5 Never Rarely Sometimes Often Very Often G 1. I have used the back of used paper for note-taking, etc. G 2. I have recycled paper as well as empty cans and bottles. G 3. I have often showered rather than taken a bath. G 4. I have waited until I had a full load before doing my laundry. G * 5. I have used a cleaner made especially for bathrooms, rather than an all-purpose cleaner. G * 6. I have bought beverages in cans. G * 7. In supermarkets, I have bought packaged fruits and vegetables. G 8. When offered a plastic bag in a store, I have refused it. G 9. I have discussed issues related to the environment with friends. G 10. I have pointed out to someone his or her unecological behavior. G 11. I have contributed financially to or volunteered with environmental organizations. G * 12. I have driven on freeways at speeds well over lOOkph. G 13. I have taken my own coffee cup to work or school. G 14. I have walked, ridden a bike, or taken public transportation to work or school. G * 15. I have eaten fast-food. G 16. I have bought organic food. G 17. I have used a compost bin. N 18. I have kept the room temperature low at home. N 19. I have made efforts to keep my overall consumption (in term of goods, fuel, etc.) low. N * 20. I have flown in airplanes (e.g., sometimes = once every six months). N * 21. I have driven a big car or truck. * - reversed item; G - adapted from the GEB; N - novel item introduced here. 67 Appendix F - New Ecological Paradigm (NEP) (Dunlap et al., 2000) Strongly Agree 2 Mildly Agree 3 Unsure 4 Mildly Disagree 5 Strongly Disagree I. We are approaching the limit of the number of people the earth can support. * 2. Humans have the right to modify the natural environment to suit their needs. 3. When humans interfere with nature it often produces disastrous consequences. * 4. Human ingenuity will insure that we do NOT make the earth unlivable. 5. Humans are severely abusing the environment. * 6. The earth has plenty of natural resources if we just learn how to develop them. 7. Plants and animals have as much right as humans to exist. ^ 8. The balance of nature is strong enough to cope with the impacts of the modern industrial nations. 9. Despite our special abilities humans are still subject to the lawsof nature. * 10. The so-called "ecological crisis" facing humankind has been greatly exaggerated. II. The earth is like a spaceship with very limited room and resources. * 12. Humans were meant to rule over the rest of nature. 13. The balance of nature is very delicate and easily upset. * 14. Humans will eventually learn enough about how nature works to be able to control it. 15. If things continue on their present course, we will soon experience a major ecological catastrophe. * - reversed items 68 Appendix G - Materialism Scale (MS) (Richins & Dawson, 1992) 1 2 3 4 5 Strongly Agree Mildly Agree Unsure Mildly Disagree Strongly Disagree 1. I admire people who own expensive homes, cars, and clothes. 2. Some of the most important achievements in life include acquiring material possessions. * 3. I don't place much emphasis on the amount of material objects people own as a sign of success. 4. The things I own say a lot about how well I'm doing in life. 5. I like to own things that impress people. * 6. I don't pay much attention to the material objects other people own. * 7. I usually buy only the things I need. * 8. I try to keep my life simple, as far as possessions are concerned. * 9. The things I own aren't all that important to me. 10. I enjoy spending money on things that aren't practical. 11. Buying things gives me a lot of pleasure. 12. I like a lot of luxury in my life. * 13. I put less emphasis on material things than most people I know. * 14. I have all the things I really need to enjoy life. 15. My life would be better if I owned certain things I don't have. * 16. I wouldn't be any happier if I owned nicer things. 17. I'd be happier if I could afford to buy more things. 18. It sometime bothers me quite a bit that I can't afford to buy all the things I'd like. * - reversed items 69 Appendix H - Balanced Inventory of Desirable Responding (BIDR) (Paulhus, 1991) 1 2 3 4 5 6 7 Not True Somewhat True Very True 1. My first impressions of people usually turn out to be right. * 2. It would be hard for me to break any of my bad habits. 3. I don't care to know what other people really think of me. * 4. I have not always been honest with myself. 5. I always know why I like things. * 6. When my emotions are aroused, it biases my thinking. 7. Once I've made up my mind, other people can seldom change my opinion. * 8. I am not a safe driver when I exceed the speed limit. 9. I am fully in control of my own fate. * 10. It's hard for me to shut off a disturbing thought. 