THS Date In'ilii'ii'iliiiiiii‘l‘iifli‘l‘l‘l’iiiiiiil 3 1293 00791 3928 This is to certify that the thesis entitled The Socialization of Affect: Implications of Parental Ideology and Authority Style For Self-Evaluations in College Students presented by Marilyn Bleiweiss Charles has been accepted towards fulfillment of the requirements for ML. degree in 2.512119% Major professor : Susan Frank, Ph.D. JUne, 18, 1992 O—7639 MS U is an Affirmative Action/Equal Opportunity Institution fl LIBRARY Michigan State University -\ A PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. —~_\ DATE DUE DATE DUE DATE DUE i372 Li ()1 ‘ Q-.. -55 h I! i‘ MSU Is An Affirmative Action/Equal Opportunity Institution “dummies THE SOCIALIZATION OF AFFECT: IMPLICATIONS OF PARENTAL IDEOLOGY AND AUTHORITY STYLE FOR SELF-EVALUATIONS IN COLLEGE STUDENTS B y Marilyn Bleiweiss Charles A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1992 ABSTRACT THE SOCIALIZATION OF AFFECT: IMPLICATIONS OF PARENTAL IDEOLOGY AND AUTHORITY STYLE FOR SELF-EVALUATION S IN COLLEGE STUDENTS By Marilyn Bleiweiss Charles There are numerous suggestions in the literature that parents' values affect their children's psychological well-being. However, the mechanism for this transmission is not clear. In this study, questionnaire data from 51 undergraduates from intact families were analyzed to evaluate the association between parental ideologies (normative, humanistic, or "balanced") and late adolescent children's self-evaluative experiences, and to assess whether this relationship was mediated by differences in the parents' parenting styles. Factor analysis of diverse measures of affective self- evaluations resulted in four primary factors describing individual differences in Shame (vs. self-esteem and mastery), Pride, Guilt, and Responsibility (prosocial guilt vs. externalization of blame and detachment/unconcern). Children of fathers describing themselves as authoritative reported greater prosocial guilt, whereas children of mothers describing themselves as permissive reported less prosocial guilt. Results of multiple regression analyses further suggest that parenting style mediated the relationship between parental ideology and adolescent self-evaluations. In memory of Justin Daniel Charles (January 18, 1980 - February 24, 1983) ACKNOWLEDGMENTS I acknowledge with gratitude the ongoing assistance and encouragement of my chairperson, Susan Frank, in clarifying the conceptualization, and facilitating the realization, of this work. I also offer sincere thanks to the other members of my committee, John Hurley and Ellen Strommen, for their guidance, optimism, and encouragement. Finally, I would like to thank my family, my greatest sources of support, inspiration, and joy, especially Bruce, Devon, and Jonathan, who have accommodated to this endeavour with resilience, grace, and humour. TABLE OF CONTENTS Page LIST OF TABLES .......................................................................................................... vi CHAPTER 1 - Introduction and Review of the Literature .......................... 1 Introduction ............................................................................................................ 1 The Role of Ideology in the Socialization Process ................................... 1 The Role of Affect in Organizing Experience ............................................. 2 The Model ................................................................................................................ 5 Parental Authority Styles ................................................................................. 7 CHAPTER 2 - Method ............................................................................................... 14 Subjects .................................................................................................................. 14 Measures ............................................................................................................... 14 Demographic Information ......................................................................... 14 Parental Ideology .......................................................................................... 14 Parental Authority Style ............................................................................ 15 Affective Self-Evaluations ......................................................................... 16 Procedure .............................................................................................................. 20 Predictions ............................................................................................................ 20 CHAPTER 3 - Results ................................................................................................ 22 Hypothesis 1: Parental Ideologies and Adolescents' Self- Evaluations ........................................................................................................... 27 Hypothesis 11: Parental Ideologies and Authority Styles ............... 27 Hypothesis III: Parental Authority Styles and Adolescents' Self-Evaluations .................................................................................................. 29 Hypothesis IV: Mediated Effects ................................................................ 29 CHAPTER 4 - Discussion....................... ................................................................... 33 REFERENCES .................................................................................................................. 39 APPENDIX A. Tables ............................................................................................................ 46 1A. 2A. LIST OF TABLES Page Factor analysis of affective self-evaluations ..................................... 24 Factor analysis of Factor 1 ........................................................................ 24 Demographic characteristics of the sample (N=51) ........................ 26 Parental ideology and late adolescents' self-evaluations: Simple correlations and beta coefficients (N=51) ........................... 28 Parental ideology and parental authority styles: Simple correlations and beta coefficients (N=51) ........................... 30 Parental authority styles and late adolescents' self- evaluations: Simple correlations and beta coefficients (N=51) ............................................................................................................... 31 Beta coefficients for predictors of parents' authority styles for total sample: Correct (C) vs. incorrect (I) instructions (N=27l) .............................................................................................................. 46 Beta coefficients for predictors of adolescents' self- evaluations for total sample: Correct (C) vs. incorrect (I) instructions (N =271) .................................................................................... 47 vii Tomkins (1963, 1965, 1979) proposes that the ideology of the parent is an important determinant in child socialization. He sees a dichotomy in our society between the "humanist": who values the human being foremost, as the measure of all things, and the "normative": who values the human being negatively, finding rules and restrictions to be essential for keeping the baser instincts in check. These orientations reflect basic ideologies with consequences in all domains of life, including how and which affects are valued and therefore maximized, minimized, or inhibited in the socialization of children. In the present study I attempt to demonstrate an association between parental ideologies and children's self- evaluative experiences (i.e. experiences of shame, guilt, and self- esteem). I also attempt to show that the relationship between parental ideologies and children's self-evaluations is largely mediated by differences in the parents' parenting styles. Tomkins argues that children of parents with differing ideologies will differ in their affective experiences (1987a). The humanist attempts to maximize the positive affects of enjoyment, interest, and surprise, while for the normative individual, norm compliance is the primary goal. From the normative position, the negative affects, such as anger, shame, and distress, are an integral part of existence, and must be controlled and/or suppressed (as in the case of anger), or endured (as in the case of shame and guilt). Suppression of affect can result in ambiguity; the child becomes less able to discern what s/he is in fact feeling. This results in further deficits in the child's ability to cope with life events. Guilt and shame may be actively encouraged by the normative parent in order to control the child's