ABSTRACT THE FAMILY AS ENVIRONMENTAL SUPPORT: A STUDY COMPARING MENTALLY DISTURBED, CARDIAC IMPAIRED, AND NORMAL CHILDREN AND PARENTS IN A TASK SITUATION BY Keith E. Lyon It has been recognized that individuals require a multilevel support structure to function in society and provide for material and psychological needs. Signs of serious psychological disturbances ranging from depression to chronic institutional recidivism have been linked to the absence of such a structure to assist the individual in c0ping with stress, and many efforts have been made to provide external support for those at highest risk for pathology. For the child, the most powerful source of support is the family unit, where he is fed, clothed, and where his personality undergoes its early development. Ideally, the family will provide contact and assistance that enhances the child's sense of self-worth and his ability to function; if this does not occur, it is likely that the child will be psychologically handicapped later in life. These messages to a child about himself need not Keith E. Lyon come directly. More often they are part of the way in which the parents help the child with his life tasks. This study examined the support structure present for children in three populations: (1) those brought to a mental health facility with some form of acting out as the primary complaint, (2) those exhibiting a heart defect serious enough to require participation or evalu- ation in a cardiac clinic, and (3) a normal population gathered from the schools. These three groups were chosen to determine if the expected differences in familial support structure between disturbed and normal families could be artifacts of the presence of a deviant child who requires Special care and attention as in the cardiac families. An interaction sample for each family agreeing to participate in the study was gathered. Each subject was asked to work for 15 minutes on a puzzle in the presence of his parents. Then, the parents were asked to teach their child a poem during the next 15 minutes (order of presentation was randomized). Sessions were videotaped, and two raters scored the frequency of occurrence of Direct Assistance, Confirmation of Self- Worth, Disconfirmation of Self-Worth, Individualization Enhancing, and Individualization Blocking for each parent separately. They also scored the child for Help Seeking, Acceptance of Assistance, Rejection of Assistance, and Keith E. Lyon Task Persistence. Each family member was also given an overall Affective Level score for each segment. Inter- rater reliabilities ranged from .45 to .97. The 14 variables remaining after combining positive-to-negative continua were examined using a dis- criminant function analysis. This analysis forms linear combinations of the original variables, maximizing the differences among the groups. The initial run generated two discriminating functions with the highest loadings on: (I) Father's Affective Level, Child's Affective Level, Mother's Affective Level, Father's Disconfirmation of Self-Worth, and Mother's Confirmation and Disconfirmation of Self-Worth, and (II) Father's Affective Level, Mother's Affective Level and Father's Disconfirmation of Self- Worth. Function I seemed to separate the Cardiac from Normal and Function II the Mental Health from Normal. When the groups were compared in pairs, the Mental Health Group was differentiated from the Normal by Negative Father's Affect, Positive Mother's Affect, and Father's Positive Self-Worth Statements, while the Cardiac was differentiated from Normal on Negative Father's Affect, Negative Mother's Affect, and Positive Child's Affect. From this, a picture emerges of an environment characterized by affective ambivalence in disturbed families. Where self-worth statements and overt affec- tive tone conflict, both within and between parents, the child has no firm base from which to build his perceptions Keith E. Lyon of self and others. This suggests ambivalence is even more destructive than congruent negative affect on the part of the parents. It follows from the above that behavioral attempts to train parents to respond more positively to their children are likely to increase ambivalence instead of reducing it unless they are com- bined with changes in the affective states of the parents. The unambivalent negative affect levels of Cardiac parents are more benign but point to the need for special external support to prevent psychological casualties in the parents. The children with cardiac problems apparently escape the strain and pressure of their condition and may, in fact, enjoy their privileged position. The findings illustrate the importance of the father's involvement in the emotional life of the family. His affective tone and resultant reSponse to the child are linked to emotional disorders in childhood. In addi- tion, his general style of relating to the child is more age specific than the mother's, implying that his role in moving the child toward independence is crucial. Thus, the father plays an important part in the interaction of the family; he cannot be ignored in successful therapeutic intervention. Finally, we found that none of the child behavioral scales discriminated the groups. This implies that the re- ported behavioral problems of client children are situation specific and, therefore, more social than dynamic. THE FAMILY AS ENVIRONMENTAL SUPPORT: A STUDY COMPARING MENTALLY DISTURBED, CARDIAC IMPAIRED, AND NORMAL CHILDREN AND PARENTS IN A TASK SITUATION BY (A «on Keith E. Lyon A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1975 DEDICATION In the examination of environmental support, the most powerful factor I discovered was the assistance and encouragement provided by my loving wife, Ann. A project of this magnitude truly becOmes a joint effort, and her contribution has been at least as great as mine as she struggled to keep our world in order and still be avail- able when the discouragements and delays threatened to overwhelm. Thus, I want to dedicate this work to Ann, whose faith and confidence in me were more important than words can express throughout my program, and to both Ann and Branden, who combine to make it all worthwhile. ii ACKNOWLEDGMENTS I would like to express my appreciation to my committee, especially to Dr. Lucy Ferguson for her assistance in formulating the problem and for continuing to serve as chairperson of my dissertation committee in spite of pressing demands from other sources. I would also like to thank Dr. Robert Zucker for his encourage- ment and careful reading of my work and Dr. Gilbert DeRath for his aid in gathering my sample. Dr. Charles Hanley's contribution to the analysis of the data was also appreciated. The project could not have been completed without the assistance of the staff and administrative personnel of St. Lawrence Community Mental Health Center, Lincoln Center for Emotionally Disturbed Children, Lansing's Family and Child Services, Ingham Medical Hospital's Cardiac Clinic, and Mt. HOpe Elementary School who actively helped in locating suitable subjects. A special note of thanks to Bob Barnes of Lincoln Center for allowing the use of the Center's facilities for testing sessions. Finally, I would like to express my appreciation to those parents and children who agreed to participate in the study. iii TABLE OF CONTENTS Chapter Page I. INTRODUCTION . . . . . . . . . . . 1 Review of Literature. . . . . . . . 4 Importance of Environmental Support . . 4 Community support . . . . . . . 4 Post-hospitalization adjustment. . . 7 Environmental Support Within the Fanily O O O O O O O O O O O 12 The pathological family . . . . . 12 Intervention . . . . . . . . . 16 The healthy family . . . . . . . l9 Individualization . . . . . . . . 21 Self-Esteem . . . . . . . . . . 24 smary O O O O O O O O O. O O 27 Statement of the Problem . . . . . . 28 II 0 METHOD 0 O O O O O O O O O O O O 33 SWjects O O O O O O O O O O O O 33 Mental Health . . . . . . . . . 33 Physical Problems . . . . . . . . 34 Control . . . . . . . . . . . 34 Demographic Variables . . . . . . . 35 Test Situation. . . . . . . . . . 36 Procedure . . . . . . . . . . . 38 Scoring . . . . . . . . . . . . 41 Raters . . . . . . . . . . . . 42 Reliabilities O O O O O O O O O O 42 iv Chapter III 0 MSULTS I O O O O O O O O O O O O Discriminant Function. . . . . . . . Discriminant Function Results . . . . . Multiple Groups . . . . . . . . . Mental Health vs. Normal Groups. . . . Cardiac vs. Normal Groups. . . . . . Cardiac vs. Mental Health Groups . . . Demographic Variables. . . . . . . . Family Participation . . . . . . . . IV. DISCUSSION 0 O C O O O O O O O O 0 Mental Health Differentiation From Normal . Cardiac Differentiation From The Normal. . Mental Health Differentiation From Cardiac. Integration . . . . . . . . . . Role of the Father. . . . . . . . . Additional Scales . . . . . . . . . Summary . . . . . . . . . . . . APPENDICES APPENDIX A. FAMILY PARTICIPATION INDEX. . . . . . . B. RATING SCALES FROM RATER'S MANUAL . . . . C. CHILDREN'S BEHAVIOR CHECKLIST. . . . . . D. SUBJECT RELEASE FORM. . . . . . . . . E. INTRAGROUP CORRELATIONS OF STUDY VARIABLES . F. GROUP MEANS AND STANDARD DEVIATIONS. . . . G. DEMOGRAPHIC VARIABLES . . . . . . . . H. TEST SITUATION POEM . . . . . . . . . I. OVERALL CORRELATION OF STUDY VARIABLES. . . LIST OF REFERENCES . . . . . . . . . . Page 45 45 47 47 50 51 51 53 55 56 56 59 60 60 65 66 68 7O 75 79 80 83 84 85 86 87 11. 12. 13. 14. LIST OF TABLES Research on Community and Hospital Adjustment Analysis of Variance of Demographic Variables Rater Reliabilities. . . . . . . . . Multiple Discriminant Functions and Loadings Prediction of Subject Group Membership . . Group Centroids on Multiple Discriminant Flmctions O O O O O O O O O O O Individually Run Discriminant Functions and Loadings O O O O O O O O O O O 0 Age Correlation with Study Variables . . . Intragroup Correlations of Study Variables . Intragroup Correlations of Study Variables . Intragroup Correlations of Study Variables . Variable Means and Standard Deviations . . Demographic Variables . . . . . . . . Overall Correlation of Study Variables . . vi Page 10 35 43 48 49 50 52 54 80 81 82 83 84 86 LIST OF FIGURES Figure Page 1. Test situation . . . . . . . . . . . 36 vii CHAPTER I INTRODUCTION The family has long been recognized as the primary focal unit involved in the care and development of chil- dren. On the most basic level it is assigned the tasks of feeding and providing shelter for the relatively help- less members of society. In addition, it helps to trans- mit the norms and values of the parental culture through its role as a socialization agent. Each child's family is also tremendously important in the develOpment of personality. For it is in interact— ing with parents that the young child learns about his universe and about himself (Coopersmith, 1967). During this time, he deve10ps the concept that the parts of his body, the responses of others to him, and the objects he receives have a common point of reference which gradually develops into an abstraction--a sense of self. This sense of self can make a great deal of difference in the later social behavior of the child, with a positive sense of self-esteem correlated with independent, creative and assertive social actions. Much of this influence seems to come indirectly--the child is not told he is or is not capable of functioning independently, rather he is treated as if he could (or could not) Operate alone; he is not told of his self-worth, others reSpond to him as if he were "good" or "bad." The child learns from these experiences as readily as he learns other societal expec- tations and adopts them into his view of the world himself. Within this framework, the family is largely responsible for providing the support for the survival needs of the child. It also can mold and shape his sense of self and his expectations of his own capabilities and the responses of others in part through the nature and quality of that support. For, the parents are in the uniquely powerful position of being the child's original source of reference. If they care for him in a way that implies that he can care for himself, then his potential for later independent operation is increased. Similarly, if the support patterns present in the home confirm the child as a worthwhile human being, this will enhance his self-esteem throughout his life. On the other hand, caretaking which encourages dependency and depreciates the child will leave him handicapped and unable to ade- quately function. The research described in the following pages was designed to examine directly the amount and quality of support present in the home within various populations. Three groups were chosen to be included in the study: families with a child exhibiting disturbed symptomatology of an acting out nature, those with a child who is or has been under treatment at a cardiac clinic with a serious enough heart murmur or malfunction to warrant evaluation by a specialist, and a normal control population gathered from the public schools. These three groups were chosen to determine if the expected differences in familial support structure between disturbed and normal families could be artifacts of the presence of a deviant child who requires care and attention as in the cardiac fami- lies. This introduction will continue with the presen- tation of a literature review divided into two major sections. First, evidence will be presented to establish the importance of environmental support to an individual's functioning, both on a practical and a theoretical level. Here, it will be shown that the presence of certain factors in the environment are vital to the adequate functioning of the individual. A prior study by this author in which the family was shown to be the most important factor in successful rehabilitation will be presented as well. Section two will examine the support structure within the family, broadening the sc0pe of inquiry to include not only the presence of stress relieving factors, but the quality of the interaction as well. Within this section, the pathological family and current strategies for coping with it will be examined. Liter- ature will also be presented on the importance of two major dimensions considered to be vital to the develop- ment of the adequately functioning individual, the capacity for individualization, and the level of self- esteem. Review of Literature Importance of Environmental Support Community suppprt. Several researchers have pointed to the support structure and culture surrounding the individual as an important determinant in nonpatho- logical community adjustment. Langner and Michael (1963) examined mental illness in a community setting through a survey