11. I never regret my decisions. * 12. I sometimes lose out on things because I can't make up my mind soon enough. 13. The reason I vote is because my vote can make a difference. * 14. My parents were not always fair when they punished me. 15. I am a completely rational person. * 16. I rarely appreciate criticism. 17. I am very confident of my judgments. * 18. I have sometimes doubted my ability as a lover. 19. It's all right with me if some people happen to dislike me. * 20. I don't always know the reasons why I do the things I do. * 21. I sometimes tell lies if I have to. 22. I never cover up my mistakes. * 23. There have been occasions when I have taken advantage of someone. 24. I never swear. * 25. I sometimes try to get even rather than forgive and forget. 26. I always obey the law, even if I'm unlikely to get caught. * 27. I have said something bad about a friend behind his or her back. . 28. When I hear people talking privately, I avoid listening. 0 * 29. I have received too much change from a salesperson without telling him or her. 30. I always declare everything at customs. * 31. When I was young I sometimes stole things. 32. I have never dropped litter on the street. * 33. I sometimes drive faster than the speed limit. 34. I never read sexy books or magazines. * 35. I have done things that I don't tell other people about. 36. I never take things that don't belong to me. * 37. I have taken sick-leave from work or school even though I wasn't really sick. 38. I have never damaged a library book or store merchandise without reporting it. * 39. I have some pretty awful habits. 40. I don't gossip about other people's business. 1-20 Self-Deception subscale items, 21-40 Impression Management subscale items; * - reversed items; O - omitted in the present study. 70 Appendix I - Conscientious Behavior Over the course of recruiting participants and scheduling and administering interviews, it became apparent that pronounced individual differences in the conscientiousness of behavior existed. While this mode of moral action was not included in the original hypotheses of this study, exploring this new variable seemed warranted. A metric of conscientious behavior (CONSC) was developed here, and evaluated for its relationship to other moral constructs as well as in terms of predictive validity. CONSC is an aggregated metric of six variables (each coded as either 0 or 1 or ranging between 0 and 1), which are summed for an aggregate score ranging from 0 to 6. To understand these variables, it is necessary to revisit the details of recruitment and interviewing. As detailed in Appendix C, participants were informed about the study, and asked to reply by email if interested. In the email, participants were instructed to include in their response their (a) name, (b) phone number, and (c) access code, the latter of which was included in the recruitment letter and served to identify the club through which the participant was recruited. The presence or absence of each of these three bits of information in the participant's original email determined the first three (dichotomous) variables: NAME, PHONE, and CODE. The fourth variable was derived from the promptness of a requested confirmation. The last page of the online questionnaires prompted respondents to indicate when they are available for an individual interview. Upon submitting the results, they promptly received an email, in which they were: (a) thanked for completing the online questionnaires, (b) offered a specific date and time for the interview, and (c) asked (in bolded text) for them to let the experimenter know if this date was satisfactory. TIME—the time that elapsed between the participant's response and the time of the eventual interview (divided by the total time between the request and the eventual interview)—is taken as the fourth variable of conscientiousness. TIME is a continuous variable, ranging between 0 and 1. For example, if a participant . waited until the very last minute to confirm the interview, his or her TIME score would be around . 