behaviour. For the humanist, on the other hand, the negative affects are to be minimized or avoided. The humanist parent, in attempting to allow the child to develop freely and make his/her own decisions, tacitly affirms a belief in the child's ability to cope with life events. However, because of a lack of structure and little emphasis on problem-solving, failures in coping are likely to ensue. Guilt and shame may result from either the child's failures to live up to vague or developmentally inappropriate parental expectations, or from the child's difficulties in dealing with challenging environmental demands outside the family context. h l f A ' n'i Ex 'nc Tomkins' script theory (1979; 1987a) provides a means for understanding how ideology can help to organize the individual parent's perceptions and memories of experiences and influence their socialization attitudes and behaviours. In Tomkins' view, ideologies can be reduced to a basic unit of experience, referred to as a "scene". Events are organized in memory as scenes, with the affect attendant to them playing a critical role in their salience and organization. The critical role of affect in memory storage (Loftus & Loftus, 1980), recall (Bower, 1981; Forgas & Bower, 1987), and organization (Isen, Daubman, & Gorgoglione, 1984; Moore, Underwood, & Rosenhan, 1984; Tomkins, 1963) has been amply demonstrated by studies of cognitive processes. Along with its role in organizing memories, affect also amplifies experience. Both type and intensity of mood appear to be critical factors in what and how well we remember (Gilligan & Bower, 1984). Intense affect amplifies the scene, making it more salient. Scenes in which the attendant affect is similar are likely to become grouped together. "Scripts" provide the rules whereby the scenes are interconnected or fused in a meaningful way. Although scripts are originally derived from scenes, they eventually come to guide the construction of later scenes. Because scripts tend to be selective, incomplete, and to some extent inaccurate, memory involves a reconstruction of facts that is based on overriding beliefs and expectations. While there are general mechanisms for script construction, and general classes of scripts, the actual scripts that individuals form are idiosyncratic. Tomkins suggests that ideological scripts are the most important class of scripts because "they endow fact with value and affect" (1987a, p.170). Over time, ideological scripts not only guide interpretations as to the important dimensions of particular types of scenes; they also provide rules for facilitating the experience of the positive affect associated with positive scenes, and for avoiding the negative affect associated with negative scenes. Tomkins (1987a) suggests that the processes involved in the magnification of positively toned and negatively toned scenes are generally quite different, and may be more or less prevalent, 4 depending on an individual's ideology. The humanist is more likely to rely on and amplify positive scenes in organizing their experience by finding variants, or detecting "differences around a stable core" (Carlson, 1981, p. 503). For example, one might become interested in a person who has attributes similar to those of a good friend. In this way, the individual expands their experiences. On the other hand, the normative individual is more likely to organize their experience in terms of avoiding negative affect by looking for analogs which are similar to previous experiences. In this way, the normative individual's focus may limit their experience. The avoidance of negative affect is facilitated by becoming aware of familiar cues within novel contexts which might signal danger. The greater the density of negative affect in one's life, the more imperative this avoidance becomes. Avoidance is unique in that it is facilitated by vigilance to cues associated with the object of the avoidance. Therefore, the focus is on the object of the avoidance. Ironically, this focus perpetuates the experiences of internal affective and cognitive stimuli associated with that object or event, and hence maintains negative affects such as fear or shame. In sum, ideological positions are associated with a differential emphasis on constructing positive versus negative scripts. The humanist is more likely to seek out variants, in order to enhance experiences of positive affect, whereas the normative is more likely to search for analogs, in order to fend off experiences of negative affect. Overreliance on either process may be nonadaptive. The humanist may rely too heavily on the pursuit of positive affect, which might result in deficits in the ability to attend to, prepare for, and cope with negative experiences. This could lead to failures in handling difficult situations and leave the individual feeling less competent, thereby increasing shame and reducing self-esteem. The normative individual, on the other hand, may rely too heavily on avoidance behaviours, with a concommitant reduction in efforts to find positive experiences. Negative affect scripts are more likely to preclude new and contradictory information than positive affect scripts, and therefore to become self-perpetuating (Hayes, Zettle, & Rosenfarb, 1989; Levis, 1989). T M 1 Parents' ideological assumptions may have far-reaching implications, not only for their own affective experiences, but also for their parenting behaviours and their children's emotional development. In particular, parental ideology may be important in determining children's experiences of positive and negative affect, especially the affects associated with self-evaluation. The empirical evidence which supports these links is primarily indirect. Tomkins (1965) has found that the humanist is more likely to anticipate positive contacts with others, while the normative is more likely to anticipate negative contacts with others. In addition, the humanist is more likely to notice and respond to positive affect, while the normative is more likely to notice and respond to negative affect (Tomkins, 1965). In this way, ideology may be a selection factor in the experiences to which a child is exposed (Carlson & Brincka, 1987). Furthermore, ideology has been found to affect how a an individual is taught to interpret their experience (Carlson, 1982), as well as the attributions he or she makes about themselves as a result (Kaufman, 1989). The present study will attempt to empirically demonstrate the link between normative and humanistic ideologies and children's self—evaluations. In particular, I hypothesize that both ideologies will be associated with relatively negative self-evaluations, although for different reasons. While the child of humanistic parents may have inadequate guidance to allow for mastery experiences, the child of normative parents may have inadequate freedom to allow for those experiences. The experience of mastery appears to be a crucial factor in positive self-evaluations (Pearlin, Menaghan, Lieberman, & Mullan, 1981). I hypothesize that a mixed ideology which combines elements of both the humanistic and normative orientations, on the other hand, will be associated with more positive self-evaluations. For example, Tomkins (in press) argues that when a child has parents of opposing ideologies, attempts by the child to reconcile these positions should enhance self-awareness. Also, attempts on the part of an individual parent to incorporate the two ideological positions Should temper them. In either case, these less polarized positions Should result in more experiences of mastery and more positive self- evaluations for the child. Finally, an important hypothesis being tested here is that in addition to its direct effects, an indirect way in which parental ideologies influence children's self-evaluations is via the impact on parents' parenting styles. EmmaLAutbeLHy—Sjlles Baumrind (1966, 1971) distinguished between three parental authority styles: authoritarian, permissive, and authoritative. The authoritarian style is defined as repressive and inhibiting, whereas the permissive style is characterized by little control, with most authority being left to the child. The authoritative style is construed as firm and rational, with a focus on active problem solving. Baumrind suggests that the authoritative parent "encourages verbal give and take, shares with the child the reasoning behind her policy, and solicits his objections when he refuses to conform" (1966, p. 891). Reason and modelling are both used as means for shaping the child's behaviour, without relinquishing parental power and authority. Baumrind found the authoritative style to be most beneficial to the child. Parental authority style has