technique, looking at social class differences in the duration of pathology. They found much variety in response styles between classes, with the lower classes utilizing much less adaptive ways of dealing with problems. Thus, the ability of the response to alleviate the stress (as in the middle class compulsive response to job-oriented worries) and avoid hospitalization (a common result of a lower class acting out response) will determine the psychOpathology's duration and the likelihood that treat- ment will be required for adequate functioning. Leff, Roach, and Bunney (1970) found unfavorable life events prior to hospitalization in the depressive patients they studied. This was true for both the acute and endogenous depressives. While at first glance this may seem unrelated to general community adjustment, it must be remembered that most individuals involved in short-term hospitalization return to essentially the same environment they experienced previously. In such a situation, high stress levels could "maintain” a chronic picture of periodic hospitalizations. Sometimes at lower stress levels the mere sug- gestion of help can produce a marked change. Kellner and Scheffield (1971) found that, with patients in both anxious, depressed, and psychophysiological categories, improvement was noted during a waiting period before treatment. This effect was greater than the effect of subsequent brief therapy, and was felt to be due to the "impact of the clinic" or the response to the ”symbols of competent care.” Dohrenwend and Dohrenwend (1969) approach the question of the role of social factors in the presence of mental illness directly. In discussing the relation- ship between stressful events and psychological symptoms (p. 129), they note that whether the symptom disappears or is maintained after the stress is alleviated depends upon the presence or absence of secondary gain. In other words, if the rewards for the symptom produced in response to the stress are sufficiently great (e.g. compensation payments, removal of responsibility, etc.) the symptom will be maintained by these rewards after the original need for it is removed. Social classes differ in the number of adequate resources present to cope with stressful situations. Lower class husbands and wives provide little mutual support (Rainwater, 1965) and a lower class individual is less likely to be involved in voluntary organizations (Cohen & Hodges, 1963). As a group, they are less able to purchase adequate medical care (Langer, 1966) and are dealt with less favorably by social agencies (Dohrenwend, 1961). The above class differences, when combined with the higher level of symptomatology present in the lower class (Hollingshead & Redlich, 1968; Dohrenwend, 1969), and the other evidence presented in this section, pro- vide strong support for the Dohrenwends' hypothesis that the absence of adequate resources with which to cope with stressful situations is a factor in the continuation of symptomatology. Thus, the studies cited seem to point to two general conditions which may lead to problems in adjust- ment. First, the stress level that produced the original response may continue unabated with no hape of relief in sight. And, second, the symptom may be reinforced by the social milieu to such an extent that it is not relinquished once the stress is removed. Either condition seems to relate directly to the environmental support picture in the individual's environment. Post-hospitalization adjustment. Several factors in the post-hospitalization environment have been cor- related with community adjustment by prior studies. Though these dealt primarily with the chronic, insti- tutionalized patient they are suggestive of forces more generally present in persons following hospitalization. Specifically, Kardiner and Spiegel (1947) examined soldiers suffering from war neuroses and found that the neurosis seemed to be prolonged by the presence of direct disability compensation (p. 392). This factor was made more intense by the role of the therapist in making the decision relating to compensation eligibility. Silverstein (1968) points to the need for agency aftercare facilities such as medication monitoring, day treatment centers, counseling, and psychotherapy. He feels that services of this sort were needed by 9 out of 10 of the patients who left the 18 state hospitals of Pennsylvania during his study. Further support for the need for aftercare facilities was suggested by the fact that 44.3% of the patients who returned to the hospital did not utilize any of the available assistance. Unfor- tunately, data were gathered only on those who failed. This study, along with the others presented in this section, makes the need for post-hospitalized help and support for successful readjustment clear. Lucas (1959) found that group support in the form of an ongoing therapy group was beneficial in increasing the self-confidence of the former patient. This was felt to be of special importance to the person labeled schizo- phrenic who must struggle with his oversensitivity and feelings of lack of belonging. Life stress has already been mentioned as a factor affecting mental health status (Dohrenwend & Dohrenwend, 1969) in the community. The difference between momentary symptoms and more chronic problems was felt to depend in part on the availability of stress alleviating community resources and also on the secondary gain afforded the symptom in the social setting. This point is supported by Silverstein's finding that 35% of all his subjects who failed in the community did so because of an inability to cope with stressful situations. The guided autonomy and social supports of a live-in ”Lodge” situation were found to be more success- ful in preventing rehospitalization than completely independent attempts at community readjustment (Fairweather, Sanders, Cressler, & Maynard, 1969). Here, peer interaction was felt to be most beneficial in stabilizing the former patient. Table l (p. 10) summarizes the studies done in the area of community and hospital adjustment. It presents the specific variables found to be related to adjustment by each researcher and the data of publi- cation of each study. It can be seen that several dif- ferent types of environmental support have been linked to the adjustment an individual will make in the com- munity. They may be formal or informal, positive or negative, but the relationship between an individual's environment and his mental health is clearly shown. In addition, a study was conducted prior to the present project (Lyon & Zucker, 1974) in which the informal interpersonal influences or supports present in the environment of 39 patients after release from a short-term psychiatric inpatient unit were examined. In particular, factors relating to the immediate day-to-day life of the expatient (e.g., amount and quality of contact with relatives and friends; employment experiences), rather than hospital related activities were utilized. A cluster analysis was performed and a stable home life (marriage, high home living involvement, older age, and many home activities) was linked with low three- month symptomatology (p §_.Ol) and improved symptom change to three months (p g .10). The presence of 10 Table 1 Research on Community and Hospital Adjustment Study Date Variables Affecting Adjustment Community Lagner and Michael 1963 Ability of response to alleviate stress Dohrenwend and 1969 Secondary gain; C0ping Dohrenwend resources Leff, Roach, and 1970 Unfavorable life events Bunney Kellner and 1971 The promise of help (screen- Scheffield ing interview) Institutional Goffman 1961 Institutional pressures into the "sick” role Stillman 1971 Patient peer group support of "well" role Post-Hospitalized Kardiner and Spiegel 1947 Direct disability compensa- tion (secondary gain) Lucas 1959 Ongoing therapy group Silverstein 1968 Agency aftercare facilities Fairweather, Sanders, 1969 Guided autonomy (live-in Cressler and Maynard facilities) ll benign visitors (high visitor involvement, little pressure from visitors, and a positive perception of visitors) was also related to three-month symptomatology (p g .05). In spite of methodological limitations, these findings imply that the most powerful environmental influence for the study population was the family. The above research tends to support the theoreti- cal model advocated by Caplan (1964). He assumes that in order to avoid mental disorder a continual source of "supplies" is needed by the individual. These supplies can be grouped into three general areas: physical, psy- chosocial, and sociocultural and a lack in any one area can be very detrimental. Physical supplies include food, shelter, sensory stimulation, and other things necessary for bodily health and development. Psychosocial supplies relate to the "stimulation of a person's cognitive and affective development through personal interaction with significant others . . . (p. 32)." And sociocultural supplies are those influences exerted by the culture and the social structure in terms of enhancing or blocking challenge or opportunity as well as in the degree of stability present in the society. These three groupings are, of course, inter- related, and it is felt that a lack in any area of supplies would greatly affect the individuals' ability to c0pe with a crisis situation. Caplan's use of the 12 term "supplies" parallels what we mean by environmental supports with some modification. First, we will be narrowing the range of supports under examination to support present within the family structure. Second, within the family, the inquiry will be broadened to include not only the presence of stress relieving factors, but the quality of the interaction as well. Environmental Support Within the Family In this study, environmental support is taken to mean assistance in the reduction of stress. Within the family, however, more seems to be involved in the concept than simple direct aid. As was mentioned briefly above, the manner in which the interaction takes p1ace--the subtle messages communicated--is at least as important in a child's development as the presence or absence of assistance. The pathological family. The largest group of theorists examining the disturbed family utilize some form of a systems approach to describe the family. Bell uses social group theory to conceptualize a family as a balancing act in which self-wishes are weighted against the imposed demands of others. This will lead to the establishment of roles in which each will bring pressure on the others to accomplish what he is unable to accom- plish for himself (1961, p. 2). Tharp and Otis, in an 13 expansion of Structural-Functional Analysis, feel that pathology results from trouble in establishing satisfying reciprocities which leads to a lower self-evaluation and thus to intrapsychic distress (1966, p. 429). Ackerman also concerns himself with the level of complementarity in his clients, working to delineate central conflicts and counteract scapegoating (1961, p. 69). In a similar vein, Jackson feels that each person is constantly attempting to define the nature of his relationships with others. But this control of one's relationships is not simply the power to tell another what to do. "Rather, it refers to the fact that every communication can be seen as»a report and a command, and the command may be of a higher order of messages than the explicit message (1968b, p. 189)." It is adherence to these subtle, indirect, and sometimes contradictory commands that constitutes being controlled by others. To Jackson, a pathological relationship involves constant attempts to control others combined with a general ”sabotaging or refusing of others' attempts to define the relationship (p. 190).” Haley sees the family as a homeostatic mechanism that is ”error activated" in the presence of attempts to change the system in much the same way that a thermostat acts to maintain a stable temperature in a house. If variations beyond a certain tolerable range are detected, forces are brought to bear to correct the situation and 14 return the system to its original state. Within the family, these forces can take the form of extreme pressure on individuals to alter their behavior. While pressure to return to a stable level always exists in a family, Haley points out that the level of behavior used as a standard can be altered as the setting of a thermostat is changed. The question of who is going to set the thermostat (decide on the behaviors to be accepted in the family) is felt to be the second level of struggle present in a family, with all activity in a disturbed family centered on the struggle at this level. Messer also speaks to the sense of homeostasis present in the family, preferring to conceptualize it in terms of an individual's physical functioning. A family trauma (or even a normal developmental shift) can be dealt with in many possible ways, some of which--like a bad set on a broken leg--wi11 forever reduce functioning. Such adaptations will become part of the system, stereo- typed and resistant to change. On the other hand, a trauma or shift can result in positive realignments that leave the system functioning adequately with its thermo- stat setting successfully altered according to Haley's model. Friedman et a1. focused on the family's inter- action with others in its environment, pointing out that 15 in the normal situation there is a great deal of inter- action with and influence from others outside of the family circle. WOrk, school, recreation, friends, and neighbors all are part of the world of the family members and no attempt is made to isolate any member from his environment. The situation in the pathological family is quite different. Using a clinic population, Friedman found a strong trend toward isolation from outside influences among disturbed families. Levels of contact were reduced greatly for family members, and, as a result, the degree: of influence and control exhibited by the family over its members increased greatly. Sometimes these families accepted influence from a single external figure (doctor, uncle, minister, etc.) but in each case the individual served to reinforce the rigid familial system, using it for his own purposes. Laing makes a strong case for the presence of such an isolated system in the families of schizophrenics. He speaks of the develOpment of the family ”nexus” or extreme sense of community as something that must be worked on to be maintained. In an analysis that appears to be bred from long years of frustrating work with psychotic families, Laing claims that "A united family exists only as long as each person acts in terms of its existence. Each person 16 may then act on the other person to coerce him (by syme