1; whereas, if an individual was prompt in confirming the date, his or her TIME score would on the order of .9. When being offered an interview time-slot, participants were provided directions to the interview room as an email attachment. On occasion, participants asked for the directions again. Thus, ASK—not asking for directions a second time—constitutes the fifth, dichotomous variable^  Finally, showing up for the scheduled interview—SHOW—constituted the sixth (dichotomous) and final variable of conscientiousness: CONSC = NAME + PHONE + CODE + TIME + ASK + SHOW Descriptive statistics for each of the contributor and aggregate variables are shown below in Table 8: 71 Table 8. Descriptive Statistics of the Conscientiousness Contributor Variables NAME PHONE CODE TIME ASK SHOW CONSC M .75 .81 .76 .70 .89 .96 4.90 SD .44 .40 .43 .35 .32 .20 1.16 As can be inferred from Table 8, the sample was on average quite conscientiousness, with the average participant succeeding in 82% of the requested tasks. Three ways of assessing the usefulness of the CONSC variable are proposed. First, the internal consistency of the CONSC variable was assessed and found to be fair (Cronbach a = .54). Second, the correlations between CONSC and other moral variables (MS, NEP, GEB-A, SRA, HON) were explored, and found to be weak, r(97) = -.06, -.04, .08, -.14, and .16, respectively (ps > .12, where the Bonferroni-corrected a = .008). CONSC also does not relate to the aggregate index of moral action (MoralBeh), r(97) = .06, p = .59. Thus, CONSC does not relate to any of the other measures of moral behavior, including the other behavioral measure (HON). And third, the gains in predictive validity in terms of prediction of moral behavior from self-content in the context of the weighted SMSU2 analysis is assessed (see p. 42 for an explanation of this analysis). While the seven schemes account for 18.5% of the variability in moral behavior, as defined as an aggregate of z-scores of SRA, GEB, and HON (p = .002), the same seven schemes predict only 14.0% (p = .03) when moral behavior is redefined as an aggregate of those same variables as well as the z-score of CONSC. That is, adding CONSC decreases the predictive validity of the self-content model. Hence, the CONSC variable fails to materialize as a useful tool in the present study in three separate analyses. For this reason, it is not included in further analyses. 72 Appendix J - The Scoring Manual for Self-Understanding 2 Introduction 1. Definition of Content: The focal notions that are communicated. Any notion that has independent significance (e.g., objects, actions, emotions, ideas) equally qualifies as content. Any element or "chunk" of content pertains to one or more of the seven "schemes," which are: a. Physical (PH)- physical body or material possessions; b. Active ( A C ) - activities or abilities; c. Social (SO)- the self in relation to others, including social-personality characteristics, . social interactions, and social relations; d. Psychological (PS)- emotions, thoughts, beliefs, or cognitive processes; e. Agency ( A G ) - statements evidencing a belief that the self is an efficacious agent; f. Continuity (CO)- statements that defend or explain the sense of self-continuity over time; and g. Distinctness (DI)- statements that defend or explain the sense of uniqueness or distinctness from others. 2. Coding for Content: Any narrative that relates to the self or one's life is codable for content. Coding for content involves the identification of any and all implicit or explicit references to the importance of any entity; coding is on a conceptual level, and not reducible to the simple recognition of "buzz" words (e.g., every active verb is n o t coded in the Active scheme). To code for content requires understanding of the deeper meaning of that which is articulated. The general strategy is to (a) break the narrative into "chunks", then (b) code each chunk for either the presence or absence of each of the seven schemes. 3. Definition of Chunk: A chunk is one characteristic of the self that is mentioned by the participant during the interview. It is comprised of both: a label or stem, which is noted by the interviewer during the interview and later rated for significance by the participant; and its associated explication. During the interview, either (a) the participant explicates the importance or relevance of the characteristic without prompting, or (b) the interviewer asks the participant to do so. a. The coder must judge which text pertains to each stem. Usually, this is a straightforward task as chunks end when the interviewer asks the next question. However, at times, participants give a single introduction to many characteristics before proceeding to explicate each. In this case, the text that serves to introduce several characteristics is coded as i f it belongs to each. For example, if the presence of the Active scheme (AC) is coded in the introduction, each characteristic to which the introduction pertains is coded as having A C present. Hence, the breaking of narrative into chunks is not necessarily a mutually exclusive procedure. Moreover, it is not always linear; in some rare instances, the narrative may switch back and forth in pertaining to different characteristics. b. The coder must then judge whether content of each scheme is present i n d e p e n d e n t of the presence or absence of the other schemes. Hence, a