AFFECTION EXCHANGE 1N MARITAL DYADS Dissertatéon for the Begree of Ph. D. MICHEGAN STME UMVERSITY KENNETH LEE VELLARD 1976 LIBRARY Michigan State This is to certify that the . thesis entitled Affection Exchange in Marital Dyads presented by Kenneth Lee Villard has been accepted towards fulfillment of the requirements for Ph. D. degree in Communication Wit/W Major professor Date 9/24/76 0-7 639 University Imam. av _ HMS & SflNS' a 800K mm INC. ii ‘ Um RY BINDEHS LIBRARY Michigan State University This is to certify that the 7, thesis entitled Affection Exchange in Marital Dyads presented by Kenneth Lee Villard has been accepted towards fulfillment of the requirements for Ph. D. degree in Communication Mil/W Major professor 9/24/76 Date 0-7 639 I)ate 0'7 639 LIBRARY Michigan State University This is to certify that the -, thesis entitled Affection Exchange in Marital Dyads presented by Kenneth Lee Villard has been accepted towards fulfillment of the requirements for Ph. D. degree in Communication Mil/W Major professor 9/24/76 ABSTRACT AFFECTION EXCHANGE IN MARITAL DYADS By Kenneth Lee Villard The affection exchange process in marital dyads is one of the most neglected areas in all of social science. However, it is hypothesized to be potentially one of the most fruitful in contributing to our understanding of dysfunc- tional family systems. The investigation focused on the marital affection exchange patterns in three types of family systems: marital conflict, symptomatic-child and normal. Data on affection exchange behavior was collected by questionnaire responses of both spouses on frequency of usage and relative importance of 24 affection exchange behaviors across 120 marital dyads. The 24 affection exchange items were derived from a conceptualization of affection exchange which posited two modes of behavioral affection: psychological and/or physi- cal intimacy, and economic resource exchange. The factor analysis of exchange items failed to support the above conceptualization identifying seven major dimensions: self- disclosing, sexual, aggressive, demonstrative, helping, gift giving and access rights. Kenneth Lee Villard Among the major findings was the discovery that the more similar the behavioral affection exchange patterns of the spouses, the more likely they are to report high levels of perceived equity and satisfaction with the relationship. Couples who reported high levels of behavioral agreement in affection behavior also tended to report high levels of accuracy in predicting the responses of their spouse. Furthermore, analysis of the data suggest that while dyadic accuracy may be necessary for the existence of equity in high agreement marriages, it may not be sufficient to ensure equity in low agreement marriages. WOmen were found to utilize initmate forms of affection exchange significantly more than men. Medium history dyads (9-17 years of marriage) were found to use the greatest amount of intimacy, followed by short and long history dyads in that order. Symptomatic-child couples were found not only to have the lowest behavioral affection agreement and accuracy scores but were also found to have the lowest intimacy usage of all three family types. Conflict couples, actively engaged in marital therapy, reported both the highest agreement and accuracy scores as well as having the highest intimacy usage. AFFECTION EXCHANGE IN MARITAL DYADS BY Kenneth Lee Villard A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1976 C) Copyright by Kenneth Lee Villard 1976 Co: Ste mer Gui Accepted by the faculty of the Department of Communication, College of Communication Arts, Michigan State University, in partial fulfillment of the require— ments for the Doctor of PhilOSOphy degree. Wren“, Director of Thesis Guidance Committee: W MMChaiman (L n . C / To Poly and Kat.... Man is periodically overwhelmed by his frailty, by his powerlessness over his ultimate destiny, by the question of meaning in human existence, by the issue of whether to connect himself with others and compound his vulnerability or isolate himself under the illusion of omnipotence and self-sufficience. It is the maddening paradox of panic in being alone and fear of being together... ii W f< ir me me ACKNOWLEDGEMENTS The author wishes to express his appreciation to Irwin Bettinghaus, Gerald Miller, Edna Rogers—Millar, and Vince Farace for their contributions and criticisms. Gratitude is also expressed to Cathy Ress and Nancy Stangelo who helped with the data collection, to Mary Fazekas and Gary Livingston for their assistance in data preparation and analysis, and to Jeannette Kaufman who typed this dissertation. I am also indebted to my wife Helen for her patience and encourage- ment, and to my parents whose esteem for education started me on this path two and one-half decades ago. iii CH} CHAPTER 1: CHAPTER 2: TABLE OF CONTENTS THE THEORETICAL MODEL: SOCIAL EXCHANGE THEORY Introduction . . . . . . . . . . . . . . Social Exchange Theory: Its Scientific Evelution O I O O O O O O O O O O O O 0 Basic Modes of Transaction in Social Exchange . . . . . . . . . . . . . . . Rewards and Costs in Social Exchanges. . Determinants of Reward in Social Exchanges Questions of Frequency, Value and Context 0 O O I O O O I I O O O I O O O Reciprocity: Issues of Form and EqUitY O O O O O O O O O I O O O O O 0 Effects of Inequity in Social Exchanges. . . . . . . . . . . . . . . MARITAL AND FAMILY AFFECTION EXCHANGE: A REVIEW OF SUBSTANTIVE FINDINGS Marital Stability, Satisfaction and Affection Exchange Patterns. . . . . . Symptom Formation in Children and Family Interaction Patterns . . . . . . . . . Schizophrenia and Family Interaction Patterns 0 O I O I O O O O I O I O I 0 Personality disorders and Family Inter- action Patterns. . . . . . . . . . . . Character Disorders and Family Inter- action Patterns. . . . . . . . . . . . Learning Disorders and Family Inter- action Patterns. . . . . . . . . . . . iv 12 18. 21 25 37 44 58 62 80 91 98 CHAPTER 3: CHAPTER 4: PROCEDURES AND DESIGN Marital Agreement and Accuracy in Affection Exchange . . . . . . . . Perceived Equity. . . . . . . . . . . . Relationship Satisfaction . . . . . . . Relationship Satisfaction: A Two— Dimensional Variable. . . . . . . . . History of the Dyad, Satisfaction with Marriage, and Affection Exchange. . . Male-Female Differences in Affection EXChange O O O O O O O O O O O O O O O Socio-Economic Status and Affection Exchange. . . . . . . . . . . . . . . Sample Populations. . . . . . . . . . . Limitations of the Investigation. . . . Validity of the Instrument. . . . . . . Reliability of the Instrument . . . . . RESEARCH‘FINDINGS Sample Descriptions . . . . . . . . . . Homeomorphism and Perceived Equity. . . Effects of Accuracy on the Relationship between Agreement and Equity. . . . . Agreement and Family Type . . . . . . . Equity and Fmily Type . . . . . . . . . Accuracy and Family Type. . . . . . . . Relationship Satisfaction as a Two- Dimensional Variable. . . . . . . . . Sex Differences in Affection Exchange . Length of Marriage and Affection Exchange Socio-Economic Status and Affection Exchange. . . . . . . . . . . . . . . 107 109 110 115 121 125 127 128 130 132 139 144 145 147 150 152 152 155 163 166 169 CHAT BIBI CHAPTER 5: SUMMARY, DISCUSSION AND RECOMMENDA- TIONS FOR FUTURE RESEARCH Summary and Discussion . . . . . . . . . . 173 Recommendations for Future Research. . . . 180 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 182 vi 1t 11 12 l3 14 15 LIST OF TABLES Table Page 1 Percentage Number of Choices of Resources, Contingent Upon Resource Given . . . . . . . . 31 2 Relationship Between Evaluation of Marital Happiness and Sources of Happiness (First- Mentioned Reasons) . . . . . . . . . . . . . . 49 3 Relationship Between Evaluation of Marital Hapiness and Sources of Unhappiness (First- Mentioned Reasons of Respondents Giving Any Reason for Unhappiness). . . . . . . . . . 50 4 Percentage of Respondents Identifying the Behavior as Used to Show Affection/Caring in Marriage. 0 O O O O O O O O O I O O O O 0 0134-135 _ 5 Factor Analysis of Affection Exchange Items. . . 137 6 Major Dimensions ofAffection Exchange. . . . . . 138 7 Reliability Coefficients of Internal Consistency. . . . . . . . . . . . . . . . . . 139 8 Inter-Correlations Among Agreement, Equity and Accuracy . . . . . . . . . . . . . . . . . 149 9 Mean Scores on Agreement by Family Type. . . . 151 10 Mean Scores on Equity by Family Type . . . . . . 153 11 Mean Scores on Accuracy by Family Type . . . . . 154 12 Mean Scores on Relationship Satisfaction by Type of Expectation/Fulfillment. . . . . . . . 159 13 Mean Scores on Economic Affection Usage by Sex and Family Type. . . . . . . . . . . . . . 164 14 Mean Scores on Intimate Affection Usage by Sex and Family Type. . . . . . . . . . . . . . 165 15 Mean Scores on Economic Affection Usage by History and Family Type. . . . . . . . . . . . 167 vii Ta' Table Page 16 Mean Scores in Intimate Affection Usage by History and Family Type. . . . . . . . . . . . 168 17 Mean Scores on Economic Affection Usage by SES and Family Type. . . . . . . . . . . . . . 170 18 Mean Scores on Intimate Affection Usage by SES and Family Type. . . . . . . . . . . . . . 172 viii Figure LIST OF FIGURES Relational Dimensions . . . . . . . . . . . . Classification of Resources by Particularism and Concreteness. . . . . . . . . . . . . . Levels of Equity and Inequity . . . . . . . . Relationship Satisfaction as a Two-Dimensional variable 0 I O O O O O O O O O O O I O O O O Hypothesized Satisfaction Levels as a Function of Realized and Expected Outcomes . . . . . Subject Responses to Relationship Satisfaction measure 0 _ O O O O O O O O O O O C O O O O 0 Marital Satisfaction and Intimacy Usage Over Time 0 O O O O O I C O O O O O O O I O O O 0 ix Page 15 111 116 117 162 179 he ti St. Cdl CHAPTER I THE THEORETICAL MODEL: SOCIAL EXCHANGE THEORY In social scientific circles, increasing attention is being given the study of family systems and the communication processes which occur between family members. More and more family researchers and practitioners alike are looking for the origins of individual and family impairment in the communication fabric which each family uniquely weaves, maintains and, theoretically, wills to future generations. With social problems such as divorce, delinquency, drug abuse and alchoholism seemingly on the constant rise and other social ailments such as schizophrenia, emotional, behavioral, and learning disorders continuing to drain the potentially productive energies of countless thousands, marital and family functioning have come under increasing scientific scrutiny, as topics of great social as well as scientific import. Although the results of most family research to date have been far from conclusive in connecting family communica- tion processes, in a causal sense, with any of the above end states, scientific research into the study of family communi- cation is still in an infant stage and as such suffers from a number of shortcomings. To be more precise, a great de ar ha SP uni to tic uti cor shi cor rerr lea as as two We eXi: 2 number of studies which have tried to relate communication processes to dysfunctional family states have been hampered by marked methodological weaknesses. The majority have employed self-report survey questionnaires, case history analyses, psychological testing and evaluation techniques and in addition have been almost always ex post facto in design. (Jacob, 1975) In addition to these methodological and design limitations, family studies have also often been hampered by a lack of conceptual and operational clarity in specifying the exact relational processes and parameters under investigation. As a result, it becomes very difficult‘ to integrate and compare the results of empirical investiga- tions and to generalize past the specific set of Operations utilized by each investigator. -Without a clear and precise conceptual framework out of which to operate, the relation- ships between relational parameters, their communication correlates, and dysfunctional dyadic and family states remain unresolved empirical questions. A review of the literature reveals the existence of at least two major relational dimensions which must be viewed as important in family and dyadic communication functioning, as well as a number of dimensions of lesser importance. The two which are mentioned in one form or another in almost every conceptual treatment or empirical investigation are control and affection. Control refers to the decision-making processes which exist and Operate within a relational system. The issue of i< Val WOQ thrc Of -. 3 control is concerned with the relative amounts of power which each individual holds within the relationship and the extent and boundaries of that influence. In effect, control is concerned with each person's relative "tOpness" or "bot- tomness" during any given interaction exchange, or across time. (Schutz, 1958, pp. 22-25) Affection, by contrast, refers to the amount of emo- tional closeness which exists within a relational system. It refers to the amount of esteem and caring individuals hold for, and exchange with, one another. In effect, affec- tion concerns the relative "closeness" or "farness" which exists between individuals within the relational system. (Schutz, 1958, pp. 23-25) The studies as well as the dimensions which have been identified and discussed are summed in Figure I.1 As Berlo (1968) suggests, family communication relation- ships are typically designed to serve both a maintenance as well as a production function. Therefore, both of these relational parameters would seem to play crucial roles in overall family functioning. Of the two, the affection dimension would appear to be more central in maintenance functions which are concerned with providing self-concept validation and support to individual members; while control would appear to be more concerned with system production through decision making processes. In terms of family system maintenance, the importance of affection is associated with the individual's need to lfirl- "II‘C ND UT sNUUX AGSUCHMCQ DUH HOV ZOHBUMEE< AOSOOHUCE QUH HOV 4028200 “NNOZBDAN iliiliiii macamsosao ancoaumaom "H mmeHm Emfiamcoflma>oum .m> kucwmuuou xuflawnflmcommom twocmmwaaoucH :ofiwsHocH ucoecfimuud msflpfld gushes uwuflamcoauoso o>fluomoz ©HOUIEHM3 coaumauomoo .m> coauuaau>m hzummam .m> xuaaauusoz oHHumomtaansoHum suwaaumomno>oq suflamaoauosm a sufiaanuoxflq ouwmno>oq suaaanuaoom coauoomma suflaanuaoom cowumumpflmcoo mmosuoncoa soauuusowuo oohoamsm mmoco>qmmHBuomummosuoauum o>ammflsndwuusacaeoo muAAdsvm .m> auanowuomsm huaocducomm .u> amouuuum soap IQUGOflHO EGHQOHM .m> HOBGOU Uwfiwalvaom >eocousfluuomm¢ scammwensmnoocmcflsoo mmoco>auuommm acupfl>wpaH Honucou cocoafifionm Hospfi>fich ouduoauum vswumwuwcH um3om coamabuomsm mo mmocomoau Anomav am pm noxuom “Homav mom Aammav gnaw Amwmav seamen a Hmmmmnom a Lawmav umwmmnom Ammmav Hmwwmnom Lemma .ommav am pm muummuom Lemmac suqu Ammmav .Hm um muummuom Ammmav upscom “emmav umuumo Ammmav means: a cmanmwwam Ammmav cm>aaa5m gamma .ommHv Hm um .NuMM mcofimcwsflo “mayo / lsgmoflsc. was “01 onaomaas «osmoaaau as” soy gemszoo lh“ mmsfiasq illlllilllll $02.5qu ..e.ucoov H mmaon uowamcou mowusooathuwuwao uoomm< ounmlo>oq auwuupwaom umsuumaz .ucmsomcnm .suaaaumom suaaflnnfloow .ucmanouuma coauflcmouom oocmuduuaz .cowuoommfi camoom now cumocoo mmoco>wuuommsm Husoauosm muaaaneumcH usuaaaneum Hoaoauoam ~.mumaouuo mcaaocoa Huwoaamabouflo o» cpsamum a mu mwmhaoem HoGOMunHouuoo sesame» no moansnoou owumauec nouoom nmsounu aaauuawum pcwmaueopa use wcuocHuxo cues o>onn pcuwo moapsum may ea confluence mosmonsm Hausa can meowmsoswp onu “ouozv cococweon cowmmfiandmloocoafisoo mauoum coauwnanxm .sOHuwnwncH mmocoanoooumm .oosouomon cowmmfisnam .ooeusfisoo auw>wuosooum now sumosoo oceamwomfln can mucoawn huouanwncH sowmuo>ouuxMIcowmno>ouucH Amnmac noose Amomav comumo Amwmdv GBOHm Amomav namzoz a muoq Avomac cousoz a oxmam Ammmav kume Avmmav msHM w meowm cowmcoawo Hosuo “camoHdsd muw How onBUmmmd Amamoaoso muw HOV nomazou mmomfibfl te IE3- ]?- I 6 feel that the self is loveable and therefore of value. If affection is given, but not returned, it is often assumed that the self is not loved, and if not loved, is perhaps not loveable, and, therefore, without value. If the affection is returned, then it is assumed that the self is loveable, has value and therefore the individual's self-concept is sustained and supported. Strangely enough, in spite of its seeming significance, little empirical inquiry has been conducted into the various means by which peOple communicate their affection or caring for one another and the kinds of relationship problems or family states which may be associated with the affection exchange process. A_purpoSe g; Ehg_present gtggy_i§_gg test several empirical hypotheses concerning the communica- tion gf affection $2 the marital context and its association with various states 9; dyadic and family functioning; Of the social psychological models available for viewing relational phenomena, two would seem to have the greatest relevance for studying the communication of affection in marital and family settings: social exhange theory and co- orientation theory. Social exchange theory as advanced by such scholars as Thibaut and Kelley (1959), Homans (1961), and Blau (1964), with its reward/cost paradigm, is an intuitively meaningful way of viewing interpersonal relations in general. It is first and foremost a behavioral theory which concerns itself with assessing end states resulting from interpersonal transactions, e.g., relationship satisfac- tion, reciprocity and perceived equity among others. 7 However, because there has been little attention given to the operational develOpment of major predictor variables in the social exchange framework, it has not lent itself easily to the generation and testing of empirical proposi- tions. That is, becuase there have been few attempts to operationalize the concepts of cost/reward in any concrete and quantifiable way, many researchers have shied away from social exchange theory as an empirical research model. Co-orientation theory on the other hand, which had its early beginnings with Newcomb and Svehla (1937), and even- tually was formalized by Newcomb (1953, 1961), is somewhat more rigorous in its theoretical framework and certainly more prolific in its generation of concepts. It therefore lends itself somewhat more readily to the generation of testable propositions. Co-orientation, which is primarily a cognitive theory, has in contrast to exchange theory, a somewhat more developed array of predictor variables, e.g., perceived congruity, cognitive agreement and accuracy. However, it tends to be weakest in its criterion measures which are abstract, mentalistic and very difficult to mea- sure. That is, the major criterion variable in co- orientation, as is the case in all cognitive theory, is cognitive imbalance or at broader levels, system imbalance. Regardless of the level of analysis, cognitive imbalance cannot be viewed as a very empirically precise or descrip- tively useful measure, especially when assessing relational outcomes. In addition, many of the alternatives available 8 for reducing system imbalance (mediated by the factors of perceived discrepancy, sign and degree of attraction, import- ance, committedness or common relevance), are largely uncon- scious, autistic and unmeasurable cognitive processes, somewhat removed from the issue of relational functioning. Thus social exchange theory, which is conceptually parsimonious and simplistic, highly behavioral in orientation and perhaps strongest in relevant criterion measures, is viewed as being the better theoretical model for studying the exchange of affection, especially if more precise and concrete measures of reward or social support can be genera- ted to index predictor variables. Quite simply, social exchange may be defined as "an exchange of activity, tangible or intangible, more or less rewarding and costly, between two persons." (Blau, 1964, p. 6) Before looking at the various parts of this definition separately, it would perhaps be useful to place social exchange theory in historical and scientific perspective. Social Exchange Theory: Its Scientific Evolution The intellectual origins of social exchange, as an emerging theoretical framework, are based on three intellec- tual planks: utilitarian economics, anthropological func- tionalism, and behavioral psychology. (Turner, 1974) The first of these, economic utilitarianism, arose between 1750 and 1850 with the work of Adam Smith, John Stuart Mills, and Jeremy Bentham. To these early economists, man was viewed 11 CI Mai Rir the Com altk impo. 9 as a rationally seeking being who attempts to maximize his gains of "utility" from exchanges with others. They sug- gested that man considers not only the potential rewards obtainable but also the relative costs to be incurred and weighs these in his exchange decisions. Although many of the early utlitarian assumptions concerning the nature of man have been retained in modern social exchange models, perhaps the greatest contribution of utilitarian thought was that it later inspired the deve10pment of psychological behaviorism. The second contributor to the evolution of social exchange is functional anthropology. Beginning with Sir James Frazer who studied the kinship and marriage practices of Australian aboriginals, this vein of thought presented the first explicit exchange analysis of social institutions. In Folklore $2 the Olg_Testament, Frazer used solely economic or material motives to explain the rise of social patterns. In effect, he suggested that social exchange processes create complex institutional patterns. Following the work of Frazer was that of Bronislaw Malinowski who studied the exchange patterns of the "Kula Ring" in the south sea islands. Particularly interested in the process of gift exchange among natives, Malinowski concluded that the exchange of necklaces and bracelets, although having a definite economic component, was, more importantly, symbolic in nature. Thus he became the first to distinguish between material or economic exchanges and e> is in< pat and It mOs1 tive PrOp indi‘ 10 symbolic, non—material ones. Because there was little utilitarian gain, Malinowski concluded that the motives behind the exchange were social psychological in nature and met societal needs only secondarily. The work of Marcell Mauss came largely as a reaction to the interpretations of Malinowski. Mauss rejected the psychological interpretations on the Kula made by Malinowski and suggested that exchanges and reciprocity were determined by the culture and not the individual. "It is groups, and not individuals, which carry on exchange, make contracts and are bound by obligation." (Mauss, translated by Cunnison, 1954, p. 1) Thus for Mauss the self-interest of the utili- tarians and the psychological needs of Malinowski were replaced by a conception of the individual as a mere repre- sentative of social groups. This type of structural approach to social exchange would later influence the writings of exchange theorists such as Blau. The culmination of the movement against the utilitarians is embodied in the writings of Claude Levi-Strauss. Respond- ing to the interpretations of Frazer on cross-cousin marriage patterns, Levi-Strauss rejected Frazer's utilitarian position and likewise the psychological interpretations of Malinowski. It is his position which is typically viewed as being the most SOphisticated of all the structural exchange perspec- tives. The main notions are embodied in the following three prOpositions: (1) "all exchange relations involve costs for individuals, but in contrast with economic or psychological a1 pr mi for Com Par: thec the F f0! e. ll explanations of exchange, such costs are attributed to society--to its customs, rules, laws and values”: (2) "for all those scarce and valued resources in society--whether material objects. . . or symbolic resources like esteem and prestige--their distribution is regulated by norms and values"; and (3) "all exchange relations are regulated by a norm of reciprocity, requiring those receiving valued re- sources to bestow on their benefactors other valued re- sources" (Turner, 1974, p. 220). The final plank upon which modern social exchange is based is that offered by psychological behaviorism. Drawing its principles from experimentally manipulated animal behav- ior, behaviorism suggests that it is necessary to study only overt behavior as a response to observable stimuli. And since it operates on the assumption that animals and humans are reward seeking organisms that pursue choices which provide the most reward and the least punishment, behaviorism might be seen as an extreme form of utilitarianism. George Homans is undoubtedly the most ardent spokesman for the psychological behaviorist position on social exchange. Coming largely as a reaction to the structuralism of Talcott Parsons and Levi-Strauss, Homans argues for an exchange theory that emphasizes face-to-face interaction, focuses on limited and direct exchanges among individuals and promotes the psychological as Opposed to the social structural need for exchange relations. (Turner, 1974, p. 233) The axioms app: and 12 of Homan's perspective on social exchange theory will be discussed later under the heading of determination of reward values through social exchanges. Thus, when one looks at the evolutionary development of social exchange, one can expect to find a curious mixture of behavioral concepts and vocabulary, intermeshed with utili- tarian influence and intertwined with many of the established thoughts of structural anthropology. Basic Modes of Transaction in Social Exchange Modern social exchange theorists utilize a wide range of exchange concepts as their fundamental units of analysis, some quite broad and abstract, others increasingly more specific and concrete. Among the first, modern-day, social exchange theorists were the psychologists John Thibaut and H. H. Kelley. According to Thibaut and Kelley (1959), the fundamental medium of exchange in all interpersonal relations is interaction, which ultimately can be reduced at an indi- vidual level to its fundamental units which they termed behavior sets. According to these authors the behavior set is the instrument by which the individual obtains his goals through exchanges with others. All the behavioral patterns which have produced successful outcomes in the past compose 'what is termed the exchange repertoire. The major concepts in Homans's (1950, 1958, 1961) approach to social exchange include activity, interaction, and sentiment. Activity refers to behaviors which are aimed Cl Pe ha' SOC tic 0th. 13 at deriving rewards. Interaction is used to refer to behav- iors in which individuals direct their activities in order to derive rewards from one another and to avoid punishments. And sentiments are defined as activities by which individuals communicate their attitudes and feelings toward another person or persons. (Homans, 1961, pp. 52-82) Of particular interest in the present study is Homans's concept of senti- ment. As he used the term, sentiment encompassed any internal state of the human body including: affection, sympathy, rage, thirst, hunger, nostalgia, scorn, etc. (Turner, 1974, p. 229) Each of the concepts in Homans's conceptualization of social exchange has two classes: descriptive terms and variables. Descriptive terms are the generic names which are applied to different types of exchange activity or behavior. Variables on the other hand are the properties of that behavior which determine the reward/cost ratio of that behavior, i.e., quantity and value. The importance of each of the variables in determining reward value will be dis- cussed in detail in a later section. The third major, modern-day social exchange theorist is Peter Blau (1964). Unlike Thibaut, Kelley, and Romans, who have attempted to explain all scoial behavior in terms of social exchanges, Blau focuses only on those social situa- tions where rewards are expected and received from designated others. He has focused on exchanges occurring within ir CO in Wh: (4) 0f 14 social relationships and thus his major exchange concept is social support. Social support is further divided into two related components: intrinsic attraction and social approval. (Blau, 1964, p. 61) Intrinsic attraction is the force which induces individuals to establish associations with one another and to expand the scope of those relationships, once formed. The emergence of attraction therefore leads to the process of social exchange with those individuals who are perceived as being the potential sources of social approval and acceptance. Among the more recent social exchange theorists are Foa and Foa (in Siegman, 1972). In their early work their primary social exchange unit of analysis is termed a resource and defined as: Anything which can be communicated or transmitted in a social situation: a material object, money, an activity performed on the body of a person (cutting his hair, giving an injection) or on things belonging to him (cleaning his clothes), an item of information, an expression of love or warmth, of respect, esteem, or appreciation. (Foa & Foa, In Siegman, 1972, p. 293). Among the resources discussed by Foa and Foa are the follow- ing: (1) love, composed of affection and warmth, (2) status, consisting of prestige and esteem, (3) information, which includes the processing of pre-existing information but which excludes information related to love or status, (4) services, consisting of activities performed on the body of a person or his belongings, (5) material goods, and (6) money. These six resources are ordered along two theore- tical dimensions: particularism, which focuses on the SOC; Offs ”hat t10n 15 reward value of a resource as a function of its being given by a particular person, and concreteness, which is concerned with the abstractness or relative symbolism of the resource.2 The classification of these resources along these two dimen- sions is illustrated in Figure 2 below. Particularism Concreteness More Less More ~ Love Services Status Goods Information Less Money (Taken from Foa and Foa, In Siegman, 1972, page 294) Figure 2. Classification of Resources by Particularism and Concreteness In their more recent work (Turner, Foa and Foa, 1971), the term resource has been changed to interpersonal rein- forcers, which although the exchange commodities have re- mained the same, places them in more of a relational context. The final and most recent theoretical addition to social exchange theory in terms of unit of analysis is that offered by Altman and Taylor (1973) within the confines of what they term social penetration theory. Social penetra- tion is somewhat unique among the social exchange approaches in that it emphasizes the joint effects of both verbal and 1‘1 Cc mo ti. Sic 16 nonverbal communication on the breadth and depth of inter- personal processes and also because it implies that satisfac- tion derived from the exchange is a function of reciprocity of interaction. Among the communication behaviors which are exchanged are information about the self shared through self-disclosure, as well as the exchange Of positive and negative affect and mutual activities. In addition, it emphasizes the significance of body posture, gestures, limb and head movements, facial expressions and eye gaze. Thus the evolutionary trends in social exchange theoriz- ing finds an increasing tendency to move from exchange commodities which are conceptually abstract, e.g., inter- action, activity and social support, to units which have increasing specificity, e.g., the interpersonal reinforcers Of Foa and Foa as well as the recent emphasis on specific communication behaviors as exemplified by the social pene- tration theorists. The present study continues that trend with the intro- duction of even more Operational specificity in the social exchange concept of relational currency (Villard and Whipple, 1976). Defined as the "communication processes by which individuals exchange affection and caring for one another," relational currency focuses on a specific relational dimen- sion (that of affection), delineates two forms of exchange commodities (the intimate and the economic), and perhaps most importantly, emphasizes the significance Of communica- tion behavior as an important factor in this crucial dimen- sion of dyadic functioning. 