pathy, blackmail, indebtedness, guilt, gratitude, or naked violence) into maintaining his interiorization of the group unchanged (p. 58).” He goes on to say that in such a family if there is not external danger, then some sort of danger or terror has to be invented and sold to the members to keep the family together. Thus, family “homeo- stasis" is in this case the product of reciprocities within a violent system. Based on the above sources, a picture emerges of the pathological family as a rigid, isolated group engaged in constant maladaptive struggles for control and domi- nance. Such families directly prevent their members from leaving the system, and are largely unable to effectively deal with internal developmental change. Further, the family members are unable to provide each other with emotional support and warmth. It is as if they are so engrossed in their battle for control that they are unable or unwilling to acknowledge the positive ties that bind them lest they be used to manipulate. This results in an emotionally barren environment for all. Intervention. Further clues to the nature of pathological families can be found in an examination of the treatment methodologies devised to deal with them. Bell, one of the first and certainly one of the most influential figures in the area, outlined several 17 clear steps of treatment (Bell, 1961, 1962). He is not primarily interested in insight, believing that changed feelings will follow changed action as a matter of course. He begins with an orientation for both parents and chil- dren and then moves to a child-centered stage in which the children are given the right to criticize and make damands. This blends naturally into a parent-child interaction which evolves into father-mother interaction. The last two stages involve sibling interaction and a family centered consolidation of therapy. Bell's goals in therapy are to improve interaction, make the family aware of its roles, and demonstrate mutual interdependence and positive ties, with the therapist's task involving situation structuring and rapport development (Bell, 1961). Ackerman works to counteract scapegoating and other disturbances in ”complementary role adaptation" (Ackerman, 1961, p. 65). He utilizes an aggressive style which "tickles the defenses" of the client family (Ackerman, 1966, p. 98) as he dares to directly shatter family myths. His style is powerful, controlling, and deliberately provocative as he mobilizes and confronts. Satir, on the other hand, works to help her families clarify and rephrase (1967) their communication and feelings, based on her feeling that anger is really a mask for hurt and defensiveness (p. 109). Using a systems analysis and a tightly structured interview 18 with specific tasks and a great deal of support, Satir has been described as a teacher and an expert in communi- cation (Beels & Ferber, 1969). She uses little direct attack, feeling that support is more effective. Jackson, utilizing a pure systems approach, also works on helping the families to relate (1967) but his main emphasis is on directly altering the family system as it presents itself before treatment, using such methods as paradoxical instructions to force a change (Beels & Ferber, 1969). Haley (1962) follows a similar line of reasoning in his view of the family as a homeostatic mechanism. The major battle within the family, and one in which the therapist becomes directly involved, is that of deciding who is going to set the "thermostat” (control the nature of interaction in the system) (p. 92). In dealing with this issue, Haley utilizes metacommunication, or communi- cation about feelings and behavior in a style he feels is parallel to individual therapy (p. 90). wynne also helps his families to utilize meta- communication. WOrking with the twin concepts of pseudo- hostility and pseudomutuality, he helps them to under- stand and clarify the confusing messages a disturbed family exchanges (1961, p. 114). His style is quite active as he assists in focusing on the important issues (1965, p. 301). 19 To Zuk, therapy entails the establishment of the therapist as a ”go-between" (1972, p. 237). His task involves structuring, directing and mediating in the family process. As in Haley's model, the family will resist the therapist's attempt to install himself in a controlling position, but such conflict is seen in a positive light, as the source of the needed energy to change the system (p. 227). Framo, working from a psychoanalytic orientation, first obtains outlines of parental introjections in an attempt to uncover the underground psychological elements in the family (1965, p. 163). He attempts to work these through in the family setting, being careful of trans- ference and counter-transference. Most important for Framo, is the creation of greater individuation and communication between family members and the improved marriage that will result (p. 206). Rakoff, Sigal, and Epstein (1972) would support Framo's use of the concept of working through (p. 306). Their interpretations are directed at the effects family members' actions have on others and the "transactional dynamics" of the family unit. Interaction patterns are considered to be parallel to character structures in the individual (p. 313). The healthy family. The healthy family, by con- trast, is described by Bell (1962) as one in which 20 complementary aims exist among the family members, aims which also support the function and structure of the group itself. This would imply not only that all members of the family are pulling in the same general direction, but also that there is a general consensus among the members as to the role of the family and the way it is to operate. Secondly, a healthy family has at its dis- posal multiple methods for accommodating the mutually incompatible demands of individual members. There is, in effect, more than one way to resolve the inevitable conflicts that will arise in the course of events of the family. Third, a means of repeatedly reevaluating the consequences of its achievements of accommodating is available to the healthy family; there exists some sort of adequate feedback system through which the family can monitor its performance. And, finally, the family in possession of the above three attributes chooses to Operate flexibly to not reject options or changes out of hand (p. 6). This final attribute of the healthy family is especially important developmentally. Such a family, in contrast to one more disturbed, is able to adapt as children grow and their needs and potentials change. It is able to capitalize on each child's strengths at each developmental stage. Thus, the general picture of the well-functioning family is one involving a level of harmony in general aims combined with an effective 21 monitoring system and the flexibility to make internal changes as needed. This is in sharp contrast to the rigid, closed system of the pathological family. Individualization One of the central themes throughout family interaction literature is the lack of individual identi- ties in members of disturbed families. Each member exerts considerable influence over each other member, forming a tight web of expectations that allows almost no freedom of movement. In contrast, the normal develop- ment of individualization is best explained by Jackson: Every child begins life in the secondary position of a complementary relationship since someone must take care of him. However, as the child grows older, in the normal family situation he is encouraged more and more to determine the nature of his own activity, and ultimately he is able to behave with his parents as one equal to another. (1968b, p. 190) {As the child begins to feel free to be an indi- vidual, he will begin to assert himself by insisting on walking by himself, trying to tie his own shoes, etc. Such behavior is encouraged within normal families, according to Jackson. Experimentation is also encouraged, with the parents standing ready to offer assistance if the child requests it. This assistance is termed ”com- plementary” by Jackson and is differentiated from that offered in the pathological family. Pathological assistance, in contrast, is given without solicitation from the child--he may even reject 22 it and attempt to escape its influence. In this case, the child's push for independence is actually punished-- often by withdrawal of the parent. He is taught that he must not operate as a separate entity, that any assistance offered must be accepted and that independent functioning will lead to loss of a love object. Brown (1967) Speaks to the same point, stating that in the families of very young children, the chil- dren's symptoms are an expression of the parent's need to deny that any developmental changes have occurred. This parental behavior is explained as an overt expres- sion of archaic object relations and attitudes within the parents (p. 313). In other words, the child is not allowed to be himself, he is forced into a mold projected on him by his parents' own needs. Often the child is not only denied access to the developmental ladder, but is used as the familial scape- goat as well. This process has been described under several labels including ”projective identification," "trading of dissociations," and ”defensive delineation" (Beels & Ferber, 1969; Ackerman, 1961) and involves the splitting off and projection onto the child of disavowed qualities of the parent's personality. Thus, the child becomes a sort of wastebasket for all the family garbage. This function becomes vital to the maintenance of a pathological system, and much of the pressure to block 23 individualization may be motivated by attempts to pre- vent loss of the scapegoat (weiner, 1966, p. 374). One mechanism through which such a system is maintained and individualization blocked may be that of reciprocal concern. Each person is expected to be con- trolled and to control the others by the reciprocal effect that each has on the other. In such a system of ”enforced debt," a family member has an investment in the others that he can collect on at any time (Laing, 1967, p. 60). Any child contemplating leaving the family would find his way barred by the unsurmountable debt he owes the "company store." Ackerman also feels that the disturbed family is built on sacrifice (1966, p. 165). In moderately dis- turbed families, the child is given the message that he must not be different, while in the case of severe dis- turbance he learns that he must not be at all! Growing up in such a climate can be severely destructive to the child, and can lead to reactions of attack, withdrawal, or an internalization of conflict and pathology. Many therapists move quickly to counteract the blocking of individualization and the use of a scapegoat (Framo, 1965). Both Ackerman (1966) and Rakoff et a1. (1972) point to the need to counteract this tendency, with Ackerman also noting the triangular pattern of punisher-victim-rescuer that can develOp in such 24 families (p. 83). He continues, stating one of the axioms of family therapy, that the child will often improve once the burden of displaced conflict is retransposed to its prime source-conflict in the parental and marital pair (p. 168). Satir also works very consciously to facilitate the develOpment of each family member as an individual. To accomplish this, she is overtly careful at the start of her sessions to see that there is a place for every- body as an individual, ending her sessions using each person's name in turn. As has been shown above, the lack of adequate individualization in disturbed families is a matter of great concern to writers in the field, both on theoretical and a practical level. The child is considered to be impaired by such a tendency and most therapists work quickly to alleviate some of its most destructive compo- nents. Based on the literature presented, we would expect to find active blocking of the child's move toward indi- vidualization in disturbed families. Self-Esteem As we have seen, attempts to reverse the blocking of individualization so prevalent in disturbed families have as their goal the increase in autonomous functioning of each family member. But providing the opportunity for independent action will not ensure freedom of movement 25 any more than opening the cage door will release a pet bird. Before the bird will venture forth, he must be convinced that no danger exists in his path of flight. Similarly, if the individual feels himself incapable of functioning outside of the protection of the family, he may not utilize his new-found freedom, even with encour- agement. Here, he is trapped not by external pressure but by his own attitudes about himself. Thus, the liberation of a disturbed family member must include two processes: the removal of obstacles to independent functioning and a corresponding increase in his self- esteem level. (The creation of a pathological family would seem to include both elements as well--blocking behavior and self-esteem reduction.) An individual's level of self-esteem (ego strength, concept of self, or self-attitude) is a central variable in most theoretical frameworks describing an individual's potential for functioning. It is also a concept vital to most practicing therapist's methodology. But, though it is a part of many theories it has only recently been studied in its own right (Coopersmith, 1967, p. 20). According to Coopersmith, during the early years a child develops the concept that the parts of his body, the responses of others to him, and the objects he receives have a common point of reference. This point of reference gradually deve10ps into an abstraction, a 26 sense of self with a degree of continuity (p. 20). The child needs this sense of continuity before he can begin to make sense out of his environment, learning about the contingencies and expectations that will later enable him to act to get his needs met. These expectations about one's own capabilities were shown to make a great deal of difference in the experiential worlds and social behaviors of individuals. Those with high self-esteem expect to be well-received and successful, they have a great deal of confidence in their perceptions and judgments and believe that they can effect a favorable resolution of most any situation. In general, such peOple are independent, creative, and more assertive and vigorous in their social action. Those with low self-esteem (measured on the basis of self-attitudes) were found to lack trust in themselves. They were apprehensive about expressing ideas and showed signs of self-consciousness and a preoccupation with their own inner problems. This, according to COOpersmith, led to a pattern of limited social intercourse, a clear behavioral difference (p. 71). In the families he studied, COOpersmith felt that these attitudes toward the self were the product of experience with the parents. In families with children exhibiting high self-esteem, he found parental patterns of acceptance. These parents exhibited interest 27 in the child and concern about his companions. They were available to the child, congenial, and participated with the child in many joint activities. The child apparently perceives and appreciates the attention and approval expressed especially by the mother, and tends to view her as favorable and supportive. Rejecting parents, on the other hand, withdraw from the child, leading to social deprivation (p. 179). Satir is the family therapist most directly con- cerned with the level of self-esteem. She makes the point that an adequate self-concept is much more impor- tant to an individual's development than the restrictive "love" traded within disturbed families (in Haley & Hoffman, 1967, p. 116). To enhance each family member's self-esteem, Satir works to relabel hostile feelings to accent the positive bonds that unite the family. Summagy Thus, for support within the family to be a posi- tive influence on the child, three factors must be taken into consideration. First, of course, is the actual presence of assistance in stress reduction. Also, to be effective the supportive act must maintain or enhance self-esteem. Third, it must allow for the individuali- zation within the family as a prerequisite to eventual independence of the children. 