given chunk can be coded as having anywhere between zero and seven schemes present. (This marks a point of divergence from existing content coding procedures such as Damon & Hart, 1988.) Coding for each chunk is done by matching the narrative to a point in the manual. • Schematically ambiguous notions are not coded. For example, / have an analytical mind. With it, I can figure out how to contribute to society is coded A G and PS but contributing to society can mean.many different things (e.g., influencing people, improving infrastructure) and hence is not coded. • Unit of analysis. As defined, content refers to the focal notions that are communicated, and that have independent significance. It is important for coders to acquire an intuitive feel for the threshold of what it means to have "independent significance". The examples included should help in that regard. What follows are some guiding principles: 73 i . Notions of independent significance are emphasized by the participant as being important. Hence, simply mentioning a possession in passing does not qualify as coding for the Physical scheme. i i . If concepts of two different schemes are mentioned in relation to one another, ask whether one concept is necessarily subsumed under the other, grander concept. If so, code only the subsuming scheme, and not the subsumed one. For example: DO YOU HA VE ANY HABITS OR UNIQUE WA YS OF DOING CERTAIN THINGS? I have to color-code my t-shirts, which is silly really. WHY DO YOU DO IT? I don't know. I think it's probably an obsessive compulsive behavior. I've never figured out the reason behind it. Here, t-shirts (physical scheme) are mentioned in relation to a habit (active scheme), both of which are explained by a psychological tendency (psychological scheme). However, t-shirts are simply the object of the activity, are incidental, and hence is subsumed under the activity. In contrast, the activity of color-coding t-shirts is framed as primary, with the psychological description acting as an explanation. Hence the act of color-coding t-shirts is not subsumed under the PS scheme. The coding is: A C and PS. Another example: WHICH OF YOUR ACTIVITIES ARE MOST IMPORTANT TO YOU? I try to go home every weekend to visit my family and my girlfriend. Ifind that pretty important to me. WHY IS THAT IMPORTANT? To just know that I'm still around... I call my girlfriend every night and that sort of thing. But it's nice to see her on the weekend. Here, going home (an activity) is mentioned in relation to his family and girlfriend (social). While the activity is framed as a means to seeing the family, the act of going home is not inherently social (unlike talking with friends, which is inherently social in nature). Hence it is not necessarily subsumed under the SO scheme and is coded both A C and SO. 74 PHYSICAL, PH Definition: Static physical existence (one's own body) or material possessions. 1. Intra-individual a) Physical characteristics: health, voice, strong arms (but not ability or activity; / can lift heavy weights, this being coded AC) b) Nominals (name, gender, race, age) c) Corporeal (eyes, size, genes, appearance) 2. Extra-individual a) Possessions (car, laptop, pictures; My favorite thing that I own is definitely my computer) b) Physical environment (city/country of residence); see note e. below. Notes and Distinctions: . a. One's appearance with explicit reference to others is coded both PH and SO. (WHATARE THE FAVORITE THINGS YOU HAVE OR OWN? My clothes. WHY ARE YOUR CLOTHES IMPORTANT TO YOU? I need to feel that I look good or Ifind that the way that I interact with people is different—I don't talk as freely.) b. For possessions that are important in relation to social implications, code both PH and SO. (My cell phone is important to me. WHY? Because it lets me stay in touch with friends.) c. For possessions that are important in relation to an activity, code both PH and A C (I really like my car. The busses here are terrible. Without my car, I don't know how I'd get to school.) d. Material existence or possessions must receive deliberate treatment, and not simply be incidentally mentioned in passing. (HOW WOULD YOU DESCRIBE YOURSELF? I have a mixed personality. In general I'm a quiet, calm person but also like to have fun. I have a belief that how I grew up sort of helpedform that. I had to drive my car 40 minutes to get to a town to see any of my friends. It sort of taught me to be quiet and to be entertained with myself. The mention of the car here is incidental and hence does not qualify for PH. In contrast, consider: YOU MENTIONED THA T READING IS IMPORTANT TO YOU. WHY? Because that's what I did when I didn't have any friends in elementary school. I read books. There was one