8: be fc Wh Se 17 The first of these two types Of affection exchange is a demonstrative, overt and largely intimate form of affection exchange in which the actual commodity Offered in order to communicate affection and caring is some level Of physical and/Or psychological identity. The communication behaviors by which intimate forms Of affection are exchanged include: aggression, sexuality, self-disclosure, physical touch and facial affect displays. These concepts are further Opera- tionally defined as follows: aggression referring to those verbally sarcastic and/or physical contact behaviors, e.g., slapping, shoving or punching which are used in a confirming or supportive manner; sexuality, referring to those physical behaviors directly associated with sexual intercourse; self- disclosure, referring to those verbal language behaviors by which one shares his thoughts and feelings concerning the self with another; physical touch referring to those inter- personal behaviors by which one individual touches the body and/or clothing of another in a supportive, but non-sexual manner; and facial affect displays, which refers to the non- verbal, disclosure Of internal emotional states through facial expressions. The second avenue Of affection exchange is a more symbolic and economic form, which uses various economically based commodities as indicators of one's positive feelings for another. Unlike the intimate forms of affection exchange which Offer physical and psychological resources of the self, economic forms Of affection exchange Offer largely a; 1.171%? so be. Hon des per der; res; POte func other 12) 18 time/energy commodities. These include access rights, money, favors and gifts. Access rights refers to permitting another person to use one's property, personal possessions and/or granting entry to one's time resources; mgpgy, referring to legal tender; favors, refering to doing tasks for another, and gifgg, referring to material Objects of varying value, which are given from one individual to another. Thus the concept of relational currency offers advantages over existing social exchange schemes in Offering greater conceptual clarity and Operational precision in specifying the exact means Of exchange which are used in a particular and highly salient area of relational functioning. Rewards and Costs in SOEial Exchanges The concepts of reward and costs are very central to social exchange theory, with the major contributions again being made by the major theorists: Thibaut and Kelley, Romans and Blau. The consequences of an interaction are described by Thibaut and Kelley in terms of the rewards the person receives and the costs he incurs as a result of emitting the behavior. Each person's reward and cost are derived not only from his own behavior, but also from the response Of the other person. That is, each person can potentially derive reward as well as cost simply as a function of making a behavioral response, regardless Of the other's response to it. (Thibaut and Kelley, 1959, pp. 10- 12) Costs and rewards are combined into a single scale Of QXCI thre 19 goodness 9: outcomes with states of high reward and low cost being assigned high outcome value and states of low reward and high cost being assigned low outcome scores. Thibaut and Kelley define rewards as pleasures, satis- factions and gratification that the person enjoys as a result Of emitting the behavior. Costs on the other hand are defined as factors that Operate to inhibit or deter the performance Of a sequence of behavior. (Thibaut and Kelley, 1959, pp. 10-12) Homans uses the word profit or psychic profit to des- cribe the desired outcome of social exchanges. Profit is further defined as reward minus costs, with costs viewed as originating from three sources. Costs can be incurred as a result of: (1) withdrawal of positive reinforcement, (2) emitting the behavior due to the energy expended or the fatigue resulting from carrying out the activity, or (3) rewards foregone from an alternative activity as a result of emitting the given one. (Homans, 1961, pp. 59-60) According to Blau, there are two types of rewards: intrinsic and extrinsic. Intrinsic rewards are made up of diffuse social support which is derived primarily from communication in intimate or personal relationships. Extrin- sic rewards in contrast come primarily in the form of eco- nomic commodities. (Blau, 1964, p. 95) Among the rewards included as either intrinsic or extrinsic, Blau lists esteem, respect, money and compliance. Blau's views on social exchange are similar to those of Homans as he discusses three types of costs. Investment costs focuses on the tl an de 19 int to tua the int] Verb 20 amount of time and energy which go into the giving of social support. Direct costs result from entering into a social exchange situation and thereby surrendering some personal power and independence as a result of rewarding another. Opportunity costs are incurred as a result Of foregoing other alternatives which are also potentially rewarding. (Blau, 1964, p. 101) Because the concept of reward has seldom had any quantifiable measure or any definable limits Of supply, it is often recommended that only the activities for which definable and exhaustible supplies exist, should be included' in the hypotheses derived from social exchange theory. For this reason some theorists recommend drOpping from the theory those rewards which would be classified as intrinsic and rely only on those extrinsic commodities that have definite supply limits and more precise values (Abrahamsson, 1970, p. 282). However, since a major portion of social exchanges involve rewards which are intrinsic, to omit them would be to seriously and prematurely limit the power of the concep- tualization. This would be particularly true in studying the sentiments Of affection and caring, which furnish largely intrinsic rewards to participants through various forms Of verbal and non-verbal communication. Therefore, instead of attempting to measure intrinsic rewards directly, it may be more valuable to assess them indirectly by analyzing the forms of communication by which th CO su§ frc rep Whit DOn. tent actiV uPOn referE over a in? the 21 they are exchanged and then inferring about the quality and quantity of social or identity support provided from an assessment Of the form and frequency of the communication forms utilized. Thus, by delineating Specific communication behaviors by which affection is communicated, the present study seeks to fill, in an Operational sense, a significant void in the theoretical framework of social exchange theory, at least as it is applied to the exchange and communication of affection. Determinants of Reward in Social Exchanges: Questions of Frequency, Value and Context Thibaut and Kelley suggest that as a behavior is re- peated, "its reward value tends to decrease over time due to the tendency Of a need state to satiate," and conversely the cost value tends to increase as fatigue sets in. They suggest that members of a dyad are likely to shift over time from one behavior set to another within their response repertoires. They further posit that those relationships in which repertoires include too few alternatives to permit non-repetition will either be "dissolved or become intermit- tent." (Thibaut and Kelley, 1959, pp. 18-19) Homans suggests, as was indicated earlier, that each activity, interaction or sentiment can be rewarding depending upon two important variables, quantity and value. Quantity refers to how frequently a particular behavior is emitted over a specified period of time, while value refers to the worth (its rewarding or punishment impact) the person receiv- ing the behavior places on it. 22 Generally speaking the less frequently a behavior is emitted, the more value it will have to the other person. For instance, if a husband frequently tells his wife, "I love you," it is not likely that the statement will have as much value as if he rarely tells her. Thus value is value per unit Of behavior emitted and consequently is always changing even as the behavior is emitted. Value therefore is determined by two factors: whether the person likes the behavior, that is, derives reward from it, and secondly, the extent to which he has been a recipient of that behavior in the immediate past. (Swenson, 1973, p. 218) The relationship between frequency of occurrence and reward value derived is posited in three Of Homans' five general propositions on social exchange, based in Operant conditions. They are as follows: (a) the more Often within a given period Of time a man's activity rewards the activity of another, the more often the other will emit the activity. (b) the more valuable to a man a unit of the activity another gives him, the more often he will emit activity rewarded by the activity of the other. (c) the more Often a man has in the recent past received a rewarding activity from another, the less valuable any further unit of that activity becomes to him. (Homans, 1961, pp. 53-55). Fundamentally, these three prOpositions contend that: (a) the more often a behavior is rewarded the more it will be emitted (continuous reinforcement), (b) the more reward- ing a unit Of behavior is the more it will be repeated (reinforcement strength), and (c) the more a behavior has been received, the less rewarding it increasingly becomes (satiation). Blau, like Homans, also recognizes the decreasing reward of exchange commodities as the frequency of those commodities increases. In technical terms, the marginal 1) ti wc re Or; her to sch. takj betw the exChe other Servi COnSE‘ Viewec resour This In exchant 23 utility of increasing amounts Of social benefits eventually diminishes. (Blau, 1964, p. 90) According to Blau, the initial value of the social support provided through social exchanges is determined by the perceived attractiveness Of the bestower. In addition, the value Of one's approval is also influenced by the indi- vidual's perception of it as genuine, and thirdly, by the individual's respect for the judgement and discrimination the bestower exhibits in furnishing approval. In other words, if the support is perceived as unauthentic or if the recipient has little respect for the sender or at least his criterion for discrimination, then the support received will have little value. (Blau, 1964, p. 64) Unlike most exchange models which only consider giving to others and receiving from others, the social exchange scheme of Foa and Foa also includes giving to self and taking from self, and attempts to determine the relationships between the mode of exchange, the Object of exchange, and the resulting supply 9: the resource. For example, in the exchange Of the resources of money and goods, giving to other precludes giving to self as in the same fashion that serving another detracts from one's ability to serve oneself. Consequently, resources with finite limits Of supply may be viewed as having higher value in relational exchanges than resources which have more loosely defined limits of supply. This may be one reason why economic forms of affection exchange are often valued above intimate forms. m in in Vie thc shi 0th in e Self qUen t0 b1 rewaz In ef to di often reCei: 24 In addition to the derived value Of any given social exchange being determined by the reward value of the behavior as held by the receiver and by the actual frequency with which that behavior is received, it is also influenced by the specific and general relational context in which those behaviors are exchanged. That is, not only is it important to assess what person A finds rewarding and to consider the frequency with which A receives those behaviors, but A's ability to actually experience reward is also dependent upon his perception of B's (the sender's) evaluation and orienta- tion towards those behaviors. ‘ Villard and Whipple (1976) suggest that individuals infer the value of a given relational behavior (particularly in the affection/caring dimension of human relationships) by viewing and assessing the relative frequency with which those behaviors are used, not only within their own relation- ship with the person, but across all relationships which the other person maintains. For example, those behaviors used in exchanging affection and caring, e.g., physical touch, self-disclosure, and sexuality, etc., which are used fre- quently and across many relationships are Often inferred not to be highly valued by the "sender" and, therefore, the reward value derived by the "receiver" is Often lessened. In effect, those behaviors are not interpreted as being used to discriminate between relationships and therefore they are Often assigned lesser amounts Of perceived value by the receiver. Thus even if person A finds self-disclosure to be p] ar di d8] the big of . 6881 but belo Occu; 25 a rewarding behavior, and even if he has 22$ been the recip- ient Of self-disclosing communication with B in the immediate past, the fact that he views person B self-disclosing to other individuals will often detract from the reward value provided to A, i£_and yh§g_B does self-disclose to him. Conversely, those affection exchange behaviors which are used infrequently are Often inferred to Operate in discriminating between relationships (primarily in terms of depth Of relationship) and, therefore, are interpreted as having greater relational significance. As a consequence the reward value derived by the receiver is likely to be high, all other things being equal. Thus the reward value of any social exchange is determined not only by the value assigned by the receiver and the frequency of the exchange, but also by the exchange networks to which both individuals belong and the quality and quantity of transactions which occur there. Reciprocity: Issues Of Form and Equity The issue of reciprocity in social exchange has been studied by numerous scholars under many different labels: Homans (1961) under the concept of distributive justice, Gouldner (1960) with the norm Of reciprocity, Jones (1964) with the concept of legitimacy and Adams (1965) with his theory of equity. In addition to these scholars, others have also recognized the importance of reciprocity, particu- larly in regard to relationship satisfaction and social Oi ir Wh (Ac dEt and ace SYS‘. have 26 stability. L. T. Hobhouse has suggested that "recip- rocity. . . is the vital principle Of society." (Hobhouse, 1951, p. 21) Thurwald also acknowledged the significance of reciprocity in stating, "this principle is almost a primor- dial imperative which pervades every relation Of primitive life. (Thurwald, 1932, p. 106) And finally, Georg Simmel suggests that social stability could not even exist without the "reciprocity of service and return service," and further- more, that "all contacts among men rest on the schema of giving and returning the equivalence." (Simmel, 1950, p. 387) Thus there is considerable support for the norm Of“ reciprocity as a cultural imperative, i.e., that society and social order are maintained by the implicit assumption that man must return the equivalent of what he has received. Taking the lead Of Homans, most investigators define reciprocity as a system state occurring whenever the profits Of each participant are prOportional to their amount of investment. In schematic form, equity exists within a dyad when: A's rewards - A's costs = B's rgwards - B's costs A's investments B's investments (Adams, 1965), p. 273 Thus it is the ratio Of profits to investments which determines the equity of the transaction for the individual and when compared against the perceived ratio for the spouse, accounts for the amount of reciprocal balance within the system. Some scholars such as Homans, Simmel, and Malinowski have suggested that the individual's profit ratios, while it 9111.; Una 27 not having to be precisely the same, nevertheless must be roughly equivalent if reciprocity is to exist. One Of the more intriguing and perhaps socially relevant issues which naturally emerges from considering the issue Of reciprocity is: how do individuals determine what consti- tutes "equivalence" in their social exchanges and, further- more, what role does the actual "form" which the exchange takes, play in that assessment. More precisely, to what extent must social exchanges be reciprocated "in-kind" for a state Of equity to exist? The emphasis on reciprocation in-kind, in human trans- actions has been written about by several scholars. Among these are Levinger (1964) who suggests that although there may be considerable task differentiation between the sexes in the marital context, socio-emotional behavior must be reciprocal if the dyad is to be maintained. Along this same line, Rausch (1974) suggests that the individual whose response repertoire is so limited as to preclude alteration from instrumental to expressive behavior, thereby not allow- ing reciprocation in—kind , may find himself interpersonally inept and emotionally unsupported. . . . the man who failed to recognize and respond to his own and other's feelings would probably experience serious difficulty in all interpersonal relationships, in tension management, and in finding the emotional support he needs to function adequately in a social world. (Rausch, 1974, p. 146) Homans (1961) also writes about the reciprocation issue emphasizing the costs incurred by the individual who is unable to respond with reciprocation in-kind-and points to at ur No fo hu. on: Co: Pa: rev the: 28 the probable feelings of ineptness and loss of self-esteem which must be suffered by such a person. Anyone who accepts from another a service he cannot repay in-kind incurs inferiority as a cost Of receiving the service. The esteem he gives becomes a loss to himself. (Homans, 1961, p. 320) Although a number of scholars have written conceptually about in—kind reciprocation, there has been little empirical research done on the reciprocity of affection exchange and the communication processes by which that is carried out. Those studies which have been done, with one major exception to be discussed below, have concerned themselves with the reciprocity of self-disclosure. In a study by Jourard and Richman (1963) analyzing college student reports of self- disclosure to and from parents and friends, the authors found an average correlation Of .70 between a student's reports of his output to other and his input from other. In a similar study by Levinger and Senn (1967) the authors obtained both non-independent and independent meas- ures of disclosure from both members of the relationship. Non-independent measures correlated .91 for husbands and .79 for wives. However, independent measures (that is between husband output or input and wife output or input correlated only about .50. Also reports Of input were more highly correlated than reports of output. This would imply that a partner's description Of the other's behavior or feelings reveals more about the joint perception of the relationship than does the individual's description of his own behavior or feelings. (Levinger, 1967, p. 247) 9: ti WC Ta re. em 29 Perhaps the most extensive study dealing with exchange resources and reciprocation patterns is the work of Foa and Foa (In Siegman, 1972, p. 294). Studying resources of exchange along the dimensions Of concreteness and particular- ism, these authors suggest that a substitution of resource will be more likely to occur or be more successful, i.e., will be received more appreciatively, when the substitution is "closer to" the one actually desired. For example, if a person desires love, he may accept either status or services more readily than he would goods, or information. (See Figure 2, p. 15) In a later study by Turner, Foa and Foa (1971), the issue of reciprocation in—kind was even more directly ad- dressed. Here the authors suggested that not only does appreciation follow reciprocation of a desired resource, but that the desired resource also correlates highly with the resource which is given. Working with the same scheme of resources (at this point called interpersonal reinforcers), subjects were presented with six exchange situations in which they were to give a particular resource to another person. They were then asked to choose from pairs of resources, the item they would prefer to receive in return. The results, summed in Table 1 support the hypothesis that individuals prefer reciprocation in-kind. The only exception, out of the six exchange commodities, was the resource "goods," with 18% choosing reciprocations in-kind, 19% choosing services and 21% choosing love. wt be re] EqL wid hus; intj rela. Cien Part 30 Thus, the data provided by Turner, et a1, support the contention that a systematic scheme Of exchange resources does indeed Operate in determining reciprocation patterns.3 Furthermore, it lends support to the emerging assertion that reciprocation in-kind is the most frequently desired response tendency. If individuals accept more appreciatively, reciprocation which is in-kind and the level of appreciation decreases the more heteromorphic (different in form) the exchange becomes, this may mean that either the assessment of what constitutes equivalence becomes more difficult to judge, and/or that individuals have value preferences for certain exchange commodities which prohibit them from evaluating positively, exchanges which are heteromorphic. Thus the central question becomes: if individuals prefer behavioral reciprocation in-kind and evaluate more positively exchange commodities which are homeomorphic then what is likely to be the outcome Of relational exchanges between individuals who have widely different exchange repertoires? More specifically, how is reciprocity and equity likely to be judged by a husband and wife who have widely different behavioral affection repertoires, e.g., a husband who is highly economic and a wife who is highly intimate. Are these relationships likely to be characterized by greater feelings of inequity, dissatisfaction and perhaps relational instability? Or are affection repertoires suffi- ciently flexible so that they can be adjusted to a marital partner's widely different affection patterns without it 31 TABLE 1 PERCENTAGE NUMBER OF CHOICES OF RESOURCES, CONTINGENT UPON RESOURCE GIVEN Resource chosen in exchange Resource Love Status Inform- Money Goods Ser- All Given ation vices Resources Love 32 24 18 2 9 15 100 Status 26 27 21 2 8 16 100 Informa- 23 21 25 5 10 16 100 tion Money 11 14 15 22 16 22 100 Goods 21 l7 l6 9 18 19 100 Services 20 17 18 9 ll 25 100 All Resources 22 20 19 8 12 19 100 (Taken from Turner, et a1, 1971, p.173) 32 affecting the reward/cost ratio of the exchange relationship? Or are marital partners able to cognitively mediate the affection exchange process and objectively assess the equity of the exchange, independently of the differences between the commodities given and received? Lack of support in the social exchange literature for the second alternative and only tenuous support for the third position would appear to make hypothesized outcomes in the first statement more probable. Assuming that individuals do have individual preferences or behavioral sets for affection exchange and that there is relational strain toward in-kind reciprocation, how are cost/reward ratios likely to be affected? In those instances where a major behavioral change, from established patterns, is necessary in order to carry out in-kind reciprocation, Thibaut and Kelley suggest that not only are larger costs likely to be incurred by the sender, but that the reward value to the receiver is likely to suffer as well. The greater the deterrence to performing a given act-- the greater the inhibition the individual has to over- come--the greater the cost of the act. Thus cost is high when great physical or mental effort is required, when embarrassment or anxiety accompany the action, or when there are conflicting forces or competing response tendencies of any sort. (Thibaut, 1959, pp. 12-13) And as Thibaut and Kelley further state: Incompatible response sets increase the actor costs and will usually impair both the quality of his performance and the other person's appreciation of the performance thus reducing the reward value of the product. (Thibaut, 1959, p. 59) ca ir. li En. lil Wit Of‘ hOrI beha 33 Thus, not only does replacement of learned patterns of exchange adversely affect the cost/reward ratio for the sender; but, by the same token, attempts to make the adjust- ment toward in-kind reciprocation may also negatively affect the satisfaciton derived by the receiver. Therefore, even if behavioral repertoires are flexible enough to carry out in-kind reciprocation, the exchange may still not be evalua- ted as worthwhile and/or rewarding by the participants. These basic cost/reward principles would also seem to hold when applied to the exchange of affection and caring in the marital context. Here as in other exchange situations, in-kind reciprocation may be important for keeping individual costs down and reward values up. However, in those marital exchanges termed transactions ip_mixed currencies (Villard & Whipple, 1976), where one individual's behavoral set predis- poses him to reciprocation with highly economic forms of affection, while the other's repertoire precipitates recipro- cation with highly intimate forms, the costs likely to be incurred as a result of forcing in-kind reciprocation are likely to be high, while the reward value provided by adher- ence to reciprocation patterns which are not in-kind is likely to be low. Thus, it is likely that two individuals with widely different affinities for affection exchange, i.e., extreme use of economic forms by one and extreme use of intimate forms by the other, are apt to be caught on the horns of a dilemma: whether to attempt to modify existing behaviors in order to bring them more into line with that of 34 the spouse, thus incurring high individual costs, or main— taining divergent dyadic patterns and suffering decreased reward value, obtainable from the relationship. In those instances where reciprocation is not in kind, and no attempt is made to move in a homeomorphic direction (perhaps because of high individual costs), this would appear to necessitate that the receiver cognitively evaluate and assess the equity of what has been given with that which has been received. Here the individual must undoubtedly take into account not only in what esteem the culture holds that commodity, but he must also assess the unique value attributed to it by the sender, within the context of that specific relationship, and perhaps equally important, the individual must evaluate it in terms of his own preferences and values. Throughout this entire process it would seem that accurate communication between exchange partners would be a necessary condition if the exchange is to be positively evaluated. And although individuals are undoubtedly capable of identifying and communicating about differences in behav- ioral affection exchange, and thereby translating and resolv- ing questions of value and meaning, the amount of relational sensitivity and sophistication required to make the identifi- cation and the amount of energy needed to cognitively medi- ate the process may be a deterrent to smooth relational functioning. 35 The necessity for cognitive mediation and/or relational negotiation is made crucial by differences both in coding and in the exchange focus of each mode, i.e., the intimate and economic. In the economic forms of affection exchange, the meaning of affection or esteem is communicated symbolic- ally, through the offering of a commodity which has funda- mentally a time/energy focus. For example, the giving of a gift is symbolic of the esteem or affection which one person holds for another; it is not the affection or esteem itself.“ By contrast, in intimate forms of affection exchange, where some level of physical or psychological identity is the commodity offered, the meaning is inherent in the behav- ioral act. For example, the demonstrative act of hugging another person carries, intrinsically, the meaning of esteem and affection. There is no need to take an inter- mediate cognitive step and assess meaning from a concrete referent. Therefore, because these forms of affection exchange i.e., the intimate and the economic, are not only fundament- ally different in terms of what each offers, but also in the way in which it is offered, i.e., the way the meaning is derived from the two modes, it would seem that it would be very difficult to determine the equity of a heteromorphic exchange. If, for example, one gives intimacy, which is identity-centered and intrinsically rewarding, but receives economics, which is time/energy centered, and extrinsically rewarding, it is (at the risk of being overly simplistic) 3.6 analogous to judging apples against oranges. It seems likely that it would be very difficult to determine the equivalence of what was received in economics with what was given in intimacy. And, as indicated earlier, in instances of heteromorphic exchanges, recognition of differences may be maximized and perceived inequity may be the product. It would seem that the only way in which this situation is likely to be positively resolved is if the individuals initially communicate about the differences in their affec- tion exchange repertoires and thus establish the meanings each holds for his own and the other's behavior and from that point on, interpret relational exchanges in terms of the other person's orientation. In summary, it is hypothesized that in heteromorphic exchanges, the dyadic outcomes are likely to be more per- ceived inequity, disenchantment with the exchange relation- ship and possible relational instability. However, it is hypothesized that these conditions may possibly be avoided if the spouses are able to communicate accurately about divergent affection repertoires. The evidence clearly indicates that attempts to make major changes in affection exchange behavior are likely to result only in increased costs and decreased rewards. Thus, the importance of reciprocation in—kind in affec- tion exchange would seem to lie in two areas: the increased probability of husbands and wives to accurately assess the equity of the transaction by measuring the rewards given against those received and, secondly, the issue of additional 37 costs incurred and reduced rewards derived by attempting to reciprocate in-kind in conflicting, affectional repertoire contexts. As long as exchanges are in—kind and particularly if they are tangible or extrinsic (economic), then equity of exchange can be more easily assessed and if equity is found to exist, then redress can be more easily resolved. On the other hand, if dyadic exchanges are carried out heteromorph- ically, then the equity of the transaction is considerably more difficult to assess, the differences are more readily noticed and identified, and if inequity is thought to exist, it is more difficult to debate and, therefore, to resolve. Effects of Inequity in Social Exchanges Given that an individual's rewards and investments become imbalanced and the exchange is perceived of as in- equitable, what are the effects on the individual and the implications for the relationship? Redressive responses to perceived inequity take basically two forms: affective and behavioral. (Blumstein and Weinstein, 1967, pp. 408-409) In the affective vein the individual may feel anger if his investments outweigh his rewards or conversely, guilt if his rewards seriously outweigh his investments. (Homans, 1961, pp. 75-76) Due to the tension which is associated with these emotions, the individual is viewed as motivated toward behavioral responses designed to correct the imbalance between investments and rewards. Along this line there are 38 several options: the individual may alter his inputs, either increasing or decreasing them; he may seek to change his outcomes; he may cognitively distort his perception of inputs and outcomes; and, finally, he may leave the relation- ship context. In addition, the individual also has the opportunity of attempting to redress through acting on the other. That is, the individual may attempt to alter or cognitively the distort other's inputs and outcomes or he may try to force the other to leave the field. (Adams, 1965, pp. 288-94) Depending on the reward/cost ratio, the individual either elects to remain in the exchange relationship or else decides to discontinue it. For as Thibaut and Kelley sug- gest: ". . . every individual voluntarily enters and stays in any relationship only as long as it is adequately satis- factory in terms of his rewards and costs." (Thibaut and Kelley, 1959, p. 37). In order to help make this decision, the individual supposedly needs some kind of criterion against which to assess his outcomes. Thibaut and Kelley call these 99m: parison Level (CL) and Comparison Level for alternatives .(CL alt). The comparison level is the standard against which the individual evaluates his present relationship in terms of its attractiveness or how satisfactory it is. (Thibaut and Kelley, 1959, p. 21) There are two kinds of comparison levels. The first of these is a function of all other relationships a person has been involved in. If the 39 current relationship is providing outcomes above that of the comparison level, then we would expect the individual to be satisfied with the relationship. If, on the other hand, the outcomes derived fall below the comparison level, then the 'logical conclusion would be that the individual would experience dissatisfaciton. A second type of comparison level is that derived from Observation of the relationships of others. If an individual observes a relationship of a friend or colleague which appears more attractive than his, the comparison level for judging his own relationship will likewise be raised. (Swensen, 1973, p. 228) The second mgjg£_level of comparison is the comparison level for alternatives. In this type of comparison, the individual contrasts his payoffs in the present relationship, with those which he might obtain from other relationships available to him. If the person perceives that the payoffs in his present relationship are higher than those in the alternatives available to him, then it is likely that he will be relatively satisfied and will remain in the relation- ship. If the payoff of an alternative relationship is perceived of as higher than that of a present one, then it is likely that he will forego the present one and become involved in the more positive one. And finally, if the perceived outcomes are approximately equal, then it is likely that he will show no great commitment or attachment to either relationship. (Swensen, 1973, p. 228) 40 By combining these two types of comparison levels with perceived outcomes, Thibaut and Kelley identify four differ- ent types of relationship categories: (a) A relationship in which the payoff in the current relationship is higher than the general comparison level and higher than the comparison level for alternatives. This type of situation would theoretically produce a relationship that is stable and satisfying for both people, given that both shared the same perspective on the relation- ship. (b) A relationship in which the payoff in the current relationship is higher than the general comparison level but lower than the comparison level for alternatives. This would provide a satisfying relationship, but it would also have to be considered unstable as the individual may elect at any time to follow the more attractive alterna- tives available. (c) A relationship in which the payoff in the current relationship is lower than the general comparison level and higher than the comparison level for alternatives. In this situation the relationship is likely to be unsatisfying but stable as the perceived alternatives are less attractive than the present one. This is also likely to be a very frustrating and dependency based type of association. (d) A relationship in which the payoff in the current relationship is lower than the general comparison level and lower than the comparison level for alternatives. 41 This type of relationship would be a very unsatisfying one which is theoretically likely to terminate. (Swensen, 1973, pp. 228-29) Applying thse exchange principles to the affection dimension of marital relatiOnships, those relationships in which the outcomes in affection are lower than both the comparison level and the comparison level for alternatives (a relationship previously described as unstable and unsatis- fying) should, theoretically, terminate in marital schism, or divorce. However, in those relationships in which the payoff in affection is lower than the comparison level but higher than the comparison level for alternatives, although being unrewarding, the relationship will remain stable. In this situation what is likely to be the effect not only on these individuals and their relationship, but on other family members as well, either indirectly through frustra- tions projected on the entire family environment or more directly through the parent's relationships with their children? The answers to these questions are the goals for the chapter which follows.5 FOOTNOTES l. The concept of relational communication as it will be used in the present investigation will expand the original definition advanced by Watzlawick, et a1. (1967) and elabor— ated on by Rogers and Farace (1974). These writers used the term relational communication to refer to the use of communi- cation to define the relative status or hierarchy between members of a dyad. The current usage broadens that defini- tion to include not only the control dimension but the affection dimension as well. In effect, relational communi- cation refers to that communication which defines the ongoing nature of all aspects of the relationship definition. The term relational dimension will be used to refer to those specific interaction dimensions which contribute to and amplify that relationship definition. 2. The concern with particularism as a quality of a resource and the exchange situation was first discussed by Talcott Parsons in 1951, in Social System. In talking about value orientations and reward systems, Parsons suggested that "the gratificational significance of an orientation can never transcend the particular relational system of which it is a part. The standard must be couched in terms of signifi- cance for this particular actor in these particular relations with these particular Objects." (Parsons, 1951, pp. 62-63). 3. The scheme referred to is the resource system classified by particularism and concreteness (see Table I) which suggests that the greater the proximity of resources that are exchanged, the greater the percieved equity and satisfaction with that exchange. 4. This is not to say that an economic form of affec- tion cannot have a very strong identity focus, as in instances of gift giving where someone makes something, utilizing his own skills, perhaps with his own hands and, therefore, "puts a lot of himself into it." In these instances it may be difficult to determine what characterizes the gift more, i.e., the amount of time/energy expenditure or the unique abilities and personality instilled in it by the giver. In these instances, the commodity may have dual-foci and there- fore be both economic and intimate. 