28 The lack of these three factors may well be related to the tendency in disturbed families to develop and maintain the destructive closeness reported by family therapists. Statement of the Problem There is direct evidence of the general importance of support and assistance in stress reduction in the pre- vention of psychopathology. It also seems clear that within the family the quality of this support as well as its presence or absence is important in determining its effect on the child. This would imply that there may be differences in the level and quality of support among the family structures of different clinical populations as well. The present study is designed to measure the support structure within three family papulations. We are comparing families of children under treatment at a local mental health center with (1) children exhibiting a heart defect serious enough to require participation or evaluation in a cardiac clinic and (2) a normal popu- lation gathered from the schools. Children who are referred for therapy at mental health facilities fall into three general categories, those who exhibit internalized symptomatology, those who tend to act out, and a third category containing autistic characteristics. These general categories were discovered 29 by Peterson (1961) in a factor analysis of 427 presenting problams from case folders for children at the University of Illinois Psychological Clinic. They break down behavior as follows: the factor "Conduct Problems" includes dis- ruptiveness, boisterousness, disobedience, attention seek- ing, and uncooperativeness; Personality Problems include inferiority feelings, tendency to be flustered, lack of self-confidence, anxiety, and self-consciousness; and Autism contains preoccupation, shortness of attention span, clumsiness, inattentiveness, and passivity. Of the three categories, conduct problems were found to be most prevalent followed by personality problems and autism. While these three patterns were found among the children studied, parents showed no such differentiation. This led Peterson to conclude that ”Whatever differenti- ation of problem.behavior five year olds diSplay may be less a function of intrapersonal tendencies, determined by Specific kinds of parental attitude, than of a general personality disturbance, shown now in one form and now in another, as situational provocations and opportunities arise (p. 160)." The family interaction literature provides the basic theoretical underpinnings of the study. This literature is not weighed either for or against any one type of symptomatology. The implication is clear that a disturbed family contains a great deal of strife and 30 conflict, and that this will affect the child's level of self-esteem and ability to individualize, but the focus of the child's response to this environment is not predetermined. This, combined with Peterson's findings and the fact that studies have Shown that both acting out (conduct problem) and withdrawn children are at risk for serious disturbance later in life (Ricks, 1970) makes the selection of one or the other symptom group for study seem arbitrary. In Spite of this, to sharpen our focus we have chosen the acting out child for study, a selection based on two considerations. First, acting out children are referred more often for help and therefore a study focus- ing on them would be most representative of current clinic pOpulations and the clinic pOpulations utilized by the family research theorists. The second rationale for use of the conduct problem child relates to the major dimensions of the study, par- ental behavior which enhances or blocks the child's move toward individualization and behavior which increases or decreases his self-esteem. Low self-esteem on the part of the child may cause ultimate withdrawal, but it could just as well precipitate a hostile defensiveness, especially when the child is young. In a similar manner, parental moves to block individualization would be more likely to elicit an initial hostile response from the developing 31 child as his scope expands and he feels the restrictions more keenly. While we are using the Conduct Problem child in the study, it should be noted that this is not a delin- quent population. Rather, these are children who have been referred for mental health services during latency and may develop a variety of adolescent and adult dis- orders ranging from schizophrenia (Roff, 1963) to delin- quency. The Cardiac group was included to determine if the expected differences in familial support structure between disturbed and normal families could be artifacts of the presence of a deviant child who requires special care and attention as in the cardiac families. To the extent that a dimension or scale differentiates the mental health from the other two groups, it could be considered to be a factor peculiar to the mental health papulation. This would be of great interest diagnosti- cally, and it would also show that the pattern of behavior was not produced in response to the presence of a physio- logically traumatic childhood. If, on the other hand, the dimension divides mental health and cardiac from control but not from each other, it would be concluded that the factor is probably the result of the Special pressures created by a deviant child in the home. A discriminant function analysis will be utilized to examine the data. Briefly, such an analysis generates 32 linear combinations of the study variables that dis- criminate among the groups. These combinations are used to predict group membership for nonclassified subjects, and it is the hypothesis of this study that accurate‘ predictive ability will be present using the following variables: Parental Support, Self-Esteem Confirmation, Individualization Enhancement, and Affective Level; Child's Help Seeking, Acceptance of Assistance, Task Persistence, and Affective Level. For the purposes of this study, the following definitions will apply: Disturbed families--families in which at least one child is an identified patient currently under treat- ment at a local mental health facility and exhibiting acting out behavior. Supportive behavior-~any active response on the part of the parent that makes the child's task easier (without doing it for him) or provides emotional encour- agement. Self-Esteem enhancing_re§ponses--those that help a child to feel adequate and competent. Individualization--the ability to operate as an independent person, to leave the shelter of parental control and protection. The above definitions are further expanded and operationalized in the raters' manual (see Appendix B). CHAPTER II METHOD Subjects Three groups of subjects were included in the study (N = 14 per group; total N = 42). Mental Health The subjects exhibiting psychological impairment were drawn from clients at St. Lawrence Community Mental Health Center, Lincoln Center for Emotionally Disturbed Children and Adolescents, and Lansing's Family and Child Services. As was explained in detail earlier, only chil- dren with symptomatology that could be classified as ”Conduct Problems” were selected. Each subject family was approached by their therapist who explained the pro- ject briefly as a "study of children and how they go about solving problems," and asked for permission to release the client's phone number to the researcher. Then, the researcher contacted the parents and made the actual request for participation. It was explained to the parents that participation in the study was voluntary and was completely independent of any treatment they 33 34 were receiving or might expect to receive at a mental health center. It was also made clear that videotapes and other materials would be handled in the strictest confidence and would be seen by study personnel only. Approximately 70% of those contacted agreed to be tested. Physical Problems The children who are physically at risk were drawn from a population of outpatients at the cardiac clinic at Ingham Medical Hospital who, on the basis of a symptom checklist (Appendix C), exhibited no overt psychological impairment coinciding with their disorder. Names were gathered from the clinic file and patients were contacted directly, using much the same format as with the mental health population except, of course, that the project was declared to be independent of treat- ment at the cardiac clinic. Eighty percent of those con- tacted agreed to participate. Control The normal control group gathered from the public school system allowed us to compare both groups of deviant families with families free of overt problems in need of intervention. A symptom checklist completed by the mother was utilized as an additional screening measure. This population was gathered from a class list of students at Mt. Hope Elementary School who had no history of 35 problem behavior. Parents were contacted directly by the researcher who accented the voluntary nature of the project but omitted any reference to any clinic. Approx- imately 90% of those contacted agreed to participate. Demographic Variables To ensure that the three subject populations were equitable, data on age, IQ, number of children in the family, and socioeconomic status were gathered on each subject. Group means on each of these variables were examined through an analysis of variance technique, and only age distributions were found to be significantly different (p §_.05) (Table 2). On closer inspection of the data, it seems likely that this significance is an artifact of the low variance in the normal population. While ages in the other two groups ranged from nearly six to nine, school restrictions demanded that we use only one grade level from which to draw our normals. Table 2 Analysis of Variance of Demographic Variables Variable F p Age 4.399 .019 IQ 1.582 .217 # of Children 1.647 .206 SES 2.419 .102 36 Test Situation In order to obtain a sample of interaction, triads consisting of mother, father, and identified sub- ject were presented with tasks which the child must solve and with which the parents could readily be of indirect assistance. Responses were videotaped. The examination room was arranged with a low table and small chair facing the camera. Two larger chairs for the parents were placed on either side of the low table, forming a semi-circle of parent-child-parent with the table at the center. This arrangement provides maximum exposure to the camera while allowing freedom of interaction (see Figure 1). Parent Child Parent Table Camera Fig. 1. Test situation Two types of tasks were presented, each during a 15-minute time period: 1. Parents were asked to teach their child a poem (Appendix H) which provided an interaction situation in which the parents were directly responsible for the child's performance. The 37 length of the poem was such that it was very difficult to teach it to the child in the 15 minutes allowed. In this situation, performance pressure was on the parents, as the child was not informed of the length of the poem. 2. In the second task, the child was asked to com- plete a puzzle before the examiner returns in order to win a prize. The puzzle was of a level of difficulty that made it impossible for the child to complete it within the time limit (CUDDLY by Milton Bradley Co., Springfield, Mass.). The use of this technique (1) presented the child with a (familiar task, (2) shifted the locus of motivation to the child, himself, and (3) provided a medium which will readily evoke helping responses_from parents. The above situation was designed to be analogous to that present in the everyday life of the family. In the child's development he is continually confronted with stressful situations in which parents can be of indirect assistance, and the assumption is made that their style and type of assistance in the testing situ- ation will parallel that given in other circumstances. 7 Natural family units were utilized where avail- _able, but problems in case finding within the disturbed 38 population required that some reconstructed families be used as well. To insure that the familial interaction patterns had stabilized, such families had to have been in existence for two or more years to be included. Procedure Mental health and cardiac subjects were selected from active cases within the clinic populations that met the criteria mentioned earlier. Normal controls were recruited from class lists within the school system. Each subject family was approached by the researcher or therapist who explained the project briefly as a "study of children and how they go about solving problems.” If they agreed to participate, an appointment was made for a testing session. Families with a child in treatment were tested at their treatment facility, while normal controls were assessed at Lincoln Center for Emotionally Disturbed Children or the Psychological Clinic, Michigan State University. Before any procedures began, parents were both asked to read a Parental Consent Form (Appendix D) that explained the project once again. After signing the consent form, each parent was given a copy of the Family Participation Index (Friedman, 1965) (Appendix A) to complete. This provided a measure of the interaction patterns within the family and also tapped the openness of the family system to interaction with its social environment. It was scored by giving 39 each symbol appearing in the "Occasionally” column a score of l and each symbol appearing in the "Often” column a score of 2. History information was gathered at this time as well. Concurrently, the child was given The Peabody Picture Vocabulary Test to provide a rough gauge of his level of intellectual functioning. Upon the parents' completion of the F.P.I., they were handed a card to read jointly: This is a study of children and how they go about solving problems. Each child is asked to do a task to the best of his ability. Your being involved will be helpful in making your child feel as comfortable as possible. If he seems to need help, then you may give it to him. Feel free to act as you would in your own home. Videotapes are being made of the childrens' responses. These will be kept confidential and will be erased when the study is over. Two tasks will be given to your child. In the first one, we would like you to teach him a poem. Any way of teaching the poem will be OK with us. Try to make the situation as much like what would happen at home as possible. The time for this part will be 15 minutes. An investigator will enter the room to let you know when the first task is over.