month when I read 42 books in a month—more than a book a day. I was burning through them. While the books are referenced as a means to an end (reading), they are given sufficient treatment to qualify as PH. This would also be coded A C and SO.) e. Mention of places or one's physical environment as a means (and not as an end) is not coded PH. (/ went to Europe to go to the beach and party it up is coded A C but not PH). However, places that are emphasized as having importance in and of themselves are coded PH (WHICH OF YOUR ACTIVITIES ARE MOST IMPORTANT TO YOU? Exploring the city. WHY IS THAT IMPORTANT TO YOU? I'm trying to understand St. Louis, get an idea of what St. Louis is. Ifind that it's quite different than a lot of other cities. Coded P H and also A C and PS.) f. Reference to pets and toy animals are coded PH unless they are presented in relational terms. I love my cat—She's like my best friend is coded SO. 75 ACTIVE, AC Definition: Dynamic physical existence: behavior (broadly speaking) or potential for action (abilities). Note the distinction between physical activities (coded AC) and mental activities (coded PS, not AC) 1. Motion a) Physical activities (play soccer, coded AC), but not mental activities (thinking in circles, learning; both coded PS) b) General activity styles or levels that have clear physical implications (Iget a lot done; I have a pretty high energy level) c) Physical habits (biting nails) but not mental habits (ruminating) 2. Persistent Activities a) Occupation or studies (getting a degree; going to class; working at the liquor store) b) Hobbies or sports c) Lifestyle defined behaviorally (I like to shop a lot) d) Passive experiences. (HOW DID YOU GET TO BE THE KIND OF PERSON THAT YOU ARE NOW? By the experiences that I had in my life. WHAT SORT OF EXPERIENCES? Oh, like when I was hit by a car.) e) Active experiences. (I definitely spend more time on my computer than anything else.) f) Roles and responsibilities defined by physical action (I have to take care of the rent every month and pay the electric bill) 3. Potential a) Abilities or Inabilities Notes and Distinctions: a. Activity styles of clear psychological origin (hardworking; learning an activity; organized; goal-oriented) are coded both AC and PS. b. Activities that could involve some psychological action are not coded PS unless the psychological action is referenced directly (I check my email 50 times a day is coded AC but not PS). Whereas, I can't stop thinking about checking my email references a psychological phenomenon (thinking) directly, and hence is coded AC and PS. c. If experiences are mentioned and emphasized without explication of their nature (AC, SO, PS, etc.), the default is that they are coded as AC. The noting of an experience in passing (i.e. not emphasized) is not coded AC (After growing up in Chicago...) d. For possessions that are important in relation to an activity, code both PH and AC (I really like my car. The busses here are terrible. Without my car, I don't know how I'd get to school.) e. Social activities are coded both SO and AC unless the activity is purely interactional. So, my brother and I read Moby Dick together is coded both SO and AC as reading is not interactional; whereas my brother and I chat a lot... we have a lot offun is coded only SO as chatting is interactional. Having fun is ambiguous in that it may entail non-interactional activities. However, schematically ambiguous notions are not coded (see introduction) f. The drawing of attention to or noting the (in)significance of something is not, in itself, coded AC, even if the language sounds active or behavioral (Ipretty much ignore my body is coded PH only; I'm more oriented towards friends than family is coded SO only; I pay attention to movies rather than theatre is coded PH only.) 76 SOCIAL, SO Definition: The self in relation to others. Attributes or aspects of the self that only make sense in social contexts 1. Extra-individual a) Relationships: family, friends (WHO ARE THE MOST SIGNIFICANT PEOPLE OR GROUPS IN YOUR LIFE? My parents.) b) Memberships and status. (I'm treasurer of the club. This qualifies as SO. If explicated in terms of tasks and so forth (e.g., I have to write cheques and balance the budget), also coded A C ; see AC2f) c) Comparison with others or social norms. Usually also coded PH, A C , or PS depending on the scheme of comparison (smarter than most, this being coded SO and PS), but not comparison against personal norms or standards (happier than I was last year, this only being coded PS) d) Interactions with (talking to friends; I do favors for people) or reactions from others (Sandra doesn't like me). 