5. »The unstable—unsatisfactory marriage has been the source of considerable investigation while the stable- unsatisfactory marriage has rarely been examined. While 42 43 marriages of the former type are easily discovered in social service and marital counseling centers, marriages of the latter type are undoubtedly more difficult to identify and locate in sufficient numbers for empirical research purposes. However, there is considerable evidence to indicate that a potential source of this type of marital relationship is the home of the emotionally disturbed child. CHAPTER II MARITAL AND FAMILY AFFECTION EXCHANGE: A REVIEW OF SUBSTANTIVE FINDINGS This chapter will concern itself with the following issues. First of all, how important is affection exchange in marriage? Is it salient enough to significantly effect: (1) the level of satisfaction derived from the relationship and (2) the actual stability of the relationship? These issues can partially be addressed by asking: Are there differences between satisfying and unsatisfying, stable and unstable marital relationships in terms of how spouses exchange affection? Secondly, what role does deficient or unsatisfying marital affection exchange play in symptom formation in children? Do deficiencies in marital affection exchange lead to maladaptive affective patterns in parent-child or family interaction? This issue can be addressed by asking: Are specific patterns of marital and family affection ex- change associated with specific types of symptom consolida- tion in children? Marital Stability, Satisfaction and Affection Exchange Patterns As one reviews the literature on affection exchange in marital settings, one is struck by two facts. The first is 44 ar. is pe: do 311C Sh: aff 45 the lack of attention which has been given to the issue of affection exchange and its relationship to satisfaction level, conflict and marital divorce. The second is the pervasive influence of Parsons and Bales (1955). Almost without exception, the studies which have been conducted since 1955 which have attempted to explore issues related to marital interaction and functioning, have either utilized wholly, the instrumental-expressive dichotomy of these writers, or at a minimum, have had their thinking structured by that influence. Briefly, Parsons and Bales assumed that in sex role differentiation in marriage, women provide socio-emotional support which is primarily ego-enhancing, conciliatory and integrative of other family roles, while men provide leader— ship, executive skills and status to the family. In effect the husband exchanges primarily occupationally derived rewards for the wife's restorative commodities. Most of the past research on marital relationships, the majority of which has been done by family sociologists, emphasizes two types of relationships: institutionalized and companionship. In the institutionalized marriage there is strict adherence to traditional role prescriptions and performance, with the instrumental aspect of the relationship dominating and being a greater determinant of happiness and success in the relationship. By contrast, in the companion- ship marriage there is greater emphasis On the expressive or affective aSpects of the relationship with variables such as 46 affection, sexual enjoyment, time spent with one another, and communication being of major significance in determining success and happiness, however these latter states are defined. With national trends seeming to swing increasingly toward an emphasis on the companionship aspects of marriage, at the same time the culture experiences a corresponding increase in divorce rates, one cannot but help be curious about the relationship between the two and the probable role of the affective or expressive dimension in marital satisfac- tion and stability. As Hawkins (1968) suggests: Companionship has often been singled out as being increasingly the primary basis for marital satisfaction in modern American marriages. Despite differing evalua- tions of the trend, virtually all Observers of the American family have noted the increasing degree to which the marital relationship has come to focus primar- ily on the affectional relationship of the Spouses. The quality of this aspect of husband-wife relations must therefore increasingly be seen as the basis of marital satisfaction. (Hawkins, 1968, p. 647) Of the studies undertaken in the last fifteen years dealing with the institutionalized marriage, most support the contention that adherence to traditional role expecta- tions and congruency between expectations and role perform— ance are likely to be associated with marital success and happiness. In particular, these studies indicate in one way or another that marital happiness is closely associated with the husband's successful performance of instrumental roles and/or the wife's adherence to female roles which do not threaten instrumental male roles. Among these are Luckey (1960a; 1960b; 19600), Blood and Wolf (1960), Hurvitz (1960; 47 1965), Westley and Epstein (1960), Nye (1961), Aller (1962), Axelson (1963), Katz, et a1. (1963), Stuckert (1963), Cutler and Dyer (1965), Kotlar (1965), Taylor (1967), Orden and Bradburn (1968), and Meyerowitz (1970). Although sometimes interesting in their findings, these studies lend little to our knowledge of the affection exchange process among marital partners and its role in the success or failure of marital relationships. The studies done on the companionship marriage, where the affection issue is more salient, are not only much fewer in number but, unfortunately, also contribute relatively little to our understanding of the affection exchange process in the marital context. Because affection exchange is fairly abstract, Often relying heavily on non-verbal communi— cation as a mode of exchange, because it is difficult to quantify and, perhaps above all else, because it is the most private of all the relational dimensions, it has rarely been the focus of empirical, scientific investigation. In looking at the variables of sexual enjoyment, com- panionship, and feelings of affection as important factors in the companionship marriage, no attempt has been made to define the major parameters of human affection as the present study has sought to do with the dimensions of economic and intimate or to specify a set of Operations for each type. In existing investigations, at best, the term affection is left conceptually undefined, grouped with other companionship factors, and then associated with various levels of marital 48 happiness and success. However, some insights can be gleaned from the studies available. Perhaps the most significant of all are those of Gurin, et a1. (1960). In sampling approximately 2,500 subjects, Gurin and his associates found that individuals who reported very happy marriages were more likely to mention relationship sources of happiness, i.e., affection, sexual compatibility, etc., while those experiencing a lesser degree of happiness, focused on the situational aspects of marriage, i.e., home, children, social life, as sources of their satisfaction. By contrast, those happiest mentioned most frequently situa- tional sources as contributing to what unhappiness there was, while unhappy spouses tended to mention relationship issues as sources of unhappiness. 49 TABLE 2 Relationship between Evaluation of Marital Happiness and Sources of Happiness (First-Mentioned Reasons) EVALUATION OF MARITAL HAPPINESS Very Above Average Not too Happy Average Happy Sources of Happiness Relationship 47% 42% 35% 18% Spouse 21 24 19 26 Situation 22 25 36 40 Other 8 7 6 2 Not happy about anything -- -- -- 8 Not ascertained . 2 2 4 6 Total 100% 100% 100% 100% Number of People* (870) (399) (546) (50) *Does not include seven people whose evaluation of marital happiness was not ascertained. (Taken from Gurin et al., 1960, p. 97) 50 TABLE 3 Relationship between Evaluation of Marital Happiness and Sources of Unhappiness (First-Mentioned Reasons of Respondents Giving any Reason for Unhappiness) Very Sources of Unhappiness Happy Relationship 12% Spouse 19 Self 9 Situation 50 Other -- Not Ascertained 10 Total 100% Number of People** (518) * Less than one-half of one percent Above Average 15% 23 8 46 8 100% (306) Average 16% 27 37 12 100% (440) EVALUATION OF MARITAL HAPPINESS Not too Happy 23% 50 23 100% (43) **Does not include seven married peOple whose evaluation of marital happiness was not ascertained, or the 553 married people who said there was nothing about their marriage that was not quite as nice as they would like it to be. (Taken from Gurin et al., 1960, p. 97) 51 These findings would tend to indicate that the affection exchange process (as well as other relational parameters) has more potential to create marital satisfaciton if present and perceived as desirable or conversely, dissatisfaction if absent, inadequate, or perceived as undesirable than situa- tional (institutionalized) aspects of marriage.1 As the authors conclude: Thus, feelings of happiness in marriage bear a clear relationship with the extent to which a person is satisfied or frustrated in the relationship aspects of his marriage. When he is happy with the relationship aspects of the marriage he tends to feel generally happy in the marriage; when unhappy with this aspect of the marriage, he tends to be unhappy. (Gurin, et al.,' 1960, p. 98) Along the same lines is the work of Levinger (1964). Studying 60 middle-class couples in terms of sources of satisfaction in their marriage, he found that both husbands and wives derived more satisfaction from the affective aspects of family tasks than from the instrumental aspects. Goode (1961), viewing marriage as a Special case of small group dynamics, attempted to assess the cohesiveness of the marital relationship in terms of what he called "bonds" and "bars."2 According to his view the cohesiveness or strength of the marital relationship is a direct function of the attractions within the marriage and the barriers around the marriage (which hold it together in the absence of attractions, e.g., religious or moral convictions against divorce, or commitments to the children), and an inverse function of such attractions and barriers from other rela- tionships. Using these two continua, i.e., the presence or 52 absence of attraction, and the strength or weakness of the barriers, Goode describes three types of marriages "full shell," "empty shell," and "no shell." In the so-called full shell marriage, not only are there strong boundaries, but also strong attractions for both spouses. It is a marriage where there is warm, emo- tional interchange. As a result, it is a cohesive unit which has the lowest probability of dissolution. In the empty shell marriage, there continues to be strong boundaries, but there exists very little attraction between the spouses. Goode describes such a marriage in the following manner: . . . The atmosphere is without laughter or fun, and a sullen gloom pervades the household. Members do not discuss their problems or experiences with each other, and communication is kept to a minimum. . . Their rationalization for avoiding a divorce is, on the part of one or both, "sacrifice for the children," "neighbor- hood respectability," and a religious conviction that divorce is morally wrong. . . The hostility in such a home is great, but arguments focus on the small issues, not the large ones. Facing the latter would, of course, lead directly to separation or divorce, but the couple has decided that staying together overrides other values, including each other's happiness and the psycho- logical health of their children. (Goode, 1961, pp. 441-442)3 As described above, the empty shell marriage lacks cohesion, but because of boundary strength, still does not dissolve. In effect, it is analogous to the unsatisfactory- stable relationship type of Thibaut and Kelley (1959). In the no shell marriage, the boundaries, as well as the attractions, have been diminished to the point that alternatives outside the marriage are eventually preferable 53 to those inside the marriage, and the union dissolves. The significance of this perSpective is that it re-emphasizes the importance of attraction and emotional bonding in mar- riage (of which affection is a major component), and in determining not only the satisfaction derived, but the stability of the marriage as well. A few studies have sought to assess the level of marital satisfaction through the analysis of mariage partners' evaluative descriptions of one another, some of which relate to affection exchange. Luckey (1964a), surveying 80 marital couples on per- ceived personality traits of the spouse, discovered that satisfied individuals described their spouses as consid- erate, helpful, tender, bighearted, friendly, neighborly and warm, all of which might be interpreted as indicating a positive affective relationship between the spouses. By contrast, unsatisfied individuals saw their partners as cruel and unkind, hard-hearted, gloomy, bitter, complaining, and slow to forgive, descriptive terms which might, by contrast, indicate a negative affective relationship. (Luckey, 1964, p. 220) This finding is also consistent with the work of Locke (1951) who reported that spouses in happy marriages des- cribed their partner's traits in a far more positive way than did divorcing persons. Happily married partners were far more likely to report the mate's traits as superior, or at least equal, to their own. 54 Investigators have also attempted to determine if satisfied and dissatisfied marital partners differ in their abilities to communicate verbally and non-verbally with one another. This is significant in the present investigation because at least the intimate forms of affection exchange are heavily dependent on the non-verbal band as a mode of exchange. Navran (1967), comparing 24 happily married couples with 24 couples who were in marital counseling, found that happily married couples (reported via paper and pencil test), had both "better" verbal as well as non-verbal commun- ication than did unhappy couples. In addition, the author also concluded that verbal communication appeared to be more important in marital adjustment than non-verbal communica- tion. The communication of happily married couples differed from that of unhappy couples in that happy couples talked more to one another, showed more sensitivity to one another's feelings, personalized their language symbols more fre- quently,“ and made more frequent use of supplementary, non-verbal communications such as glances and affect displays. (Navran, 1967, p. 182) Along the same lines Kahn (1970), in an ingenious study comparing 21 maritally satisfied couples with 21 dissatisfied ones, and using the Locke-Wallace Marital Adjustment Scale, found that dissatisfied husbands and wives were particularly prone to misinterpret each other's non-verbal behaviors. In particular, "the dissatisfied husbands were clearly more 55 inclined than the satisfied husbands to attribute negative connotation to their wives' attempts to communicate affec- tion, happiness, and playfulness." (Kahn, 1970, p. 455) This finding tends to suggest that a potentially significant aspect of marital disharmony may lie in the misunderstanding of affectional intentions that are communicated non-verbally. In addition, several studies have attempted to associate sexual incompatibility and maladjustment with marital con- flict and divorce. Terman (1938), found that the most adjusted couples in his sample had the highest ratio of actual and preferred frequency of sexual relations. Locke‘ (1951) reported that happy and divorced spouses differed significantly, both in their enjoyment of actual intercourse, as well as in their desire for it. And finally, Kephart (1954) reported that amOng divorce applicants, concern for sexual compatibility was primarily among the higher SES couples. This last finding would be in line with Gurin, et al. (1960), who found that the better educated placed greater emphasis on the relationship aspects of marriage. Along the same lines, Levinger and Senn (1967), in their study of conflict and non-conflict couples, found that the conflict couples (those actively involved in some form of marital counseling) had a lower prOportion of self- disclosure than did the other couples. Self-disclosure was found not only to be positively correlated with general marital satisfaction, but it was also found to be even more associated with positive evaluations and feelings for the other person. Di: 56 And finally, Young (1966), emphasizing the importance of affection exchange in marital relationships, noted the great frequency with which couples in marital counseling complained of affection deprivatiOn from their spouses as a primary source of their marital dissatisfaction. We can summarize our findings concerning marital affec- tion exchange, satisfaction, and stability along the follow- ing lines: 1. There is some evidence to indicate that the rela- tional aSpects of marriage, e.g., affection/caring, control, etc., are the major determinants of marital satisfaction. This tendency appears to be magnified among the better educated. ‘ 2. There is some evidence to indicate that satisfied marital partners have more accurate non-verbal communication, use more facial affect diSplays and self-disclose more to one another than do dis- satisfied marital partners. 3. There is some evidence to indicate that dissatis- fied marital partners are more likely to mis- interpret their spouse's non-verbal attempts to communicate affection, playfulness, etc., than are satisfied spouses. 4. There is also evidence which indicates that happily married couples tend to have higher rates of sexual intercourse, enjoy sexuality more, and are sexually more compatible than are dissatisfied partners. Discussion and Hypotheses One of the major themes emerging from the previous chapter's review of social exchange theory was the general notion that the more homeomorphic the social exchange, the greater the likelihood of perceived equity and satisfaction with the exchange.5 The present review of affection exchange 57 in marital contexts has suggested that satisfied marital partners tend to use more demonstrative (or what the present study would term "intimate" forms of affection exchange than do dissatisfied spouses. These marriages were also found to be more stable. Implicit in these findings was the assump- tion that these exchanges were reciprocated homeomorhically. Furthermore, it has been suggested that dissatisfied spouses are more inclined to misinterpret each other's non- verbal attempts to communicate affection than are satisfied marital partners. These marriages have also been found to be unstable. In light of the previous discussion concerning coding in Chapter 1, it might be reasoned that such misinter- pretations may be associated with widely different behavioral repertoires, with resulting difficulty in determining rela- tional significance and meaning. Therefore, it might be reasoned that affection exchange processes which produce the perception of inequity as well as marital dissatisfaction and instability are those which are transacted heteromorphic- ally. Thus the first major hypothesis of the study: In marital dyads, the more homeomorphic the relational patterns of affection exchange, the greater the per- ceived equity. Closely related is the sub hypothesis: Normal dyads will be more homeomorphic in their behav- ioral affection exchange patterns and will report higher levels of perceived equity than will conflict dyads. 58 Symptom Fogmation in Children and Family Interaction Patterns If inequities in affection exchange are thought to exist and the perceived outcomes are viewed as being lower than both the comparison level and the comparison level for alternatives, then theoretically that relationship should terminate. However, in the stable but unsatisfying relation- ship, where the individual's payoffs are lower than his comparison level, but higher than that of his perceived alternatives, continued inequity, particularly in the affec- tion exchange area, would seem likely to have deleterious effects not only on the individual and other aspects of that marital relationship, but on other family members and the family system as well. As Charney (1969) states: . . . with our increasing understanding of emotional disturbances not only as statements of the weaknesses with which people move into life from their own child- hoods, but as interactional responses in their current key family relationships, we now think of so many symptoms of mental illness in adults--psychosis, alco- holism--also so many symptoms of disturbance on other levels, such as psychosomatic problems and acute neur- osis, as referable in significant part to a strain, often covert, in the marital relationship of the patient. (Charney, 1969, pp. 12-13) Renne (1970) also voices the same opinion and extends the implications to other family members and to relationships which the individual maintains with others. Satisfaction with one's marriage is, in our monoga— mous society, an important component of individual well being. The independence of each "nuclear" family (parents and dependent children) from its own kin, and the relative social isolation of urban households, makes the spouse the primary companion and source of emotional sustenance. Relations with the spouse are so central a feature of an individual's social and emo- tional life that an unhappy marriage may impair the def inun in 1 the: tot not I in 1 they the furt tion 03m, mari 59 capacity of both partners for satisfactory relations with their children and others outside the family. (Renne, 1970, p. 54) The hypothesized causality of chronic marital dissatis- faction leading to symptom formation in children which in turn only intensifies marital difficulties is described by Renne as the "sick family syndrome." After studying 4,452 households she concluded: Parents who indicated a problem (or problems) with their children and also said they had talked with a professional person (school principal, minister, judge, policeman, social worker or psychologist) about their children, were decidedly more likely to be unhappy than those who reported a problem but no professional consul- tation. This suggests a "sick family syndrome" in which the unsatisfactory marriage not only affects the child's behavior but also is made more unhappy by the apparent failure in childrearing. (Renne, 1970, p. 62) Furthermore, there are theorists who hypothesize that deficiencies experienced in childhood, if not leading to immediate and observable problems, manifest themselves later in maladjustments in the person's own marriage. Among these are Foa and Foa (1972) who suggest that children tend to be deficient in those resources scarce to their parents, not only in such resources as goods, money and services, but in love, esteem, and information as well. And, furthermore, they suggest that deficits at early ages preclude or impair the ability for exchanges at later points in life. This is further supported by Young (1966) who reported that "affec- tion deprived spouses" consistently complained that their own childhood homes lacked warmth and affection. Thus we have a hypothesized chain of events in which marital maladjustment can contribute not only to childhood a1 ti in ca le 50 st) 60 and adolescent disorders, but can also precipitate maladap— tive behavior in the child's own marriage, which in turn may subsequently affect his or her own children. It is a cycle which theoretically may influence and impair the develOpment of virtually generations of family members. The direct tie between marital frustrations and con- flicts (not necessarily specific to the affection dimension) and symptom formation in the dyad's children has long been posited by researchers and practitioners alike. Among them are Ackerman (1957, 1965), Hawkes (1962), Jackson and Yalom (1964), Miller, et a1. (1966), Young (1966), Bell and Vogel (1968), and Charney (1969). As Charney suggests: It certainly is almost a universal experience in child guidance work that overt and more often covert disturb- ances in the marriage is found to be playing a huge role in symptom consolidation in the child. (Charney, 1970, p. 12) Looking at the implications of these views for the affection exchange process in family functioning, the percep- tion of inequity in affection exchange, given its importance in at least the companionship marriage, may be sufficient to cause dissatisfaction with the exchange relationship, and lead to its termination. However, depending on the compari- son level for alternatives, the relationship may remain stable and intact with perhaps severe implications for the children, particulary if the inequity continues and is covertly managed by the couples. The range of childhood disorders which are often hypo- thesized to be associated with marital maladjustment include schizophrenia, character disorders, personality disorders 61 and some psychologically based learning disorders, although the evidence for the latter two is far less convincing than for the former. In addition, there is also considerable evidence to indicate that certain interaction patterns in the parent— child relationship (particularly in the affection realm, which may result from cultural anomalies in the affective relationship of the husband and wife) may be associated with cognitive and behavioral disorders as severe as the psycho- ses. Among those who hold this view are Hess and Handel (1959), Geisman (1960), and Horner (1964). While the present investigation will focus only on the relationship of affec- tion exchange between husbands and wives, and the incidence of personality disorders in their children, the boundaries which distinguish these disorders from the other psycho- genetic disorders mentioned above are, at times, extremely blurred, often differing only in terms of degree of severity or predominant symptom at time Of diagnosis. For these reasons, a review of the affection exchange process both in terms of marital exchanges and parent-child exchanges, for each of the family disordered types will follow in the remainder of the chapter. The most severe of all the disorders is schiZOphrenia. Although the present study will not include a sample of schiZOphrenic families, it is important to the present investigation for two reasons. First of all, being the most severe, it has received the greatest research interest, tl' wl'. 1'10 Id It!) 9. m of to Syn the of and 62 particularly in terms of the marital relationship of the parents of the schizophrenic. Data derived from the analysis of the marital relationships of parents of schizophrenics may have relevance for assessing the marital relationship of parents of personality-disordered children. G And secondly, the symptoms which distinguish schizo- phrenia, although differing quantitatively, parallel fairly closely the qualitative symptoms of emotionally disturbed children (personality-disordered children), e.g., shy, withdrawing behavior is common for some types of emotionally disturbed children, while total withdrawal is typical of schizophrenic patients. This view would be consistent with the perspective of many family researchers, who feel that what distinguishes the normal from the abnormal family is not a gross distortion in interaction parameters, but rather a subtle (and often unnoticeable) quantitative and qualitative difference. Schizophrenia and Family Interaction Patterns Schizophrenia is typically viewed as "an extreme form of social withdrawal characterized specifically by efforts to modify reality into a tenable form by distorting the symbolization of reality, or through extreme limitation of the interpersonal environment." (Lidz, et. al., 1957a, p. 241) Among the most common symptoms can be included loss of interest in pe0ple and events, autism, emotional blunting and distortion, delusions and hallucinations, anomalies in 63 non-verbal communicative behavior, and the lowering of impulse controls. (Coleman, 1964, p. 275) Perhaps the most significant characteristics of schizophrenic behavior, from a communication standpoint, are that there is marked devia- tion from shared meanings and constriction of the inter- personal environment. Being the most debilitating of all the psychogenetic disorders, schiZOphrenia has drawn the greatest attention and has the most extensive research and theoretical history. Although the early investigators like Bleuler and Kraepelin posited on organic cause for schizophrenia, they nonetheless believed in the importance of psychological factors in the develOpment of the disorder, particularly in the role of frustration and conflict. Since the time of Bleuler, both researchers and practitioners have placed increasing emphasis on the signficance of early childhood experiences in the etiology of the disorder, as well as stressing the importance of the intrafamilial milieu in which the child is raised. The means by which schizophrenia and such family inter- active variables as conflict, dominance, affect and communi- cation clarity have been studied, range from the descriptive methodologies of case study, psychiatric interview and evaluation, and projective testing, to more recent attempts at empirical measurement. Together they have produced a number of theoretical camps on the etiology of schizophrenia and its relationship to family interaction. Perhaps the most widely-known theory of schizoPhrenic etiology is that of the "schizophrenogenic mother" (Fromm- 64 Reichman, 1948). This theory posits that the interaction effect of a strong, dominant mother, and a weak passive father deprives the child, particularly the son, of an adequate model of indentification and he becomes schizo- phrenic in response to the atypical cultural situation. In the theory of the schizophrenic mother, dominance or control has been the interactional dimension that has been focused on to the greatest extent, and due to the influence of this particular theory, control has been the parameter which has received the greatest attention in subsequent studies. Although the schiZOphrenOgenic mother is also typified as cold and rejecting, impervious to the feelings of others, as well as being seductive to the son and condescending toward the father, the importance of affection has been of only secondary concern. This is a tendency which is preva- lent throughout most of the family interaction literature. A number of investigations have previously and since revealed findings which may be interpreted as lending support to the theory of the schizophrenogenic mother. Despert (1938) found that 50% of the mothers of schiZOphrenic child- ren could be described as aggressive, overanxious, over- solicitous and considered to be the dominant parent. Clardy (1951), found that 50% (n = 30) of schizophrenic children had families characterized as overprotective yet rejecting. Mark (1953), reported after comparing mothers of schizo- phrenics with normal mothers, that schizophrenic mothers were inconsistent in their controlling behaviors (vacil- lating between overrestrictive and lax), with their emotional 65 tOne toward the child being one of "cool detachment." Freeman and Grayson (1955), in another comparison of normal and schizophrenic mothers, found that schizophrenic mothers portrayed themselves as somewhat more possessive, but inher- ently more rejecting than normal mothers. And, finally, in interviews with schizophrenic patients, Bolles, et a1. (1941), Lane and Singer (1959), and Singer (1954), all reported that schiZOphrenic patients viewed their mothers as interpersonally rejecting. Thus there would appear to be two main themes running through the theory of the schizophrenogenic mother: (1) a major deviation from the cultural norm in the distribution of power in the husband-wife relationship along with some evidence of a negative affection relationship between the spouses, and (2) a strong, emotional dependency between mother and child where affection and acceptance is largely dependent upon behavioral conformity on the part of the child. The theory of the schizophrenogenic mother also trig- gered a number of investigations into the role of the father in the development of the schiZOphrenic patient. Most of these studies have described the father of the schizophrenic as inadequate, indifferent, detached and similar to the mother in terms of being insensitive to the feelings and needs of others. Among the investigators who have proposed this portrayal of the father are Frazee (1953), Ellison and Hamilton (1949), and Gerard and Siegel (1950). 