‘ He will give your child a puzzle and will give him 15 minutes in which to solve it. At the end of 15 minutes, he will return to the room. The only restriction during this portion of the test is our request that the children move the puzzle pieces themselves. After the parents read the cards, the experimenter took them into the taping room, seating each one in the proper seat. They were told that they could begin at any time, and were left alone in the room. The experimenter returned in 15 minutes with a box containing several small new toys. The child was 40 shown the toys and invited to select one of them that he would like for his own. He was told that he will win the toy if he is able to solve the puzzle placed in front of him by the experimenter but that he must com- plete the puzzle before the examiner returns. Specifically, the child was told: First of all, I want you to choose one of these toys for yourself. [Child selects.) You can have this if you can put together a puzzle for me, but you'll have to work fast if you want to win the prize. Here's the puzzle. I'd like to have you work on. When it's done it will look like this [points to picture of completed puzzle]. I'm going to leave the room and your parents will be here with you. If you hurry and finish the puzzle before I get back, you get to keep the prize. The experimenter returned in 15 minutes. To terminate the stress situation he said: That's all the time that we can give you to do the puzzle. It is such a hard one that almost nobody can finish it in the time we give. But you did very well and did enough of it that I'm going to give you your prize anyway. The order of task presentation was randomized, with the instructions and procedure varying accordingly. At the end of the session, the experimenter turned to the parents and said: "Now, I'd like to talk to you, Mr. and Mrs. while spends a few minutes with ." The assistant observed the child in a free play situation while the parents were interviewed about ways they handle the child when he is under stress. 41 Scoring Each 30-minute session was videotaped, and a technique utilized by Alexander (1973) was adopted to facilitate scoring. Scoring_units were divided into 35- second intervals with 5 seconds devoted to watching the tape and 30 seconds used for coding the response. This produced a total of 50 scoring units per session, divided evenly between puzzle and poem segments. A parallel audio tape was used for timing. To reduce variability, the audio tape was started on a signal dubbed into each videotape segment. Each unit was scored separately for each of the three individuals involved. Parents were scored for (1) Presence of direct assistance in cOping with and reducing stress, (2) Confirmation of self-worth of child, (3) Disconfirmation of self-worth of child, (4) Presence of statements or nonverbal behavior which block the individualization of family members. Such statements would include comparisons of the subject with relatives (i.e., that's just the way your brother did puzzles) and impositions of family standards and rationalizations (i.e., to the examiner--we've never been good with our hands) as well as direct orders to perform, (5) Presence of statements or nonverbal behavior which enhances indi- vidualization such as encouragement to work independently and maintaining physical distance from the child, and (6) Affective level. 42 The child was scored for (1) Amount of help seeking behavior, (2) Acceptance of parental assistance and direction, (3) Rejection of parental assistance, (4) Task persistence, and (5) Affective level. (See Appendix B for Raters' Manual.) Raters Four raters were trained using two training tapes and instructions. After training, initial interrater reliabilities were determined through the use of a Pearson Product-Moment Correlation analysis of scores on a sample tape. The two raters with the best cor- relations were chosen to participate in the study. They rated the tapes independently, and their scores were pooled for further analysis. Reliabilities Each scale's reliability was computed using the Spearman-Brown formula for multiple raters. Reliabili- ties based upon all study data are reported in Table 3 and ranged from .45 to .97. In a review of these figures, it can be seen that the raters were in closest accord when scoring DIRECT ASSISTANCE, a simple measure of .any overture made to the child. The scale AFFECTIVE LEVEL produced less agreement, implying that it is a much more complex measure. However, in Spite of its lower correlation, it remained a powerful factor in 43 Table 3 Rater Reliabilities Scale Reliability Father's Direct Assistance .97 Father's Confirmation of Self-Worth .67 Father's Disconfirmation of Self-Worth .88 Father's Individualization Responses .91 Mother's Direct Assistance .96 Mother's Confirmation of Self-WOrth .59 Mother's Disconfirmation of Self-Worth .79 Mother's Individualization Responses .89 Child's Help Seeking .45 Child's Assistance Responses .94 Child's Task Persistance .93 Affective Levels--Father .63 Mother .60 Child .88 44 the analysis. CHILD HELP SEEKING was the least reliable scale, probably because help seeking was followed imme- diately by some form of assistance. It seems that the parental rating took precedence over the child's as it was usually a longer response and, hence, easier to score. Because of its low reliability, HELP SEEKING was omitted from the analysis. CHAPTER III RESULTS Discriminant Function Scoring procedures outlined above measure-the frequency of occurrence of 17 scores for each family. Each score was derived by summing the Puzzle and Poem frequency counts for both raters. As the maximum score on any segment was 25, the resulting potential range for the dimensions used in the study was 0-100. The data were analyzed with a discriminant function analysis which is designed to combine all raw scales to form general factors. The mechanics of the analysis involve a mathe- matical search for linear discriminant functions (Y) which discriminate best among the groups by minimizing overlap. Using the formula: number of discriminant functions a number of groups minus one, our data will generate two orthogonal (independent) functions. These will be rank ordered according to ability to discriminate, and the percentage of the total variance of all scores 45 46 accounted for by each discriminant can be calculated. Discriminant selection is based on the maximizing of the F ratio: variance between means on Y. variance within groups on Y. The individual variables are given weights on each discriminant function, with the same set of weights being applied to the scores of each person in each group. This will result in two new scores for each subject, 'Y1 and Y2, which can be plotted on a two-dimensional graph. Centroids, the points representing the group average profiles, can be plotted as well. They are obtained by using the group means on each dimension plotted on the coordinates. This process is similar to that of factor analysis, but using discriminant functions provides the option to retain the various treatment groups as separate entities. The main strength of a discriminant function analysis lies in its ability to predict the group member- ship of uncategorized subjects based on their discriminant scores in relation to group centroids. In making these predictions, centours (contours drawn around the centroid of each group) are used and a new subject is given a probability score for placement in each group based on his centour position. Actual prediction of membership is made to the group of highest probability. 47 The evaluation of a discriminant analysis is discussed by Nunnally (1967), who feels that the technique: . . . simply has not worked very well in studies to date. The amount of overlap between groups in the discriminant Space tends to overshadow the separ- ation between groups. Discriminant analysis is more useful in UNDER- STANDING the major differences between groups than it is in placing individuals in groups. (pp. 399-400) Given this inherent problem in psychological research using the discriminant function, the results discussed below and the predictive strength of the dis- criminant functions produced seem even more significant. Discriminant Function Results Multiple Groupg’ In beginning the analysis, all variables that represented positive and negative components of the same dimension as reflected both in face validity and in a reasonable negative correlation were combined (Appendix I). This was done through a simple summation, with the nega- tive values subtracted. The following variables were combined: FATHER'S INDIVIDUALIZATION BLOCKING AND ENHANCING; MOTHER'S INDIVIDUALIZATION BLOCKING AND ENHANCING; and CHILD'S ACCEPTANCE AND REJECTION OF ASSISTANCE. The above procedure reduced the number of variables in the study to 14. This portion of the analysis produced two functions summarized in Table 4. Though there is no decision rule for determining the significance of a 48 Table 4 Multiple Discriminant Functions and Loadings Scale Loadings Discriminant l Discriminant 2 Father's Direct Assistance Father's Confirmation of Self-WOrth Father's Disconfirmation of Self-WOrth Father's Independence Enhancing Mother's Direct Assistance Mother's Confirmation of Self-WOrth Mother's Disconfirmation of Self-WOrth Mother's Independence Enhancing Child Acceptance of Assistance Child Task Persistance Father's Positive Affect Level Mother's Positive Affect Level Child's Positive Affect Level .01554 .02358 .17716 .01758 .00425 .25224 .29491 .02494 .01322 .10070 .20677 .13388 .35606 .01723 .30851 .34100 .01348 .13616 .14027 .07669 .04986 .01227 .00975 .90999 .96812 .13455 49 specific function loading, the following criteria were used throughout the study: the largest loading on each function was taken as a base, and loadings were included if they were one-third of the base or larger. According to this rule, Discriminant Function I included CHILD'S AFFECT LEVEL, MOTHER'S DISCONFIRMATION OF SELF-WORTH, MOTHER'S CONFIRMATION OF SELF-WORTH, FATHER'S AFFECTIVE LEVEL, MOTHER'S AFFECTIVE LEVEL, and FATHER'S DISCON- FIRMATION OF SELF-WORTH. Discriminant Function II included MOTHER'S AFFECTIVE LEVEL, FATHER'S AFFECTIVE LEVEL, and FATHER'S DISCONFIRMATION OF SELF-WORTH. Since the primary test of a discriminant function analysis is its predictive ability, six subjects were held out in forming the function (two for each group) and were projected into the discrim- inant space. Of these Six, group membership of three was correctly predicted (Table 5). Table 5 Prediction of Subject Group Membership Subject Actual Group ’ Predicted Group aim-book)!“ MUNNHH QHWNWH 50 Group centroids on the two functions are given in Table 6. It is clear from an examination of these means that one discriminant function is primarily reSponsible for splitting the Mental Health vs. Normal groups while the other divides the Normals from the Cardiac Patients. Table 6 Group Centroids on Multiple Discriminant Functions Centroids Group Discriminant l Discriminant 2 Cardiac 1.56327 .13678 Normal Controls -l.07672 .47506 Mental Health Clients -.48655 -.61184 The above suggests the usefulness of two separate discriminant analyses contrasting each study group with the control group to sharpen the focus on the Specific factors separating the Mental Health and Cardiac popu- lations from the Normals and to increase predictability by reducing the level of variability in each analysis. Mental Health vs. Normal Groupg The following factors were of primary importance in separating the Mental Health group from the Normals (in order of importance): NEGATIVE FATHER'S AFFECT, POSITIVE MOTHER'S AFFECT, FATHER'S POSITIVE SELF-WORTH 51 STATEMENTS, and a 1225 of FATHER'S NEGATIVE SELF-WORTH STATEMENTS. MOTHER'S NEGATIVE SELF-WORTH STATEMENTS just missed being included according to the criteria outlined earlier (loading of at least one-third the largest loading in the function). Specific function loadings are given in Table 7. As in the initial analysis, the same two subjects per group were projected into the discriminant Space formed by the other 24 subjects. All four subjects were placed in their correct groups in this stage of the analysis, implying a strong predictive ability for the function. Cardiac vs. Normal Groupg When compared to the Normal group, the Cardiac Patients were discriminated by NEGATIVE FATHER'S AFFECT, NEGATIVE MOTHER'S AFFECT, POSITIVE CHILD'S AFFECT: FATHER'S NEGATIVE SELF-WORTH STATEMENTS, and _b_o_t_h_ MOTHER'S POSITIVE and NEGATIVE SELF-WORTH STATEMENTS (Table 7). Two of the four subjects held out of the initial analysis were correctly placed after the for- mation of the function in one-dimensional Space. One Normal and one Cardiac subject were missed. gardiac vs. Mental Health Groups Table 7 also outlines the results of the analysis contrasting the Cardiac and Mental Health groups. The 52 Table 7 Individually Run Discriminant Functions and Loadings Group Combinations Scale Mental Cardiac Mental Health VS Health VS Normal vs Normal Cardiac Father's Direct Assistance .02429 .02661 -.02033 Father's Confirmation of Self-Worth .70759 .03252 .07565 Father's Disconfirmation of Self-Worth -.69547 .20529 -.20869 Father's Individualization Enhancing -.06385 .02169 .03409 Mother's Direct Assistance -.ll995 .05160 -.01213 Mother's Confirmation of Self-Worth -.28112 .27494 -.28272 Mother's Disconfirmation of Self-Worth .43932 .19520 -.36146 Mother's Individualization Enhancing -.l3776 .06171 -.04824 Child's Acceptance of Assistance -.25390 -.04029 “-- Child's Task Persistance --- -.l6061 .10546 Father's Positive Affective Level -1.3852l -.25088 .10356 Mother's Positive Affective Level ' 1.13804 -.36969 .48926 Child's Positive Affective Level .36816 .50268 -.64672 53 Mental Health group exhibited more NEGATIVE CHILD AFFECT and POSITIVE MOTHER'S AFFECT and .1235}. of both MOTHER'S POSITIVE Egg NEGATIVE SELF-ESTEEM COMMENTS, a finding which essentially confirms the results of the other two separate computer runs. Again, two out of four of the test subjects were correctly placed by group. Thus, the above discriminant functions confirm the original analysis using all three groups, in terms of the specific functions selected to divide the groups. The separate runs increased predictability from 3 of 6 to 8 of 12, with most of the predictability centered on the MENTAL HEALTH vs. NORMAL analysis. Demographic Variables Since a significant group difference on age had been found, a Pearson Correlation analysis was run including age and each of the other variables in the study. The results can be found in Table 8 and Show no Significant correlation with any of the scales upon which the discriminant functions are based. Age wag found to be Significantly correlated with FATHER'S INDIVIDUALIZATION ENHANCING (r = .43, p :_.01) and negatively correlated with FATHER'S DIRECT ASSISTANCE (r = -.47, p < .01). 