2. Intra-individual a) Personality characteristics relevant to social relations (friendly but not hardworking, unless defined interpersonal ly); also coded PS b) Style of interacting with others (shy, good communicator) c) Roles and responsibilities defined interpersonally (I take care of my sister. If followed with specific examples of activities that are involved, also coded AC.) d) Antisocial tendencies (7 keep to myself; I don't have any friends) Notes and Distinctions: a. Psychological action in relation to social contexts is coded both SO and PS (learning from others; attitudes towards women; I'm developing leadership skills). b. One's appearance with explicit reference to others is coded both P H and SO. (WHATARE THE FA VORITE THINGS YOU HA VE OR OWN? My clothes. WHY ARE YOUR CLOTHES IMPORTANT TO YOU? I need to feel that I look good or Ifind that the way that I interact with people is different—I don't talk as freely.) c. For possessions that are important in relation to social implications, code both PH and SO. (My cell phone is important to me. WHY? Because it lets me stay in touch with friends.) d. Activities performed in social contexts are coded A C and SO only if the activity has independent meaning (I talk a lot and my friends know that about me. Talking a lot (e.g., to oneself) can be done without the friends; I watch movies with my brother. Watching movies can be done without the brother.). In contrast, inherently social activities are coded SO but not A C (talking with friends. Talking with friends cannot be done without friends; caring about other people. Caring for others makes no sense without people.). e. Explicitly social language used as a proxy for the first person is not coded SO. (You 're responsible for contributing your part for a global environment. So you just do what you can. I start by recycling. This would be coded both A C and PS.) f. If experiences are mentioned and emphasized without explication of their nature (AC, SO, PS, etc.), the default is that they are coded as A C . g. Reference to pets and toy animals are coded PH unless they are presented in relational terms. I love my cat—She's like my best friend is coded SO. 77 PSYCHOLOGICAL, PS Definition: Direct reference to intrapsychic existence and functioning. Note the distinction between physical activities (coded AC) and mental activities (coded PS) Capabilities a) Memory b) Intelligence c) Attentiveness Cumulative Features a) Knowledge (learned a lot) b) Memories c) Mental habits (thinking in circles, ruminating) Temporary Aspects and States a) Moods, feelings, emotional dispositions (I'd describe myself as a happy, content person.) b) Intrapsychic action (think a lot, learning, making a decision) Orientations (if described in systematic terms and as affecting change (of the physical, social, or intrapsychic variety), also coded AG) a) Beliefs (religious, existential, relativist) b) Values including lifestyle defined psychologically (I try to be calm and reflective in the way I go about things; I'm materialistic: I have my wants and desires.) c) Personal Philosophy d) Epistemology (process of understanding the world) e) Attitudes (I have an pretty bad attitude about having to go to school; also coded A C . goals) Notes and Distinctions: a. Interpersonal comparisons of psychological aspects are coded both PS and SO. (I'm more driven than any of my siblings.) b. Psychological action in relation to social contexts is coded both SO and PS (learning from others; attitudes towards women; Hove my brother; I'm developing leadership skills). c. Activity styles of clear psychological origin (hardworking; learning an activity; organized; goal-oriented) are coded both A C and PS. d. Expressing a preference for something is not coded PS (I would rather play soccer than volleyball but it's the off-season for soccer right now is coded A C only; I just like interacting with people is coded SO only). To qualify as PS, there must be more talk of the intrapsychic process (e.g., related emotions). e. The drawing of attention to or noting the (in)significance of something is not, in itself, coded PS, even if the language sounds psychological (I care about my body; Hove skiing). f. If experiences are mentioned and emphasized without explication of their nature (AC, SO, PS, etc.), the default is that they are coded as A C . g. Activities that could involve some psychological action are not coded PS unless the psychological action is referenced directly (I check my email 50 times a day is coded A C but not PS). Whereas, I can't stop thinking about checking my email references a psychological phenomenon (thinking) directly, and hence is coded A C and PS. 78 A G E N C Y , A G Definition: Implicitly or explicitly seeing oneself as an agent: Volitional action of the self that affects actual change of any kind, be it change of the physical, active, social, or intrapsychic nature. This definition is more expansive than and diverges from the notion on agency concerning the formation or existence of the self. That is, this definition does not require change to affect oneself but it does require the I-self to be the source of the change. There must be explicit (or strongly implicit) indication that the volitional action of the self causes the change. 