66 More specifically in terms of his relationship with his children, the father is typically viewed as rejecting toward the son and seductive toward the daughter. In those families where there is a schizophrenic daughter, it is often found that the father, in addition to being seductive to the daughter, is Openly derogatory and contemptuous of the mother, the exact mirror image of the mother-son relationship described in the theory of the schizOphrenogenic mother. In instances where there is a schizophrenic son, it is often found that the father is rejecting of the son, who is viewed as competing for the affection and esteem of the mother. Thus far, from the theoretical evidence, we begin to see an interesting pattern develOping, the incidence of schizOphrenia associated with the worst of all possible communication worlds, in terms of family control and affec- tion behaviors: weak and rejecting behavior by the father, and cold, domineering behavior by the mother. As Lidz, et al., (1957b) indicate: Perhaps in the most general terms one can follow Zelditch and Parsons in pointing out that the father is the leader of the family in adaptive-instrumental roles, as against the integrative-expressive primacy of the mother and, therefore, a "weak, ineffectual" father is more significant than a "weak, ineffectual" mother, whereas a "cold, unyielding" mother is more of a problem than a "cold, unyielding" father. (Lidz, et al., 1957, p. 333) In addition to the significance attributed to the parent-child relationship in the development of the schizo- phrenic individual, the marital relationship of the parents of schizophrenics has also been the focus of attention. The 67 research evidence strongly suggests that the marriage rela- tionships of parents of schizophrenics are typically char- acterized by unhappiness, conflict, and mutual derision. However, they are also surprisingly stable. Deficiencies in the relationships of parents of schizo- phrenics have been studied by a number of investigators, with these findings having perhaps the greatest application to the present investigation. Lidz and Lidz (1949), found a high frequency of broken homes and marital instability with 61% of their sample (n = 33) of schizophrenic patients coming from homes marked by marital conflict. Tietze (1949) reported that approximately half of a sample (n = 25) Of mothers of schizophrenic patients indicated that their marriages were very unhappy. Incidentally, another nine reSpondents reported that their marriages were "perfect," but upon further investigation, were actually found to have strained marriages, being far from perfect. Gerard and Siegel (1950) found marital conflict existing between 87% (n = 71) of parents of male schizophrenics, compared with only 13% in the control group. Frazee (1953), found that over half of her sample (n = 23) of marital couples were in severe conflict, with none having a conflict level which might be described as "normal" or "moderate." Plank (1953) found in a sample of schizophrenics, that 63% came from families where one parent was absent due to death or separa- tion. And Wahl (1954, 1956), found a greater incidence of parental loss and rejection early in life for schizophrenics than for normals. tI' P1 69 6) 68 Based primarily on these early findings, a number of researchers began theorizing about the marital relationship and its role in the development of schizophrenia in the child. Lidz, et al. (1957, 1958), in an oft cited study of families (n = 14) of schizophrenic patients, failed to find a single marriage that they could describe as "reasonably well integrated." Eight of the couples lived in a state of severe and chronic marital discord in which the continuation of the marriage was constantly uncertain, a condition which was labeled marital schism. Of particular interest in this type of marital relationship was the constant verbal sniping between the spouses, coupled with competition for the affection of the child. A particularly malignant feature in these marriages is the chronic "undercutting" Of the worth of one partner to the children by the other. The tendency to compete for the children's loyalty and affection is prominent; at times to gain a substitute to replace the affection missing from the spouse, but at times perhaps simply to hurt and spite the marital partner. (Lidz, 1957a, p. 243) In addition, in seven out of eight of these families, the instrumental role of the husband was limited only to providing financial support, with the wives being described, again, as emotionally cold, distant, and sexually aloof. In the remaining six families of the Lidz study, there existed serious psychopathology in one marital partner, which dominated the home environment, a condition termed marital skew. 69 In all of these families, one partner who was ex- tremely dependent or masochistic had married a spouse who had appeared to be a strong and protecting parental figure. The dependent partner would go along with or even support the weaknesses or psychopathologic distor- tions of the parental partner because dependency or masochistic needs were met. In contrast to the mar- riages with overt schism, one partner could gratify rather than combat a spouse's narcissistic needs. It may be significant that no member of these six marriages had intense emotional bonds to the parental family, and it is possible that the absence of such alternative sources of gratification tended to hold these Spouses together. A striking feature in all cases was the psychOpathology of the partner who appeared to be dominant, creating an abnormal enviromnent which, being accepted by the "healthier" spouse, may have seemed to be a normal environment to the children. (Lidz, et al., 1957a, p. 246) In skewed marital relationships there was a tendency for wives to dominate the family and to again relegate the husband to a minimal and ineffectual instrumental role. In the minority of cases, however, the husband was found to be dominant to an abnormal and exteme extent. According to the research of Lidz and his associates, female schizophrenics were more likely to come from schismatic families, while male schizOphrenics were more likely to come from skewed families. While Lidz, et al., were emphasizing the possible effects of overt marital conflict on the emergence of schizo— phrenia, another group of investigators began assessing the effects of covertly managed marital conflicts. Wynn, et a1. (1958), in studying families of reactive schizOphrenics, found that marital conflicts between parents, rather than being openly dealt with and conscious attempts 70 made to resolve them, were more often covertly managed, denied and kept from emerging. This the authors labeld pseudomutuality. Charney (1969) aptly describes this phenom- enon as situations . . . where a couple unconsciously work together to sweep under the rug, their various disagreements and antagonisms so as to dovetail with each other in a seemingly smooth-flowing way that wipes out the exper- iencing of problems, and, more importantly, the exper- iencing of separate individuality. . . (Charney, 1969, p. 13) Bowen (1959, 1960), in studying the family backgrounds of schizophrenic patients (n = 12), continued the trend toward theory building by noting the "striking emotional distance" between the parents of schizophrenic patients. This he referred to as emotional divorce. A phenomenon of particular interest in the emotionally divorced marriage was the seeming importance of gift giving between the spouses, inferred by many of the investigators to be an attempt to maintain a facade of an affectionate and loving relationship when, in fact, it did not exist. In support of the theory of emotional divorce is also the work of Behrens and Goldfarb (1958) who, in one of the first family interaction observa- tion studies, described the relationship of parents of schizophrenics as one of "basic isolation." These largely "second generation" family investigations into the etiology of schizophrenia, although not specifically supporting the theory of the schizOphrenogenic mother, did succeed in bringing into focus the seeming importance of the marital relationship as the cornerstone for either functional 71 or maladaptive family develOpment. Although producing some very interesting theoretical positions, most of the studies were largely descriptive, clinical case studies, utilizing few controls, often with very small sample sizes and data bases. For these and other reasons their conclusions must be viewed very carefully in terms of inferring a causal relationship between such family interaction patterns and the disorder. There are other widely held theories on schiZOphrenic etiology concerned primarily with clarity of family inter- action. Among these are Wynne and Singer's (1963a, 1963b) assertion that the disorder results from blurred meanings, encountered as the parent shifts to increasingly tangential issues during conversation with the child. Along the same lines, there is the double-blind theory of Bateson, et a1. (1956), who suggested a similar etiologic origin by positing quite convincingly that conflicting communication messages at different levels of abstraction was the cause of the disorder. Although these positions are intriguing and potentially relevant, especially in terms of the implications of double-binding communication during affection exchange between parent and child, they are perhaps the least fruitful in their direct contributions to the present study. The investigations and subsequent theoretical specula- tions of Lidz, et a1. (1957) , Wynn, et a1. (1958), and Bowen (1959), were followed by yet another series of investi- gations into the relationship between marital conflict and schizOphrenia. 72 Fisher, et a1. (1959), reported that parents of both schizophrenic and neurotic children disagreed more than the parents of normals when discussing their children, as well as When working jointly on TAT stories. Farina (1960), coding parental interaction on ten indices of conflict found that parents Of poor pre-morbid patients manifested more conflict than parents of normals on seven of the measures, while parents of normals and good pre-morbid patients were found not to differ significantly on any of the ten conflict measures. A comparison between the parents of good pre-morbid patients and the parents of poor pre-morbid patients found the latter higher on three measures. A later replication by Farina and Dunham (1963) also supported these findings. Caputo (1963), reported that parents Of schiZOphrenics disagreed more and displayed more hostility toward one another than did parents of normals and Lerner (1965), in a similar investigation, found that the parents of the control group of normals compromised their differences more often than either of two schizophrenic groups. And finally, Farina and Holzberg (1968) reported that the fathers of poor pre-morbid schizophrenics were found to disagree with their ‘wives more than fathers of disturbed but non-schizophrenic patients. Although supporting previous research, these studies lend little to our etiological knowlege of schizophrenia, 73 primarily because one can explain the relationship out of a responsive model as well as a causative one, i.e., the marital conflict arising as a response to the pressures of dealing with a schizOphrenic child, or schiZOphrenic symptom- Ology arising from marital conflict. Logical cases can be constructed for either position. In a later study, Gartner and Goldstein (1972), after surveying mothers (n = 19) of severely disturbed children (chi1dren with either psychosis or severe behavior dis- orders), found that 78% reported serious marital conflict, and 68% reported prolonged depression during the first year of the child's life. The significance of the depression issue rests on the authors' hypothesis concerning the result- ant affection deprivation to the child. . . . depression in a mother during the crucial first year of the child's life may result in her psychological unavailability to her infant. The resultant affectional deprivation may then be expressed in a variety of ways, depending on the degree of deprivation, the temperament of the child, and the extent to which other family members compensate for the deficiency. (Gartner and Goldstein, 1972, p. 902) The significance of this particular study lies in the fact that it is among the first (although inadvertently) to begin to address the time-order problem which confounds virtually every schizophrenic family study. Although still contaminated by being self-report data, the study does establish somewhat more precisely the sequence of events, i.e., marital conflict or depression in the mother during the first year of the child's life (with hypothesized affec- tion deprivation to the child resulting), following by 74 extremely high incidence of schizophrenia in those children. The lack of control of other variables, however, still leaves the hypothesized relationship open to charges of spuriousness. In the last ten years, the study of family interaction and its relationship to schizophrenia has shifted its focus from the marital relationship of the parental dyad to the larger arena of the entire family system, including not only other children, but often extended family as well. Increas- ingly there has been greater utilization of empirical, direct observation techniques. The dimensions which have thus far received the greatest attention parallel the areas scrutinized by earlier investigators. Of particular rele- vance are the dimensions of conflict, affect (affection) and, to some extent, control.6 Although far from conclusive, and often difficult to compare, the family interaction studies lend some additional and interesting insights. A few studies have examined the amount of conflict in schizophrenic and control families through the use of direct observation techniques and by means of verbal frequency measures. Among these are Lennard, et a1. (1965), Stabenau, et a1. (1965), Ferreira, et a1. (1966), and Farina and Holzberg (1968). The major findings which have emerged from these studies have been the indications that normal families interrupt more frequently, but have less conflict than schiZOphrenic families. For instance, Lennard, et a1. (1965), reported that normal mothers interrupted more often 75 during family interaction than did schizophrenic mothers. And Mishler and Waxler (1968) found that normal families and, in particular, normal fathers, interrupted more often than good pre-morbid schizophrenic families and fathers. This may imply that normal families are more spontaneous, more adaptive, and, again, less conflict ridden than schizo- phrenic families. Other studies have sought to distinguish between schizo- phrenic and normal families in terms of certain qualitative conflict measures, i.e., relative agreement, disagreement, failure to agree, and conflict. Among these are Cheek (1964a, 1964b, and 1965b), Lennard, et al. (1965), Farina and Holzberg (1968), and Friedman and Friedman (1970). As reported earlier, Farina and Hozberg (1968) found that fathers of poor pre-morbid schizophrenic patients disagreed with their wives more Often than did the fathers of non- schizophrenic, yet severely disturbed patients; and also that the parents of poor pre-morbids failed to agree more often than did disturbed parents. Cheek (1964a, 1964b, and 1965) found that normal families almost always expressed greater disagreement than schizophrenic families. Cheek concluded that in schizophrenic families there is a breakdown in the mother-father dyad that is characterized by an inabil- ity to reach spontaneous agreement, greater conflict with the resulting tendency toward the develOpment of a symbiotic relationship between mother and child. And finally, Friedman and Friedman (1970), also found that schizophrenic families expressed greater conflict than normal families. 76 From these studies we might safely infer that normal families are more spontaneous and less rigid in their inter- action styles, based on their willingness to interrupt one another; and, secondly, that normal families have less conflict and experience less disagreement than schiZOphrenic families. FollOwing this vein of thought, we might, there- fore, hypothesize that normal families may also be more spontaneous in their affection exchange repertoires, perhaps using greater amounts of physical touching, facial affect displays, and other intimate forms of interpersonal affective behavior, as spontaneity would seem to have a natural association with demonstrative forms of affection. Whether there is support for this hypothesis might best be determined by examining affective dimensions of family interaction in available schizophrenic family studies. Although not directly labeled in any of the studies to be cited, the dimension of affect or expressiveness is roughly anologous to the concept of affection. As Mishler and Waxler suggest in their study of schizophrenic families: The large number of interaction indices . . . were reduced to five conceptual domains . . . expressive- ness, strategies of attention and person control, speech disruption and responsiveness. For purposes of this discussion, it is convenient to group these further into two large c1asses--variables that are substantive in the sense that they refer to qualities of relation- ships and interaction, and structural variables that denote the more formal or stylistic aspects of inter- action. Under the former heading we would place expres— siveness and the strategies of attention and person control; these are analogous to the typical and fre- quently observed general dimensions of social relation- ships that traditionally are referred to as affection 77 and power. In the second class of structural variables are the domains of communication style and responsive- ness. (Mishler and Waxler, 1968, p. 283) Two family interaction studies have explored the differ- ences between schizOphrenic and control families in terms of amount of affective expression: Morris and Wynne (1965), and Mishler and Waxler (1968). The Mishler and Waxler study in particular is the most methodoligically sophisticated of all the family interaction studies, and has produced the most substantive findings, particularly in terms of affection exchange in schizophrenic families. In their study comparing families, they found that normal families reflected more (total) expressive behavior than good pre-morbid families with poor pre-morbids occupying an intermediate position. This was particularly the case in families with sons, although the same pattern existed for families with daughters, however, it was less marked. In regard to the quality of affective behavior, that is, whether it is predominantly positive or negative, there have been four investigations which have been concerned with family differences in terms of positive affect: McCord, et a1. (1962), Cheek (1964a, 1964b, 1965), Winter and Ferriera (1967), and Mishler and Waxler (1968). Mishler and Waxler reported normal fathers greater than both poor and good pre- morbid fathers, and normal mothers greater than poor pre- morbid mothers, in indirect positive affective communication. In addition, normal families and, in particular, normal mothers, expressed more laughter than good pre-morbid fami- lies and mothers; and lastly, normal families expressed more 78 tension release behavior than good pre-morbid families. (All findings were significant with male children only.) With reference to negative affective behavior, several studies have compared schizophrenic and control families on negative affective communication, i.e., antagonism, depressive moods, etc.: Cheek (1964a, 1964b, 1965), Winter, et a1. (1966), Winter and Ferreira (1967), Friedman and Friedman (1970), and Mishler and Waxler (1968). There are some conflicting findings in this area, yet some interesting trends do emerge. Winter, et a1. (1966), in assessing hostility from family TAT stories, found that disturbed but non-schizophrenic families produced greater amounts of overt hostility than either schizophrenic or normal families. Friedman and Friedman (1970), reported depressive moods to be more frequently expressed by schizophrenic families than normal families and hostility to be more frequently expressed by schizophrenic mothers than normal mothers. Mishler and Waxler reported that normal families and normal sons used more negative affective communication than did good pre-morbid families and good pre-morbid sons, however, since the latter groups were affectively blunted, i.e., expressing very little affective behavior either positive or negaitve, these findings are somewhat misleading. Placed more in perspective by comparison with positive affective and total amount of expressed affect: . . . the higher level of expressiveness in normal families is associated with disporportionately more positive feelings; the lower level of expressiveness in good pre-morbid families is associated with dispropor- tionately more negative feelings. Poor premorbid 79 families are in between on both level and affective quality of expressiveness. (Mishler and Waxler, 1968, p. 93) In addition to having low levels of affective communica- tion, good pre-morbid families were also found to have high levels of instrumental behavior. This tends to support the interpretation that these families are defending themselves against the expression of affect and feeling, either positive or negative. As Mishler and Waxler suggest: . . . there seems to be an implicit model in which non- expressiveness is viewed as a mechanism of defense, a defense against real-but-denied feelings. Thus, for the patient, the lack of expressed feelings signified to others a lack Of interest and concern; this is likely to dampen their efforts to engage him in a meaningful but potentially threatening relationship. The mother's distance guards her from the dangers of intimacy that would follow a more openly expressive relationship. In the family there is a mutual protec- tion pact in which feelings are collectively denied. (Mishler and Waxler, 1968, p. 81) And finally, one additional finding is of significance which relates to interpersonal orientations in affective behavior, i.e., whether affect is expressed directly or indirectly. Normal families were found to be indiscriminate in their expressions to family members, with little prefer- ence for either direct or indirect expression. Poor pre- morbids, who are also found to be both relatively expressive and positive, had higher rates of affect expressed directly toward specific family members. In contrast, good pre- morbids who had the lowest expression and, proportionately, the highest negative affect, showed very high rates of indirect behavior, i.e., feelings not expressed directly 80 toward specific individuals, but found to "reappear in— directly in abstract personal references." (Mishler and Waxler, 1968, p. 105) We can summarize our findings of schizophrenic and normal families along the following lines: (1) (2) (3) (4) (5) (6) there is some evidence to indicate cultural var- iance in role governed affection and control behaviors for parents of schizophrenics with the father assuming an ineffectual instrumental role, and the mother being ineffectual in the expressive role; there is overwhelming evidence to indicate that there is more marital conflict and emotional distance between the parents of schizophrenics than between the parents of normals; there appears to be more conflict and disagreement within schizophrenic families (among all family members) than in normal families. There also appears to be a fairly high level of covert hostil- ity in abnormal families, when compared with normal families; interaction in schizOphrenic families appears to be less spontaneous and somewhat more highly structured in terms Of who talks to whom and about what. Overall there is more rigidity in the interaction patterns of schizophrenic families; schizophrenic families appear to be less expressive of affection and emotion generally than normal families. Closely related is the conclusion that schizophrenic families show more concern with instrumental behavior and role relationships than normal families; and finally, expressive behavior in schiZOphrenic families tends to be more negative, and more indirect than expressive behavior in normal families. Personality Disorders and Famin InteractIOn'Patterns The majority of behavioral disturbances occuring in children and young adolescents are psychiatrically classed 81 either as transient or situational personality disorders (if associated with a particular event or adjustment situation; common examples include sleep disturbances, particularly nightmares, and occasional hyperactivity’), psychoneurosis (if associated with regressive behavior such as irrational fears, withdrawal and/or psychosomatic complaints; common examples include school phobias, depression and crying spells), or as personaligy trait disorders (if associated with chronic aggressive or ineffectual behavior; common examples include conflicts with teachers and peers, truancy, learning problems and neurosis). Usually, all these dis-- orders in children are grouped under one heading and the possessor labeled "emotionally disturbed." Thus, the term emotionally disturbed encompasses both shy and regressive behavior, as well as aggressive, acting out behavior. In effect, it is a catch-all for a very wide range of often drastically different disorders. All of which is to suggest that, if one is using the label "emotionally disturbed" as a criterion measure, either for purposes of identifying relevant literature or in data gathering, one should be aware of the fact that he is at the nominal level of thought, and at times barely that.8 The emotionally disturbed child and his family have not attracted nearly the amount of research interest that the schizOphrenic family has. However, there are some interest- ing theoretical findings and speculations concerning the marital relationship of the emotionally disturbed family as 82 well as some valuable insights concerning the affection exchange process between family members in such families. As indicated earlier in the introduction to the section on symptom formation in children, the role of the child as a family scapegoat is one of the most widely held and accepted views in clinical and psychiatric circles. It is particu- larly ascribed to by those who treat the emotionally dis- turbed child and his family. In essence, it suggests that parents with marital conflicts which cannot be directly addressed and confronted, use the child as an object for the displacement of their hostility for one another, thus making it possible to live together. Bell and Vogel (1960), in one of the earlier studies (n = 9 families) dealing with family scapegoating, concluded: In all the disturbed families it was found that a particular child had become involved in tensions exist- ing between the parents. In the 'well' families used for control purposes, either the tensions between the families were not so severe or else the tensions were handled in such a way that the children did not become pathologically involved. In general, both parents of the emotionally disturbed child had many of the same underlying conflicts, but in relationship to each other, they felt themselves to be at opposite poles, so that one spouse would act out one side of the conflict and the other would act out the other side of the conflict. They had developed an equilibrium in which they minimized contact with each other and minimized expressions Of affect, particularly hostility, which they strongly felt for each other, and this made it possible for them to live with each other. But this equilibrium had many difficulties, the most serious of which was the scapegoating of a child. (Vogel and Bell, 1960, p. 383) Spiegel (1968) when comparing emotionally disturbed families with normal families has made similar observations: 83 In the first group (disturbed) the children inevitably become involved in a conflict or disequilibrium situa- tion which exists between the parents. Most frequently, neither the child nor the parents are aware of this fact, nor are they aware of the ways in which it comes about. In the second (healthy) group of families, although there may be sources of tension between the parents, the children are minimally involved in them. (Spiegel, 1968, p. 392) In many instances, scapegoating (which can occur as a result of dissatisfaction with any family or marital area of functioning) is associated with the affection exchange process, particuarly in the sexual area. For instance, in two of the clinical cases studies of Bell and Vogel, the spouses reported many problems in their own marital sex lives. However, "rather than face these maladjustments directly, the problems were eXpressed through concern about the masculinity of a slender, graceful son." (Bell and Vogel, 1960, p. 389) In these instances the son becomes the focus of attention, thus drawing off the tension which is really associated with an unsatisfying marital sexual relationship. Thus the theory of the family scapegoat would suggest that symptom consolidation in the emotionally disturbed child is a largely unconscious, homeostatic family mechanism which functions in reducing tension resulting from unresolved marital conflicts. And, as has been implied, this may or may not, depending on the particular family dynamics, center around the affection exchange process. Also, as was the case in schiZOphrenic families, there is often a deviation from normal cultural patterns in terms 84 of how affection is exchanged within the parent-child rela- tionship, largely as a result of an unsatisfying marital relationship. Frequently, the mother [referring to mothers of emotionally disturbed children] expressed, although inconsistently, unusually strong affection for a son. They justified this almost invariably in the same way: the child had problems and difficulties and thus needed more help and care than the other children. However, what they considered care and protection far exceeded the usual limits. This can be seen, for example, in the mother who carried her twelve yearold son from the bed to the bathroom so that he could avoid bed-wetting; in the mother who continually fondled her adolescent son and called him lovie; and in the frequent slips of the tongues by a variety of family members which identified the mother and son as spouses. Fathers on the other hand Often had special attachments to, and fondness for, daughters. (Vogel and Bell, 1960, p. 394) In addition to the marital relationship being of direct or indirect significance in the etiology of emotional distur- bances, the affection exchange process within the parent- child relationship has also been viewed as contributory. Early studies have produced some rather familiar findings. Some researchers, Silberpfennig (1941), Newell (1934, 1936), and Ingham (1949), have suggested that childhood neuroses are associated with maternal rejection. Others have sug- gested that the other extreme--very close mother-child relationships, where the mother discourages any degree of emotional separation of the child and herself is likely to be associated with either psychosomatic disorders, Sperling (1949), and Miller and Baruch (1950), or school phobias, Wilson (1955), Talbot (1957), Eisenberg (1958) and Davidson (1961). 85 More recent investigations report similar findings, particularly in terms of parental rejection and emotional distancing in the parent-child relationship. For example, Cox (1962), using TAT scores to assess the closeness of the parent-child relationship of emotionally disturbed boys,. found that a significant number rejected one or both of their parents. Schulman, Shoemaker, and Moelis (1962), reported, from observed interaction, that the parents of emotionally disturbed children (specifically children with behavioral conduct problems) were more hostile and rejecting than parents of normal children. In a very interesting study comparing emotionally disturbed boys with normal boys, Weinstein (1965), found that when given felt figures, both human and non-human, to arrange on a board, the emotionally disturbed children placed the child figure closer to the adult male figure than the adult female figure. This was in contrast to the normal children who placed the child figure closer to the female adult figure.9 This finding supported the interpretation of rejection of the mother in the social schemata (the way in which the individual congitively arranges social stimuli and experiences, i.e., response sets) of disturbed children. The author concluded: the schemata of emotionally disturbed children appear to single out the woman-child relationship as particu- larly distant and negative. (Weinstein, 1965, p. 461) 86 In the second phase of the same study, the two groups were asked to reproduce an already prepared arrangement from memory. The majority of the emotionally disturbed group placed the human figures farther apart than the non-human objects, whereas the normal group made no distinction between the two groups of objects. The results indicated that: . . . emotionally disturbed children have not developed the normal adult schemata which organizes humans as a close units. Whereas adults replace humans closer than rectangles, emotionally disturbed children replace the rectangles closer. It is uncertain whether the dis- turbed children have developed an opposite schema, one which conceives humans as separate and unfriendly, or have simply not developed a positive human relationship schema. (Weinstein, 1965, p. 460) Fisher (1967), using a similar procedure to assess the social schemata of elementary school boys who could not adjust to the school environment, found that emotionally disturbed boys placed greater distance between all figures in the social schemata, than did a group of normal boys from the same school. And finally, DuHamel and Jarmon (1971), comparing the social schemata of emotionally disturbed boys, with that of their male siblings and a control group, found that dis- turbed boys did schematize all human relationships as more distant than normal control children, but not significantly more distant than their well siblings. Disturbed boys were found to place child figures between the mother and father figures, whereas well and control siblings placed the child figure apart from these parental figures. Also, emotionally 87 disturbed boys schematized the mother-father relationship as more distant than did the other two groups, who depicted it essentially the same way. These findings were interpreted as to suggest that: . . . emotional disturbance in children is an outgrowth of some idosyncratic interactional patterns between parents as well as between parents and children, rather than as a product of a disturbance in any specific child-parent relationship. (DuHamel and Jarmon, 1971, p. 285) This again would tend to support the theory of the emotional- ly disturbed as the family scapegoat. In addition, there have also been other projective technique studies as well as direct Observation studies which have shed additional light on the intrafamilial milieu of the emotionally disturbed child, relating specifically to the conflict and affectional themes in these families. Winter, et al. (1966), found through assessing hostility themes in TAT stories, that emotionally disturbed families produced more total and overt hostility than both normal and schizophrenic families. Schizophrenic families, as reported earlier, had the lowest weighted and overt hostility, and delinquent families had the highest weighted scores for hostility, however the level of overtness was close to that of normals. Becker and Iwakami (1969), and O'Connor and Stachowiak (1971), both reported that normal, or non-clinic families engaged in more simultaneous speech and interrupting behavior than did emotionally disturbed families. Becker and Iwakami interpreted these findings in the following manner: 88 The finding that non-clinic parents interacted somewhat more conflictually support the contention that the qualitative aspects of intrafamilial conflict may be more related to psychopathology than quantitative ones. (Becker and Iwakami, 1969, p. 334) O'Connor and Stachowiak (1971), in comparing the inter- action patterns of families of low adjusted, high adjusted, and retarded, children, in addition to reporting the con- flict findings mentioned above, also found that high adjusted families produced the greatest number of emotionally support- ive speeches, and had the greatest cohesion. Retarded families were second on both dimensions with the cohesion score being very near the high adjustment families. Low adjustment families were significantly lower on both dimension. In a descriptive, non statistical investigation, Odom, et a1. (1971), compared families with highly adjusted child- ren with family units having low adjusted children, and made the following observations about affective and maintenance behavior: The families in the low group seemed characterized by less emotional closeness. Rather than showing the affection and warmth that passed between the members of the high families, these families tended to be vigilant, hostile, intellectual and/or indifferent. (Odom, et al., 1971, p. 283) In addition, the authors also indicated that in affec- tive or maintenance behaviors, the mothers in highly adjusted families were "key figures," while in low adjusted families, maintenance functions were not assumed by the mother, but if they were assumed, they were taken over by other family members. 