54 Table 8 Age Correlation with Study Variables ._. 62:31:13?“ Father's Direct Assistance -.47 Father's Confirmation of Self-Worth -.35 Father's Disconfirmation of Self-Worth -.14 Father's Independence Enhancing .43 Mother's Direct Assistance -.05 Mother's Confirmation of Self-Worth -.20 Mother's Disconfirmation of Self-Worth -.15 Mother's Independence Enhancing .23 Child's Help Seeking -.l4 Child's Acceptance of Assistance .ll Child's Task Persistance .43 Father's Affective Level .13 Mother's Affective Level -.22 Child's Affective Level -.37 55 Familprarticipation The Family Participation Index was scored by giving each symbol appearing in the "Occasionally" column a score of l and each symbol appearing in the "Often" column a score of 2. Using an analysis of variance, no significant difference was found among groups. CHAPTER IV DISCUSSION Mental Health Differentiation From Normal To answer the specific research question with which we began this study, a discriminant function was generated which proved capable of dividing mental health clients and normals within the stressful, structured environment generated by the test Situation. The mental health families included fathers exhibiting more negative affect, mothers whose affect level was more positive than in the Normals, and the presence of responses from the father which tended to confirm the self-worth of the child. The predictive ability of this function appears to be strong though given the small sample size and the difficulty encountered in prediction when the Cardiac group was included in the analysis, a replication would be needed before the results could be considered to be definitive. The Mental Health was distinguished from the Normal group in two separate forms of the analysis: 1. The initial run in which variables on positive- to-negative continuums were combined. 56 57 2. When Mental Health vs. Normal was run directly, the same discriminant function emerged with the variables in the same rank order of importance and with a 100% predictive capacity. Thus, the above stability and predictive ability of the discriminant function establish it as a relevant dimension separating our client population from the controls. The discriminant can be characterized as extreme ambivalence, primarily on an affective level. The client fathers exhibited an overtly negative affect (at least in this situation) though appearing to compensate for it with positive self-esteem comments directed to the child. From this it is concluded that such positive self-esteem comments delivered in a negative atmOSphere are not con- structive and, in fact, may be destructive to the child's emotional development. It may be argued that these fathers were on their best behavior, attempting to give the positive responses that an ongoing therapy contact had taught were most apprOpriate. However, such an hypothesis does not do damage to the overall interpretation in light of the following considerations. First, most cases were included in the study early in treatment, while the therapist was still involved in history-taking and/or catharsis and before any formal instruction or modeling 58 could occur to a significant degree. Second, even if the father were feeding back the ”correct” response, his elevation of self-esteem reSponseS beyond the normal level shows a characterological tendency to mask his true feelings in an inappropriate display of positive concern. Such a response style would parallel the pseudomutual responses described by Wynne, where overt harmony and affection serves as a mask for an underlying hostility and unacknowledged anger. The mother's positive emotional level adds further ambivalence to the child's environment. Most Obviously, there is the contrast between father and mother's affect and the confusion it must produce in a child attempting to create a stable self- and familial image. But perhaps more importantly, even the mother's positive affective responses become suspect when the presence of MOTHER'S NEGATIVE SELF-ESTEEM comments in the discriminant function is considered. Though this scale's loadings were not significant enough to add to the overall predictive ability, (Table 7) they could be interpreted to Show an ambivalent trend in the interaction with her child when combined with her POSITIVE AFFECTIVE LEVEL. In addition, there is no logical explanation for the mother's ele- vated affective level as compared to Normals. It seems likely that the affective score represents a reaction formation on the mother's part-~her response to the 59 taping session, but also her general response to stress situations. Thus, the mother's responses toward the child contribute to rather than contradict the paternal ambivalence. Cardiac Differentiation From The Normal The cardiac patients were included in the analysis to examine the possibility that some of the interactional factors in the study could be related to the presence of a high risk child in the home. Such a child would intro- duce stresses on the family network. Though the function generated proved weak in pre- dicting group membership, this population was found to include fathers and mothers who tended to exhibit nega- tive affect, while the child's affective level was con- sistently higher than in the control group. Cardiac fathers were disconfirming of the child's self-worth, while mothers consistently made more confirming 35g disconfirming comments to the child when compared to normals. Such a finding suggests that the burden of a child at risk for heart problems rests primarily on the parents, and that the mothers may be attempting to com- pensate and adjust by fostering an intense relationship with the child. The cardiac child apparently is not seriously affected by the condition--at least in its ambulatory form--and may, in fact, enjoy a position of 60 privilege within the family sphere. The child's elevated affective level may also be a result of the continued close involvement with the mother. Given the level of predictability, more research, including longitudinal studies, needs to be conducted to test the above conceptualization. But what can be deter- mined from the data presented here is the negative effect of a physiologically high risk child on the parents' general emotional tone. This implies that such parents, while perhaps not generating a mentally disturbed child, are themselves at risk psychologically. In OOping with the added stress, they are likely to need a more complete support structure than normal parents, including contact with other parents in a similar position, or some form of direct counseling. Mental Healtthifferentiation From Cardiac As expected, the Mental Health vs. Cardiac function served to reinforce the patterns that dif- ferentiated each group from the Normal. Each factor working strongly in the function could have been pre- dicted using the other two distributions. Integration The extent to which the Cardiac and Mental Health groups are similar provides a clue to the origins Of the disturbed interaction present in the client population. 61 For, any component Shared by the two groups could be considered the result of coping with a problem child and, subsequently, one requiring a Special adaptation. Care must be taken in interpreting these similarities in view of the weakness of the Cardiac function, but it is felt that one trend Egg apparent in the analysis and is worthy of mention. A similarity was found in the father's affective level. When this is combined with the high maternal negative affect in the cardiac group, it becomes evident that negative parental affect is characteristic Of fami- lies with a "problem" child in either the social or medical sphere. Negative parental affect is not neces- sarily destructive in itself. Apparently it can be triggered by external circumstances and need not affect the child either by reducing his own affective level or by producing psychological or behavioral symptomatology. The fact that Simple negative affect in the Cardiac group has little effect on the child enhances the importance of the ambivalence in the affective environment of the Mental Health group. AS we discuss in more detail above, the ambivalence found in the client population Operates on two levels. First, the father's negative affect combined with his positive ego building communication produce a double message which undoubtedly is confusing to the child. As out- lined in Appendix B, such communication includes direct 62 verbal encouragement ("that's fine," "good," "you're doing good") as well as any especially warm nonverbal response (such as hugging, combing hair out of eyes, smiling directly at the child). The mother, on the other hand, exhibits a positive affective tone, while tending to depreciate the child's self-esteem through angry glances or such comments as: "hurry up," "here, let me do it" (puzzle), or "get busy" delivered in a negative context. Thus, the child is caught in a web in which father is affectively negative but delivers self-esteem rewards, while mother is overtly more posi- tive though belittling. The child in such an environ- ment is left with no consistent, uniform source of either feedback or validation Of his own perception of his parents' positions. Instead, he receives from his parents a higher level of both positive Egg negative feedback which he must attempt to integrate. Fathers were able to be more positive in the relatively instrumental self-esteem reSponses while mothers seemed to elevate their general affective tone over the normal population in their attempts to compen- sate; it may be concluded that this is the result of differing sex role response styles. The finding of affective ambivalence is con- sonant with the work of Satir and other family therapists who directly assist families in rephrasing their 63 communication of feelings. It is Satir's contention that anger and hostility are really a mask for hurt and defensiveness, and she and her followers have had a great deal of success in putting family interaction on a less ambivalent positive plane. Though less directive, Wynne's approach to the problem through the twin concepts of Pseudohostility (anger masking affection) and Pseudomutuality (affection masking anger) also leads to a restructuring and clarifi- cation Of affective messages in a troubled family. Affective ambivalence may also be addressed within a more psycho-dynamic orientation. Framo's use of parental introjects (those perceptions and expec- tations which are internalized by the parent as a child and then projected into his own family situation) is a good example of such an approach. The parents, once freed of old introjects, will then be able to be more congruent in their relationships with the child. Role of the Father Most of the scales loading heavily on the dis- criminant functions were measures of the father's behavior or emotional state. His general affective level was, in fact, the most powerful element in the entire study. Further evidence of the father's important role is found in the correlations of the scales with the age 64 of the child. While essentially unrelated to the more powerful discriminations, age was found to correlate with two measures of independence in interaction with the father. Fathers tended to give their Older children less direct assistance and to enhance their individuali- zation more frequently; this correlation was not found for maternal behaviors. Thus, the father seems to have two important functions within the family. While we can only speculate as to whether he is responsible for his child's growing independence or responds to it more directly when it occurs, it is clear that he does play a role in his child's movement from the Sphere of familial influence into the extended social environment. In addition, as will be discussed in more detail later, his affective tone as well as his efforts to compensate for it have been strongly implicated in the types of childhood psy- chopathology which come to the attention of mental health workers. These two findings raise serious questions about the advisability of solely involving the mother/child dyad in treatment when a father is present. To do so would be to select out and ignore a major aSpect of the problem. In summary, ambivalence in the affective and self-worth messages given to the child has been shown 65 to be an important factor to consider in treating the mental health client. Negative affect alone does not seem to harm the child unless it occurs in a context of confusion and negation of messages, leaving him with no clear understanding of his parent's position. This issue can be dealt with using a variety of approaches and conceptualizations, and the research has established that the father, far from being an isolated nonmember of the family, is an important element both in the problem and the ultimate cure. Additional Scales The two measures of DIRECT ASSISTANCE were ineffective in discriminating the groups. The important dimension proved to be the emotional climate in which the assistance was offered. In retrospect it seems clear that the parental INDIVIDUALIZATION scales were fundamentally contaminated. Specifically, Individualization Enhancing was designed as a positive scale, but it could also have been scored on the basis of rejecting withdrawal. Support for this explanation of the