1. Deliberate efforts for change (I've made adjustments at work to try to find some balance in my life — more time for tennis. This example is also coded AC.) 2. Desire or motivation for change (I have a lot that I want to accomplish. LIKE WHA T? Right now I just decided to do a double major... and I want to have a really full social life. This example is also coded A C (doing a double major) and SO (social life).). 3. Personal or moral evaluations of life possibilities that affect change (I'm reflective. WHY ARE YOU REFLECTIVE? During my life, I have met different kinds of people, and I talked with them. And I found out that different persons will go through different stages. And if you go through these stages without thinking before, or thinking after, you will get nothing in your life. And so I started to be observing and to think about what I really want. Here the personal evaluation is expressed in the first two sentences while the change is expressed in the third as starting to be observing. This is also coded SO (meeting different kinds of people) and PS (thinking, learning).) Notes and Distinctions: a. Necessary criteria checklist (must satisfy all 5) o Is there volition for change? o Is it explicit or strongly implicit? o Does it come from the self? o Does it necessarily cause change (and not the other way around)? o Is the change actual as not only desired? b. Changes arising out of social relationships (parents) are coded A G if and only if the social relationship causes change in some intrapersonal aspect which, in turn, affects change. That is, the self must mediate the change. c. In that agency entails change of some sort, the specific nature of the change is always also coded (PH, A C , SO, or PS). Hence, all chunks coded A G are also coded in at least one other scheme. d. The mere expression of a desired state does not, in itself, qualify as A G . There must be reference to actual (and opposed to desired) resultant change. I want to have some of the personality characteristics of my grandfather fails in this regard. But if followed with and so I'm trying to become more dedicated like he was, it would then qualify as A G . Had the follow-up been and so I'd like to become more dedicated like he was, it would not qualify as A G as the action is framed as desired and not actual. Note that this example is also coded PS and SO. e. Change affected directly by talents, abilities, and personality characteristics do not qualify as A G (Neither of these examples qualify as A G : If it weren't for me being outgoing, I never would have gone into Engineering; I'm really good at Math and so I won the state-wide competition). Change must originate from a subjective and volitional core. f. If the relationship between volition and action may be reversed, it does not qualify as A G . I want to live every day to its fullest. WHY? You just never know what might happen or how long you 11 live. I was once in a motorcycle accident and nearly died. I had a lot of time to myself to think about this while recovering... to decide what I wanted to do with my life. 79 CONTINUITY, CO Definition: referencing one's own connection through time. Defending one's continuity based on those aspects of the self that are connected to or continuous with the past or future. Note that many individuals deny the question. In these cases, if the interviewer does record a stem, code for presence of the source of their change (e.g., personality changes PS, body changes PH). 1. Physical. The selfs continuity over time is defended by reference to externally observable physical characteristics. Also coded PH. (GIVEN THAT YOU CHANGE FROM YEAR TO YEAR, HOW DO YOU KNOW IT'S STILL AL WA YS YOU? I've always had the exact same genes.) a) Corporeal existence or properties b) Possessions 2. Active. The selfs continuity over time is defended behaviorally. Also coded AC. (GIVEN THAT YOU CHANGE FROM YEAR TO YEAR, HOW DO YOU KNOW IT'S STILL ALWA YS YOU? I know it's Still me because I've always had the same talents... like being an excellent violin player.) a) Behaviors or behavioral tendencies b) Abilities c) Activities 3. Social. The selfs continuity is defended in relation to interactions with others. Also coded SO. GIVEN THAT YOU CHANGE FROM YEAR TO YEAR, HOW