89 In a study which adds even more importance to the expressive role of the mother, Bugental, et a1. (1971), found that the mothers of emotionally disturbed children give, simultaneously, more conflicting evaluative messages in different communication channels than do mothers of normal children. And finally, in a study exploring the reciprocity of defensive and supportive communication behaviors between parents and aggressive sons (aggression not severe enough to be disruptive to the son's school environment), Alexander (1973a), found that these families tended to be reciprocal in supportiveness but not in defensive communication.1° Also, the author reported that the son's aggression in school seemed related to the mother's defensiveness.ll We can thus summarize our findings of emotionally disturbed families along the following lines: 1. there is logical and empirical evidence to support the role of the emotionally disturbed child as a buffer between parents who are unable to resolve significant relational issues between themselves, with the symptoms acting to unfocus latent rational issue(s); 2. the emotionally disturbed child tends to view and portray all human relationships, especially parent- child and parent-parent relationships as more distant and, in some cases, as more rejecting than does the normal child; 3. families with emotionally disturbed children, when compared with normal families, tend to be lower on cohesion and emotionally supportive messages, at the same time being lower on overt conflict and spontaneous interruption indices; 4. there is some evidence to suggest that the mother of the emotionally disturbed child does not assume a normal maintenance or emotional role in the family, when compared with the mother of the 90 normal family and, furthermore, that she is more likely to engage in affective double-binding communication or in behavior likely to precipitate defensive (aggressive responses). Discussion and Hypotheses: The review of the literature on the intra-familial milieu of the emotionally disturbed child's family would suggest that, although the amount of overt conflict between the parents of emotionally disturbed children tends to be low, the amount of covertly managed conflict tends to be relatively high. Thus, there is likely to be a constant undercurrent of tension which pervades the family's communi- cation climate, tension which may have debilitating conse- quences for weaker family members, or members who play a more central role in conflict management, as in the case of the family scapegoat. Although the actual source of this marital conflict may be focused on any relational issue, it seems likely in the case of the emotionally disturbed child's family, that it would originate from the affection dimension, more so than from any other relational area. One of the commonly held rules of thumb in psychiatric circles suggests that psych- iatric symptoms are always symbolic of their etiology, and as has often been demonstrated, the symptoms of emotionally disturbed children often relate to affection-inclusion themes. Thus it would seem plausible and is supported substan- tially by the literature that emotional disorders in children 91 are likely to be associated with affection issues either between the parents or between the parent and the child. By the same reasoning, family disturbances such as character disorders, e.g., delinquency, may be associated with control problems within the marital relationship, or within the parent-child relationship. Thus, the second sub-hypothesis: Normal dyads will be more homeomorphic in their behav- ioral affection exchange patterns and will report higher levels of perceived equity than will symptomatic child dyads. Thus, the only differences between conflict dyads and symptomatic child dyads is hypothesized to be the manner in which the perceived inequity is managed. In conflict dyads the exchange inequity which introduces imbalance into the family system is managed overtly with no symptom consolida- tion in the child. In symptomatic child dyads, the exchange inequity is hypothesized to be covertly managed, with the system imbalance leading to maladaptive behaviors in the child. Character Disorders and Familyglnteraction Patterns Character disorders refer to a wide range of relatively stable and pathological behavior patterns which precipitate, not from stressful situations, but rather as "developmental defects" in the personality structure. Included among character disorders are personality pattern disturbances, including inadequate, schizoid, cyclothymic and paranoid personality types; personality trait disturbances, including emotionally unstable, passive-aggressive and compulsive 92 personalities; and sociopathic personality disturbances, including antisocial, dyssocial, and sexually deviant person- ality types. (Coleman, 1964, pp. 352-419) Generally speaking, if transient situational personality disorders in childhood and early adolescence go untreated or unmodified, they generally evolve into the dominant response mode and become labeled character disorders. Thus, in addition to the two categories of disorders, i.e., personal- ity and character, being distinguishable by severity, they are also differentiated by history of the symptom and degree of disruption to the individual's social system. Unfortunately our knowledge of the etiologic relation- ship between family interaction patterns and the incidence of character disorders is fairly limited, restricted primar- ily to sociOpathic personality disorders, manifested by relatively severe, aggressive, acting out behaviors. They are most commonly manifest in childhood and adolescence by instances of delinquency, status offenses and other anti- social behavior. It is fairly well established that the home enviornment of the delinquent or antisocial personality is characterized by broken marriages and a high "turnover" in parental figures. For instance, Barker and Adams (1962), reported only one- third of delinquent children in a Colorado institution as coming from intact families. And Morris, et a1. (1956), indicated in their sample of delinquent children, a prepon- derance of multiple figures in the home lives of such children.12 93 While marital schism in character-disordered families is well-documented, the intrafamilial milieu has been less widely studied than that of either the schiZOphrenic or emotionally disturbed child. However, many of the same themes emerge in terms of affection and conflict. Andry (1962), reported that a majority of his sample of delinquent boys felt rejected by their father, but loved by their mother, in comparison with non-delinquents who felt equally loved by both parents. Along the same lines, a number of authors, who have studied the family relationships of the delinquent child, have commented on a high incidence of rejection in these families, coupled with inconsistent disciplining by both parents. Among these are Watson (1934), Hewitt and Jenkins (1946), Lewis (1954), Sears, et a1. (1953), Bandura and Walters (1959), McCord, et a1. (1961). Direct observation data on delinquent families is considerably more sparse. Alexander (1973b) in a study similar to the one reported in the previous section, found that "disintegrating" families (families with a runaway child) displayed reciprocation of defensive communication, but not of supportive communication. Normal families, by contrast, reciprocated supportive communcation but not defensive patterns. The author suggested the following insights concerning the relationship of the parents of the delinquent child: Normal parents, while behaving reciprocally with their children in supportiveness, did so less with each other. In contrast, abnormal parents did not behave 94 reciprocally with their children in supportiveness, but they were strongly reciprocal with each other. Numerous authors have commented on apparently the same phenom- enon; abonormal parents "use" the children to deal with their own relationship. . ., and in the process, do not deal with their own problems directly. Normal parents, on the other hand, do not "need" to use this indirect vehicle. (Alexander, 1973b, pp. 227-228) Hetherington, et a1. (1971), in comparing normal fam- ilies with those having either a neurotic, psychopathic or socially delinquent child, found no group differences in the expression of warmth on the part of mothers or fathers separately, but when combined, normal parents were found to express more warmth than neurotic delinquent parents (only in interaction with male children) or psychOpathic delinquent parents (only in interactions with female children). In a study comparing normal families with families possessing an aggressive, socially withdrawn, or poor atten- tion control child, Bugental, et al. (1972), compared the family types in terms of activity, controlling behavior and positive or negative evaluation. While no differences were found between the mothers of any family type (all mothers were found to be controlling, talkative and evaluatively extreme), the fathers did demonstrate differences. Fathers of normal children were found to talk infre- quently, were evaluatively neutral (in terms of affect) and also non-directing. By contrast, the fathers of aggressive children were evaluatively extreme and negative as well as being controlling. The fathers of socially withdrawn child- ren demonstrated a behavior pattern which was Opposite to 95 that of the fathers of aggressive children. They were non- controlling and evaluatively neutral. Fathers of distract- able children were found to be evaluatively extreme in terms of affective expression and very talkative. And finally, in a significant study, which grouped families in terms Of type and amount of maladaptive behavior, Riskin and Faunce (1970) assessed the communication corre- lates of each of four family types. The study, using the Family Interaction Scales, compared the interaction patterns of four types of families, differentiated in terms of prob- lems and diagnosis. One group consisted of multiple problem families, who had at least three labeled problems, including marital difficulty. A second group consisted of families who were described as constricted, with two or three problems which were diagnosed as psychoneurotic. The third group was composed of families where only the child was labeled as maladaptive, usually as an underachiever, or as being overly aggressive. (In the present investigation, this type of family will be termed a symptomatic child family.) The remaining two groups were classified as “normal families" with one group being viewed as somewhat more adaptive than the other. The interaction data derived from the multiple-problem families depict them as follows: They tended to disagree slightly more than they agreed. . . They tended to express slightly more affect outward than holding it inward. On relationship, there was, in general, an absence of support, especially from child to parent. And there was a very high amount of attack across all the who-to-whom categories. The 96 atmosphere, then, was distinctly the most unfriendly one. Their communication network was characterized by high child-to-child interaction with minimal parent-to- parent interaction. The child-to-parent interaction was also minimal. . . . Their many disagreements were not balanced by friendly statements, suggesting that the absence of support, concurrent with high disagree- ment could lead to quite explosive situations. (Riskin and Faunce, 1970, p. 532) The interaction recorded from constricted families provide a family profile which portrays them affectively as follows: In terms of intensity of affect, there is both a marked absence of enthusiasm and a strong presence of a depressive tone. The atmosphere is generally a somewhat unfriendly one. The parents are critical of the children, without balancing supportive comments.' The children are unfriendly with one another and with their parents. Quite unexpected was the finding that the parents are quite friendly to each other (as well as talking a great deal to each other). There is at least one scapegoat in each family who is very strongly attacked. . . There is a relatively large amount of attacking without much open disagreeing, which suggests a quality of covert bickering. (Riskin and Faunce, 1970, p. 533) In the third family type, the symptomatic child family, there exists . . . an argumentative, non-collaborative atmosphere, all rather low-keyed emotionally. One person is very stongly disagreed with, suggesting a possible scapegoat role. They are affectively constricted, with little intensity-up and much intensity-down, which suggests a depressive tinge. There is an absence of positive support. . . but also a striking absence of attacks. That is, the 'relationship' atmosphere is bland, neutral-- they do not give clear relationship messages. . . There is a great deal of disagreement without positive friendly statements to balance it. This supports the impression of low-keyed bickering, which could become explosive. (Riskin and Faunce, 1970, p. 534.) Perhaps the most significant characteristic of the symptomatic child family is the marked absence of affective exchanges of any qualitative nature. 97 In contrast to the other families, the normal families tended to . . . display a wide range of intensity of feelings, both up and down, with a tendency for outward-directed (intensity up) feelings to predominate. They have by far the friendliest atmosphere. There is much positive support, including parent-to-parent, parent-to-child, and child-to-parent. There is a relative absence of attack. . ., but they can show some anger. . . mostly between children and from children to parents. There is a relative absence of attack from parent to child. It would seem that the friendly atmOSphere is not simply a defense against angry feeling. (Riskin and Faunce, 1970, p. 535) We can summarize our findings of families with charac- ter-disordered children along the following lines: 1. there is considerable evidence to indicate a high incidence Of marital conflict and divorce among families with a child diagnosed as character- disordered. There is also some evidence to sup- port the role of the delinquent child as a scape- goat, in those families which are still intact; 2. there is considerable evidence to indicate that the parent-child relationships of the character- disordered family are characterized by parental rejection, particularly by the father, along with control behaviors which are both harsh and incon- sistent; 3. there is some evidence to suggest that character— disordered families tend to be disintegrating in that they reciprocate defensive communication. Normal families by contrast are more likely to be integrative, in that they express more warmth, reciprocate supportive communication, and do not perpetuate destructive cycles by reciprocating defensive communication; 4. the interaction of character-disordered families may, consequently, be viewed as arugmentative, non-cooperative, and as providing very little emotional support to family members. 98 Learning Disorders and Family Interaction Patterns Another relevant, but somewhat tangential area of symptom consolidation in children lies in psychologically based learning disorders. Because there tends to be a high degree of association between childhood personality dis- orders, adolescent character disorders, and low levels of scholastic achievement, this would seem to be a related and potentially fruitful area to be examined.13 There are some theorists who would suggest that the inability to achieve scholastically is very much related to a home environment characterized by tension and strain which is never resolved, i.e., homes where marital conflict is chronic and primarily covertly managed. This type of marital setting Wynn, et al. (1958) has labeled pseudomutuality and, as described previously, it is a relationship where the partners attempt to deny their differences and to a large extent their own separate individualities in order to pre- serve an image of tranquility and harmony. The association Of pseudomutuality to childhood learning disorders is aptly described by Charney (1969). Thus with learning disordered children, we often find that the family psychology is one in which family members have stOpped learning how to grow out of their differences, and conflicts, and that at least one important aspect of the child's inability to learn is that he is not free to experience aggressiveness, separateness, conflict and differences, and accordingly must give up being able to be a creative participant in the business of exPloring life. (Charney, 1969, p. 13) 99 Most of the studies which have sought to explore the relationship between family interaction patterns and schol- astic achievement or the need for achievement, have chosen students (mostly males) of high school age or Older as subjects, and have examined the mother-child relationship primarily in terms of nurturance-affection, and the indoc- trination of the mother's intellectual values. The findings are, at best, sketchy and are quite often conflicting. For example, McClelland, et a1. (1953), in samples of college age and high school males, found that the college age males with high needs for achievement rated both parents as more rejecting than male students with low needs fOr achievement. In the second study, the authors found just the opposite, that high school males with high needs for achievement rated their fathers as less rejecting than those students with low achievement needs. Crandall, et a1. (1960), working with pre-nursery school children, explored the relationship between a number of relevant achievement variables; amount of achievement effort demonstrated by the child in a free play situation, as well as nurturance, general affectionateness, and achieve- ment rewarding behaviors by the mother. The study found that mothers who rewarded achievement efforts were less nurturant toward their children (less accepting and rewarding of help seeking and emotional support seeking overtures) than mothers who did less rewarding of achievement efforts. Also, mothers who rewarded achievement efforts were found to 100 be somewhat less affectionate than mothers who did not reward such efforts. The results were interpreted to mean that mothers who discouraged emotional dependency were the most likely to encourage achievement efforts from their children. In a study which assessed the relationship between achievment motivation in women, the intellectual values of the father and the affection aspect of the daughter-father relationship, Grebow (1973), found that among college age women, whose fathers stressed academic achievement, and whose relationship with their father was described as warm, had higher academic values (though not a corresponding difference in performance) than those whose fathers placed less emphasis on intellectual values, and whose relationship was less intimate. Although it appears that the joint effect of a warm parent-child relationship, coupled with parental pressure for academic achievement, as suggested by a number of authors, i.e., Winterbottom (1958), Rosen and D'Andrade (1959), Crandall, et al. (1962), and Crandall, et al. (1964), is the best combination for creating achievement motivation in children, there is also evidence that an overly nurturant or affectionate parent-child relationship may be as counter- productive by fostering dependency, as a lack of support may be in spawning indifferences to academic pursuits. Since most of the above studies have hypothesized a relationship between affective states of family interaction, specifically those between parent and child, e.g., affection, 101 nurturance, rejection and either actual achievement or need for achievement, it would seem likely that the affective climate between the parents and within the family generally, would also affect learning variables. That is, it seems only logical that in schismatic families, which are openly conflictual, that the tension and anxiety associated with such families would function in reducing if not the need for achievement, then the actual ability to perform. By the same token, in those families where conflicts are managed covertly, where the tension is never really acknowledged or allowed to surface, or in those families who scapegoat one of their children and thus maintain homeostasis by this means, it would seem probable that the child's ability to focus on scholastic achievement would be reduced. However, because family studies have generally investigated the more socially disruptive symptoms in children, e.g., schizo- phrenia, delinquency and emotional disturbance, it is not surprising that the literature is lacking in this area. Consequently, this area may be next in line to be scrutinized by family researchers, hopefully with more improved method- ologies developed in the other areas. We can summarize our rather sketchy findings of psycho- logical or family based learning disorders along the follow- ing lines: 1 1. there is some speculation which suggests that learning disabilities which are pschologically or family based are associated with family interac- tion patterns where parental conflict is covertly managed, and where individuality is sacrificed for dyadic stability and the illusion of harmony. 102 2. there is some evidence to suggest that the parents of children who achieve scholastically tend to encourage independence and to discourage emotional dependency on the part of their children by not rewarding help-seeking and support-seeking behaviors. 3. there is some evidence to indicate that a warm parent-child relationship, coupled with parental pressure for achievement, is likely to produce if not high academic performance, at least high intellectual values.‘“ The present chapter has examined: (1) the affection exchange process in marriage and its relationship to marital satisfaction and stability, (2) the marital relationships of parents of four types of symptomatic children for evidence of overt or covert conflict, which may be directly or indir- ectly associated with marital affection issues, and (3) the parent-child relationship for anomalies in affection exchange which might be directly or indirectly associated with an unsatisfying marital affection exchange relationship. Fundamentally what can be concluded following this review of relational interaction variables, specifically affection, and its association with marital and family functioning?15 The most relevant issue in attempting to interpret the substantive findings of the present chapter is what kind of model(s) fits the data. The literature has supported the fact that stable and satisfying marriages do differ from other types of marital relationships. By the same token it has been demonstrated that normal families and abnormal families are different in terms of the quantitative and qualitative interaction parameters which characterize their relationships. However, the major problem facing the re- searcher attempting to eXplain these differences is what 103 model is most appropriate: etiologic, responsive, situa- tional, a combination, or all three. In analyzing much of the research literature, although we would like to be able to use a causal model, many of the findings can be explained as validly from a reSponsive stance, e.g., specific marital and family affection exchange patterns arising as a reSponse to the onset of schizophrenia in a family member. Likewise, many of the differences in symptomatic or delinquent family affection patterns can be explained quite logically with a situational model, e.g., the differences in affection exchange resulting from the family having to deal with a clinic or family court system which has some influence over them and thereby alters their normal interaction patterns. In very few investigations is there some attempt made to establish time-order which would enable us to infer from an etiological model, e.g., an extremely high incidence of schizophrenia associated with maternal depression resulting in affection deprivation to the child in the critical first year of the child's life. However, even here the hypothe- sized relationship is still open to charges of spuriousness, because of any number of uncontrolled, extraneous variables. This issue will be discussed further in the next chapter on methodology and limitations of the present investigation. FOOTNOTES 1. These issues will be discussed further in Chapter 3 under the heading of marital satisfaction as a two- dimensional variable. 2. This theoretical perspective is similar to Kurt Lewin's force field theory of personality, applied to the dyadic context. 3. Goode's description of the empty shell marriage will become significant later on in the chapter when the author reviews pathological family types for this depiction mirrors rather precisely the findings of many family researchers. 4. Refers to using words which have a private meaning for the spouses. 5. Chapter 3 will suggest that perceived equity is an antecedent and necessary yet perhaps not sufficient condition for the existence of satisfaction with the affection exchange relationship. 6. The dimension of control is included because of the increasing realization of its inseparability from the affection parameter. 7. Other than that which is diagnosed as organically based hyperkinesis. 8. The point that is being made here is that the particular diagnostic label which is applied depends not only on the symptoms, history of the symptoms, and often age of the child, but also on who is doing the labeling and what that person views as the dominant symptom (from his particu- lar perspective) at the time of the evaluation. For example, the author once viewed a situation where a field psychiatrist diagnosed a 17 year old enlisted man as suffering acute stress reaction (a situational personality disorder), placing him on the psychiatric ward. He was in turn seen by the hospital psychiatrist, and later by a clinical psychologist (Ph.D.) for testing. The psychiatrist evaluated the man, diagnosed him as passive-aggressive, and recommended "return to duty." The clinician diagnosed him as paranoid schizo- phrenic and recommended separation from the service. 104 105 9. Previous research, Kuethe (1962a, 1962b, 1964), Kuethe and Stricker (1963), and Kuethe and Weingartner (1964), support the assertion that normals place child figures closer to mother than father figures, and that human figures are kept together, unseparated by non-human objects. 10. The terms defensive and supportive communication are used according to the conceptualization of Gibb (1961). Defensive communications are those which are threatening or punishing to others, e.g., behaviors that are evaluative, controlling, involve strategies, indifference, etc. Support- ive communications are those which lead to lowered anxiety, clearer communications, and are exemplified by behaviors which seek information, are spontaneous, empathetic, problem oriented, etc. 11. This finding will have greater significance and relevance when compared with supportive and defensive recip- rocity patterns in families having a more severly aggressive, i.e., character-disordered child. Comparisons of Alexander's findings will be made in the next section on character disorders. 12. In support of these findings, the author was forced to discontinue data collection (for this particular study) among delinquent child dyads, who were referrals to the Stark County Family Court System. After contacting nearly 75 families (parents of children referred to family court must be interviewed by court appointed probation officers) it was found that over half were single parent families (divorced or unmarried), with another large segment being separated. The remainder either refused to come in, refused to OOOperate even when they did come (even under the implied pressure of the study being endorsed by the court system, which had some means of control over their children) or were unable to read well enough to complete the questionnaire. 13. In personal telephone conversations with Dr. Mario Pascal, Director, Division of Teacher Training, Marianne Frostig Center of Education Therapy, Los Angeles, California, it was suggested that the lack of literature in this area is due to the fact that most professionals and researchers approach the problem from a medical perspective, and have been concerned primarily with organic causes. Dr. Pascal did suggest that what available literature there was on psychological based learning disorders would probably be found under the heading of emotionally disturbed children. This is consistent with the findings of this researcher who found that some of the same writers who research emotional disturbances also concern themselves with learning disabili- ties, e.g., Wynn (1958) and Charney (1969). 106 14. The apparent conflict in 2 and 3 might be inter— preted to suggest that parents of children with high needs for achievement, make clear distinctions between family maintenance functions, i.e., affectionate and supportive behaviors and individual task functions, i.e., academic achievement and therefore do not allow the child to confuse the two. 15. One of the more interesting conclusions which can be drawn from the present chapter's review of marital and family interaction parameters is the inseparability of affection and control. Throughout the review, these two appear to be inextricably woven together as in such family descriptions as conditionally loving, possessive, rejecting, etc. To consider one parameter singularly without studying its interaction with the other is to artificially restrict our view of marital and family communication and functioning. CHAPTER III PROCEDURES AND DESIGN Marital Agreement and Accuracygin Affection Exchange The major scientific question asked by the present investigation is: what ig the relationship between the form g: affection exchange and the perceived equity pf the ex- change within marital relationships? In response to this question the following hypothesis has been advanced: in marital dyads, the more homeomorphic the affection exchange patterns, the greater the perceived equity of the exchange relationship. Homeomorphic exchange will be measured by the amount of relative agreement which the partners report in their affec- tion exchange repertoires. This will be indexed by their responses to 24 items describing various ways by which husbands and wives can or do show affection and caring for one another, along with a measure of the relative importance of each of those means to them. Independent measures of the wife's perceived outputs will be compared with the husband's perceived outputs to give an overall dyadic agreement score. In instances where agreement is high there is perhaps less need for accuracy in communicating about affection exchange, as individuals are likely to infer similar meanings and values from manifest behaviors. However, where agreement is low, where the affection exchange is heteromorphic, then 1017 108 it would seem important, if not perhaps necessary that the dyad discuss the differences in affection repertoires and agree on issues of meaning and value connected to divergent behavioral patterns. In effect it would seem necessary for marital partners to have a high degree of accuracy in being able to recognize affection responses of their spouse and to assess the value and relational significance of those behav- iors. Accuracy is thus defined as the extent to which B's affection repertoire and the relative importance of each behavior in it is similar to A's prediction of B's affection repertoire. Therefore, based upon the above rationale the following hypothesis is advanced: Under conditions of heteromorphic affection exchange, the higher the accuracy, the higher the perceived equity. Thus while high agreement (homeomorphic affection exchange) is hypothesized to be associated with the absence of marital discord and symptom consolidation in children, low agreement (heteromorphic affection exchange) is hypo- thesized to be related to the presence of marital discord and/or symptomatic children, especially in those instances where accuracy is low. Based on our previous review of the marital relation- ships of the parents of emotionally disturbed children, specifically their inability to deal overtly with their differences and sources of tension, the following hypothesis is advanced: 109 Normal dyads will have higher accuracy than either conflict or symptomatic child dyads with symptomatic child dyads having the lowest level of accuracy. Perceived Equity There are two major relational system states which will be considered: perceived equity (reciprocal balance) and relationship satisfaction. Perceived equity in affection exchange is defined as the extent to which each spouse feels that what he has given (in investments), approaches in value what he has received (in profits). To the extent to which one of the spouses feels that he has given more than he has received, or received more than he has given, then the relationship is defined as inequitable or reciprocally imbalanced. A marital relationship may reach the threshold of imbalance without the individual(s) attempting to restore the balance either by altering their role or the role of other in the relationship. In likelihood, all marriage relationships go through periods when they are reciprocally imbalanced, due to illness, time demands on one of the members, because of vocational pressures or for other reasons. It is only when the imbalance reaches prOportions sufficient to cross a second threshold, that of felt injustice, that the individual is likely to move to redress the inequity through any number of the means discussed previously. Eventually if the inequity continues and ongoing attempts to redress are unsuccessful, the situation may culminate in reaching the third threshOld--dissolution, with the marital 110 relationship ending in separation or divorce. It is inequity at these latter levels that the present study seeks to compare with heteromorphic affection exchange patterns. (These thresholds and various levels of balance and imbalance are depicted in Figure 3.) In the present investigation equity will be measured by the spouses' responses to three Likert type items which ask about the equity of the marital exchange. Relationship Satisfaction Perceived equity may be a necessary, yet insufficient condition for the occurrence of another important relatiOnal state, that of relationship satisfaction. That is, the perception of equity in affection exchange, may not be sufficient in and of itself to produce satisfaction with the relationship. Actual satisfaction may be associated more with the mode of exchange than the perceived equity of it. Relationship satisfaction may be defined as the extent to which one's expectations (eXpected outcomes) are congruent with one's actual outcomes (realized outcomes). However, in addition to satisfaction level being influenced by one's expectations and one's realized outcomes, it may also be influenced by the actual forms of affection which are ex- changed. Some forms of affection exchange may be more readily associated with actual relational satisfaction than are other forms of exchange, mediated, of course, by individ- ual expectations. For example, it is hypothesized that intimate forms of affection exchange may be more readily 111 Redress of Injustice Dissolution of the Relationship Reciprocally Reciprocally Point of Maximum Perceived Equity Balanced Threshold of Felt Injustice Threshold of Perceived Imbalanced Inequity Figure 3. Levels of Equity and Inequity 112 associated with actual relationship satisfaction than are economic forms of affection exchange. There is considerable empirical evidence to indicate that intimate forms of affection exchange are prime sources of psychological and self validation which is so necessary for emotional and mental health. In the literature on self- disclosure, one of the most intimate forms of affection exchange, there is evidence with suggests that marital satisfaction is associated with the occurrence of self- disclosure between spouses. For instance, Levinger (1963) found that highly satisfied dyads exceeded less satisfied dyads in their frequency of discussing topics with one another. Katz et al. (1963) found that the disclosure of personal information was associated with marital satisfaction for wives, but not for husbands. Levinger and Senn (1967) found a positive correlation between both perceived favor- ability of marital communication (a specific measure of marital satisfaction), overall marital satisfaction (a general measure of marital satisfaction) and the disclosure of one's feelings to the marital partner. Finally, Worthy, Gary and Kahn (1966) have suggested that reception of self-disclosing information from another is rewarding, since reception of such information implies that one is trusted and furthermore that the more intimate the information, the greater the validation for the receiver. In addition, there is also considerable evidence to suggest that physical touch behavior, another intimate form of affection exchange, is psychologically validating and 113 necessary for maintaining natural development and health. In support of this position is the work of Harlow (1963) who found that physical touch deprivation in baby monkeys is detrimental to the development of normal adult relations and the ability to nurture offspring. The generalization to human infants has been supported by considerable evidence suggesting that institutionalized children, who are deprived of physical stimulation suffer arrested physical and mental development and have higher incidences of unexplained