relationship between the individuali- zation scales and other parental behaviors is found in each scale's correlation with the reSpective parental NEGATIVE affective level (r = .23 for both). Thus, it is evident that individualization as a positive move 66 toward independence was not measured by the scale developed within this project. The lack of discriminant ability Of the CHILD behavior scales would imply that task behavior, at least in such a structured environment, does not necessarily reflect psychological disturbances. The behavioral checklist clearly differentiated the Mental Health population from the other two groups (F = 12.78, p 5 .001) although no significant differences were found in interview behavior. It is difficult to gen- eralize to home or school from a structured interview situation, but this result, in conjunction with the child therapist's frequent inability to duplicate those behaviors which precipitate the referral, would lend support to a situational determinant of childhood disorders. Summagy This research has directed attention to the role of emotional ambivalence in the psychologically dis- turbed family. Where self-worth statements and overt affective tone conflict, both within and between parents, the child has no firm base from which to build his perceptions of self and others. This suggests ambivalence is even more destructive than congruent negative affect on the part of the parents. It follows from the above that behavioral attempts to train parents 67 to respond more positively to the children are likely to increase ambivalence instead of reducing it unless they are also combined with changes in the affective states of the parents. The unambivalent negative affective levels of Cardiac parents are more benign but point to the need for Special support structures to prevent psychological casualties in the parents. The children with cardiac problems apparently escape the strain and pressure of their condition and may, in fact, enjoy their privileged position. Further, the importance of the father's involve- ment in the emotional life of the family has been estab- lished. His affective tone and resultant reSponse to the child have been linked to emotional disorders in childhood. In addition, his general style of relating to the child is more age specific than the mother's, implying that his role in moving the child toward independence is crucial. Thus, the father plays an important part in the interaction of the family; he can- not be ignored in successful therapeutic intervention. Finally, we found that none of the child behavioral scales discriminated the groups. This implies that the reported behavioral problems of client children are situation Specific and, therefore, more social than dynamic. APPENDICES APPENDIX A FAMILY PARTICIPATION INDEX Name APPENDIX A FAMILY PARTICIPATION INDEX Date Specify the activitIes in which you participate alone or together with other members of your family. An activity in which you have participated at least once in the past Six months would be checked as "occasionally." If, how- ever, you participated more frequently than this, check it for "fairly often." List the following symbols in the apprOpriate Spaces below to indicate what you do with whom? P = Marriage Partner * S = Son D = Daughter R: Rater Himself * If more than one, rank according to age by using number after the symbol: for example, Dl for oldest Daughter, 52 for second Oldest Son, etc. Activity Family Members Who Par- ticipate with You Fairly Often Occasionally l. 2. 3. Movies Dances Play baseball, foot- ball, etc. Watch baseball, foot- ball, etc. Outdoor activities, hikes, picnics, etc. Social gatherings with friends, parties 68 69 Family Members Who Par- ticipate with You Activity Fairly Often Occasionally 7. Politics 8. Hobby clubs and classes 9. Membership in clubs and organizations 10. Business or professional activities 11. Meals out 12. Attend church 13. Listening to radio, watching TV with friends 14. Meetings and parties with family members not living in your home APPENDIX B RATING SCALES FROM RATER'S MANUAL APPENDIX B RATING SCALES-PARENT 1. PRESENCE OF DIRECT ASSISTANCE IN COPING HITH.AND REDUCING STRESS. Any active response on the part of the parent that makes the child's task easier or provides emotional support. Examples: POEMdverbal encouragement, mouthing the words with the child, providing hints when the child seems to be forgetting a line. PUZZLEdverbal encouragement, directing the child's attention to pieces that might fit specific sections, directing the child to move pieces. 2. CONFORMATION OF SELF NORTH OF THE CHILD. A parental behavior that enhances self-esteem; makes the child feel good about himself, adequatg, etc. . Examples: POEM~positive verbal encouragement delivered with positive affect ("you're doing fine," "there you go" after a success, "good." PUZZLE-positive verbal encouragement delivered with positive affect. Additional Examples: In addition to reSponses including verbal encourage— ment, any especially warm nonverbal response (such as hugging, combing hair out of eyes, smiling at the child) should be scored as well. 3. DISCONFORNATION OF SELF NORTH OF THE CHILD. A parental behavior that reduces self-esteem r tends to make the child feel like a "bad person," defective, inadequate, etc. Examples: POEM-negative affect in responses to the child, angry glances and other nonverbal signs of displeasure as parents try to enhance per— formance. PUZZLE-all of the above, parent taking over the task for the Child a This category is a very difficult one to find examples of for two reasons. First, the parents in the testing session have little invest- 'ment in their child's performance in a completely experimental situation. Thus, they are under no real pressure to increase their child's task oriented behavior and this allows-them to stay somewhat detached from the proceedings. Second, and probably most important, is the effect of the videotape on the parents. They are aware that they are being taped, and this seems to put them on their best behavior. Thus, this category should be scored strictly and when a question exists the benefit of the doubt should not be given to the parents. As direct, verbal disconformation is not likely to occur in sessions, this item will be determined primarily on the basis of the affective tone of the parent's responses to the child. Any hint of IRRITATION, FRUSTRATION, or ANGER should be scored. These affective states could be revealed in a sharpness of the voice, an increase in volume or perhaps raising of the tone level. Such occurrences should be scored whether or not they are combined with nonverbal signs of displeasure. 70 71 4. PRESENCE OF STATEMENTS 0R HON-VERBAL BEHAVIOR WHICH BLOCK THE INDIVIDUALIZATION OF FAMILY MEMBERS. Parental responses which tend to prevent the child's operating as an independent person, responses which keep him eithin the family circle. Example: BOTH-comparison of the subject with relatives ("that's just the way your brother would do it"), impositions of family standards and rationalizations (to the examiner-we've never been good at this"), direct orders to perform, excessive physical closeness 5. PRESENCE OF STATEMENTS 0R NON-VERBAL BEHAVIOR WHICH ENHANCES INDI- VIDUALIZATION. Parental responses which tend to encourage the child to Operate as an independent person. Examples: BOTH-encouragement to work independently "you can do that all by-yourself" "I'm not supposed to help you" and such comments as "don't be lazy, you can do it yourself!" POEM-anytime that the child is handed the card (or the parent holding it allows it to be taken) I.E. will be scored for the parent on THE FIRST RATING SEQUENCE THE CHILD HOLDS THE CARD ONLY. For example, if the child is given the card between sequence 5 and 6 and holds the card for 6, 7, 8, 9 and 10, the parent giving him the card is scored for I.E. on #6 ONLY. Since the task necessitates a great deal of interaction and parental involvement, the following discrimination will apply to memorization sequences. BLOCKING-Parents reading the poem to the child either with the child remaining silent or reading along with them. Here, the major discriminating factor is the parent's failure to pause to allow the child time to respond on his own. NEUTRAL-Parent reading a line and then pausing to allow the child time to repeat it back to him. Here, the child is guided in his productions, but he is also allowed time in which to respond on his own. ENHANCINGrParent provides only key words to help the child begin or finish a phrase or corrections when the child makes a mistake. The spotlight is on the child in this category, and the parent is acting as the facilitator for the child's responses. Here, it might be noted that the categories of the scales are not designed to be mutually exclusive within a rating sequence. For example, during a poem sequence one parent could alternate between conformation and disconformation of his child's self worth and be- tween blocking and enhancing of his individualization. This is expected, especially with the poem where the task is progressive and the child gains expertise during the session. On either task, anytime that BOTH parents refrain from any in- teraction with the child for the 5 second rating sequence, they BOTH receive a score for individualization enhancing during that se- quence. Direct Asistance is not scored for either parent. 72 RATING SCALES-CHILD 1. AMOUNT OF HELP SEEKING BEHAVIOR. Requests to either parent for assistance in the task. Examples: "it's too hard for me," whining, eye contact. 2. ACCEPTANCE or PARENTAL ASSISTANCE AND DIRECTION. Willingness to follow directions and respond to advice. Examples: POEM-parent "now try it from the beginning" child Responds, parent - "listen to it again" child listens attentively. PUZZLE-child responds to suggestion of pieces fitting specific slots, parent - "find the flat peices first for the outside" child follows(thié response also blocks individualization because it gives the child no choice.) 3. REJECTION 0F PARENTAL ASSISTANCE AND DIRECTION. Unwillingness to follow'directions and respond to advice. Examples: BOTH—child either insists in doing the tabk.himself or refuses to follow specific directives from parents. Both 2 and 3 are contingent on a parental suggestion or directive. 4. TASK PERSISTENCE. Presence of an attempt on the child's part to cope with the task in a manner directed toward its completion. AFFECTIVE LEVEL - PARENT AND CHILD 73 This category is designed to be a global measure of the general affective tone of each of the participants. It will replace the Child's Self Esteem dimension in future rating. Individual scores will be recorded for mother, father and child at the end of each 15 minute rating segment. To simplify scoring, each participant's score should be entered to the left of his scales on the rating sheet for that segment. While the affective tone in a session is, in part, an interdependent function of all the participants, for our purposes each family member needs to be raced independently. It is quite possible within one session for. example, to rate the father at 1 (much positive affect) and the mother at 4 (much negative affect) with the child's score anywhere between 1 and 4. General cues to be used in scoring this category positively include the presence of smiles, laughter, enthusiasm, confidence and assurance. Expressions.of anger should be scored for negative affect as should negative verbal behavior such as whining and refusing to perform.(child) as well as depressed or impatient disinterest. Often facial clues can he's great deal of help in making a determina- tion of affective level. For example, the mouth turned downward slightly combined with a downcast gaze would indicate negative affect. On the other hand, an active, mobile face, with eyes focused either on the task or on 1 other objects in the room would reveal a level of self-assured behavior sufficient to be scored high on positive affect. Other useful clues could be found in the child's body posture. "Slouching" in the chair or Sitting with shoulders drOpped could indicate negative affect, eSpecially when combined with the facial determinants 74 mentioned above. Thus, an alert posture, sitting straight in the chair or leaning forward into the task would indicate confidence and enthusiasm and would be scored positively. SCALE 1. Clear evidence that the subject is happy, pleased with his performance, confident and pleased with others. In general, he seems to be having a good time. 2. The subject is not enjoying himself quite as much as in #1, but he performs his task.without frustration or boredom.and with some confidence. 3. Here, subjects are not overtly depressed or hostile, but they do show some hints of boredom, dissatisfaction or mild unhappiness. 4. A subject scored at this level has revealed a negative affective level on one of two dimensions, anger or unhappiness. The judgment could be made on the basis of tone of voice, nonverbal posture or verbal content. Anger could take the form of frustration or impatience and may not be directed at anyone in particular. Unhappiness could be seen in whining and tears (usually the child!) or in a generally flat emotional tone (either parent or child). APPENDIX C CHILDREN'S BEHAVIOR CHECKLIST APPENDIX C CHILD STUDY PROJECT CHILDREN’S BEHAVIOR CHECKLIST Name of child: Age: Date: w—— Name of person filling out checklist: Relationship to child (mother, father, teacher, clinician, etc.): 'Situation in which child has been observed (home, school, clinic, .etc.): This is a list of items describing many aspects of children's behavior-~things that children do or ways they have been described by others. Not all of the items will apply to the particular child you are describing, but quite a few of them will.. First, go through the list and put a checkmark (j) in the first column by each item which applies to this child. If there are some items which you do not check because you do not know whether they apply or not, or have never had the opportunity to observe them (for instance, "He (she) is a finicky eater,” if you see this child only in school and don't know anything about his (her) eating habits), put an (O) in the first column. After you have gone through the list, please go back through those items you have checked and put another checkmark (x/) in .the second column opposite those that are now most characteristic of this child, that describe how he (she) is most of the time. Does this Is it char- apply at all?- acteristic? 1. Is tidy and neat, perhaps even a 1 little bit fussy about it. 2. Is concerned about feelings of others. 3. Can't wait - must have things immediately. 4. Gets irritated or angry easily. 