been found to be a critical factor in the development of self-esteem in children (Demo, Small, & Savin- Williams, 1987; Gecas & Schwalbe, 1986; Harris & Howard, 1984), as well as an important factor in adolescent development more generally (Fauber, Forehand, Thomas, & Wierson, 1990). However, while there is some empirical support for a link between parenting styles and children's self-evaluations (Baumrind, 1966,1971; Buri, Louiselle, Misukanis, & Mueller, 1988; Hoffman, 1970, 1977; Stierlin, 1974), there is little empirical data elucidating the links between parental ideology and styles of parenting. Luster, Rhoades, & Haas (1989) did find that parental values - such as the extent to which conformity or self-direction are valued in the child - do help to determine which aspects of a child's behaviour are considered important, as well as prescribing how behaviour should be shaped and/or managed. While their findings support the importance of parental values in determining factors relating to authority style, the role of humanistic versus normative parental ideologies in forming different parenting orientations has yet to be studied. It is likely that the three parental authority styles described by Baumrind (1966, 1971) are related to the three types of ideological orientations depicted by Tomkins (1963, 1965, 1979). The humanist values the person foremost, much as the permissive parent focuses on the expressed needs and desires of the child. The person with a normative ideology values rules and control, consistent with the focus of the authoritarian parent on rules and obedience. A less polarized position should be associated with parenting oriented toward taking both the child's needs and the constraints of the situation into account. From the mixed ideological position, both the expressed needs of the child and the needs of others, including societal and familial rules, are valued. This ideology is more conceptually consistent with the authoritative style, in which the parent exercises control while being sensitive to the effects of that control on the child's own growth. In short, parenting style may be the mechanism via which parental ideologies impact on child outcomes, especially self- evaluations. For the humanist, it may be most important for the child to grow up feeling good about herself. Rules may be seen as less important than providing opportunities for the child's enjoyment and self-expression. The humanist is most likely to use a permissive style in an attempt to provide experiences of positive affect, and to avoid negative affect. However, this style may encourage a child to focus exclusively on pleasure, resulting in diminished enjoyment over time due to habituation. Furthermore, this style may not encourage a child to develop strategies for dealing with negative affect, thereby leaving the child more vulnerable in the long run. Optimal parenting appears to provide strategies that facilitate the child's ability to actively seek out pleasant experiences as well as to minimize discomfort (Baumrind, 1966). Insufficient guidance and structure, along with inadequate attempts to teach and model problem-solving skills, may actually impede the child's attempts to make sense of, and be effective in her environment. Consequently, the child may perceive herself as unable to meet unclear parental expectations or generally inadequate to life tasks. The normative parent is more likely to take an assertive role in parenting and to exhibit an authoritarian style. This individual might focus on obedience and conformity in an attempt to prepare the child for the harsh realities of existence. From this framework, it may be as important to learn to exercise moderation with regard to 10 positive affect, as it is to control, suppress, and tolerate negative affect. Contempt and shame are maximized from this orientation, and are deemed to be appropriate responses to extremes of positive or negative affect. Affective expressions such as laughing, crying, or flairs of temper would tend to be condemned as evidence of loss of control from the normative position, while accepted as a normal part of childhood from the humanist position. Each of these positions encourages an overly narrow affective focus, which can impede emotional development (Kaufman, 1989; Solomon, 1980; Tomkins, 1987a). Parents with a permissive or authoritarian style may not actively teach the child how to conceptualize problems in constructive ways. Rather, dissent would tend to be suppressed by authoritarian parents and bypassed by permissive parents. The child of the humanistic, and presumably overly permissive, parent who has not been taught to attend to and work through difficult experiences, is likely to be overwhelmed with negative affect. This may encourage negative self-evaluations as well as negative expectations more generally. On the other hand, the child of the normative, and presumably authoritarian, parent who is critical of the expression of strong affect, may learn that it is wrong to feel it, as well. Power assertion techniques, such as those often associated with the authoritarian style, impede the child's ability to acknowledge and tolerate, and therefore to self-regulate and effectively recover from intense affective experiences. Without the ability to self-regulate affect 11 which is so important to healthy development (Kobak, 1992), the child is easily overwhelmed, and more likely to utilize avoidant strategies (Hoffman, 1982). High levels of parental dominance also encourage avoidant and repressing strategies, such as externalization of blame (Hoffman, 1982), detachment, stoicism, and the inhibition of negative affect (Kobak, Cole, Ferenz-Gillies, Fleming, & Gamble, 1992). Critical parental attitudes, such as those often associated with the authoritarian style, encourage feelings of helplessness and discourage further attempts at mastery (Stierlin, Levi, & Savard, 1974). In the extreme case, the major concern is to avoid pain, and pleasure may be seen only as a relief from discomfort (Solomon, 1980). Normative and humanistic ideologies may both be more constructive when they temper one another and allow for a more authoritative parenting style, characterized by flexibility, guidance, and age-appropriate expectations. To be able to experience both pleasure and pain, and to be effective in accomodating to both of these kinds of experiences, it is important to have strategies for actively dealing with discomfort, as well as strategies that seek pleasure through proactive mastery. The authoritative style may be most beneficial to children because they have an opportunity to experience dissent and resolve it (Baumrind, 1966). Parents who engage their children in active problem-solving experiences give them the opportunity to experience the actual consequences of their actions, thereby facilitating their ability to make realistic appraisals and to be 12 effective in further actions (Hayes, 1987; Hayes, Brownstein, Zettle, Rosenfarb, & Korn, 1986). Experiences of mastery appear to serve both facilitative and protective functions. While enhancing the child's potential for being effective in the environment (Buri, Louiselle, Misukanis, & Mueller, 1988) by encouraging persistence in the face of adversity (Maier, Laudenslager, & Ryan, 1985), they tend to also immunize children against later failures. In sum, the ideological positions of parents appear to have pervasive effects on their children. To a large extent the effects of these ideologies on children's self—evaluations may be mediated by the parent's authority style. From previous research on parenting styles (Baumrind, 1966, 1971) we can predict that self-evaluations of children will be most negative for children of authoritarian parents and most positive for children of authoritative parents, with children of permissive parents falling in between. In the current study I addressed these issues with late adolescents. In adolescence, the emerging cognitive capacity for formal operations is accompanied by an increased emphasis on self- initiated, goal-directed behaviours rather than behaviours which are motivated by identiifications with parental values and the desire to avoid feelings of guilt and Shame (Josselson, 1980). The perceived ability to both effectively manage one's external environment, and to regulate internal affective experiences may be of particular importance to self-evaluations. There is an increasing focus on self- directedness through internalization of values, controls, and self- 13 esteem; Josselson suggests that adolescents begin to develop the capacity to internally regulate self-esteem as they give up beliefs in omnipotent others, or "deidealize" their parents. Although deidealization increases adolescents' sense of autonomy (Frank, Pirsch, & Wright, 1990), it also makes them vulnerable to experiences of shame and guilt (Blos, 1967). Differences in parental ideology and parenting styles may exacerbate or buffer against this developmental vulnerability. Method Subjects The subjects were 119 male and 152 female students between the ages of 18 and 22 at a large midwestern university, as well as their parents. Most were white (96%), from middle to upper-middle class intact families. They were recruited from introductory psychology courses, and received credit for their participation. They were informed that participation in this study necessitated the participation of their parents as well. Measures WW Demographic material was collected from both students and their parents, including age, religion, social status, and ethnic background. 