DO YOU KNOW IT'S STILL ALWAYS YOU? I Still know it's me because I still respect my family. a) Recognition from others (note: relationships are not coded CO) . b) Style of interacting with others 4. Psychological. The selfs continuity is defended in relation to persisting psychological attributes. Also coded PS. GIVEN THAT YOU CHANGE FROM YEAR TO YEAR, HOW DO YOU KNOW IT'S STILL ALWAYS YOU? I think the usual philosophical answer is memory. But I think it's more of an emotion and continuity of consciousness. a) Preferences b) Stable knowledge or memory c) Cognitive capabilities d) Beliefs, values, goals e) World-view or epistemology f) Spirit or conscience Notes and Distinctions: a. Referencing a connection with others or anticipated/recalled activities do not qualify as CO as it refers to the continuity of a particular relationship/activity, and not one's own connection through time. b. The denial of continuity precludes coding as CO. If anything, coded elsewhere. c. The restating or paraphrasing of the question is not coded (I think that it's still me because my parents tell me that I'm the same is coded CO and SO. But thinking that it's still me is not coded PS.). Note that a restatement of the question can come at the end of the explication as well (Always being a great soccer player reminds me that it's still me is coded CO and AC. But being reminded is not coded PS.) 80 DISTINCTNESS, DI Definition: The sense that one's self is unique or different from all other selves. This sense of distinctness can entail a single or combination of subjective (values) or objective (genetics) features. 1. Physical attributes and possessions; also coded PH (genetics, my face) 2. Abi lities or activities; also coded AC. (HOW DO YOU KNOW THA T YOU ARE UNIQUE OR DIFFERENT FROM EVERYBODY ELSE IN THE WORLD WHEN THERE ARE MANY PEOPLE THAT YOU DON'T KNOW? I'm unique only in that I've experienced different sort of things than other people. I think that when it comes down to it, a lot of people are similar to me. But we've taken different experiences, different routes. That's what makes us different.) 3. Social relationships; also coded SO (7 am the son of my parents. If someone were exactly like me, they would have to have been born to my parents at the exact same minute. That didn't happen.) 4. Psychological; also coded PS (conscience/spirit,personality, intelligence) Note and Distinctions: a. The nature of the question suggests social comparison. To also qualify as SO, explicit social characteristics or relations must act as the source of the sense of distinctness. Mentioning that one is different from others is taken as merely restating the question. (I'm unique only in that I've experienced different sort of things than other people is coded DI and AC but not SO. Whereas I'm unique because of a combination of things. LIKE WHAT? People that I've met. There was this one mentor that I had... is coded DI and SO.) b. Denial of the distinctness of the self precludes coding as DI. Possibly codable elsewhere. 81 \ Appendix K - Schwartz' Values Typology (Schwartz, 1992) • Power: social status and prestige, control or dominance over people and resources (social power, wealth, authority, preserving public image). • Achievement: personal success through demonstrating competence according to social standards (successful, capable, ambitious) • Hedonism: pleasure or sensuous gratification for oneself (pleasure, enjoying life) • Stimulation: excitement, novelty, and challenge in life (daring, a varied life, an exciting life) • Self-direction: independent thought and action—choosing, creating, exploring (creativity, freedom, curious, independent, choosing own goals) • Universalism: understanding, appreciation, tolerance, and protection for the welfare of all people and for nature (social justice, broadminded, world at peace, wisdom, a world of beauty, unity with nature, protecting the environment, equality) • Benevolence: preservation and enhancement of the welfare of people with whom one is in frequent personal contact (helpful, forgiving, honest, loyal) • Tradition: respect, commitment, and acceptance of the customs and ideas that traditional culture or religion impose on the self (accepting of my portion in life, devout, respect for tradition, humble, moderate) • Conformity: restraint of actions, inclinations, and impulses likely to upset or harm others and violate social expectations or norms (obedient, self-discipline, politeness, honoring parents and elders) • Security: safety, harmony, and stability of society, or relationships, and of self (family security, national security, social order, clean, reciprocal favors, sense of belonging). 82 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items