5. Doesn't pay attention to what grown-up says to him (her). 6. Looks awkward when he (she) moves around. 75 Children's Behav. List 76 Does this Is it char- apply at all? acteristic? 12. 13. 14. 15. 16. 17. 18m 19. '20. 21. 23. 24. 225. Feelings are apparent in facial : expression. 8 l 'Handles small objects skillfully. i Is left out of things and ignored by others. Can be depended on to do what he (she) is supposed_to do without reminders. Likes to play with girls instead of boys. Can accept new ideas without getting upset. Appears stiff in walking or moving about. Shows pride in accomplishment. Seems comfortable in new situations. Does what other adults ask him (her) to. Has trouble finding the right words tO' say what he (she) means. Moves gracefully - is well coordinated. Plays to win. ‘ Starts things off when with others. Others seem to want to be with him (her). 1 Has a characteristic mannerism or nervous habit. Specify: Makes friends quickly and easily. Quickly loses interest in an activity. Self confident. -—~ woom‘-’ " ' Children's Behav. List 77 Does this Is it char- apply at all? acteristic? 26. Plays mostly with younger or smaller children - even when children of own age are around. 27. Seems sad and unhappy. 28. Tends to go too far unless frequently reminded of rules. 29. Talks all the time. 30. Often has to be reminded of what he (she) can and cannot do. 31. Seems out of touch with what is going on around him (her) - off in his (her) own world. 32. Has uncontrollable outbursts of temper. 33. Able to stand up for himself (herself). 34. Polite and cooperative with others. 35. Easily embarrassed. 36. Careful in explanations — precise. 37. When told to do something he (she) doesn't want to do, he (she) becomes very angry. 38. Play is aimless, doesn't seem to make or accomplish anything. 39. Is curious about things. 40. Shows appreciation when others help or do _ things for him (her). : 41. Will lie to get out of a tight spot. 42. Energetic. Children's Behav. List 78 Does this Is it char- apply at all? acteristic? 43. Asks sensible questions in new situation. 44. Shows pleasure and involvement in most things he (she) does - enthusiastic. 45. Seems selfish, always wants own way. 46. Doesn't seem to care about hOW'he (she) looks - often looks sloppy. 47. Prefers following others to taking the iniative. 48. Competes with other children. 49. Pitches in when things are to be done. 50. Fidgety and restless. 51. Acts as if everyone were against him. 52. Quick and clever. 53. Learns quickly from others. 54. Retains composure even when those around him (her) are acting in a boisterous way. 55. Prefers playing with older or bigger children even when child of own age are around. APPENDIX D SUBJECT RELEASE FORM APPENDIX D FAMILY STUDY PROJECT DEPARTMENT OF PSYCHOLOGY_ MICHIGAN STATE UNIVERSITY We would like your help in a study of children and how they go about solving problems at home. If you agree to participate, it will involve only one session of less than an hour in which your child will be asked to perform two tasks to the best of his ability. we are asking both parents to attend the session to make it as much like home as possible. Families that agree to participate will be paid the sum of five dollars ($5) to compensate for their time. Payment will be made at the close of the testing session. The project is completely separate from whatever treatment your child will receive at the Center, and is designed to assist us in better understanding the children who come to us fOr help. It is being conducted.by Mr. Keith Lyon, who will be available to answer any further questions you might have. Of course, all information Obtained in the study and the videotapes made of the sessions will be kept strictly confidential. I, , parent of , have read the above description of the Family Study Project and agree to participate in it. Signed 79 tVitness Date APPENDIX E INTRAGROUP CORRELATIONS OF STUDY VARIABLES .va mum .ma .HH .oa nuance ou mmflammm Oman coauofiuomcw mesa «am>mq 0>Huoommd m.Hmcuoz n Adz “H0>mq 0>Huoowm¢ m.umnumm u Adm “oucmumwmumm xmca m.©aw£0 “moamumammd mo oucmummood m.oHHnU .Hm>oq m>auomume m.oaeeo n qeo mBU mm<0 «nuuozumamm mo coHumEHHmcoomHQ m.Hmnuoz u 3mm: anuuoz Imamm mo coaumeuflmcoo m.Hwnuoz n zmoz “cowumNHHmsofl>aocH m.Hmnuoz n Qsz «mocmumwmm< uowuwa m.umzuoz u do: “nanozumamm mo coaumauflmaoomflo m.Hmnumm u Swan «nuuosumamm Mo Godumfiuwmcoo m.Hm£umm n 3m0m «coaumNHHmsofl>aocH m.Hm£umm n asz «mocmumflmmd uomuwn m.Hmnumm n avsum mo mCOHumHoHHOU msoummuucH 8O 81 oo.H mm.l mo.l mo. mm. mm. mm. mo. vo.l mH. Hm: mH. Hm.l 3sz oo.H hN.I mo. mo. mo.l 0H. mN.I mo. Hm.l mm. mo. mo.l zmoz oo.H mm. mH.I mm.l mm.| vH. MN. mo.l oH.I mN.I hm. Smoh oo.H mm. 00. mo.| hv.l VH. HN.I vo.l o~.I mm. 3m0m oo.H mm. Hh. mm.l Hm.l mH. No.l mo.l vo.l HdU oo.H mm. mN.I vm.l mH. mH.I MN.I mH. Adz oo.H hm.l mm.l wH. «H. 00.! «N.I Hém oo.H mm. 00. mm.l Ho. mm. mBU oo.H mm.l mv. Nh.l mv. mmdU oo.H hm.l Hm. mN.I QZHZ oo.H om.l Hv.l 4Q: oo.H mm.l Qsz oo.H dam 3mm: ZmUS Zmam Zmum 440 442 44% mBU mmdo QZHS dds QZHh dam moonw Hmeuoz OH mHnma mmHnmwum> mosum mo mcoHuMHmHHou muoummuucH 82 oo.H 5v.l mm. mm.l mm. mm. mm. hH.I em.l mH. Ho. mo. mN.I BmQZ oo.H om.l mm. mm.l mv.l NH.I mm.l mm. Nm.l mm. 00. mm.l Smoz oo.H vN.l mm. Hm. ww. mo. wH.I no. mN.I mm.l mm. 3mnm oo.H mN.I mN.I mN.I mH. mm. mm.| h¢.l mm.l om. Bmom oo.H mm. 5w. mh.l Nm.l mo. HH. mm. vN.I H¢U oo.H Nb. mw.l mm.l Hm. mm.l Hm. mv.l Adz oo.H mm.l mm.l mm. HN. mo. mm.l Adm oo.H om. mo.l mo. hN.I mo. mBU . oo.H om.l mo. mm.l Nv. mm40 oo.H mm.l om. mN.I QZHE oo.H hm.l NM.I dflz oo.H Hm.l Qsz oo.H dam BMQE Smoz Bmam Bmom Ado H<2 Adm mBU mmdo QZHZ €92 Qth dam msouo nuHmwm Hounds HH OHQMB mmHnmwum> modum mo mcoHuMHouHoo msoummuucH APPENDIX E GROUP MEANS AND STANDARD DEVIATIONS o.H ~.N h.H m.N m.v m.m m.m o.m m.mH v.N o.NN n.5H m.o~ «.mH om nUHmmz h. o.~ m. m.m m.m v.m m.m o.Hm N.mm m.v H.m m.mv N.vH h.mm z mCOHomH>oQ ouoocmum one name: oHanHm> NH OHQMB h Xanmmmd 83 APPENDIX G DEMOGRAPHIC VARIABLES N. N.N H N N N N N N N N N N N N N couoHHso no N N.N N.NN N NN NN N NN NN N NH HN NH NN NH oN NN uoH>mrmm o>HumNoz N.NH N.NN NN HH NH NN NN NH NN Ne NH NN NN NN NN «N uoH>mamm o>HuHmom N.NH H.NNH NNH NN NN NN NNH NoH oHH NNH NN NHH NNH NNH NN NHH oH N. N.N N N N N N H N N N e N N N N on N.H N.N N N N N N N N N N N N N N N oNN NN HN NN NN NN NN NN NN NN NN NN HN NN NN somanm macaw nuHoom House: N.H N.N N N N N N N N N N N N N N N eonNHHno no N N.N N.N NH N N N N N H NH O N NH N N N uoH>NeoN o>HumNoz N.N H.NN Ne Ne Ne NN Na NN NN Ne NN NN Ne NN Ne NN uoH>Nnom o>HuHNoN N.NH N.NHH NNH NNH NNH NNH ooH NNH NN NNH NHH NoH NHH NHH NNH NNH oH N. N.N N N N N N N N N N N N N N N sow N. H.N N N N N N N N N N N N N N N mNN NN NN NN NN NN NN NN HN NN NH NH NH NH NH soonnsm msouo Houusou H.N N.N N N N N N N N N N N N N N N emucHHro no N H.N N.N N N N HN N N NH «N 0 HH H N N Ho noH>mnom o>HumNoz N.N N.NH NH NN NN Nv NN NN Ne NN Ne NN NN NN HN NN uoH>mnom o>HuHNoN N.NH N.NN NHH HNH HNH oHH NHH NN NN NN NNH NN NNH NoH HNH NN oH N. N.N N N N N N N N N N N N N N N on N.H N.N N N N N N N N N N N N N N N 0N4 NH NH NH HH NH N N N N N N N N H uooanN msouo ooHpHmo om zen: mmHQMHHm> OHnmmanEoo NH «Home 0 xHozmmmd 84 APPENDIX H TEST SITUATION POEM APPENDIX H TEST SITUATION POEM Whenever I walk in a London street, I'm ever so careful to watch my feet; And I keep in the squares, And the masses of bears, Who wait at the corners all ready to eat The sillies who tread on the lines of the street, Go back to their lairs, And I say to them, "Bears, Just look how I‘m walking in all of the squares." And the little bears growl to each other, "He's mine, As soon as he's silly and steps on a line." And some of the bigger bears try to pretend That they came round the corner to look for a friend; And they try to pretend that nobody cares Whether you walk on the lines or squares. But only the sillies believe their talk; It's ever so portant how you walk. And it's ever so jolly to call out, "Bears, Just watch me walking in all the squares." 85 APPENDIX I OVERALL CORRELATION OF STUDY VARIABLES oo.H H5. mm. vm.! hm.! 0H. NH. v0. ¢0.! N0. hm. mo. m0.! 440 00.H Hm. Nv.! mv.! vN. 5N. 0N.! mm.! HH. mm. m0. m0. 442 00.H 0N.! Hm.! mH. 0N. VH. VN. MN. mv. Nm.! Hm.! Adm 00.H Nb. m0. m0.! NN.! mo. 00. NH.! mN.! 0H.! mBU 00.H 0m.! m0.! m0.! hm. hv.! mH.! H0. NH. mmdu 00.H m0.! NH.! 00.! 55. N0.! NN.! 0N.! QZHZ 00.H MH. 0H. m0.! mm. NH. 0H.! 3mg: oo.H 0v. 00.! MN. mN. hH.! 3muz oo.H 0N.! H0.! 00.! vv.! flax oo.H 0H.! 0m.! Nm.! Qsz oo.H no. 0H. 3mam oo.H mo. Zmom oo.H dam 440 H¢2 Adm mBU mmdo QZHE 3m02 ZmUZ £62 Qsz Smom Smoh 4am mmanHHm> mosum mo cONuMHmMHoo HHmum>o VH OHQMB H Nanmmmd 86 LIST OF REFERENCES LIST OF REFERENCES Ackerman, N. W. (Ed.) Exploring the base for family therapy. New York: Family Service Association of America, 1961. Ackerman, N. Family process. New York: Basic Books, Inc., 1970. Ackerman, N. W. Treating the troubled family. New York: Basic Books, 1966. Alexander, J. Defensive and supportive communications in normal and deviant families. Journal of Consulting and Clinical Psychology, 1973, 32, 223-231. Beels, C., & Ferber, A. Family therapy: A view. Bell, J. B. Family group therapy. Public Health Mono- graph, 1961, 61, HEW. Bell, J. E. Recent advances in family group therapy. Journal of Child Psychology and Psychiatry, 1962, 3, 1-15. Brown, S. L. Breakdown in family developmental process and the psychiatric preschools center as a therapeutic agent. American Journal of Orthqpsychiatry, 1967, 31, 312-313. Caplan, G. Principles of preventive psychiatry. New York: Basic Books, Inc., 1964. Committee on the Family of the Group for the Advancement of Psychiatry. Treatment of families in conflict: The clinical study of family process. New York: Science House, 1970. Coopersmith, S. The antecedents of self-esteem. San Francisco: W. H. Freeman, 1967. 87 88 Dohrenwend, B. P. The social psychological nature of stress: A framework for casual inquiry. Journal of Abnormal Social Psychology, 1961, pg, 293-362. Dohrenwend, B. P., & Dohrenwend, B. S. Social status and psychological disorder. New York: John Wiley and Sons, Inc., 1969. Fairweather, G. W., Sanders, D. H., Cressler, M. L., & Maynard, H. Community life for the mentally_i11. Chicago: Aldine Publishing Co., 1969. Fox, R. B. Effect of psychotherapy on the spouse. Family Process, 1968, 1(1), 7-16. Framo, J. L. Rationale and techniques of intensive family therapy. In I. BOSgormenyi-Nagy and J. Framo (Eds.), Intensive family therapy. New York: Harper and Row, 1965. Friedman, A., Boszormenyi-Nagy, I., Jungreis, J., Lincoln, 6., Mitchell, H., Sonne, J., Speck, R., & Spivack, G. Psychotherapy for the whole family. New York: Springer Publishing Co., Inc., 1965. Greenblatt, M., Levinson, D., & Klerman, G. (Eds.) Mental patients in transition. Springfield, 111.: Charles C. Thomas, 1961. Haley, J. Whither family therapy? Family Process, 1962, "l, 69-101. Haley, J., & Hoffman, L. (Eds.) Techniques of family therapy. New York: Basic Books, 1967. Hollingshead, A. B., & Redlich, F. C. Social class and mental illness. New York: John Wiley and Sons, Inc., 1958. ' Horvitz, N. Marital problems following psychotherapy with one spouse. Journal of Consulting ngé 01101092, 1967' 11:, 38-17. Jackson, D. D. (Ed.) Communicatiop,family and marriage: Human communication, Vol. I. Palo Alto, Cali .: Science and Behavioral Books, Inc., 1968. Jackson, D. D. (Ed.) Therapy, communication and chan e: Human communicatiOn, Vol. II. Palo Alto, Ca11f.: Science and Behavioral Books, Inc., 1968. 89 Kardiner, A., & Spiegel, H. War stress and neurotic illness. New York: Paul B. Hoeber, 1947. Kellner, R., & Sheffield, B. F. The relief of distress following attendance at a clinic. British Journal of Psychiatry, 1971, 118, 195-198. Kohl, R. Pathological reactions of marital partners to improvements of patients. American Journal of Psychiatry, 1962, 118(11), 1036-1041. Laing, R. D. The politics of experience. New York: Pantheon Books of Random House, Inc., 1967. Langner, T., & Michael, S. Life stress and mental health. London: The Free Press of Glencoe- Collier MacMillian Ltd., 1963. Langsley, D., Pittman, F., Machotka, P., & Flomenhaft, K. Family crisis therapy-results and implications. Family Process, 1968, 1, 145-158. Leff, M., Roach, J., & Bunney, W. Environmental factors preceding the onset of severe depressions. Psychiatry, 1970, 33(3), 293-311. Lucas, L. The Detroit Group. Detroit: United Community Services of Metropolitan Detroit, 1959. Lyon, K. B., & Zucker, R. A. Environmental supports and post-hospital adjustment. Journal of Clinical Psychology, 1974, 33(4), 460f465. Lytton, H. Observational studies of parent-child inter- action: A methodological review. Child Develop- Messer, A. The individual in his family: Ag adaptational stud . Springfield, 111.: Charles C. Thomas, 1970. Michaux, W., Katz, M., Kurland, A., & Gansereit, K. The first year outi:_Mentalpatients after hospitalization. Baltimore: JOhns Hopkins Press,’1969. Midelfort, C. F. .The family_inypsychotherapy. New York: McGraw-Hill, Inc., 1957. Nunnally, J. Psychometric theory. New York: McGraw- Hill, 1967. 90 Peterson, D., Becker, W., Shoemaker, D., Luria, 2., & Hellmer, L. Child behavior problems and parental attitudes. Child Development, 1961, 33, 151-162. Rainwater, L. Family design: Marital sexuality, family size, and contraception. Chicago: Aldine Pub- lishing Co., Inc., 1965. Rakoff, V., Sigal, J., & Epstein, N. Working through in conjoint family therapy. In G. Erickson and T. Hogan (Eds.), Family therapy: An introduction to theory and technigue. Monterey, Ca11f.: Brooks/Cole Publishing Co., 1972. Ricks, B., & Berry, J. In M. Roff and D. Ricks (Eds.), Life history research in psychopathology. Minneapolis: University of Minnesota Press, 1970. Roff, M. Childhood social interactions and young adult psychosis. Journal of Clinical Psychology, 1963, l3, 152-157. Satir, V. A family of angels. In J. Haley and L. Hoff- man (Eds.), Techniques of family therapy. New York: Basic Books, 1967. Silverstein, M. Psychiatric aftercare. Philadelphia: University of Pennsylvania Press, 1968. Strodtbeck, F. The family as a three-person group. American Social Review, 1954, 33, 23-29. Tharp, R., & Otis, G. Toward a theory for therapeutic intervention in families. Journal of Consulting Waxler, W., & Mishler, E. Experimental studies of fami- lies. In H. Berkowitz, Advances in experimental social psychology, Vol. 57 Academic Press. Weiner, R. Adolescent problems: A symptom of family dysfunction. Social Casework, 1966, 31, 373-377. Wynne, C. C., & Singer, M. T. Thought disorder and family relations of schizophrenics. Archives of General Psychiatry, 1963, 3, 191-206. Ilil Ipifl I! i- . I..1.{I»n|..L!III. "IT'E‘Tflfilflji'flllilflll”Hjifll'1'“