11mm Tomkins' (1987b) Polarity Scale was used to assess parents' ideo- affective postures. The scale consists of 59 pairs of items in which either, neither, or both of the ideas in the pair can be endorsed. Items sample beliefs about human nature, the nature of reality, child rearing, and emotional expression, among others. Since the choices are seen as independent, the normative and humanistic scores are often calculated independently. In this study, however, scoring will be based on the percentage of humanistic responses, such that H% = 1L X 100, as suggested by Stone and Schaffner N+H 14 15 (1988). A high score is indicative of a humanistic ideology, whereas a low score is indicative of a normative ideology. Tomkins' early work supported the theoretical distinction between these two orientations. For example, he found that humanistic individuals are more likely to hold interpersonal values, whereas normative individuals are more likely to favour individualistic values (Tomkins, 1965). This measure was also validated by later studies looking at relationships between ideology and affect (Carlson & Brincka, 1987; Stone, 1986). For example, in a review of the literature, Stone reported that ideology, as defined by the Polarity scale, predicted facial affect, sociophilia, and the tendency to blame victims. Although there has been much theoretical speculation as to the effects of ideology on actual parenting behaviours, this measure has not been explicitly tested in regard to parenting style. The reliability coefficient (alpha) for this scale was .86 in the current study. Pategtgl Authority Style The Parental Authority Questionnaire (PAQ: Buri, 1989) was given to each parent. It consists of 10 permissive, 10 authoritarian, and 10 authoritative statements, each in a 5-point Likert format, based on Baumrind's (1966) delineation of these authority styles. The items are measured on a 5-point scale from 1 = n l ' r to 5 = w. Test-retest reliabilities have been found to be from .77 (father's permissiveness) to .92 (father's authoritativeness), and Buri reported coefficient alphas of .74 (father's permissiveness) 16 to .87 (father's authoritarianism). This measure was face-validated using 21 professionals in education and the social sciences, who indicated whether they felt each item was clearly indicative of one of the three parenting styles. Items were only retained if more than 95% of the judges agreed (Buri). Studies by Buri and his colleagues together yield evidence of criterion-related validity; they found that both the authoritative style and parental nurturance predict higher self-esteem in adolescent children (Buri, Kirchner, & Walsh, 1988; Buri, Louiselle, Misukanis, & Mueller, 1988). This measure also has adequate discriminant-related validity; Buri found that authoritarianism was inversely related to permissiveness and authoritativeness, while permissiveness was not significantly related to authoritativeness. EEE . S IE-E l . Proneness to guilt and shame was measured using the Test of Self—Conscious Affect (TOSCA: Tangney, Wagner, & Gramzow, 1989), a revision of the Selt-ansgigus Affegt and Atttibntign layentgry (SCAAI: Tangney, Burggraf, Hamme, & Domingos, 1988). The TOSCA has four main subscales: Shame, Guilt, Externalization of blame, and Detachment/Unconcern. The TOSCA consists of 15 brief scenarios, 10 negatively valenced (e.g. "While out with a group of friends, you make fun of a friend who's not there"), and 5 positively valenced (e.g. "You and a group of co-workers worked very hard on a project. Your boss singles you out for a bonus because the project was such a success"). Four responses are offered for each negatively valenced l7 scenario. The options indicate shame, guilt, externalization of cause or blame, and detachment/unconcern. Five responses are offered for each positively valenced scenario, indicating shame, guilt, externalization, and two types of pride: those stemming from an evaluation of the entire self, and those stemming from an evaluation of a specific behaviour. Tangney (1990a) reports internal consistencies (Cronbach's alpha) of .46 to .82, and test-retest reliabilities of .71 to .79 for the four main subscales of the SCAAI. Because test-retest correlations and internal consistency were substantially lower for the two pride subscales, Tangney advocated interpreting results for these scales with caution. Preliminary analyses for the TOSCA show it to be at least equivalent to, and possibly superior to the SCAAI in terms of reliability and validity (Tangney, Wagner & Gramzow, 1990). In the current sample, alpha coefficients of .73 (shame), .63 (guilt) .64 (externalization of blame), and .68 (detachment) were found for the four main subscales. The reliability coefficient for the two pride scales combined was .68. Additional measures of self-evaluations were included in an Affect Inventory composed of items from the following five scales: the Rosenberg Self-Esteem Scale, the Mastery Scale, the Buss and Durkee Guilt Scale, Chang and Hunter's Guilt Scale, and the Internalized Shame Scale. All of these measures have high face validity. The Rosenberg Self-Esteem Scale and the Buss and Durkee Guilt Scale have been used in a large number of research studies. The Internalized Shame Scale and the Mastery Scale have each been 18 found to possess satisfactory psychometric properties (Cook, 1987; Marshall & Lang, 1990; Turner & Noh, 1988). The Chang and Hunter Guilt Scale was empirically derived. Items were chosen to be consistent with Friedman's (1985) definition of the construct as "the appraisal, conscious or unconscious of one's plans, thoughts, actions, etc. as damaging, through commission or omission, to someone for whom one feels responsible" (p. 529). The scale was validated in an empirical study, in which Chang (1988) found correlations of .52 between guilt and shame, and correlations of .65 between guilt and fear of exposure. Self-esteem was measured with the Rosenberg Self-Esteem Scale (Rosenberg, 1965) which is of general use as a measure of self- esteem. Items are endorsed on a 5-point scale from 1 = 51.129.11.211 m to 5 = W. Items from the scale include "on the whole, I am satisfied with myself", and "I certainly feel useless at times". Alpha, a measure of internal reliability, was .86 for the present sample. Mastery was assessed using the 7 Mastery Scale items from the Self Attitude Questionnaire as described in Pearlin, Menaghan, Lieberman, and Mullan (1981). Items are endorsed on a 5-point scale from 1 = W to 5 = W. Items from this scale include "what happens to me in the future mostly depends on me", and "there is little I can do to change many of the important things in my life". This scale was statistically derived. Pearlin and 19 his colleagues reported test-retest reliabilities of .44, and internal reliability of .78. In the current sample, alpha was .78. The Buss and Durkee Guilt Scale (1957) and Chang and Hunter's Guilt Scale (Chang, 1988) provided additional measures. Global guilt was measured by items from the Buss and Durkee Guilt Scale, as revised by Chang (1988). The Modified Buss and Durkee Guilt Scale consists of 9 items which have been found to measure guilt most specifically, eliminating items which have been found to measure other constructs. The Chang and Hunter Guilt Scale consists of 9 items based on a definition of guilt in terms of perceptions of self as causing harm to others and efforts toward making reparations when harm has been done (Chang). Items include "Sometimes I cannot forgive myself for having caused deep pain in those I love or care '1 for . Items from both guilt scales are rated on a 5-point scale from 1 = strengly disagree to 5 = strongly agree. These measures have been found to have adequate internal consistency (Chang, 1988). Internal consistency for the composite