death. This condition called marasmus or maternal deprivation syndrome is described by Coleman (1964) as: "a condition in which, despite adequate physical care, infants seem to 'waste away' for lack Of warm, personal contact and the meeting of other psychological needs" (Coleman, 1964, p. 130). The existence of maternal deprivation syndrome and its association with the absence of physical touch stimuli is well documented and supported by the work of Spitz (1945), Bakwin (1949), Bowlby (1951), Yarrow (1961) and Provence and Lipton (1062). Stimulus deprivation in children appears to deter the normal develOpment of several processes. Ainsworth (1962) has suggested that particularly vulnerable are intel- lectual develOpment including language develOpment and the ability to abstract, emotional develOpment including the ability to form warm affectional ties, and control processes, including the ability to control inner impulses and to delay gratification. Although Gray (1958) has suggested that vulnerability in children seems to be maximum between three 114 and six months to seven years, the potential effects appear to be much more long standing. For example, Beres and Obers (1950) in a sample of 38 adolescents who had previously been institutionalized between the ages of three weeks and three years, found that four were psychotic, 21 were character disordered, four were mentally retarded and two were psycho- neurotic at the time of the survey.) Although the effects of physical touch deprivation may be the most pronounced as the result of their absence in childhood, it would only seem logical that individuals would continue to need physical touch stimuli in order to ensure healthy psychological and social functioning as adults. Along these lines Frank (1957) suggests that all human relationships are learned through physical interaction in such behaviors as wrestling, holding, kissing, and hugging and that failure to experience such contact impairs one's ability to establish normal relationships and receive subse- quent support. Montague (1971) has found that some individ- uals rely heavily on physical touch behavior as a source of affection meaning and suggests that when deprived of such contact, suffer deleterious effects. And finally, Bardeen (1971) has found that some individuals actually prefer touching behavior over other forms of communication, namely verbal and visual forms. Thus, there is considerable evidence to support the hypothesis that the intimacy mode of affection exchange is 115 potentially the most validating and rewarding interpersonally and the most crucial for normal individual functioning. Relationship Satisfaction: A Two Dimensional Variable Building on the work of Herzberg (1959), relationship satisfaction, as he suggested with job satisfaction, may not be a unidimensional variable. Relationship satisfaction may not be the Opposite of relationship gigsatisfaction. The opposite of relationship satisfaction would more likely seem to be the absence of satisfaction, while the opposite of relationship dissatisfaction may well be the absence of dissatisfaction. The relationships among these four states are illustrated in Figure 4. Consistent with a two-dimensional view of relationship satisfaction, those behaviors involved in producing relation- ship satisfaction may be distinct from those producing relationship dissatisfaction. Extrinsic forms of affection exchange, such as those manifested in the economic forms of affection exchange, may not be interpreted as meeting self- concept needs, but rather fulfilling only physical needs and therefore largely reducing physiological deprivation. If, for example, the individual expects intimacy, but receives economics, then this condition is likely to precipitate diffuse, unanchored feelings of dissatisfaction or more likely to produce the absence gt satisfaction, in that he; has still received a concrete, quantifiable commodity which is implicitly presented as valuable, while his real (and 116 often largely unconscious) desire was the reception of intimate forms of affection, which are admittedly less concrete, less easily identified, evaluated and more diffi- cult tO quantify. However, even if the individual expects economics and that expectation is fulfilled, that exchange may still prove to be inadequate for self-concept validation and the perception of that relationship as a satisfying source of identity support. This condition is likely to produce the absence pf dissatisfaction as an evaluation of the relationship. E; F‘ o '8 ’6‘ :3 Relationship Absence of ,fl 4.) m . . . . o 3 Satisfaction Satisfaction :2 75 (U A m Jfg El .2 " . . ma 'ggj Absence of Relationship m -a (I) . u ‘H o o o o I o '81; Dissatisfaction Dissatisfaction g 216 C.m _g 9. Figure 4. Relationship Satisfaction as a Two-Dimensional Variable On the other hand, intrinsic forms of affection ex- change, such as those embodied in the more intimate forms of affection exchange, may be more readily identified as di- rectly providing identity support, meeting self-concept needs and, therefore, reducing psycholgoical deprivation. Thus, this means may be more readily associated with produc- ing feelings of relationship satisfaction, especially if one's relational expectations are for intimate forms of 117 affection exchange. However, if the individual's expecta- tions (expected outcomes) are for highly economic forms of affection exchange, but he receives relatively more intimacy, this situation is likely to precipitate feelings of dissatis- faction that are likely to be readily connected with the relational exchange. This is because the absence of these concrete and quantifiable economic commodities is easily recognized and at a conscious level of recognition. Thus, unfulfilled expectations in economics may be more easily identified and connected to the relational transaction itself than are unfulfilled expectations in intimacy. We might then expect the marital partner whose outcomes are in economics to manifest more overt forms of dissatisfaction, when his expectations are violated, while we might conversely expect unfulfilled expectations in intimacy to be associated with vague, diffuse and unanchored feelings which closely approach a state of absence gt relational satisfaction. The hypothesized relationship between expected outcomes, realized outcome, and satisfaction with the relationship are illustrated in Figure 5 below. Expected Outcomes Intimacy Economics Intimacy Satisfaction Dissatisfaction Realized Outcomes Absence of Absence of Economics Satisfaction Dissatisfaction Figure 5. Hypothesized Satisfaction Levels as a Function of Realized and Expected Outcomes 118 The absence gf_satisfaction and the absence gt dissatis- faction as individual states do not initially appear to be conceptually distinct. However, it is believed that rela- tively important, although subtle differences in the two do exist and indeed operate in the assessment of satisfaction derived from relational exhanges. The differences are primarily ones of emphasis as a result of: (l) the expecta- tions of the individuals, (2) the relative tangibility or intangibility of the two forms of affection exchange, and (3) the level of need fulfillment each form of exchange is capable of providing. In the case of absence g£_dissatisfaction, one's expec- tations have been fulfilled, given that the individual expects economics and receives economics in the exchange. Therefore, logically, he should experience the feeling of satisfaction, given that his expected outcomes are congruent with his realized outcomes. However, if our assumption of self validation being provided by intimate exchange, and physical support, being provided by economic exchange is valid, then this kind of exchange may not produce the ulti- mate relational outcome at all, i.e., identify support and subsequent satisfaction with the relationships as a source of self-concept validation. In this instance, it may merely preclude the occurrence of a negative instance, namely physical deprivation and subsequent dissatisfaction with the relationship as a source of positive outcomes. In this instance the individual has succeeded in avoiding the nega- tive occurrence and should, therefore, feel that he has 119 successfully "avoided" rather than actually "gained." Therefore, the emotional feeling that is likely to be exper- ienced in terms of evaluating the relationship, should more closely approximate the absence gt dissatisfaction than anything else. In the case of absence g: satisfaction, the individual's expectations for intimacy have not been fulfilled, which should preclude him from feeling satisfied with the exchange. However, in that he has actually gained economic commodities, consensually validated as valuable, this should also preclude his being dissatisfied with the relationship as a source of positive outcomes. In a culture which places high value on economic commodities, there is extreme social pressure toward this interpretation and assessment of economic ex- changes. In effect, he has gained an undesired goal, which may not validate the self, and which should produce the relational feeling most closely approximating the absence g; satisfaction. The only available literature which would support this view of relationship satisfaction is that provided by Renne (1970) in her concept of the "adequate marriage," defined as "not necessarily 'happy' but certainly not 'unhappy'" (Renne, 1970, p. 54). This is a level of relationship satisfaction which the present study would label as absence gf dissatis- faction (the second highest level Of satisfaction) and hypothesize as resulting from fulfilled expectations in economics. 120 In the present investigation, expected outcomes will be indexed by responses to 24 affection expectation items. Realized outcomes will be measured by the same 24 items posed in terms of how frequently one's spouse performs each of these behaviors. The relationship of expected outcomes to realized outcomes will then be compared to the level of relationship satisfaction experienced by that individual. Level of relationship satisfaction will be measured by each spouse's repsonse to the question which asks them to select from four levels of satisfaction the one which best describes their feelings about their marriage relationship. Based on the above discussion the following hypothesis is advanced: A relationship between expected outcomes, realized outcomes and relational satisfaction is hypothesized such that: a. fulfilled expectations in intimacy will produce the greatest degree of relationship satisfaction; b. fulfilled expectations in economics will produce the second highest level of relationship satisfac- tion, i.e., the absence of disstisfaction. c. unfulfilled expectations in intimacy will produce the third highest level of relationship satisfac- tion, i.e., the absence of satisfaction. d. unfulfilled expectations in economics will produce the lowest level of relationship satisfaction, i.e., actual dissatisfaction. Based on a review of the available literature, the following additional variables would appear to be associated with the affection exchange process in marital dyads: history of the relationship, sex, and socio-economic status. 121 History of the Dyad, Satisfaction with Marriage, and Affection Exchange There is considerable evidence to indicate that women experience more dissatisfaction in marriage than do men. In support of this position are the findings of Hamilton (1929), Bossard and B011 (1955), Blood and Wolfe (1960), Paris and Luckey (1966) and Marlowe (1968). In addition there is also evidence to suggest that women are more likely to attribute the cause of their dissatisfaction to their spouse than are men. In support of this finding is the work of Katz, et al. (1963) and Gurin, et a1. (1960). Gurin offers an explanation for this phenomenon based on the differences in instrumental and expressive roles assumed by the spouses: If we think of the usual male-female role distinctions for marriage--the man as initiator and the woman as supporter--it is, in a way, "easier" to think of prob- lems occurring in the active as opposed to the passive role. It is easier to pinpoint a man's lack of material support than the more tenuous emotional support that is a major contribution of the woman to marriage (Gurin, 1960, PP. 111-112). This position is consistent with the previous discus- sion of relationship satisfaction which suggested that satisfaction in marriage is more likely to result from intimate forms of affection exchange which are much more a part of the expressive role of women, than to accrue from economic exchanges which are more central to the instrumental role of men. Men have also been found to experience a decrease in marital satisfaction with the passage of time. This position 122 is supported by Burgess and Wallin (1953) and Pineo (1961) who suggested that this so-called "disenchantment" actually occurs earlier for men than it does for women. Aside from the question of which spouse experiences the greatest dissatisfaction with marriage, for which there are somewhat conflicting findings, perhaps the most relevant question is how does relationship satisfaction change over the life cycle of a marriage? Lang (1932) found a gradual decline over the 16-year period which he studied. Bossard and Boll (1955) found that the women's age group 20-29 years was relatively happier than their older counterparts. Blood _and Wolfe (1960) found in their sample of 900 women, a continual decline in marital satisfaction until the children were launched and then a slight rise followed by a decline in the reitered group. Pineo (1961) reported a general decline in marital satisfaction and adjustment with, inter- estingly enough, one of the areas of greatest decline being in demonstrative affection. Gurin, et a1. (1960) found a curvilinear trend with decreasing marital satisfaction during the early stages, a leveling off and an increase during later stages. Mathews and Milhanovich (1963) found no tendency for the number of marital problems to decrease with the length of time married. Paris and Luckey (1966) in studying satisfied and unsatisfied couples found that satis- fied couples tended to decrease and unsatisfied couples to increase in satisfaction over time; but that in general the trend was to decrease. Luckey (1966) found that length of 123 marriage is accompanied by an increase in unfavorable percep- tions of the spouse, regardless of whether the couple is happily or unhappily married. Rollins and Feldman (1970), studying approximately 850 couples, found a sharp decline in satisfaction from the beginning of marriage to the preschool stage, followed by a leveling Off to the "empty nest" stage and then a rapid increase to the retired stage. And finally, Burr (1970), reported a decrease for women and an increase for men in satisfaction during the early stages of marriage. This was followed by a discernable decline into the school- age stage of marriage. There was then a gradual increase as the oldest child moved through adolescence, with continued until the retirement stage. In terms of which stage is the most difficult and is characterized by the least amount of satisfaction, the research consensus is nearly unanimous. Bernard (1934), Bossard and Boll (1955), Deutscher (1959), Blood and Wolfe (1960), and Rollins and Feldman (1970), all suggest that the launching of the children or the period immediately prior to launching is the most troublesome. Terman (1938) suggests that preschool is the most difficult while Burr (1970) contends that school-age is the most difficult. In summary, it is perhaps safe to conclude that marital satisfaction decreases rather steadily after the first year or two years of marriage, levels off during the middle or school-age years, bottoms out during the launching period, and then recovers slightly through the retirement stage. 124 This pattern is similar for both men and women, but is perhaps more marked for the latter. The question which the literature raises is, since affection exchange is such an important factor in a marital relationship, how does the affection exchange process covary with these apparent shifts in relationship satisfaction during the marriage life cycle? Based on the previous discussion of relationship satisfaction as a two-dimensional variable and the above research, the following hypotheses are advanced: Short history dyads will utilize economic forms of. affection exchange more than long history dyads, with medium history dyads utilizing the least amount of economics. Short history dyads will utilize intimate forms of affection exchange more than long history dyads, with medium history dyads utilizing the least amount of intimacy. Short history dyads will be operationally defined aS‘ couples who have been married one to eight years, medium history dyads as couples who have been married more than eight years but less than 17 years, and long history dyads as couples who have been married 18 years or longer. Forms of affection exchange utilized by each dyad will be indexed by summing the perceived importance of the 12 intimacy and 12 economic items into a dyadic measure for each affection mode. F" "‘- 125 Male-Female Differences In Affection Exchange Building most fundamentally on the instrumental- expressive dichotomy first discussed by Parsons and Bales (1955), we would expect males to utilize, more heavily, economic forms of affection in their marital exchanges, while females might be expected to utilize a greater rela- tive amount of intimacy. In addition there is considerable empirical evidence to indiCate that differences do exist between the sexes and that these differences are largely attributable to developmental processes occurring in the family. In regard to the intimate form of affection exchange-- physical touch, Clay (1966) found that female babies receive more demonstrative acts of affection than do male babies. Goldberg and Lewis (1969) found that six-month-old girls are already being touched more often by their mothers than are infant boys. And Hollander, et al. (1973), in attempting to explain variability in interpersonal distancing, found that the more maternal affection the individual received, the closer he permitted persons to approach. Together these findings would suggest that adults respond differently to male and female infants in terms of touch behavior and as a result perhaps predispose males to interpersonal repertoires containing lesser degrees of physical touch intimacy. These conclusions are also consistent with Jourard's (1966, 1968) oft-cited findings that females touch and are touched more 126 than males, and that friends of the Opposite sex and mothers do most of the touching, while fathers limit most of their touch to the hands of others. In regard to self-disclosure, another intimate form Of affection exchange, Komarovsky (1964) found that wives tend to be more self-disclosing than husbands. Katz, et a1. (1963) reported that wives revealed their anxieties much more frequently than did husbands in return. In addition, Levinger and Senn (1967), in a study of conflict and non- conflict dyads found partial support for their hypotheses that wives would be higher than husbands in the proportion of feelings which they disclose to their spouses. Gurin (1960), Levinger and Senn (1967), and Raush, et a1. (1974), all found that wives are much more inclined to disclose their marital disappointments and to verbalize their inter- personal hostilities than are their husbands. And finally, Jourard (1961), Jourard and Landsman (1960), Jourard and Lasakow (1958), all found that men reveal less personal information about themselves than do women. Based on these findings, the following hypotheses are advanced: Males will utilize economic forms of affection exchange to a greater extent than females (and) Females will utilize intimate forms of affection ex- change to a greater extent than males. Forms of affection exchange for both male and female will again be indexed by a measure Of each Spouse's response 127 to the 24 affection items in the context of, "Do you use it to show affection and caring and, if so, how important is it to you?" Socio-Economic Status and Affection Exchange There is some evidence to indicate that higher educated couples are more inclined toward companionship marriages and to attribute greater significance to the "intrinsic" aspects of marriage than are couples with less education. Renne (1970) and Gurin, et a1. (1960) also report that the better educated are more likely to experience greater relationship satisfaction. As Gurin indicates: . . . the findings with respect to education suggest that the more educated respondents are more sensitive to both the positive and negative aspects of the mar- riage relationship. Their greater sensitivity to the positive aspects is reflected in the fact that expressed marital happiness increases progressively with increas- ing education. This finding is undoubtedly a reflection of many factors: The increased involvement in the more "intrinsic" interpersonal aspects of marriage in the more educated groups; the lesser economic strains on the marriage that one might expect in these groups; perhaps a greater concern with "happiness" as an ideal in marriage (Gurin, 1960, p. 114). Along these same lines Levinger (1966) found that spouses in middle class marriages were more concerned with psychological and emotional interaction, especially lack of "love," while lower class spouses were more inclined to report financial issues and unpredictable behavior from one's mate as more important. With a heavier reliance on the relationship aspects of marriage among higher SES couples there should be an increased 128 emphasis on all aspects of affection exchange though perhaps a somewhat greater emphasis on intimate forms. Based on the above rationale, the following hypotheses are advanced: Dyads with higher SES will utilize economic forms of affection exchange more than dyads with middle SES, with dyads lower in SES hypothesized to use the lowest amount of economic affection exchange. Dyads with higher SES will utilize intimate forms of affection exchange more than dyads with middle SES, with dyads lower in SES hypothesized to use the lowest amount of intimate affection exchange. In the present investigation, SES will be measured by a modified form of Hollingshead's (1957, 1958) SES indices. Perhaps the most widely used measure of socio-economic status, the Hollingshead model combines measures of occupa- tion, education and housing. The present investigation will utilize only occupation and education, which will give a range from two (executives with graduate degrees) to four- teen (unskilled workers with less than seven years of school). Forms of affection exchange utilized by each dyad will be indexed by the same dyadic measure used in the analysis of affection exchange and dyadic history, i.e., perceived importance across all 12 items for each affection mode. Sample Populations The samples selected for testing the study's hypotheses consisted of three groups of 40 marital dyads each. Group 1 consisted of an accidental sample of self-identified con- flict dyads, who were self referrals to social agencies for 129 purposes of marital counseling. In these dyads the presence of marital conflict, discord or some level of relationship dissatisfaction was overt and mutually recognized. Criteria for inclusion into this sample population were: Caucasian, at least one child between the ages of 1 and 18 years, marriage intact (spouses living together) and having been married for at least one year.1 Group two consisted of an accidental sample of symptom- atic child dyads whose children were referrals to a childhood and adolescent treatment center for treatment of emotional, behavioral or learning disorders. As indicated previously this sample of dyads was included in the present investiga- tion partially because it was thought that the parents of the emotionally disturbed child were the best source of data on the stable-unsatisfactory marriage relationship. Criteria for inclusion in this sample population were: Caucasian, the couple must have at least one emotionally disturbed child (in the age range of pre-school to 18 years, with a range of symptomology to include either: adjustment prob- lems, mild-moderate acting out behavior, emotional instab- ility, or hyperactivity which is not of organic origin), the marriage must be intact and the spouses must have been married for at least one year. Group 3 consisted of a random sample of normal marital dyads selected from the area telephone directory. Criteria for inclusion in this sample pOpulation were: Caucasian, at least one child between the ages of l and 18 years, the dyad 130 must ggt have a child who had been seen for behavioral, emotional or adjustment problems during the past year, the dyad must ggt have sought the consultation of a marriage counselor or other professsional for marital difficulties during the past year, the marriage must be intact and the couple must have been married for at least one year. All sample populations consisted of marital dyads who were residents of the Massillon-Canton area (western Stark County) of Ohio. Marital conflict dyads were provided by the following public and private agencies: Catholic Com- munity League, Canton, Ohio; Family Service, Canton, Ohio; Massillon Community Mental Health Center, Massillon, Ohio; and Psychological Services, North Canton, Ohio. Symptomatic Child Dyads were provided by Children's Services of Canton, Ohio. Limitations of the Investigation Undoubtedly the most limiting feature of the present investigation is its ex post facto design. Without the ability to establish time-order, an investigator's ability to infer a causal relationship between affection exchange patterns and various dyadic system states is seriously attenuated. Ideally a longitudinal approach would enable one to measure the hypothesized antecedent conditions and given those, predict future dyadic and relational events, which when measured at a future point in time, would then allow for the drawing of causal inferences. 131 The present investigation falls far short of that goal in that data collection was restricted to a one time, cross-sectional survey. However, the establishment of covariation is still very much within the realm of the present investigation and if supported must be considered a healthy stride toward our understanding of the dynamics of affection exchange and its association with marital satisfac- tion and stability. A second limitation lies in the ability to generalize the findings past the sample populations surveyed. Although the normal dyads were drawn randomly from the telephone directory, only 19 percent of these participated in the study. A considerable number did not meet the necessary criteria; however, an equally large number simply refused to participate. This self selection process consequently must be viewed as introducing an unknown bias into the data. The accidental samples of symptomatic child dyads and conflict dyads, by virtue of their voluntary self-referral status, may also be questionable as to their representativeness. One final point deserves consideration. Data collection for the conflict dyads and symptomatic child dyads was conducted inside participating public and private social agencies by individual case workers. Data collection for the sample of normal dyads was conducted in the family's own home. Although there are few instructions needed to be given by the experimenter, one cannot dismiss altogether the effects of differential settings and rapport establishment. 132 It seems likely that responding to a questionnaire in one's home with familiar surroundings is more likely to generate valid responses than an administration in an outside agency where there is typically some social influence exerted on the client. However, the exact effects cannot be predicted aside from the conclusion that some bias is likely to have been introduced. Validity of the Instrument Perhaps the most important issue, especially in develop- ing items to index such a social scientific concept as dyadic affection exchange, where little conceptual as well as empirical investigation has been done, must be content valida- tion. Essentially this issue focuses on whether or not the questionnaire items measure affection/caring behaviors which are characteristic of dyadic marital exchanges. Although the present conceptualization gleans some support from such social exchange theorists as Foa and Foa (1971) and Altman and Taylor (1973) with their respective emphasis on exchange resources and non verbal communicative behavior, there is a real paucity of available instruments from which one might draw valid indices from the spectrum of affection behavior. The present version of the questionnaire with its 12 intimacy items and 12 economic items was based on the concept- ualization of affection/caring described in chapter 1. Evidence of the relevance of the items derived from that conceptualization, to the dyadic exchange process was provided by this initial 133 investigation. In fact there is considerable data to indi- cate that the majority of items are extremely relevant to marital affection exchange as demonstrated by participant responses to whether or not the behaviors are used to show affection and caring in respondent marriages. The results of the tabulation are summed below in Table 4. The percentages indicating relevance are high across both affection modes and across most forms within each mode. The only systematic exceptions are the two verbal aggression items (no. 7 and 9) where 28% and 23% of respondents, respectively, indicated that these were affection exchange behaviors used in their marriage. These items were included however, not because they were believed to be universally used by all couples, as were for example the sexuality items, but because of their potential relevance in two specific family contexts -- conflict and symptomatic-child dyads. A much related issue is that of construct validation. One of the major questions which the study sought to answer was what are the relevant dimensions of dyadic affection exchange as indexed by the present instrument? What percentage of variance can be explained by the conceptual categories of intimacy and economics? Are these relevant constructs in affection exchange? Are they the only relevant constructs? Or are there others? In seeking answers to these questions, a factor analysis of the items was performed. The analysis was based on the fre- quencies with which each respondent perceived his or her spouse 134 Table 4: Percentage of Respondents Identifying the Behavior as Used to Show Affection/ Caring in Marriage Mode and Form of Affection % Respond- N Respond- ing Yes ing Yes Intimacy Self-Disclosure l. Talks to me about job or things that happen at home/work 85% 202 3. Talks to me about innermost thoughts and feelings 78% 187 5. Talks to me about fears and apprehen- sions for the future 75% 179 Aggression 7. Cuts me down in kidding kinds of ways 28% 68 9. Teases me about things which he/she knows bother me 23% 54 Physical Touch 11. Holds hands with me in public places 82% 197 13. Spontaneously touches me while talking 65% 156 15. Hugs me 90% 216 Facial Affect Displayg 17. Shows feelings through facial expres- sions 72% 172 19. Openly displays emotions 73% 176 Sexuality (Identity Centered) 21. Wrestles and fools around with me 64% 153 23 Shows warmth and tenderness in sexual relationship 96% 231 Economics Money_or Time/Energy Expenditure 2. Shares money for hobbies and interests 86% 207 4. Fixes or makes things for me 85% 206 6. Gives of time and energy in helping me with my interests and activities 88% 211 12. Spends free time with me 94% 225 Access Rights 8. Allows me to use the things that he/she considers his/her personal property 76% 182 10. Allows me to open personal mail without his/her being there 58% 138 135 Table 4 (Cont'd) Mode and Form of Affection % Respond- N Respond- ing Yes ing Yes Favors 14. Does me favors so that I can have some free time to myself 93% 223 16. Does the jobs around the house that he/she knows I don't like to do 80% 193 Gift Giving 18. Gives little gifts as surprises 69% 166 20. Buys expensive gifts on special occasions 60% 145 22. Buys gifts to "make up" after arguments 20% 47 Sexuality (Time/Energy Centered) 24. Satisfies me in our sexual relationship 97% 232 136 perform each affection behavior (page 3 of the questionnaire). The factor analysis performed was a Principal Components, Varimax Rotation with an Eigenvalue 1 criteria cutoff. The results of the analysis (see Table 5) show seven quite distinct dimensions of affection exchange with few multiple loadings across dimensions. The dimensions are ordered in Table 6 by variance explained. Together these seven factors accounted for 63.3% of the variance. Although the factor analysis identified seven fac- tors in affection exchange instead of the two dimensions expli- cated in this study, the broader affection dimensions of inti- macy and economics should be viewed as higher level abstrac- tions each of which encompasses various forms of affection which themselves may differ from one another.2 Intimacy with its focus being affection exchange through the sharing of commodi- ties of self (physical and psychologicalidentities) can take several forms: self-disclosing, sexual, aggressive and demonstrative. Economics with its vehicle of affection exchange being commitments of time/energy or their derivatives, takes the form of helping, gift giving and access rights. The data analysis will reflect an adherence to the two dimensional perspective on the affection exchange process, asserting that the communality of these forms at the higher level presupposes differences at this, the more Operational level. 137 ~m~mo.o nmooa.o oaaoo.o onmno.ou mnmoo.o «Noso.o mmova.o non ca oases» Suwanee v~o m¢> maqq~.o naumo.o1 novo~.o omena.ou noaoo.o moaoo.o vuaaa.o son as nauauooaou can guano: zoom moo m<> nonvo.o monon.o nooo~.o nomoo.on noooo.o hocoo.ot commo.o uaoaaouu sauna a: one: cu uuam son -o m<> Anson.o ooama.o monom.o moeoo.o noono.o maooa.o ~nvmo.o oaaoua deco one cannons duo m¢> Hom~H.o nomno.o «advoaot somva.o osmoo.o «oomo.o Honno.o acoaaaooo anacoAa co mamas maauaoaxo mam oNo m<> oamna.o mammo.o mmamn.o movod.os Ano~o.o muono.o oo~n6.o aaoauoao saunaao adaoao aao «as enamo.on Hanaa.o ao~H~.o Hmooo.ou naea~.o momoo.o Hoaou.o nuaauanan an nuuao oaao oao m¢> aoov~.o Hoo~a.o onna4.o oawoo.ou nvNNo.o voavo.ou madam.o naoannouaxo Augean aoaounu amaaaoou zoom aao ¢4> -ouo.o mqaoo.o manna.o onoao.o HooH5.o naoao.ou mammo.o oxaa u.:nooo mucosa one» who“ an ego ¢¢> moooa.o mmavo.o ovono.o moooH.on oaoau.o nooo~.o aoHn~.o . unseen nous mao m¢> NAHHH.o ooaoo.o mann~.o Hoon~.ou ovaao.o oomoa.o «mmao.o new» oouu an: unseen on nuoaou oo vao m¢> omoao.on Hoomo.o ooo~o.o nnooo.ou aoooa.o nnnwo.o ~ovo~.o moaxaau «Haas nose» saaaooauuooam Mao m¢> «nono.o anaoo.o acuan.o nanno.ou mnomn.o engov.o anooH.o unseen ova: use» mous ocean «Ho :45 aoa~o.ou nouva.o oao-.o masco.ou umona.o oaooa.o ooono.o mounds oaanaa ea noon: vac: Ado m¢> navaa.o ammno.o amona.o oaa-.o amnva.o «Han~.o moano.ou Haas uncommon uo oaaaoao saga: oao m¢> Homao.o onvmo.o o~a¢o.on monno.o voooa.ou oaood.o ove~o.o- moanna maauaaaou ozone canoe moo mas onawo.o nnamo.o vonoo.o mnaoo.ou oumoa.o oov~o.ot amnaa.o .suuaooua announce no on: aoaa« moo m<> ooooo.ou vooao.o oavvo.ou avooa.o nvnmo.ou ovooo.ot maaoo.ou an: «0 coax oneness ca asoouso noo m¢> «undo.ou neaoa.o «AHHH.o cameo.ou ~mmn>.o «mona.o ooonH.o nunououaa ova: magnum: ca assuaaxuaau «0 oaao ooo m¢> mmaoo.ou anamo.o ovona.o nanao.o nmvo~.o «oono.o noncn.o nooancosouaaa can auuou noon. sans moo s¢> Hooao.o Hmmao.ou HHoNo.ou noona.o- naoan.o oaoao.ou moon~.o uneasy one: no nan aoo «as oaomo.o- maoma.o moaoaso «undo.o Hoonn.o oHnHH.o «noon.o aucaaoou on. augusonu unusuaaaa ozone sans moo m¢> naan~.o aboao.o oouoo.o wamoo.o oomam.o vooma.o vaooo.o nonououaa can «cannon uou noses mason «oo m¢> vamno.o aanao.o ohmoa.ou ~oaao.ou snooa.o ovm-.o nunoo.o one: us some»: uoau manage can no“ anon: sacs Hoo m¢> h uouoam o uouoau m uououm v uouosu n uauouh N uouoah a Hannah summed osd>wo cowuooum¢ soduoouud codaoouut usdoououwo mascot uuwo o>du o>wnmouom¢ undone: «usuom Inflow ldhugafla alouH soauoomud no nauaans¢ uoouom um wanna 138 mm.mo New Hmuoe wh.v oa.m munmwm mmmoo< munmwm mmmood n wh.o mm.om~oa mcfl>fio page mofi>flo “moo o Amcwaummu3v coauowmmd mo.m H~.mH.MH.HH mafiamsxmm noses accommnm m>fiumuumcoEmo m coauommm< w>.m m.h cowmmmummfi Hmnum> m>wmmmumm¢ v wa.n ma.va.m.v.m muo>mm mmumcm\mefla mo mcfl>aw mcwmawm m muwamaxmm vmumucmo scan wm.o om.m~.