of these two guilt scales combined was .84 for the present sample. Global shame was measured by items from the Internalized Shame Scale (188: Cook, 1985) as selected by Chang and Hunter (Chang, 1988). The revised Shame Scale consists of the 11 items which have been found to measure shame (e.g. "I feel like I am never quite good enough"), eliminating those which have been found to measure other constructs (Chang). The items are negatively worded, and are rated on a 5-point scale ranging from geyer (1) to 20 almest always (5). Alpha, a measure of internal reliability, was .94 for the current sample. Ereeeeute Participants were recruited from introductory psychology classes, and advised that participation in this study entailed the participation of their parents as well. They were first informed as to the nature of the study and their rights as subjects. They were then asked to fill out questionnaire data, and received extra credit for their participation. Participants were also asked to arrange to have their parents complete parental questionnaire data.» To ensure anonymity, all data was identified by code numbers only, and students received credit only for those parental forms which were returned in sealed envelopes which had been initialed by the parent. Parents were given the option of returning their forms by mail to further ensure confidentiality. Students received additional credit for their parents' participation. E 1° . Predictions were as follows: 1. Parental ideology will be related to late adolescent childrens' self- evaluations in the following ways: Parental ideology will have a curvilinear relation to adolescents' self-evaluations; more extreme parental ideology scores (defined as the quadratic nonlinear component) will be associated with more negative self-evaluations, and more moderate, or balanced, parental ideology scores will be associated with more positive self-evaluations. 21 2. Parental ideology will be related to parents' descriptions of their parenting styles in the following ways: a) More moderate, or balanced, parental ideology scores will be associated with an authoritative style b) More normative parental ideology scores will be associated with an authoritarian parenting style. 0) More humanistic parental ideology scores will be associated with a permissive parenting style. 3. Parental authority styles will be linked to their adolescents' self- evaluations in the following ways: a) the authoritative parenting style will be associated with more positive self-evaluations b) the authoritarian style will be most strongly associated with negative self-evaluations. c) the permissive style will also be linked to more negative self-evaluations than the authoritative style, although this relationship may be somewhat weaker than that for the authoritarian style. 4. Parental authority styles largely mediate the relationship between parental ideology and adolescent self-evaluations. In other words, controlling for parenting style will decrease the magnitude of the relationship between ideology and adolescents' self-evaluations. Results Parents' ideology was defined by the relative humanism (vs. normative stance) of the parent, and the relative polarity of the ideological position. Moderate scores were considered indicative of a mixed ideology; extreme scores were considered indicative of a polarized position. The effect of a mixed ideology on each of the dependent variables was assessed by testing the significance of the quadratic trend, after controlling for the main effect (Cohen, 1978); in other words, I tested whether the quadratic nonlinear component added significant variance to the model after the linear association had been tested. Positive beta weights for the quadratic term would indicate a positive relationship between extreme ideologies and the outcome variable, and a negative beta weight would indicate an association between a mixed ideology and an outcome variable. Mothers' and fathers' ideology and parenting style scores were computed separately in order to determine their relative importance in predicting scores of their children. The scores from the TOSCA and the scales in the Affect Inventory were subjected to a factor analysis using a varimax rotation to reduce the data to a smaller number of self-evaluation dimensions. This identified three primary factors (See Table l). Shame and guilt both loaded on the first factor. Because shame and guilt are highly correlated, but have been found to be separate constructs (Chang, 1988; Tangney, 1990a), the first factor was subjected to a further factor analysis (see Table 2). This second 22 23 factor analysis resulted in four primary factors describing individual differences in Shame (Factor 1A), Pride (Factor 2), and Guilt (Factor 1B). The fourth factor described individual differences in prosocial guilt. I inverted this scale to consist of positive loadings for the TOSCA measures of prosocial guilt (i.e. guilt associated with attempts toward making retribution and/or maintaining relationships) and negative loadings for the TOSCA measures of externalization of blame and detachment/unconcern, and labeled it "Responsibility." Both self-esteem and mastery loaded on the shame factor. Internal reliability (Cronbach's alpha) for these factors were .63, .68, .84, and .66 respectively. Scales were created based on this factor structure. Unfortunately, due to an error in wording on the large majority of the questionnaires, I could only be certain that the parenting styles reported by the parents on the FAQ for 51 subjects had been responded to according to the intended instructions. Whereas I had intended that the parent answer questions in terms of the relationship with his or her child, all but 51 of the questionnaires asked the parent to answer in terms of the relationship with his or her father. When the data were analyzed separately for parents with incorrect versus correct instructions, the results for the regression of each of the parenting styles on the humanism of the parents for the samples with correct versus incorrect instructions were relatively consistent (See Appendix A). However, when parents' authority styles were regressed on the parents' ideology, there were discrepancies. Therefore, the results presented in this 24 Table 1 F i f f iv lf- v 1 i n . FACTORS _Eaeter_L .1321le Variable Shame .81 -.19 Self-Esteem -.79 .27 Tosca Shame .75 -.03 Mastery -.68 .21 Chang & Hunter Guilt .64 .20 Buss & Durkee Guilt .61 .17 Alpha Pride —.04 .88 Beta Pride -.06 .88 Detachment -.04 .30 Tosca Guilt .32 .32 Externalization of Blame .42 .16 Table 2 EaeteLanalxsiLeLEaeteLL FACPORS WA. EaeteLlfl Variable Self-Esteem -.88 -.18 Mastery -.83 -.O4 Shame .83 .29 Tosca Shame .59 .36 Buss & Durkee Guilt .16 .85 Chang & Hunter Guilt .20 .84 25 section only concern the 51 subjects (40 females and 11 males) whose parents received the correct instructions on the PAQ. This sample consisted of 40 female and 11 male undergraduates between the ages of 17 and 22 (See Table 3). All subjects were from intact families, and most were white. Internal reliabilities (Cronbach's alpha) for the 51 mothers and 51 fathers for the PAQ scale were quite adequate. In particular, coefficient alphas in the current sample for authoritarian styles were .67 for mothers and .83 for fathers, for permissive styles were .71 for mothers and .72 for fathers, and for authoritative styles were .77 for mothers and .87 for fathers, respectively. Preliminary analyses revealed a non-significant relationship between mothers' and fathers' ideology (r = .21, p, n.s.). There was also no significant relationship between the endorsement of an authoritarian parenting style by one parent and the endorsement of the same style by the other parent (r = .23, p, n.s.), nor was the relationship between an authoritative style on the part of mothers and an authoritative style on the part of fathers statistically significant (r = .16, p, n.s.). However, mothers' and fathers' endorsement of the permissive style correlated significantly (r = .47, p < .001). In testing the various hypotheses, mothers' and fathers' data was analyzed separately to test for parent gender differences in predicted associations. The results for mothers and fathers combined are not reported, as they did not add any unique information. Table 3 l N: 1 Variable Percent Variable Percent Sex Religiea Female 78.4 Protestant 43.1 Male 21.6 Baptist 2.0 Roman Catholic 33.3 Age Jewish 3.9 1 8 9.8 None 5.9 19 54.9 Other 11.8 20 23.5 21-22 9.8 MetheL'Lflueatien Class High School or less 29.4 Freshmen 41.2 Some College 21.6 Sophomores 43.1 4 year Degree 27.5 Juniors 7.8 Advanced Degree 21.6 Seniors 7.8 W W High School or less 11.8 Black 5.9 Some College 15.7 White 92.2 4 year Degree 39.2 Hispanic 2.0 Advanced Degree 31.4 27 Because of the small sample size, the results for sons and daughters were not computed separately. ,gHt' ' .e-lt choc" it‘u“ ' - ago; I predicted that parental ideology would have a curvilinear relation to adolescents' self-evaluations: more extreme parental ideology scores would be associated with more negative self- evaluations, and more moderate parental ideology scores would be associated with more positive self-evaluations. The data did not support Hypothesis 1. In particular, when the adolescents' self-perceptions were regressed onto parental ideology, only mothers' ideology was directly linked to adolescents' self- evaluations (See Table 4). Maternal humanism was associated with adolescents' reports of less responsibility (i.e. less prosocial guilt, greater externalization of blame and detachment; B = -.29, p < .05), and more extreme (rather than balanced) maternal ideology was associated with adolescents' reports of less shame (as well as higher self-esteem and greater sense of mastery; B = .33, p < .05). ,1 Ht‘ ' ° -, 1914091, «.10 Pros , l I predicted that more balanced (i.e., less extreme) parental ideology scores would be associated with an authoritative style, whereas more normative parental ideology scores would be associated with an authoritarian parenting style. I also predicted that more humanistic parental ideology scores would be associated with endorsements of a permissive parenting style. Consistent with predictions, greater humanism in mothers was associated with 28 Table 4 i lat dol sc nts' s lf- v 1 i ns: im l rr 1 i n ffii N=l SELF-EVALUATIONS Shme Pride Quilt Respensibility IDEOIDGY r Beta r Beta r Beta r Beta Methers_QnIx Humanism -.26 .33a .06 .08 .00 .07 -.31 -.29a Extreme Ideology .50 .10 .27 .07 Total R2 .15‘11 .01 .07 .10 EntheLLinx Humanism -.11 -.09 -.16 -.15 .08 -.08 .13 .10 Extreme Ideology .17 .09 -.01 -.25 Total R2 .04 .04 .01 .08 ap<.05 29 endorsements of a more permissive parenting style (B = .31, p < .05; See Table 5). In addition, greater humanism in fathers was associated with endorsements of a less authoritarian parenting style (B = -.66, p < .001). Contrary to predictions, more balanced ideology scores for mothers were associated with a more authoritarian style (B = -.39, p < .01). 9 99 !‘ ° ' Pr 1 A_9hri ‘ gn9 A991‘ 5' f- Eraluatiens I predicted that the authoritative parenting style would be associated with more positive self-evaluations, whereas the authoritarian style and (to a lesser extent) the permissive style would be associated with more negative self-evaluations. Consistent with predictions, the authoritative style was associated with positive self-evaluations (See Table 6). In particular, children of fathers endorsing an authoritative style described themselves as higher in reponsibility (i.e. they tended to be less detached, were less likely to extemalize blame, and had higher levels of prosocial guilt; B = .31, p < .05). To the contrary, children of mothers endorsing a permissive style described themselves as lower in reponsibility (B = -.50, p < .001). H l . III' 111' lEfE I predicted parenting authority styles would largely mediate the relationship between parental ideology and adolescent self- evaluations. In other words, controlling for parenting styles would 30 Table 5 '. ‘1 t 9'919! in. um -I .uh9ri l ' im91 9rrli9n tn9 9‘- = 1 PARENTAL AUTHORITY STYLES Aatheritative Autheritarian P rmi i IDEOLOGY r Beta r Beta r Beta MetheLLinx Humanism .09 .09 -333 -.42b 27 .313 Extreme Ideology .08 -.39b .03 Total R2 .01 .24b .09 W Humanism .21 .19 -.65° -.66° .07 .08 Extreme Ideology -.22 .09 .16 Total R2 .09 .43C .03 ap<.05 bp<.01 cp<.001 31 Table 6 .9 ‘! l 1. ‘9“ 1 an! 1" 9.01. II ' If’ 1 9.1.11 . .1113] r1 i n n ffi i n N: 1 SELF-EVALUATIONS Jim _Pr1_'de_ .9413 Reseensihility r Beta r Beta r Beta r Beta Mmhers_Qn1x Authoritative .16 .20 -.02 .00 .03 -.01 .11 -.03 Authoritarian -.26 -.17 -.09 -.09 -.20 .20 .20 .08 Permissive .20 .24 .10 .06 .04 .01 -.50c -.50c Total R2 .13 .01 .04 .27b EathetLinx Authoritative -.16 -.06 -.27 -.24 -.12 -.06 .32“1 .313 Authoritarian .21 .22 .02 -.01 .06 .06 -.10 -.06 Permissive .22 .21 .18 .11 .18 .16 -.12 -.02 Total R2 .10 .09 .04 .11 ap<.05 bp<.01 32 decrease the relationship between ideology and adolescents' self- evaluations. In order to say that variable B mediates variables A and C, three conditions must be met: 1) The relationship between A and C must be statistically significant. 2) The relationships between A and B, and B and C must be statistically significant. 3) Controlling for B must reduce the magnitude of the relationship between A & C (Baron & Kenny, 1986). The only relationshps which met conditions 1 and 2 were the relationships between mothers' humanism, mothers' permissiveness and responsibility, i.e., children of more humanistic mothers reported less responsibility (condition 1), more humanistic mothers described themselves as more permissive, and children of more permissive others reported less responsibility (condition 2). When I controlled for mothers' permissiveness, the magnitude of the relationship between mothers' humanism and responsibility was reduced (condition 3). In particular, when responsibility was regressed on mothers' humanism, mothers' humanism accounted for 10% of the variance in responsibility; this was reduced to 3% after controlling for mothers' permissiveness. Hence, 7% of the variance in responsibility seemingly explained by mothers' humanism could be accounted for by mothers' permissiveness. Discussion The results of this study are largely in contrast with the research predictions. Parental ideology was directly linked to the adolescents' self-evaluations, but only for mothers. In addition, mothers' humanism was associated with Less prosocial guilt and mm externalization of blame and detachment. In addition, mothers' endorsement of a more balanced ideological position was associated with greater shame, lower self-esteem and a diminished sense of mastery. I had predicted that more moderate parental ideology scores would be associated with endorsements of an authoritative style, whereas more humanistic parental ideology scores would be associated with endorsements of a permissive parenting style, and more normative parental ideology scores would be associated with endorsements of a more authoritarian parenting style. When significant, associations between parental ideology and parenting style were in the expected directions. For fathers, greater humanism was associated with a less authoritarian style, and for mothers, greater humanism was linked to a more permissive style. However, contrary to predictions, a balanced ideological position tended to be associated with less permissiveness. This latter finding is difficult to interpret. I also had expected an authoritative parenting style to be associated with more positive adolescent self-evaluations, and the authoritarian and permissive styles to be associated with less positive self-evaluations. Several findings were in contrast with 33 34 these predictions: late adolescent children of fathers endorsing an authoritative style reported higher levels of prosocial guilt and described themselves as less detached and less likely to externalize blame. These results support and extend findings from previous studies which have linked the authoritative parenting style and adolescent self-esteem (Buri, Louiselle, Misukanis, & Mueller, 1988; Lamborn, Mounts, Steinberg, & Dornbusch, 1991), by using the parents' own reports, rather than the adolescents' perceptions of the parents' parenting style. In addition, a related finding that was consistent with that just discussed, was that children of mothers endorsing a permissive style reported lower levels of prosocial guilt, and described themselves as more detached, and more likely to externalize blame. Finally, I had expected parenting style to mediate the effects of parental ideology on adolescents' self-evaluations. There was only one example of a mediated relationship, and this was for mothers. Mothers' permissiveness mediated the relationship between ideology and responsibility, in that mothers' permissiveness accounted for a good part of the relationship between a humanistic maternal ideology and lower levels of responsibility. One important limitation of this study was that the Buri instrument does not allow for a distinction between two very different types of "permissiveness". In the current study, mothers' permissiveness was associated with Less responsibility in adolescents, whereas fathers' permissiveness did not predict self-evaluations in 35 adolescent children. Baumrind (1971), in tune with Maccoby