~a mmumomstfia can muflucmoH sowmma Hmsxom N ma.ha mhmammwo pommmd HMflomm wh.mm m.m.a musmoHomHnnmamm Hmnum> mewmoHomHQImamm H cmcwmamxm madmanm> mo mEmz Hmnfisz cowuuomoum .oz EmuH madame; Euom Mona: Houowm uouomm wmcmnoxm noduommmd mo mcoflmcmsflo Honmz no GHQMB 139 Reliability of the Instrument In the present investigation, the most appropriate measure of instrument reliability was thought to be a measure of inter- nal consistency, specifically a coefficient based on a split- half analysis of respondent scores. Four separate split-half analyses were computed by sex and for each type of affection item. For example, the first coefficient represents a comparison of the first six intimacy items with the second six intimacy items for 120 male respondents. The results of the reliability analysis are summed in Table 7. Table 7: Reliability Coefficients of Internal Consistency Respondent and Between Forms Spearman Brown Guttman Type of Item (Correlation) Formula Split-Half Reliability for Hus- bands on Intimacy Items. . . . . . . . . . .65 .79 .78 Reliability for Hus- bands on Economic Items. . . . . . .‘. . . .66 .80 .80 Reliability for Wives on Intimacy Items. . . . .69 .82 .82 Reliability for Wives on Economic Items. . . . .69 .82 .82 The moderate reliability coefficients produced by split- half comparisons is probably partially attributable to the small number of items being compared. Application of the Spearman Brown PrOphecy Formula which projects instrument 140 reliability if the number of questionnaire items were increased reflects the importance of item number. This projected measure reflected a more acceptable level of instrument reliability. 141 Hypotheses: H1 H2 In marital dyads, the more homeomorphic the relational patterns of affection exchange, the greater the per- ceived equity. Normal dyads will be more homeomorphic in their behav- ioral affection exchange patterns than will conflict and symptomatic child dyads. Normal dyads will have higher levels of perceived equity than will conflict and symptomatic child dyads with conflict dyads having the least amount of perceived equity. Under conditions of heteromorphic affection exchange, the higher the accuracy, the higher the perceived equity. Normal dyads will have higher accuracy than either conflict or symptomatic child dyads with symptomatic child dyads having the lowest level of accuracy. A relationship between expected outcomes, realized outcomes and relationship satisfaction is hypothesized such that: a. fulfilled expectations in intimacy will produce the greatest degree of relationship satisfaction. b. fulfilled expectations in economics will produce the second highest level of relationship satisfac- tion i.e., the absence of dissatisfaction. c. unfulfilled expectations in intimacy will produce the third highest level of relationship satisfac- tion, i.e., the absence of satisfaction. d. unfulfilled expectations in economics will produce the lowest level of relationship satisfaction, i.e., actual dissatisfaction. Males will utilize economic forms of affection exchange to a greater extent than females. Females will utilize intimate forms of affection ex- change to a greater extent than males. Short history dyads will utilize economic forms of affection exchange more than long history dyads, with medium history dyads utilizing the least amount of economics. Short history dyads will utilize intimate forms of affection exchange more than long history dyads, with medium history dyads utilizing the least amount of intimacy. H11 142 Dyads with high SES will utilize economic forms of affection exchange more than dyads with middle SES, with dyads low in SES hypothesized to use the lowest amount of economic affection exchange. Dyads with high SES will utilize intimate forms of affection exchange more than dyads with middle SES, with dyads low in SES hypothesized to use the lowest amount of intimate affection exchange. Footnotes 1. The reason non caucasian families were excluded from the study is because it was felt that race was likely to be a relevant theoretical variable. This would have necessitated enlarging the sample population by at least 75%. 2. The concepts of intimacy and economics are analogous to what Homans (1950) would call "first order abstractions"—- being one level from "what people really do" in affection exchange. 143 CHAPTER IV RESEARCH FINDINGS Sample Descriptions The analysis of demographic data for the three subgroups produced the following profiles: Normal Dyads The length of marriage for normal dyads ranged from two to 37 years with a mode of three years and a mean of 14.7 years. The average income for normal dyads was $15,290.00. The average number of years of education completed by spouses in the normal groups was 13.3 years. 11.2% (n = 9) of the respond- ents had been divorced at least once. Consistent with criteria for inclusion into this sample, none of the dyads' children had been seen for either emotional, behavioral or adjust- ment problems during the past year and none of the couples had been seen for marital difficulties during that same time period. Symptomatic-Child Dyads The length of marriage for symptomatic-child dyads ranged from one to 25 years with a mode of ten years and a mean of 13.7 years. The average income for symptomatic dyads was $13,160.00. The average number of years of education com- pleted by spouses in this group was 12.5 years. 23.7% (n = 19) 144 145 of the group had been divorced at least once. And finally, 22.5% (n = 9) of the symptomatic-child dyads had been seen for marital difficulties during the past year. Conflictygyads The length of marriage for conflict dyads ranged from one to 30 years with a mode of eight years and a mean of 11.3 years. The average income for conflict dyads was $14,180.00. The average number of years of education completed by spouses in this group was 13.7 years. 12.5% (n = 10) of the individuals had been divorced at least once. And in addtion, 15% (n = 6) of the conflict dyads had had at least one child seen for emotional, behavioral or adjustment problems during the past year. Homeomorphismand Perceived Equity Hypothesis 1: In_marital dyads, the more homeomorphic the relational patterns of affection exchange, the greater the percEived equity, In this, the major hypothesis of the investigation, the degree of homeomorphism was indexed by the amount of relative agreement which the spouses indicated in responding to the question: "Do you use the following behaviors to show affection and caring for your spouse and if so, how important is it to you in your marriage?" In this case the agreement score is actually a measure of discrepancy between the spouses. For example, if the husband checks Yes to a given behavior and then indicates that it is very important to him, while the wife checks N9, the resulting score is a maximum 146 discrepancy score as compared to a zero discrepancy score if the spouses respond in an identical manner. Thus a low score on the agreement measure actually indicates a high agreement or homeomorphism, while a high score on the agree- ment measure actually indicates a low agreement or heteromorphism. The relationship between homeomorphism and equity is therefore hypothesized to be an inverse relationship, where a low score on agreement (indicating high agreement) is likely to be associated with a high score on equity. The amount of perceived equity was indexed by responses to the three Likert-type items listed below: 1. Generally speaking, how often do you feel that you give more in your marriage than your receive? 2. How often to you feel taken advantage 9: in your marriage? 3. How often do you feel that you give more affec- tion than you receive? The items were placed on a five point scale from "Always" to "Never," with "Always" being assigned a value of l and "Never" being assigned a value of 5. The inter-correlations among the three items were relatively high and thus allowed them to be summed together into one overall measure of equity with a dyadic score range of 6-30. The correlations among these three equity items, across all 240 subjects were as follows: (a) items 1 and 2 = .68; (b) items 1 and 3 = .58; and (c) items 2 and 3 = .56. 147 When computed across all 120 dyads the relationship between agreement (homeomorphism) and perceived equity was indeed found to be inverse (-.2736) and significant (.001), i.e., the higher the agreement (low agreement score) the higher the equity. When this relationship was analyzed for each example, it was found to be significant at beyond the .05 level for each group: (a) normal (-.35); (b) symptomatic- child (-.31); and (c) conflict (-.39). Effects of Accuragy on the Relationship between Agreement and Equity Hypothesis 4: Under conditions 9: heteromorphic affection exchange, the higher the accuracy, the higher theequity. The central issue in this hypothesis concerns the role which accuracy plays in determining marital equity under various conditions of behavioral affection agreement. It was previously suggested that accuracy would not be as crucial in homeomorphic instances as it would be in heteromorphic cases, i.e., that behavioral agreement (homeomorphism) might well be a sufficient condition for the perception of equity. The accuracy score as in the case of agreement is a mea- sure of discrepancy between one spouse's response to the question of whether or not a given behavior is used to commu- nicate affection/caring and, if so, how important is it and the other spouse's prediction of that response. The sample of dyads was divided into three groups based on level of agreement and zero-order correlations between agreement, accuracy and equity were computed for each level. 148 Then first-order partial correlations for each of these three groups was computed holding the effects of accuracy constant. The results are summed below in Table 8. The results of the analysis shows that the inverse rela- tionship between agreement and equity became weaker across all agreement levels when the effects of accuracy were held constant, suggesting that accuracy is indeed important in determining equity. However, contrary to the hypothesis, accuracy appears to be more important in the highest agree- ment or most homeomorphic group than it is in the most heteromorphic or medium agreement group. Also, the zero- order correlations between accuracy and equity for the heteromorphic group was the lowest of all three groups. That is, while the relationship between accuracy and equity was significant (.05) for the medium agreement group and approached significance in the high agreement group, it was not found to be significant in the hypothesized and lowest agreement group. Therefore, we might tentatively infer, contrary to our hypothesis, that accuracy may be necessary for the existence of equity in homeomorphic (high agreement) cases, but may not be necessary or perhaps even sufficient under heteromorphic (low agreement) conditions. 149 ma. n w Hoe. n m no. u m mo. u m mH.- Hm. o~.- o~.- refinance IooEomv game Iwmnm< swam mm. H m mo. n m no. u m mo. n m NH. mm. mm.: oo. pomsmmuma Eswpmz no. u m Hoo. n m mm. H m no. u m oo.- om. so.- a~.- Announce Ioumummv acme Immumd 30A Aaomusoocv huwswm momunoom muwsvm muwovm can ucmfimoumd can ucosmmumfl can momnnood 0cm ucosoonmfl macfluumm Hopuo umuam H “mono ouoN H Hopuo Ohms H Hopuo ouou moonsood can muwsvm .ucofiomum¢ mcoad msoflumHouHoonumusH “m manna 150 Agreement and Family Type Hypothesis 2: Normal ands will bg_more homeomorphic ip_their behavioral affection exchange patterns than will conflict and symptomatic-child dyads. In testing this hypothesis, the degree of homeomorphism was again indexed by the amount of agreement which the spouses indicated in responding to the question: "Do you use the following behaviors to show affection and caring for your spouse and if so, how important is it to you in your marriage?” A one-way analysis of variance was used to compare agree- ment scores of each of the three family types.' The results of the analysis are found in Table 9 below. As the three family types were found to differ significantly at beyond the .05 level, selected comparisons between groups using Scheffe procedure was utilized. A minimum alpha level of .05 was set for the Scheffe comparisons. Of the three comparisons, only conflict and symptomatic-child dyads were found to differ significantly.‘ Although the scores for normal dyads and symptomatic-child dyads were in the predicted direction, they were not found to be significantly different. The comparison of normal and conflict dyads were not only non- significant, but the scores were opposite of the predicted direction. 151 hmmm.m¢mvw mHH Hmuoa omom.ono oooo.aommh has masons canons mmo. moo.m mmm~.~¢o~ boos.oo~m N maoouo cmmzumm ammo Ho>oq mosmwum> was: oocmowmwsmwm Owumm m mumsvm coo: moumsqm mo saw me no mousom oncommom usmEomumm on» uo3oH on» whoom on» mono“: one « ooom.~n ow . momma uoaamcoo oomw.>m ov momwo nafinonowumEoumahm aoomm.mh co mocha HmEHoz mouoom ucmEmoumm com: 2 omha mawsmm mama maweom ma usmfiooumd co mouoom com: um manna 152 Equity and Family Type Hypothesis 3: Normal dyads will have higher levels pf per- ceived equity than will conflict and sympto- matic-child dyads with conflict dyads havipg the least amount g£_perceived equity. In this hypothesis, perceived equity was again indexed by the three items which asked about the equity of the exchange relationship. Individual scores for each spouse were then totaled into a dyadic measure for purposes of analysis. A one-way analysis of variance was again used to compare the equity scores of the three family types. The results of the analysis are found in Table 10 below. :1- As the three family types were found to differ signifi- cantly at beyond the .001 level, Scheffe procedure, with an alpha level of .05 was again utilized for selected comparisons. As hypothesized, normal dyads were found to have the highest level of perceived equity, symtomatic-child dyads were found to have the second highest level, and conflict dyads the lowest level of perceived equity. Comparisons between the three family types showed that each of the groups differed from one another at beyond the .05 level of confidence. Accuracy and Family Type Hypothesis 5: Normal dyads will have higher accuracy than either conflict 2E symptomatic-child dyads with symptomatic-child dyads havipg the lowest level g£_accuracy. As was the case with the agreement measure, the accuracy score is actually a measure of discrepancy between the spouses. Thus a high score on accuracy indicates a low level of accurate 153 hwov.HmmH mad Hmuoa oooh.ma oomo.~ooa FHH museum casuaz Hoo. «Ho.ma mom~.ooa noam.osm a mmsouo ommsumm omso Hm>mq mOGMwum> ass: mocMOflwfismwm Owumm m mumsvm com: mmucsqm «0 saw me no mousom uncommom muwsvo may stmfln on» whoom onu “mamas one: omom.oa ov momma uoaamsoo ooow.ma co momma pawnuuowumeoumeam ommm.o~ ow momma Hushoz monoom muflsqm com: z mama maaamm maze hawfimm an >uasqm so mouoom com: "OH wanna 154 ooo~.m>moo~ mad Hopes oama.oooa oooo.amaooa has masons canons «mo. ~oo.m ooom.oa~o oooo.a~oma N museum cmmsumm ooso am>mq . mocmwum> uwco mosm0fimwcmflm ofiumm m mumwm :mmz mmumsvm «0 85m mo «0 mousom mmcommmm momusoom mnu Hmzoa may muoom ms» Hmnmfis mas « oomm.~va ow momma uoaamcou oomm.mma . ov momma pawnolowumeoumfixm aooom.mva ow momma Hmauoz mmuoom womusoom com: 2 maze maaEmm mama waafimm an momusoom so mmuoom smmz “Ha manna 155 prediction between the spouses, while a low score indicates a high level of dyadic accuracy. A one-way analysis of variance was used to compare the accuracy scores of the three family types. The results of the analysis are found in Table 11 below. As the three groups were again found to be significantly different at beyond the .05 level, selected comparisons between groups using Scheffe procedure was utilized. As was the case on the agreement measure, only conflict and symptomatic- child dyads were found to differ significantly, using an alpha level of .05 as criteria. Although the accuracy scores for normal dyads and symptomatic-child dyads were in predicted direction, they were not found to be significantly different. The comparison of normal versus conflict dyads were again not only non-significant, but the scores were Opposite of the prediction. Relationship Satisfaction as a Two-Dimensional Variable. Hypothesis 6: A relationship between expected outcomes, realized outcomes and relationship satisfac- tion was hypothesized such that: (a) fulfilled expectations in intimacy will produce the greatest level of relation- ship satisfaction, (b) fulfilled expectations in economics will produce the second highest level of relationship satisfaction, i.e., the absence of dissatisfaction, (o) unfulfilled expectations in intimacy will produce the third highest level of rela— tionship satisfaction, i.e., the absence of satisfaction, and 156 (d) unfulfilled expectations in economics will produce the lowest level of relationship satisfaction, i.e., actual dissatisfaction. The level of expectation in each affection mode, i.e., the intimate and the economic, was determined from page two of the questionnaire which asked about the role expectations in affec- tion of each person. The issue was posed in the following con- text: "In showing that he likes his wife and cares about her, a Husband should:" The 24 behavioral items then followed and were placed on a five point scale from Strongly Agree to Strongly Disagree, with the former being assigned a value of 1 and the latter a value of 5. If the respondent checked either Strongly Agree or Agree on an item, that was sufficient to con- stitute an expectation for that behavior. Using this criteria, the expectation level for each of the 12 intimacy behaviors and 12 economic behaviors was calculated to give an overall expec- tation score for each affection mode. The mode with the highest score was then labeled the primary mode of affection expectation. If an individual had identical intimacy and economic expectation scores, he or she was deleted from the analysis. There were 28 such expectation ties out of 240 respondents. Realized outcomes or level of expectation fulfillment was tmaudetermined for each behavioral expectation item, across both modes of affection. This was determined from page 3 of the questionnaire which asked how frequently the person's spouse performs each behavior. The same items were scaled 1 to 5 from Very Often to Never. For each item for which an expectation existed, the level of fulfillment was determined by assessing 157 how frequently the individual was the recipient of that behav- ior from his or her spouse. If the individual checked Always, Often, or Now and Then, that was deemed sufficient to constitute fulfillment on that item. The criteria were extended to the third category, Now and Then, based on the previously discussed assumption that infrequently performed behaviors may have as much, if not more, reward value than frequently performed behaviors. The total number of expectations which were fulfilled was calculated for each mode using this procedure. Then taking the respondent's primary mode of expectation (his higher expectation score) and comparing it with his level of fulfill- ment in that mode, each respondent was assigned to one of four groups: Group (1) fulfilled expectations in intimacy (n = 97) Group (2) fulfilled expectations in economics (n = 32) Group (3) unfulfilled expectations in intimacy (n = 72) Group (4) unfulfilled expectations in economics (n = 11) A respondent's expectations were judged to be fulfilled if all the expectations in his primary affection mode were fulfilled or if the expectation and fulfillment discrepancy was no greater than 1. For example, if a respondent had expectation scores in intimacy of 8 and economics of 5 and had corresponding fulfill- ment scores of 7 in intimacy and 5 in economics, he or she would be assigned to group 1 -— fulfilled expectations lg intimacy. If he or she had a fulfillment score of 6 in intimacy, he or she would be assigned to group 3 -- unfulfilled expecta— tions in intimacy. (The fulfillment score in the secondary 15, will; .n I I 158 mode, in the above example, economics, was not considered in the group assignment decision.) Using these criteria, the 212 respondents (28 deleted due to tie scores in expectations) were assigned to the four groups according to expectation and level of fulfillment. (The distribution across all four groups is indicated above for each group.) The relationship satisfaction scores of each respondent were then entered as the dependent measure -- relationship satisfaction being measured by each individual's response to the questionnaire items which asked: Read each of the following statements very closely and then check the one which most accurately describes your feelings about your marriage. "Most of the time, I feel": Generally Satisfied with my marriage (Scale Score = 1) Not Satisfied, but not really Dissatisfied (Scale Score = 3) Generally Dissatisfied with my marriage (Scale Score = 4) Not Dissatisfied, but not really Satisfied (Scale Score = 2) (The scaling order used for the dependent measure, since each theoretically represents a different ordinal level of relationship satisfaction, is listed to the left of each description.) The relationship satisfaction scores of the four groups were then analyzed with planned comparisons for trends using the method of orthogonal polynomials. The results of the one- way analysis of variance with tests for linearity and t-tests for comparison of groups means are summed in Table 12. 159 mocoaum> mo mammamco m>ono mnu ca Bump Homcwa mom on msomoncm ma aufiaa annoua more . *1.u.ov oao. oom ~hp.H nouns Hauum>o ooo. mom moo.~r o moouo o m ozone Hoo. oom ado.vn m moons a N moose A.u.:c moo. mom ~oh.ou m moons a H ozone muaaannnoum a so moam> a umouucoo mononucou fluowua 4 ooa~.om~ ado Hopes Homo. moao.noa ooN masons canvas Hoo. ooa.o~ oooo.o~ oaoo.ma u Humans EOHM GOHumwxme 1.m.oc «so. oma.m omom.~ owoo.~ H sums saunas Hoo. noo.oa oomm.ha mooo.~m m mmsouo owmsumm Hanoa>nocH H0>QH 00GMHHM> UflGD mocMOfiwflcmwm owumm a mumsvm cum: mmumsvm mo 85m an no mousom mucoammm coauomwmwumm mam umnuwn may .muoom mnu Hm30H maa « oomo.a as .moaeoooom on maoauuuomaxm omaasmasmooc o maouo Homv.~ Nb AaomEMucH cw mcowumuomaxm omaawmasmcsv m asouu mamm.a mm Amoflaocoou cw mcowumuomaxm omaafimasmv m anouo «oaom.a so AmomsauoH an mooaumuomoxm omaasmasmc a msouo mmuoom acauoommsumm masmcoaumamm z acmsaflamaom\coaumuomaxm no mass com: unmadaflmaaa\c0flumuomaxm mo maze wn cofluomwmflunm awnmcofiumamm co mmuoom com: "NH manna 160 The results of the analysis of variance show a significant F ratio between groups (.001) which in this instance is only of passing interest. The items of real interest are the F ratios for linear trends and deviations from linearity. The orthog- onal polynomial weights for linear trends which take into account a progressive increase or decrease in means, shows a near, although non-significant (.07) linear relationship among group means. The deviation from linearity term shows signifi- cance at the (.001) level and can be attributed to the scores of group 4 which portrays the trend as more quadratic or curvi- linear. Thus, in this case the small cell size (n = 11) of group 4 tends to misrepresent the underlying and hypothesized linear trend demonstrated by the other three groups. The contrasts between groups shows an insignificant T value for groups 1 and 2; a significant difference (.001) be- tween groups 2 and 3; and a significant, although opposite direction difference between groups 3 and 4. The overall trend contrast (which is analogous to the F ratio for linearity) among all four groups again approached significance (.05) with a T probability of .078. One of the major issues which this hypothesis sought to explore concerned the dimensionality of the construct -- relationship satisfaction. Is relationship satisfaction uni- dimensional, i.e., is the Opposite of relationship satisfaction simply dissatisfaction or is it the absence of satisfaction with the absence of dissatisfaction being the antithesis of dissat- isfaction? In effect is there evidence of two overlapping dimensions comprising the construct? 161 The responses of the subjects indicate that individuals do make distinctions in terms of feelings of satisfaction, dissatisfaction, or their respective absence. As evidence, while 148 subjects checked generally satisfied as describing their feelings about their marriage, 34 subjects checked not satisfied, but not really dissatisfied (the absence of satis- faction). On the other dimension, while only 26 subjects checked generally dissatisfied, 32 subjects checked not dis- satisfied, but not really satisfied (the absence of dissatisfaction). The overwhelming majority of subjects who checked either absence of dissatisfaction or absence of satisfaction, as representing their feelings, came from either symptomatic-child couples or conflict couples, couples who would be expected to check lesser degrees of satisfaction. Thus it appears that the often nebulous and difficult to describe relational feelings which marital partners often wrestle to define for themselves, their confidants, or their therapists, may be more accurately described by these two overlapping continua, than by a single, unidimensional one. (See Figure 6.) The second issue concerns the question of which affection mode produces the greatest amount of relationship satisfaction in relational affection exchanges. The data show that fulfilled expectations in intimacy produces the highest satisfaction scores (mean = 1.39). The second highest level (mean = 1.53) was produced by fulfilled expectations in economics. (These values, however, did not differ significantly.) The lowest 162 level of relationship satisfaction (mean = 2.49) was produced by unfulfilled expectations in intimacy. This ordering of the means is the order previously predicted. The fourth alternative, unfulfilled expectations in economics, with a small cell size, produced an intermediate level of satisfaction (mean = 1.64). Scale Score = (l) (3) Satisfaction F {AbsenCe of Satisfaction (n = 148) (n = 34) Fulfilled Expecta- Unfulfilled Expectations. tions. Intimacy Intimacy (mean = 1.39) (mean = 2.49) (2) (4) Absence of Dissatis-i : DissatisfaCtion faction (n = 32) (n = 26) Fulfilled Expectations Unfulfilled Expecta- Economics tions. Economics (mean = 1.53) (mean = 1.64) Figure 6. Subject ReSponses to Relationship Satisfaction Measure Based on this initial data it might be tentatively con- cluded that relationship satisfaction can be viewed as a two- dimensional construct composed of two overlapping continua, one representing adequate or inadequate intimacy, the other, adequate or inadequate economics. In terms of which mode possesses the largest capacity for producing the greatest degree of satisfac- tion from relational exchanges, that inference cannot be made from the present data. However, there is at least some data to suggest that intimacy may have the greater capability. 163 Sex Differences in Affection Exchange Hypothesis 7: Males will utilize economic forms cf affection exchange tg_g greater extent than females. The level of economic affection usage was indexed from the same data used to generate the agreement measure, i.e., "Do you use the following behaviors to show affection and caring for your spouse and if so how important is it to you in your marriage?" However, in these analyses, only the 12 economic items were summed to produce an individual measure of economic usage. The data analysis for testing this hypothesis as well as the five which follow were designed as a two-way analysis of variance to include family type to check for possible interaction as well as main effects among sex, family type and economic affection usage. The results of the analysis are found in Table 13 below. No significant main or interaction effects for economic affection usage by either sex or family type were found. Hypothesis 8: Females will utilize intimate forms cf affec- tion exchange £3 E greater extent than males. The level of intimate affection usage was indexed by the same procedures described in the above economic hypothesis, this time with the 12 intimacy items being summed to produce an individual measure of intimacy usage. The results of the two- way analysis of variance are found in Table 14 below. Men and women were found to differ in terms of intimate forms of affection usage with women being significantly more intimate in their reported affection behavior than men. There 164 _ mmv.mmh~m mmN Hmuoa ~mm.~mm evw.ahoam mmm cwcuwz .m.c oho.H mHH.o~¢ m-.o¢m N cowuomnmucH .m.s hem. mom.oNH cmH.H¢~ N hawfima .m.c Hoe. o>¢.~ha ohm.~ha a xmm Hmoow>aocH Hm>mq muosvm mocmaum> muwco monomemacmflm ofiumm a com: mmumsum mo Edm an no mousom mucoammm mo.oo om momma uosamcoo mm.oo oo momma cannonosumeouoesm mm.vm om woman HmEHoz mama hHHEma oa.mo oma mmHmEma mm . om oma mmamz mmuoom masocoom com: 2 xmm maaa meEma tam xmm an momma cOAuomwmd oweocoom co mmuoom cmmz "ma manna 165 mmv.mmw>~a mmm Hmuoa oma.oam ovo.mmmoaa mmN cwnuwz .m.c mmm. omo.~ma Hma.va ~ cowuomumucH 1.m.oc «oo. aoo.~ omm.aooa hoo.moo~ m saaamm Hoo. mom.HH mvv.ommm www.5wmm H xmm Hmsow>wocH Hm>mn mumsvm mmumsvm mocmwnm> pass mocmoawflcmwm oaumm a com: no How an no mousom mucoammm ww.mw om momma uoaamcoo mm.om om momma oaacouoaumsouasam mm.Hw om mocha HMEhoz mama aafiama mm.wm ONH mmHmEma Nh.mm QNH mmamz muoom momeucH com: 2 xmm maae hawEma ocm xmm an momma cowuommmm mumfiaucH co mmnoom com: "ea magma 166 was also a near significant main effect for family type with conflict dyads being the most intimate and symptomatic-child dyads the least intimate in their reported intimacy usage in marriage. Length of Marriage and Affection Exchange Hypothesis 9: Short history dyads will utilize economic forms 9: affection exchange more than long history dyads, with medium history dyads utilizing the least amount 9: economics. The same measures of economic and intimate affection usage used in hypotheses 7 and 8 were again utilized. However in the following instances, individual measures of the spouses were combined into a dyadic measure for each couple. Again the data analysis was designed as a factorial to check for possible interaction effects. The results of the analysis are summed in Table 15 below. No significant main or interaction effects for economic affection usage for either history or family type were found. Hypothesis 10: Short history dyads will utilize intimate forms of affection exchange more than long history dyads, with medium history dyads utilizing the least amount cf intimacy. Again individual measures of intimacy usage were combined into dyadic scores and the data analyzed by factorial design. The results of the analysis are summed below in Table 16. The analysis revealed a significant difference (.05) in intimate forms of affection usage by dyadic history and a near significant difference by family type. However, contrary to the hypothesized ordering of short history dyads being the most 167 amm.amawoa maa Hmuoa www.mam mow.ooa~oa HHH sanuwz .m.c mom. Hom.aam mom.aama a coauomumucH .m.c mam. «mo.mwm vw~.amHH N aaaEma .m.c vam.H oem.am~a Hmm.mwmm N amoumam puma Hm>mq mucsvm mmumsvm mocMfium> macs mocmowwwcmflm oflumm a com: mo saw an no mousom mmsoammm mm.mmH ow momma uowamcou mm.ama ow momma oaflcouoflumsouasam mo.m~H co momma Hueuoz mama aHflEma va.HmH am Amumma animav moon ha.mma Ho Amumms saumc guano: oo.¢~a «a Amumms ouflo uuoam mmnoom owEocoom cum: 2 aHOpmam maaa maaama pom muoumflm an momma cowuommmm oweocoom co mmuoom com: "ma manna 168 oom.movama maa Hmuoa www.mooa mvv.voomaa HHH canuwz .m.c mmm.a mom.~ama Nam.amma v coHuomumucH A.m.cv coo. mom.~ acc.mmm~ mam.mwmm m mHHEma «oo. opa.m omo.ammm moo.mooo m muoumam coma Hm>mq mumsvm mocmanm> vac: mocMOflMflcmam oflumm a cum: mmumowm mo Esm an no monoom mucoammm mv.~ma ov momma moaamcoo om.maa ow momma pawnUIoaumEouaEmm om.m~a oo momma Hmsuoz mama mHaEma mm.NHH mm Amummm mmnmav moon o~.~ma Ho Amummm naumc fiasco: om.oma mo Amuoms also uuoom mmuoom momfiwucH com: 2 muoumflm mama maflama can muoumam ma momma c0wuommm< momfiauca so mmuoom com: ”ma magma 169 intimate, medium history dyads the least intimate, and long history dyads occupying an intermediate position, medium his- tory couples were found to be the most intimate and long history couples found to be the least intimate with short history couples occupying the intermediate position. Socio-economic Status and Affection Exchange Hypothesis 11: Dyads with high socio-economic status will utilize economic forms of affection exchange more than dyads with middle SES, with dyads low in SES hypothesized to use the lowest amount of economic affection exchange. The same measure of economic affection usage used in hypo- thesis 9 was again used here. SES was measured by the Alba Edwards Occupational Scale as modified by Hollingshead (1957, 1958) to include educational skills. With a possible range of two (highest SES) to 14, the sample of dyads was divided into three levels of SES: Highest SES (scale score 2 - 5), middle SES (scale score 6 - 9), and lowest SES (scale score 10 - 13). The results of the factorial analysis are summed below in Table 17. No significant main effects for either SES or family type were found for economic affection usage. However, the main effect of SES which approached significance at the (.10) level is interesting to note in passing and will be discussed later in chapter V. moa.ammwoa mad Hmuoa 170 Nvm.oom me.ooamm HHH casua3 .m.c mam. mmo.awv Nmm.wmmH q coauomumuca .m.o Noe. oom.omv moo.ooo m Nassau A.m.:v moH. omN.N Hmm.mmON Noo.moav N mam puma Hm>mq mumsvm mocmaum> macs mocmowmwcmwm oaumm a cum: mmumsqm no How an no mousom mmcoammm mm.mma ow momma mowamcou mm.mNH co momma oaaoonowuchuaEmm mo.m~a ow momma Hosuoz mama mawama Ha.mma mm Amauoa m manomv mmm 30A vo.mNH «a Amlo m manomv mmm mauve: mm.a~a oo Amum m manomc mum roam mmuoom OHEOcoom com: 2 mmm mama mHaEma com mmm mn momma GOHuomwmm OHEocoom co mmuoom com: and magma 171 Hypothesis 12: Dyads with high SES will utilize intimate forms of affection exchange more than dyads with middle SES, with dyads low in SES h othesized to use the lowest amount EffintifiatggaffEEEiEH'Eichafige. The same measure of intimacy usage used in hypothesis 10 was again used here, along with the previously described SES indices. The results of the analysis are summed below in Table 18. No significant main effects for either SES or family type were found for intimate affection usage. However, the effect of family type again approached significance (.05). 