and Martin (1983), who found that the most important factors in parenting are warmth and demandingness, suggests that two very different parenting styles may come under the rubric of "permissive." Permissive or "low control" families may be classified as indulgent or neglectful depending on their degree of warmth. Permissiveness in these two types of families may differ not only in quality but in the ideological foundations from which the parenting behaviours derive. While an indulgent style appears to have its roots in humanistic values, such as trust and democracy, a neglectful style has different roots (Lamborn et al., 1991; Maccoby & Martin), and results in very different outcomes for the adolescent. For example, Lambom et al. found the poorest outcomes in terms of self—evaluations in children who described their parents as neglectful. Children from indulgent homes had more favourable self— evaluations than children from neglectful homes, particularly in regard to self-reliance (Lamborn et al., 1991). Prior research indicates that the present measure of permissiveness is unrelated to nurturance or warmth (Buri, 1990). Because I was unable to distinguish between the indulgent and neglectful styles in the current study, it is difficult to interpret the findings linking ideology to this parenting style. A second important weakness in the current study is that I was unable to evaluate the effects of parental values and parenting styles separately for sons and daughters. Not only are males and females 36 socialized differently (Kaufman, 1989; Mosher & Tomkins, 1988), but the implications of parents' behaviours depend on the respective genders of the parent and child. For example, there is evidence that parental modeling and an authoritative style may be more important in the socialization of prosocial behaviours for boys than for girls (Eisenberg-Berg & Mussen, 1978). In addition, sex-linked, socialized definitions of role-appropriate behaviours influence not only our perceptions (Liben & Signorella, 1980; Martin & Halverson), but also our evaluations and behaviours (Deaux, 1977; Martin & Halverson, 1981). For example, Kobak and his colleagues found that in situations characterized by high maternal dominance, males exhibit more anger, whereas females exhibit greater passivity (Kobak, Cole, Ferenz-Gillies, Fleming, & Gamble, 1992). They suggested that current socialization practices tend to motivate males to respond to conflict by focusing on autonomy, whereas females tend to focus on maintaining relationships. As a result, sons may have to struggle harder for connection while daughters must struggle harder to gain autonomy in order to attain a healthy balance (Cooper & Grotevant, 1987). These differences imply that boys and girls will respond differently to parents' emphasis on control versus support. Notably, neither parental ideology nor parents' parenting styles predicted the late adolescents' experiences of shame, pride, or guilt. This was surprising, and possibly is due to the fact that the parents, rather than the adolescents themselves, reported on the parents' styles of parenting. Buri (1988) found a stronger association 37 between adolescents' perceptions of parenting styles and adolescent self-esteem, than between the parents' own reports of their parenting styles and the adolescents' self-esteem. Nevertheless, the self-evaluation factors were interesting in and of themselves. Consistent with previous studies (Chang, 1988; Tangney, 1990a), factor analysis of shame and guilt measures resulted in several separate factors in the current study. However, while the various shame measures used in this study formed a coherent cluster with the self-esteem measures, the guilt measures did not form one cluster. This suggests that guilt is a complex construct with both positive and negative aspects which have not been sufficiently delineated. While guilt is often viewed as a negative experience, there have been suggestions in the literature that moderate levels of guilt may have adaptive functions (Hoffman, 1982; Kaufman, 1989). In fact, guilt induction is one method for developing pro-social behaviour in children (Maccoby & Martin 1983). Consistent with this, Tangney (1989) has found that the TOSCA guilt-proneness scale is correlated positively with empathy and inversely with anger and hostility. In addition, previous findings link shame-proneness with greater externalization of blame (Tangney, 1990b), anger, and hostility, as well as less empathy (Tangney, 1989). Longitudinal data would be needed to understand the long-term consequences of these interrelationships. Due to the small sample size in the current study, any conclusions must be tentative and subject to further examination. 38 However, the results do suggest that parental endorsement of ideological positions and parenting style are, to a large extent, independent. Previous research indicated that the parent's view of their parenting style is often inconsistent with the child's view. Furthermore, the child's description is often a better predictor than the parent's of the child's self-evaluations (Buri, 1989). Further studies should take into account these inconsistencies in reporting as well as evidence that the implications of parental ideology and parenting styles are different depending on the respective genders of the parent and child (Charles, 1992). The same parental behaviours may have different meanings for a son than for a daughter (Gjerde, Block, & Block, 1991). These relationships must be studied over time to better understand their implications for the adolescent's well- being. REFERENCES REFERENCES Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. LoamaLgLf Betseuality aud Seeial Psyehdlegy, 51, 1173-1182. Baumrind, D. (1966). Effects of authoritative parental control on child behavior. Child Qeveldumeut,31, 887-907. Baumrind, D. (1971). Current patterns of parental authority. Develeumental Psyehelegy Mendgrauhs,i(2, Pt. 2). Baumrind, D. (1983). Rejoinder to Lewis's reinterpretation of parental firm control effects: Are authoritative families really harmonious? Wilma 9_£l-.(1). 132-142. Blos, P. (1967). The second individuation process of adolescence. Psyeheaualytie Study 9f the Child, 22, 162-196. Bower, G. H. (1981). Mood and memory. Amerieau Psyehelegist, 16(2), 129-148. Buri, J. R. (1989). Self-esteem and appraisals of parental behavior. WAG). 33-49. Buri, J. R. (1990). An instrument for the measurement of parental authority types. Jeumal ef Petsenality Assessment. Buri, J. R, Kirchner, P. A., & Walsh, J. M. (1988). Familial correlates of self-esteem in young American adults, Jeurnal df Seeial Buehelewmw), 583 588. Buri, J. R., Louiselle, P. A., Misukanis, T. M., & Mueller, R. A. (1988). Effects of parental authoritarianism and authoritativeness on self-68tcem WWW 1_(2) 271- 282. Buss, A. H., & Durkee, A. (1957). An inventory for assessing different kinds of hostility. Jeurual ef Qeusultiug Psyehdlegy,;_1_, 343- 349. 39 40 Carlson, R. (1981). Studies in script theory: 1. Adult analogs of a childhood nuclear scene. Jdurnal ef Persenality and Seeial Emmi._fl(3), 501 -510. Carlson, R. (1982). Studies in script theory: 11. Altruistic nuclear scripts. Eereeutual and Metor Skills, 11, 595-610. Carlson, R., & Brincka, J. (1987). Studies in script theory: III. Ideology and political imagination. Edlitieal Psyehelegy.§(4), 563-574. Chang, S. S. (1988). m as rem n f h m n i r l i te ether perseuality ednstruets. Unpublished doctoral dissertation, Michigan State University. Charles, M. B. (1992). 1' i f ' f adeleseeuthelgeyaluaLdus. Poster presented at the Biennial Meeting of the Society for Research on Adolescence, Washington, DC, March. Cohen, J. (1978). Partialed products ate interactions; Partialed powers are curve components. Psyehdlegieal Bulletin, 85(4), 858-866. Cook. D. R. (1985). W. Unpublished manuscript, University of Wisconsin, Menomonie. Cook, D. R. (1987). Measuring shame: The Internalized Shame Scale. Aleehelism fIfteattuent Quarterly, M2), 197-215. Cooper, C. R., & Grotevant, H. D. (1987). Gender issues in the interface of family experience and adolescents' friendship and dating identity. WWW. 16(3). 247-264. Deaux, K. (1977). Sex differences. In T. Blass (Ed), mm W (pp 357- 377) Hillsdale NJ: Erlbaum. Demo, D. H., Small, S. A., & Savin-Williams, R. C. (1987). Family relations and the self-esteem of adolescents and their parents. f ° h F '1,12, 705-715. Eisenberg-Berg, N., & Mussen, P. (1978). Empathy and moral development in adolescence. QeyeleumeutaLflsyehdjegy, 13(2), 1 8 5 - 1 8 6 . 41 Fauber, R., Forehand, R., Thomas, A. M., & Wierson, M. (1990). A mediational model of the impact of marital conflict on adolescent development in intact and divorced families: The role of disrupted parenting. Child Develdument, d1, 1112-1123. Forgas, J. P., & Bower, G. H. (1987). Mood effects on person- perception judgements. leutaal ef Perseaa11ty and Seeial Embelegx, 53(1), 53-60. Frank, S. J., Pirsch, L. A., & Wright, V. C. (1990). Late adolescents' perceptions of their relationships with their parents: Relationships among deidealization, autonomy, relatedness, and insecurity and implications for adolescent adjustment and ego identity status. ,leurnal ef Yeuth and Adeleseeuee, _1_2, 571-588. Friedman, M. (1985). Toward a reconceptualization of guilt. Ceutemperary Psyeheanalysis,_2_1_(4), 501-547. Gecas, V., & Schwalbe, M. L. (1986). Parental behavior and adolescent self-esteem. Jeumal ef Marriage and the Eam1'ly,4_8_, 37-46. Gilligan, S. G, & Bower, G. H. (1984). Cognitive consequences of emotional arousal. In C. E. Izard, J. Kagan, & R. B. Zajonc (Eds), Emetjen ns, eegaiti ea, an nd behavier (pp. 547- 588). Cambridge. Cambridge University Press. Gjerde, P. R, Block, J., & Block, J. H. (1991). The preschool family context of 18 year olds with depressive symptoms: A PIOSPective study. WW l(1). 63-91. Harris, I. D., & Howard, K. I. (1984). Parental criticism and the adolescent experience. I f Y h , _1_3_(2), 1 1 3 -1 2 l . Hayes, S. C. (1987). A contextual approach to therapeutic change. In N. S. Jacobson, (Ed.), Pisyehetherau 1sts ia elinjeal utaetiee; W New York: Guilford 42 Hayes, S. C, Brownstein, A., Zettle, R. D., Rosenfarb, I., & Korn, Z. (1986). Rule- -governed behavior and sensitivity to changing contingencies WW _5_, 237 256. Hayes, S. C., Zettle, R. D., & Rosenfarb, I. (1989). Rule-following. In S. C. Hayes (Ed.), R l - v r h vi r° n'i n eentiageneies, and insuuctional eentrel. New York: Plenum. Hoffman, M. L. (1970). Conscience, personality, and socialization techniques. Human Development, 1.1, 90-126 Hoffman, M. L. (1977). Moral internalization. Current theory and research In L Berkowitz (Ed) Wm psyehdldgy (Vol. 10). New York. Academic Press. Hoffman, M. L. (1982). Development of prosocial motivation: Empathy and guilt. In N. Eisenberg (Ed.),lhe_dey_e_LQum_ent_ef_ur_Qs_o_c_i_a1 1191111191 (pp. 281- 313). New York: Academic Press. Isen, A. M, Daubman, K. A., & Gorgoglione, J. M. (1984). The influence of positive affect on cognitive organization. In R. Snow & M. Farr (Eds.),Aut1tude, learning an ndl ir11st uet1du; Affeetiye and W- Hillsdale, NJ. Lawrence Erlbaum Associates. Josselson, R. (1980). Ego development in adolescence. In J. Adelson (Ed.). WW (1313- 188-210). New York: Wiley. Kaufman G (1989) WWW eLshameasedendremes New York: Springer Kobak. R. (1992). W W- Paper to be presented at the Biannual Convention of the Society for Research on Adolescence, Washington, DC, March, 1992. Kobak, R. R., Cole, H. E., Ferenz-Gillies, R., Fleming, W. S, & Gamble, W. (1992). A ' i l i l W. Unpublished manuscript, University of Delaware. 43 Lamborn, S. D., Mounts, N. S., Steinberg, L., & Dornbusch, S. M. (1991). Patterns of competence and adjustment among adolescents from authoritative, authoritarian, indulgent, and neglectful families. Child Develdpment,6;, 1049-1065. Liben, L. S., & Signorella, M. L. (1980). Gender-related schemata and constructive memory in children. Child Develeumeut, 51(1), 11- 18. Loftus, E. F., & Loftus, G. R. (1980). On the permanence of stored information in the human brain. Amer1ea11 Psyeheldgt'st, 31(5), 409-420. Luster, T., Rhoades, K., & Haas, B. (1989). The relation between parental values and parenting behavior: A test of the Kohn hypothesis. Jeumal ef Marriage and the Family, 11, 139-147. Maccoby, E, & Martin, J. (1983). Socialization in the context of the family: Parent- child interaction. In E. M. Hetherington (Ed.), P. H. Mussen (Series Ed), Handheek df eh 1d usyehelegy; yo]. &. Sdejaljzatien, petsenality, an nd s99! a1_de_v_el_onmeut (PP 1 101) New York: Wiley. Maier, S. F., Loudenslager, M. L., & Ryan, S. M. (1985). Stressor controllability, immune function, and endogenous opiates. In F. R. Brush & J. B. Overmeier (Eds.), Wmd edgnitida (pp. 183-202). Hillsdale, NJ: Lawrence Erlbaum Associates. Marshall, G. N., & Lang, E. L. (1990). Optimism, self-mastery, and symptoms of depression in women professionals. ,LQuma]_gf PersonalitLanLSeeiaLflsxeuologx 52(1), 132-139. Martin, C. L, & Halverson, C. F. (1983). The effects of sex- typing schemas on young children's memory. Chj]d__11e_yel_dp_ment, LG), 11- 18. Moore, B., Underwood, B., & Rosenhan, D. L. (1984). Emotion, self, and others. In C. E. Izard, J. Kagan, & R. B. Zajonc (Eds.,) Emetjens, W (pp. 464- 483). Cambridge. Cambridge University Press. 44 Mosher, D. L., & Tomkins, S. S. (1988). Scripting the macho man: Hypermasculine socialization and enculturation. Ihe_1_o_uma1_o_f SeLReseereh. 25(1). 60-84. Pearlin, L. 1., Menaghan, E. G., Lieberman, M. A., & Mullan, J. T. (1981). The stress process. Jeurnal ef Health and See1a1 Behavier, 22, 337- 356. Rosenberg, M. (1965). i l l-' Princeton: Princeton University Press. Ryan, R. M, & Lynch, J. H. (1989). Emotional autonomy versus detachment: Revisiting the vicissitudes of adolescence and young adulthood. Child Develeument,d0_, 340-356. Solomon, R. L. (1980). The opponent-process theory of acquired motivation: The costs of pleasure and the benefits of pain. Ametiean Psyehdldgist,31(8), 691-712. Stierlin, H. (1974). Shame and guilt in family relations. Memes ef Cenetal Psyehiauy, 311, 381-389. Stierlin, H., Levi, L. D., & Savard, R. J. (1971). Parental perceptions of separating children. Family Prdeess,_1_0_, 411-427. Stone, W. F. (1986). Personality and ideology: Empirical support for Tompkins' polarity theory, Pelitieal Psyehelegy. 1(4), 689-708. Stone, W. F., & Schaffner, P. E. (1988). IhepsyeheldngLpehtjes (2nd Ed.). New York: Springer-Verlag. Tangney, J. P. (1989). Mural affeet; The gded, the had, and the ugly. Manuscript under review. Tangney, J. P. (1990a). Assessing individual differences in proneness to shame and guilt. Development of the self-conscious affect and attribution inventory LeumeLQLRersQnelitxJntLSeeial Eueheleex 12(1). 306- 329 Tangney, J. P. (1990b). Sharing Shame and guilt: Another social clinical interface. Centemuetary Seeial Psychem.fl(2), 83- 88. 45 Tangney, J. P., Burggraf, S. A., Hamme, H., & Domingos, B. (1988). A 'n inivi l iffrn in rnn ~il'Th lf-9n i9_sAf ._9_n Ar'- 91 ‘ 9 Poster presented at the meeting of the Eastern Psychological Association, Buffalo, NY. Tangney, J. P., Wagner, P, & Gramzow, R. (.1989) Ihe_S_elf_-_C_Q11se1dus A ffeet and Attr1hutien Inventery Reyi sed. George Mason University, Fairfax, VA.. Tangney, J. P., Wagner, P., & Gramzow, R. (1990). Ereaeness te shame, preneaess td guilt, and psyehepathelegy. Expanded summary of a poster presented at the meeting of the American Psychological Society, Dallas. Tomkins, S. S. (1963). Left and right: A basic dimension of ideology and personality. In R. W. White (Ed.), The study ef liyes (pp. 389-411). Chicago: Aldine-Atherton. Tomkins, S. S. (1965). Affect and the psychology of knowledge. In S. S. Tomkins & C. E. Izard (Eds.,) WM (pp. 72- 97). New York: Springer. Tomkins, S. S. (1979). Script theory. Differential magnification of affects. In H. E. Howe, Jr. & R. A. Dienstbier (Eds.), Nehtaska symuds1um ea metivat1en (Vol. 26). Lincoln: University of Nebraska Press. Tomkins, S. S. (1987a). Script theory. In J. Aronoff, A. I. Rabin, & R. A. Zucker (Eds.), flfhe emergenee ef persenality (pp. 147-216). New York: Springer. Tomkins, S. S. (1987b). Pelam'ty Seale (Rev. Ed.). Springer: New York. Tomkins, S. S. (in press). Affeet, Imagery, Ceaseidusuess (Vol. 3). New York: Springer. Turner, R. J., & Noh, S. (1988). Physical disability and depression: A longitudinal analysis. Journal of Health and Social Behavior, 22, 23-37. APPENDIX 46 Table 1A 9 t f9r 9 9 9 9 9 .1" I ' 119 9 9 99 9 rr V In rr I in t i n N=2 l AUTHORITY STYLE P . . 5 l . . i l . . IDFDLOGY C I C I C I Mothers' Humanism .30a .11 -.323 -.47c .07 .14a R2 .0931 .01 .103 .22c .01 .02a Fathers' Humanism .06 .11 .65c -.52c .22 .153 R2 .00 .01 .42° .27“ .05 .02a ap<.05 bp<.01 Cp<.001 47 Table 2A 3' ._ 9‘ ' f9r 9'991‘ 9f .991' ' ' ‘ - ._ :91 9 9. 1 rr I in n = SELF-EVALUATIONS 51 E .1 G 'l E 'l 'l' AUTHORITY STYLE c I c I c I c I Melhers Permissive .24 .14 .06 .20b .01 -.01 -.50° -.24° Authoritarian -.17 .04 -.09 .15a -.20 .16a .08 -.13 Authoritative .20 -.06 .00 .26C -.01 -.01 -.03 .00 Total R2 .13 .02 .01 .09C .04 .03 .27b .05b Esther: Permissive .21 .08 .11 .06 .16 -.07 -.02 -.19a Authoritarian .22 .03 -.01 .15 .06 .15a -.06 -.14 Authoritative -.06 -.09 -.24 .18a -.06 -.01 .31a -.03 Total R2 .10 .01 .09 .04a .04 .03 .11 .04 ap<.05 bp<.01 Cp<.001 "‘llililiiliiiiiES