172 oom.movmma mad Houoa omN.ooHH oao.amaNNH HHH swap“: .m.c Nmo.a mmm.HoNH cam.mom¢ w :ofluooumucH A.m.cv Hoo. mwm.N mva.omam amN.omNm N mawema .m.c mmm.H omo.aama mmo.NN¢m N mum puma Hm>mq munnvm moccaum> macs mocooamacmam owunm a com: mmuosvm mo saw an no mousom mmcoammm mv.NmH ov momma uoaamcou oa.maa ow momma oaaaOIOHuchuaEmm om.mNH ov momma HMEMOZ mama mHHEma aH.mNH mm Amaloa m manomv mam 3oq ma.maa «o “mum m manomv mum maooaz o~.o~H oo Amnm m oamomc mum some mmuoom momefiucH com: 2 mum mama maaema ocm mum mo momma cowuommmm mumEHucH co mmuoom com: "ma magma CHAPTER V SUMMARY, DISCUSSION AND RECOMMENDATIONS . FOR FUTURE RESEARCH Summary and Discussion As discussed previously, the study found that homeomor- phic patterns of affection exchange are likely to be associated with higher levels of dyadic equity. That is, couples who behave similarly in their affection exchange behaviors tend to report higher levels of marital equity. Furthermore, it was found that equity was Significantly (.001) associated with rela- tionship satisfaction (-.22). (In this instance the higher the equity, the lower the satisfaction score, which actually indicates high satisfaction.) Therefore, we find initial evi- dence of significant covariation for the overall hypothesized causal relationship of high agreement leading to high equity, which in turn leads to higher levels of relationship satisfaction. In addition, accuracy was found to be positively associ- ated with agreement. That is, couples who had high levels of agreement were also found to have high levels of predictive accuracy. Conversely, couples low on agreement were found to have corresponding low levels of accuracy. From a functional standpoint this is indeed unfortunate, as a higher level of 173 174 accuracy might allow heteromorphic couples to achieve the higher levels of dyadic equity demonstrated by homeomorphic couples. These are significant findings for they imply that unless couples have homeomorphic patterns of affection exchange, that their relationships may be marked by higher degrees of instability and/or dissatisfaciton, with consequences not only for the spouses themselves, but perhaps for other family members as well. This inference is based onthe finding that symptomatic- child dyads had not only the lowest agreement scores but also the lowest accuracy scores. One of the most surprising findings of the investigation was that conflict couples were the most homeomorphic of the three family types. And while there was overall significance (.02) between the three groups, only symptomatic-child and conflict dyads were found to differ significantly (.05) with Scheffe procedure. In addition, the conflict couples were also found to have the highest accuracy. And while there was again an overall significant difference (.02) in accuracy among the groups, only conflict and symptomatic-child couples were found to differ at .05 Scheffe. How can these unpredicted findings be explained? The conflict couples were intact couples, i.e., they were still living together and therefore we must assume, maintaining some commitment to the relationship. In addition they were also engaged actively in marital counseling which, in various degrees, attempts to focus on relevent aspects of the relationship definition. Thus, while we did not expect high agreement from 175 this group, we might well have anticipated high accuracy as a result of treatment efforts of the therapist. That is, as a result of therapy sessions, we might well expect spouses to become increasingly able to pinpoint their spouse's feelings or position on relational issues. It also seems plausible that conflict couples, with some "‘4“qu remaining commitment to the relationship may attempt to make behavioral concessions to one another as a result of therapy, which may also effect their affection exchange patterns in the direction of increasing their level of agreement. One cannot help but wonder if the results would have differed if the conflict couples had not been intact couples and had been com- posed instead of divorcing couples who had not undergone any type of marital intervention. All three family types differed as predicted on level of dyadic equity with normal couples having the highest equity, followed by symptomatic-child and conflict couples in that order. Although men and women were not found to differ in economic affection usage, they were found to differ significantly (.001) on their reported usage of intimate forms of affection exchange. This is an important finding, because coupled with our previous inferences concerning agreement, equity and relationship satis- faction, this may mean that sexual differences in intimacy repertoires may doom or adversely predispose a great number of heterosexual relationships to the largely unconscious influences 176 of heteromorphic exchange, perceived ineqity and eventually relationship dissatisfaction and/or instability. The findings of sexual differences in intimacy usage also support Hawkins' assertion that socio-emotional factors are more crucial to the wives' marital satsifaction than to the husbands'. (Hawkins, 1968, p. 650.) That is, while males may accept intimacy from their spouses, they may not, as our data suggest, "be able" to provide adequate socio-emotional support through intimacy to their wives. Thus it may take on added significance to the wives, by virtue of its relative absence. The nearly significant main effect (.06) for family type is also extremely important. Conflict dyads were found to have the highest intimacy scores with symptomatic-child dyads report- ing the lowest usage. This is an interesting finding because it tends to support the previous descriptions of families of the emotionally disturbed child as lacking cohesion and emotionally supportive communication. It also tends to support the emotionally disturbed child's portrayal of his parents marital relationship as distant (non-intimate) and rejecting. However, a lack of significant interaction effect between sex and family type on intimacy usage failed to support the view of the mother of the emotionally disturbed child as not performing normal maintenance or emotional roles in the family. The most important issue, however, remains unanswered. What is the sequence of events? Do marriages simply succumb to the pressures of an emotionally disturbed child, with less mari- tal intimacy? Or is lack of intimacy an antecedent condition 177 which predisposes children to these functional disorders? Unfortunately, the present investigation does not provide that answer. One of the most unanticipated findings was that conflict dyads, while having the lowest dyadic equity also had the high- est intimacy usage. This finding would appear to fly in the face of our assertion that the intimacy mode is the greatest provider of validation and satisfaction in relational exchanges. However, the conflict dyads also had the highest level of educa- tion of the three groups and as the literature review revealed, the highly educated tend to put more emphasis on the rela- tional aspects of marriage, as opposed to the instrumental aspects. They also tend to have higher expectations of marriage and thus may become disenchanted more easily. Also, couples actively engaged in marital counseling may feel some pressure to respond in the socially desirable manner to an affection exchange questionnaire administered in a private or public clinic. While there were no significant main or interaction effects for history or family type on economic usage, there was a sig- nificant main effect for history and again the near significant effect for family type on intimacy usage. While short history dyads were hypothesized to use the most intimacy and medium history couples the least, the data reveal that medium history couples reported the greatest intimacy usage, with long history couples utilizing the least. Thus intimacy usage does not appear to follow the marital satisfaction trends which all the 178 previous literature supports. (See Figure 7.) This intimacy trend may be interpreted as to suggest that it may take several years of marital living for couples to feel sufficiently secure and comfortable with one another, to be able to engage more actively in intimate forms of affection exchange. In regard to long history dyads having the lowest intimacy usage, it may suggest, as does Hawkins (1968) that the instrumental aspects of marriage may be as important as relational aspects and may become increasingly important in the later years of marriage. With regard to the relationship of socio-economic status and economic affection usage, no significant main effect was found with an alpha criteria of .05. Although the significance level for the main effect of SES only approached .10, the mean scores of high, medium and low SES couples are interesting and deserve further attention. Interestingly enough, low SES couples had the highest economic usage. Following the vein of thought which suggests that people generally value those things which are rare, this might be interpreted to mean that those families who possess the lowest amount of socio-economic resource attribute the greatest amount of value to these time/ energy commodities in their relational affection exchanges. And finally, no significant main effect for SES on intimacy usage was found. It is interesting to note in pass- ing that medium history dyads reported the greatest amount of affection exchange behavior across both affection modes, while long history couples reported the least amount. Also, low SES 179 High Marital Satisfac- tion LOW short medium long Dyadic History 140} 130 Intimacy Usage 120) 110, 100r short medium long Dyadic History Figure 7. Marital Satisfaction and Intimacy Usage Over Time 180 couples had the highest overall affection exchange score with middle SES couples having the lowest affection scores across both combined modes. And finally, conflict dyads reported the greatest amount of overall affection exchange with symptomatic-child couples having by far the lowest amount of com- bined affection. Recommendations for Future Research First of all, there are considerable analyses which can and should be conducted on the present data. It would be very interesting to analyze the data along the seven affeciton dimen— sions identified in the factor analysis by sex and family type. For example, it would be extremely interesting to find out if the three family types differ on their usage of aggressive affection, as slightly over 25% of the sample identified them- selves as communicating affection through that avenue. In addition, it would be interesting to see if more men than women relied on aggressive affection behavior exchange. The same kind of analysis could be done on the dimensions of self-disclosing, sexuality and helping. Another interesting and necessary analysis would be to test for curvilinearity inthe relationship of dyadic history and intimacy usage. Although it appears that the relationship may be curvilinear, a test for degree of significance should be undertaken. Aside from future analyses of the present data, more refinement in the instrument and scaling needs to be done. In 181 hindsight, I believe that the use of multi-dimensional scaling might have been a preferred, or at least a secondary data collection device, which might have provided more graphic descriptions of orientations toward forms of affection exchange among various family types, sexes, social classes, etc. In addition, it would seem desirable to perfect the present instrumentation with another, perhaps larger sample of each family type, this time including delinquent-child couples. And finally, we should attempt to move toward the point where both major relational dimensions, affection and control, can be used to analyze the interaction patterns of dysfunctional family types. For at least this observer believes that may of the keys to unravelling the mysteries of dysfunctional family systems lie hidden within the family's interaction fabric and its inter- woven threads of affection and control. BIBLIOGRAPHY BIBLIOGRAPHY Abrahamson, Bengt. "Homans on Exchange: Hedonism Revived." American Journal gf_SociologY. Vol. 76, No. 2, 1970, pp. 273-285. Ackerman, N. W. "An Orientation to Psychiatric Research on the Family." Journal cf Marriage and Family Living, Vol. 19, 1957, pp. 103-110. . "The Family Approach to Marital Disorders," in Greene, B. L. (ed.), The Psychotherapies pf Marital Disharmony. New York: Free Press, 1965. . Family Process. New York: Basic Books, 1970. Adams, J. Stacy. "Inequity in Social Exchange." Advances ig Experimental Social Psycholcgy, Vol. 2, 1965, pp. 267- 299. Ainsworth, M. D. "The Effects of Maternal Deprivation: A Review of Findings and Controversy in the Context of Research Strategy," in Deprivation cf Maternal Care: .5 Reassessment of Its Effects. Geneva: World Health Organ- ization 1196217 Alexander, J. F. "Defensive and Supportive Communications in Family Systems." Journal g; Marriage and;the Family, Vol. 35, 1973a, pp. 613-617. "Defensive and Supportive Communications in Normal and Deviant Families." Journal cf Clinical and Consulting Psychology, Vol. 40, 1973b, pp. 223-231. Aller, Florence A. "Role of the Self-Concept in Student Marital Adjustment." Family Life Coordinator, Vol. 11, 1962, pp. 43-45. Altman, I. and Taylor, D. Social Penetration: The Development cf Interpersonal Relationships. New York: Holt, Rinehart & Winston, 1973. Andry, R. G. "Paternal and Maternal Roles and Delinquency." WHO Public Health Paper, Vol. 14, 1962, pp. 31-44. Axelson, Leland. "Personal Adjustment in the Postparental Period. Marriage and Family LiViEg) Vol. 22, 1960, pp. 66-68. 182 183 . "Marital Adjustment and Marital Role Definitions of Husbands of Working and Non-Working Wives." Marriage and Family Living, Vol. 25, 1963, pp. 189-195. Bandura, A. and Walters, R.H. Adolescent Aggression. New York: Ronald, 1959. Bakwin, H. "Emotional Deprivation in Infants." Journal pf Pediatrics, Vol. 35, 1949, pp. 512-521. Bardeen, J. P. "Interpersonal Perception Through the Tactile, Verbal, and Visual Modes." Paper Presented at ICA, Phoenix, 1971. Barker, G. H. and Adams, W. R. "Comparison of the Delinquencies of Boys and Girls." Journal cf Criminal Law, Vol. 53, Bateson, G.; Jackson, 0.; Haley, J.; and Weakland, J. "Toward a Theory of Schizophrenia." Behavioral Sciences, Vol. 1, 1956, pp. 251-264. Becker, J., and Iwakami, E. "Conflict and Dominance within Families of Disturbed Children.” Journal gf_Abnormal Psychology, Vol. 74, 1969, pp. 330-335. Becker, W. C.; Peterson, D. R.; Luria, Z.; Shoemaker, D. J.; and Hellmer, L.A. "Relations of Factors Derived from Interview Ratings to Behavior Problems of Five Year Olds." Child Development, Vol. 33, 1962, pp. 509-553. Becker, W. C., and Krug, R. S. "A Circumplex Model for Social Behavior in Children." Child Develcpment, Vol. 35, 1964, pp. 371-396. Behrens, M. L., and Goldfarb, W. A. "A Study of Patterns of Interaction of Families of Schizophrenic Children in Resi- dential Treatment." American Journal pf Orthopsychiatry, Vol. 28, 1958, pp. 300-312. Bell, Norman, and Vogel, Ezra. “A Modern Introduction tg_the Family. New York: The Free Press, 1960. . "The Emotionally Disturbed Child as the Family Scape- Goat." In A Modern Introduction t9_the Family. New York: Free Press, 1968. Beres, D., and Obers, S.J. "The Effects of Extreme Deprivation in Infancy on Psychic Structure in Adolescence: A Study in Ego Development." In R. S. Eissler, et. al. (Eds.) The Psychoanalytic Study of the Child. New York: International University Press, 1950? 184 Berlo, D. Four Essays gg_Communication, Human Communication: The Basic Propositions. East Lansing: Michigan State University, 1968. Bernard, Jessie. "Factors in the Distribution of Success in Marriage." American Journal 2; Sociology, Vol. 40, 1934, pp. 49-600 Birdwhistell, Ray. Kinesics and Context. Philadelphia: University of Pennsylvania Pres, 1970. Blake, R., and Mouton, Jane. The Managerial Grid. Texas: Gulf Publishing, 1964. Blau, Peter. Exchange and Power lg Social Life. New York: Wiley, 1964. _ . ”Ala-“q .- Blood, Robert O., and Wolfe, Donald M. Husbands and Wives: Dynamics gf_Married Living. Glencoe: The Free Press, 1960. -9 Blumstein, Philip, and Weinstein, Eugene. "The Redress of Distributive Injustice." American Journal gf Sociology, 1967, pp. 408-418. Bolles, M. M.; Metzger, H. F.; and Pitts, M. W. "Early Home Background and Personal Adjustment." American Journal 2: Orthopsychiatgyj Vol. 11, 1941, pp. 530-534. Borgatta, E. F. "Rankings and Self-Assessments: Some Behavioral Characteristics of Replication Studies." Journal gf Social Psychology, Vol. 52, 1960, pp. 297-307. . "The Structure of Personality Characteristics." Behavioral Science, Vol. 9, 1964, pp. 8-17. Borgatta, E. F.; Cottrell, L. S.; and Mann, J. M. "The Spectrum of Individual Interaction Characteristics: An Inter- dimensional Analysis." Psychological Reports, Vol. 4, 1958, pp. 279-319. Bossard, James H.S., and Boll, Eleanore, S. "Marital Unhappi- ness in the Life Cycle of Marriage." Marriage and Family Living, Vol. 17, 1955, pp. 10-14. Bowen, M. "A Family Concept of Schizophrenia." In Jackson, D. D., ed. The Etiology 9f Schizophrenia. New York: Basic Books, 1960. Bowen, M.; Dysinger, R. H.; and Besamanie, B. "The Role of the Father in Families with a Schizophrenic Patient." American Journal gf Psychiatry, Vol. 115, 1959, pp. 1017-1020. 185 Bowlby, J. "Maternal Care and Mental Health." WOrld Health Organization Monograph Series, No. 2 [1951]. Brown, Roger. Social Psychology. New York: Free Press, 1965. Bugental, 0.; Love, L.; and Kaswan, J. "Verbal-Nonverbal Con- flict in Parental Messages to Normal and Disturbed Children." Journal 9: Abnormal PsycholOQY: Vol. 77, 1971, pp. 6-10. Bugental, D.; Love, L.; Kaswan, J.; and April, C. "Videotaped Family Interaction: Differences Reflecting Presence and Type of Child Disturbance." Journal g£_Abnorma1 Psychology, Vol. 79, 1972, pp. 285-290. Burgess, E. W., and Locke, H. J. The Family from Institution tg Companionship. New York: American Book Company, 1953. Burgess, E. W., and Wallin, P. Engagement and Marrige. Philadelphia: Lippincott, 1953. Burr, Wesley R. "Satisfaction with Various Aspects of Marriage Over the Life Cycle: A Random Middle-Class Sample.” Journal g£_Marriage and the Family, Vol. 32, No. l, 1970, pp. 29-37. Caputo, D. V. "The Parents of the Schizophrenic." Family Process, Vol. 2, 1963, pp. 339-356. Carson, Robert. Interaction Concepts g£_Personality. Chicago: Aldine, 1969. Carter, L. F. "Evaluating the Performance of Individuals as Members of Small Groups." Personnel Psychology, Vol. 7, 1954, pp. 477-484. Charny, Israel W. "Marital Love and Hate." Familprrocess, V01. 8' NO. 1’ 1969, pp. 1-240 Cheek, F. "The 'Schizophrenic Mother' in Word and Deed." Family Process, Vol. 3, 1964a, pp. 155-177. . "A Serendipitous Finding: Sex Roles and Schizophrenia." Journal g£_Abnormal and Social Psychology, Vol. 69, 1964b, pp. 392-400. . "The Father of the Schizophrenic." Archives 9: General Psychiatry, Vol. 13, 1965, pp. 336-345. Clardy, E. R. "A Study of the Development and Course of Schizo- phrenia in Children." Psychiatric Quarterly, Vol. 25, 1951, pp. 81-90. 186 Coleman, James. Abnormal Psychology and Modern Life. Chicago: Scott, Foresman & Company, 1964. Cox, F. N. "An Assessment of Children's Attitudes Towards Parent Figures." Child Development, Vol. 33, 1962, pp. 821-830. Crandall, V.; Preston, A.; and Rabson, A. "Maternal Reactions and the Development of Independence and Achievement Behav- ior in Young Children." Child Development, Vol. 31, 1960, pp. 243-251. Crandon, V.; Katkovsky, W.; and Preston, A. "Motivational and Ability Determinants of Young Children's Intellectual Achievement Behaviors." Child Development, Vol. 33, 1962, pp. 643-661. Crandall, V.; Dewey R.; Katkovsky, W.; and Preston, A. "Parent's Attitudes and Behaviors and Grade School Child- ren's Academic Achievement." Journal g£_Genetic Peychology, Vol. 104, 1964, pp. 53-66. Cutler, B. R., and Dyer, W. G. "Initial Adjustment Processes in Young Married Couples." Social Forces, Vol. 44, 1965, pp. 195-201. Davidson, S. "Schol Phobia as a Manifestation of Family Dis- turbance: Its Structure and Treatment." Journal of Child Pyschology and Psychiatry, Vol. 1, 1961, pp. 270-287. Dentler, Robert A., and Pineo, Peter. "Sexual Adjustment, Marital Adjustment and Personality Growth of Husbands: A Panel Analysis." Marriage and Family Living, Vol. 22, 1960, pp. 45-48. Despert, Louise J. "Schizophrenia in Children." Psychiatric Quarterly, Vol. 12, 1938, pp. 366-371. Deutscher, Irwin. Married Life lg the Middle Years. Kansas City: Community Studies, 1959. DuHamel, T. R., and Jarmon, H. "Social Schemata of Emotionally Disturbed Boys and their Male Siblings." Journal g£_Con- sulting and Clinical Psychology, Vol. 36, 1971, pp. 281- 285. Eisenberg, L. "School Phobia: A Study in the Communication of Anxiety." American Journal g£_Psychiatry, Vol. 114, 1958, pp. 712-718. Ekman, P., and Friesen, W. V. "The Repertoir of Nonverbal Behavior: Categories, Origins, Usage and Coding." Semiotica, Vol. 1, 1969, pp. 49-98. 187 Ellison, E. A., and Hamilton, D. M. "The Hospital Treatment of Dementia Praecox, Part II." American Journal gfi Farina, A. "Patterns of Role Dominance and Conflicts in Parents of Schizophrenic Patients." Journal sf Abnormal and Social Psyhcology, Vol. 61, 1960, pp. 31-38. Farina, A., and Dunham, R. M. "Measurement of Family Relation- ships and Their Effects." Archives g£_General Psychology, V01. 9, 1963' pp. 64-73. Farina, A., and Holzberg, J. "Interaction Patterns of Parents and Hospitalized Sons Diagnosed as Schizophrenic and Non- schizophrenic." Journal sf Abnormal Psychology, Vol. 73, 1968, pp. 114-118. Ferreira, A.; Winter, W.; and Poindexter, E. "Some Interactional Variables in Normal and Abnormal Families." Family Process, Vol. 5, 1966, pp. 60-75. Fisher, R. "Social Schema of Normal and Disturbed School Children." Journal sf Educational Psychology, Vol. 58, 1967, pp. 82-92. Fisher, 5.; Boyd, 1.; Walker, D.; and Sheer, D. "Parents of Schizophrenics, Neurotics, and Normals." Archives 9: General Psychiatry, Vol. 1, 1959, pp. 149-167. Fleishman, E., and Harris, E. "Human Relations Training and the Stability of Leadership Patterns." Journal sf Applied Psychology, Vol. 39, 1955, pp. 20-25. Foa, U. C. "Convergences in the Analysis of the Structure of Interpersonal Behavior." Psychological Review, Vol. 68, 1961, pp. 341-353. Foa, U. C., and Foa, E. "Toward a Structural Theory of Inter- personal Communication." In Studies is_Dyadic Communica- tion, ed. A. W. Siegman and B. Pope. New York: Pergamon Press, 1972. Frank, G. H. "The Role of the Family in the Development of Psychopathology." Psychological Bulletin, Vol. 64, 1965, pp. 191-205. Frank, Lawrence. "Tactile Communication." Genetic Psychology Monographs, Vol. 56, 1957, pp. 209-255. Frazee, H. E. "Children Who Later Become Schizophrenic." Smith gollege Studies is Social Work, Vol. 23, 1953, pp. 125-149. 188 Frazer, James G. Folklore is the Old Testament, Vol. 2. New York: MacMillan Company, 1919. Freeman, R. V., and Grayson, H. M. "Maternal Attitudes in Schizophrenia." Journal g£_Abnormal Social Psychology, Vol. 50, 1955. PP. 45-52. Freidman, C., and Friedman, A. "Characteristics of Schizo- genic Families During a Joint Family Storytelling Task." Family Process, Vol. 9, 1970, pp. 333-354. Fromm-Reichmann, F. "Notes on the Development of Treatment of Schizophrenics by Psychoanalytic Psychotherapy." Psychiatry, Vol. 11, 1948, pp. 263-273. Gartner, Dorothy, and Goldstein, Harris. "Some Characteristics of Mothers of Severly Disturbed Children in a Therapeutic Nursery." Psychological Reports, Vol. 30, 1972, pp. 901-902. Geisman, L. L. "The Multi-Problem Family: Significance of Research Findings." In Social Welfare Forum. New York: Columbia University Press, 1960. ' Gerard, Donald L., and Siegel, Joseph. "The Family Background of Schizophrenia." Psychiatric Quarterly, Vol. 24, 1950, pp. 47-73. Gibb, Jack. "Defensive Communications." Journal sf Communica- Goldberg, S., and Lewis, M. "Play Behavior in the Year Old Infant: Early Sex Differences." Child Development, Vol. 40, 1969, pp. 21-32. Goode, William J. "Family Disorganization." In Contemporary Social Problems, ed. by R. K. Merton and R. A. Nisbet. New York: Harcourt Brace, 1961. Gouldner, Alvin. "The Norm of Reciprocity: A Preliminary Statement." American Sociological Review, Vol. 25, 1960, pp. 161-178. Gray, P. H. "Theory and Evidence of Imprinting in Human Infants." Journal gf Psychology, Vol. 46, 1958, pp. 155-166. Grebow, H. "The Relationship of Some Parental Variables to Achievement and Values in College Women." Journal g; Educational Research, Vol. 66, No. 5, 1973, pp. 203-209. Greenberg, Martin, and Shapiro, Solomon. "Indebtedness: An Adverse Aspect of Asking for and Receiving Help." Sociometry, Vol. 34, 1971, pp, 59-73, 189 Gurin, G.; Veroff, J.; and Feld, S. Americans View Their Mental Health. New York: Basic Books, 1960. Hamilton, Gilbert V. A. Research is_Marriage. New York: Boni, 1929. Harlow, H. F., and Harlow, M. K. "Social Deprivation in Monkeys.” Scientific American, Vol. 34, 1963, pp. 136-146. Hawkes, R. W., and Albert, R. S. "Social Psychological Dimensions of Mental Illness in the Family." Psychiatric “ Quarterly Suppliment, Vol. 36, 1962. “I Hawkins, James L. "Associations between Companionship, . Hostility and Marital SatisfactiOn." Journal g£_Marriage h and the Family, Vol. 30, 1968, pp. 6471330. 5“ Herzberg, F.; Mauserner; and Synderman, B. The Motivation ts Work. New York: John Wiley, 1959. Hess, R. D., and Handel, G. Family Worlds: A_Psychosocial Approach tg_Family Life. Chicago: University of Chicago Press, 1959. Hetherington, E.; Stouwie, R.; and Ridberg, E. "Patterns of Family Interaction and Child Rearing Attitudes Related to Three Dimensions of Juvenile Delinquency." Journal g£_ Abnormal Psychology, Vol. 78, 1971, pp. 160-176. Hewitt, L. E. and Jenkins, R. L. Fundamental Patterns g£_Mal- adjustment: The Dynamics g£_Their Origin. Springfield: State of Illinois, 1946. Hicks, Mary, W., and Platt, Marilyn. "Marital Happiness and Stability: A Review of the Research in the Sixties." Journal g£_Marriage and the Family. Nov. 1970, pp. 553-574. Hobhouse, L. T. Morals is EvOlution: s_Study is Comparative Ethics. London: Chapman and Hall, 1951. Hollander, J.; Duke, M.; and Nowicki, S. "Interpersonal Dis- tance: Sibling Structure and Parental Affection Antece- dents." Journal 9: Genetic Psychology, Vol. 123, 1973, pp. 35-45. Hollingshead, A. B., and Redlich, F. C. Social Class and Mental Illness. New York: John Wiley & Sons, 1958. Hollingshead, A. B. Two Factor Index sf Social Position. New Haven: Yale University, 1957. 190 Homans, George. The Human Grogp. New York: Harcourt, Brace and World, 1950. . "Social Behavior as Exchange." American Journal sf Sociology, Vol. LXII, 1958, pp. 597-606. . Social Behavior: Its Elementary Forms. New York: Harcourt, Brace & World, 1961. Horner, R. F. "Important Stumulus Variables in the Early Family Relationships of Schizophrenic Patients." Journal sf Clinical Psychology, Vol. 20, 1964, pp. 344-346. Hurvitz, Nathan. "The Marital Roles Inventory and the Measure- ment of Marital Adjustment." Journal sf Clinical Psycho- logy, Vol. 16, 1960, pP. 378-380. . "Control Roles, Marital Strain, Role Deviation, and Marital Adjustment." Journal sf Marriage and the Family, Vol. 27, 1965, pp. 29-31. Ingham, H. V. "A Statistical Study of Family Relationships in Psychoneurosis." American Journal sf Psychiatry, Vol. 106, 1949, pp. 91-98. Jackson, D., and Yalom, I. "Family Homeostasis and Patient Change." In Current Psychiatric Techniques. Ed. J. H. Masserman, Vol. 4, 1964. Jackson, D. "The Study of the Family." Family Process, Vol. 4, 1965, pp. 1-20. Jacob, Theodore. "Family Interaction in Disturbed and Normal Families: A Methodological and Substantive Review." Psychological Bulletin, Vol. 82, No. l, 1975, pp. 33-65. Jones, E. E. Ingratiation: 5 Social Psychological Analysis. New York: Appleton-Century-Crofts, 1964. Jourard, Sidney. "Self-Disclosure and Other Cathexis." Journal s£_Abnormal Psychology, Vol. 59, 1959, pp. 428-431. . "Age and Self-Disclosure." Merrill-Palmer Quarterly sf_Behavior and Development, Vol. 7, 1961, pp. 191-197. . "An Exploratory Study of Body-Accessibility." British Journal sf Social and Clinical Psychology, Vol. 5, 1966, pp. 221-231. Jourard, Sidney, and Lasakow, P. "Some Factors in Self-Disclosure." Journal sf Abnormal Social Psychology, Vol. 56, 1958, pp. 91-98. Jourard, Sidney, and Landsman, M. J. "Cognition, Cathexis, and the 'Dyadic Effect' in Men's Self-Disclosing Behavior." Merrill- Palmer Quarterly sf Behavior and Development, Vol. 6, 1960, pp. 178-186. 191 Jourard, S. M., and Richman, P. "Disclosure Output and Input in College Students." Merrill-Palmer Quarterly 2; Behavior and Development, Vol. 9, 1963, pp. 141-148. Jourard, Sidney, and Rubin, J. E. "Self-Disclosure and Touching: A Study of Two Modes of Interpersonal Encounter and Their Inter-Reaction." Journal sf Humanistic Psychology, Vol. 8, 1968, pp. 39-48. Kahn, Malcolm. "Nonverbal Communication and Marital Satisfaction." Family Process, Vol. 9, 1970, pp. 449-456. Katz, D.; Maccoby, N.; and Morse, N. Productiyity,Supervision and Morale is_ss Office Situation: Part l, Ann Arbor: Institute for Social Research, 1950. Katz, 0.; Maccoby, N.; Gurin, G.; and Floor, L. G. Productivity, Supervision and Morale Among Railroad WOrkers, Ann Arbor: Institute for Social Research, 1951. Katz, Irwin; Goldstein, J.; Cohen, M.; and Stucker, S. "Need, Satisfaction, Perception and Cooperative Interaction in Mar- ried Couples." Marriage and Family Living, Vol. 25, 1963, pp. 209-213. Kephart, William M. "Some Variables in Cases of Reported Sexual Maladjustment." Marriage and Family Livisg, Vol. 16, 1954, pp. 241-243. Knapp, M. Nonverbal Communication is Human Interaction. New York: Holt, Rinehart and Winston, 1972. Komarovsky, Mirra. Blue-Collar Marriage. New York: Random House, 1964. Kotlar, S. L. "Middle-Class Marital Role Perceptions and Marital Adjustment." Sociological and Social Research, Vol. 49, 1965, pp. 283-293. Keuthe, J. L. "Social Schemas." Journal sf Abnormal and Social Psychology, Vol. 64, 1962a, pp. 31-38. . "Social Schemas and the Reconstruction of Social Object Displays from Memory." Journal gf_Abnormal and Social Psycho- logy, Vol. 65, 1962b. pp. 71-74. . "The Pervasive Influence of Social Schemata." Journal cf Abnormal and Social PsychologY. Vol. 68, 1964, pp. 248-254. Keuthe, J. L., and Stricker, G. "Man and WOman: Social Schemata of Males and Females." Psychological Reports,Vbl.13,]363, pp. 655-661. Keuthe, J. L., and Weingartner, N. "Male-Female Schemata of Homo- sexual and Non Homosexual Penitentiary Inmates." Journal sf Personality, Vol. 32, 1964, pp. 23-31. 192 Lane, R. C., and Singer, J. L. "Familial Attitudes in Para- noid Schizophrenics and Normals from two Socioeconomic Classes." Journal 93 Abnormal and Social Psychology, Vol. 59, 1959, pP. 328-339. Lane, Richard O. "A Study of the Degree of Unhappiness or Unhappiness in Marriage." Unpublished Master's Thesis. University of Chicago, 1932. Leary, T. Interpersonal Diagnosis cg Personality. New York: Ronald Publishing, 1957. Lennard, H.; Beaulieu, M.; and Embry, N. "Interaction in Families with a Schizophrenic Child." Archives pg General Psychiatry, Vol. 12, 1965, pp. 166-183. Lerner, P. M. "Resolution of Intrafamilial Role Conflict in Families of Schizophrenic Patients. I: Thought Disturbance." Journal 2: Nervous and Mentachisease, Vol. 141, 1965, pp. 342-351. Levinger, G. "Instrumental and Expressive Functions in Marriage." (Working Paper No. 1), January 1963. . "Marital Cohesiveness and Dissolution: An Integrative Review." Journal pf Marriage and the Family, Vol. 27, 1965, pp. 19-28. . "Sources of Marital Dissatisfaction Among Applicants for Divorce." American Journal pf_Orthopsychiatry, Vol. 36, 1966, pp. 803-807. Levinger, G., and Senn, D. "Disclosure of Feelings in Marriage." Merrill-Palmer Quarterly g: Behavior and Development, Vol. 13, 1967, pp. 237-249. Levi-Strauss, Claude. The Elementary Structures g: Kinship. Boston: Beacon Press, 1969 (1949 Revision). Lewin, Kurt. A Dynamic Theory g: Personality. New York: McGraw-Hill, 1935. Lewis, B. Deprived Children. London: Oxford University Press, 1954. Lidz, R. W., and Lidz, T. "The Family Environment of Schizo- phrenic Patients." American Journal 9: Psychiatry, Vol. 106, 1949, pp. 332-345. Lidz, T.; Cornelison, A.; Fleck, S.; and Terry, D. "The Intra- Familial Environment of Schizophrenic Patients: II Marital Schism and Marital Skew." American Journal g£_Psychiatry, Vol. 114, 1957a, pp. 241-248. 193 . "The Intra-Familial Environment of the SchiZOphrenic Patient: I The Father." Psychiatry, Vol. 20, 1957b, pp. 329-342. Locke, Harvey. Predicting Adjustment llearriage: 5 Comparison of a Divorced and Happiiy Married Grosp. New York: Holt, I‘9’5I. Lorr, M., and McNair, D. M. "Expansion of the Interpersonal Behavior Circle." Journal of Personality and Social Psychology, Vol. 2, 1965, pp? 823-830. Luckey, Eleanore, B. "Marital Satisfaction and Its Association with Congruency of Perception." Marriage and Family Living, Vol. 22, 1960a, pp. 49-54. . "Marital Satisfaction and Parent Concepts." Journal 9: Consulting Psychology, Vol. 24, 1960b, pp. 195-204. . "Marital Satisfaction and Congruent Self-Spouse Con- cepts." Social Forces, Vol. 39, 1960c, pp. 153-157. . "Marital Satisfaction and Personality Correlates of Spouse." Journal g£_Marriage and the Family, Vol. 26, 1964, pp. 7-8. ‘ . "Number of Years Married as Related to Personality Perception and Marital Satisfaction." Journal sf Marriage and the Family, Vol. 28, 1966, pp. 44-48. Malinowski, Bronislaw. Argonauts g£_the Western Pacific. London: Routledge & Kegan, 1922. Mark, J. C. "The Attitudes of the Mothers of Male Schizophrenics Toward Child Behavior." Journal g£_Abnormal and Social Psychology, Vol. 48, 1953, pp. 185-189. Marlow, Roy H. "Development of Marital Dissatisfaction of Mormon College Couples over the Early Stages of the Family Life Cycle." Unpublished Masters Thesis, Brigham Young University, 1968. Mathews, V. C., and Milhanovich, C. S. "New Orientations on Marital Maladjustment." Marriage and Family Living, Vol. 25, 1963, pp. 300-304. Mauss, Marcel. The Gift. Translated by Ian Cunnison. New York: Free Press, 1954. McClelland, D. C.; Atkinson, J. W.; Clark, R. A.; and Lowell, E. L. The Achievement Motive. New York: Appleton- Century-Crofts, 1953. 194 McCord, W.; McCord, J.; and Howard, A. "Familial Correlates of Aggression in Nondelinquent Male Children." Journal g: Abnormal and Social PsycholOgY, V01. 62, No. 1, 1961, pp. 79-93. McCord, W.; Porta, J.; and McCord, J. "The Familial Genesis of Psychoses." Psychiatry, Vol. 25, 1062, pp. 60-71. Meyerowitz, J. H. "Satisfaction During Pregnancy." Journal pg Marriage and the Family, Vol. 32, 1970, pp. 38-42. Em: Miller, D., and Westman, J. C. "Family Teamwork and Psycho- therapy." Family Process, Vol. 5, 1966. Miller, H., and Baruch, D. "A Study of Hostility in Allergic Children." American Journal sf Orthopsychiatry, Vol. 20, 1950, pp. 506-519. Mishler, Elliot G., and Waxler, Nancy, E. Interaction ls Fami- lies: Ag Experimental Study 9: Family Processes and .1: Schizophrenia. New York: John Wiley & Sons, 1968. Montagu, M. F. Touching: The Human Significance cf the Skin. New York: Columbia University Press, 1971. Morris, G., and Wynne, L. "Schizophrenic Offspring and Parental Styles of Communication." Psychiatry, Vol. 28, 1965, pp. 19-44. Morris, H. H.; Escoll, P. J.; and Wexler, R. "Aggressive Behav- ior Disorders of Childhood: A Follow-up Study." American Journal sf Psychiatry, Vol. 112, 1956, pp. 991-997. Navran, Leslie. "Communication and Adjustment in Marriage." Family Process, Vol. 6, 1967, pp. 173-184. Newcomb,1L M- "An Approach to Communicative Acts." Psychological Review, Vol. 60, 1053, pp. 393-404. Newcomb, T- M., The Acquaintance Process. New York: Holt, Rinehart & Winston, 1961. - Newcomb, T. M., and Svehla, G. "Intra-Family Relationships in Attitudes." Sociometry, Vol. 1, 1937, pp. 180-205. Newell, H. W. "The Psycho-Dynamics of Maternal Rejection." American Journal sf Orthopsychiatry, Vol. 4, 1954, pp. 387- 401. . "A Further Study of Maternal Rejection." American Journal pg Othopsychiatry, Vol. 6, 1936, pp. 576-589. 195 Nye, Ivan F. "Maternal Employment and Marital Interaction: Some Contingent Conditions." Social Forces, Vol. 40, 1961, pp. 113-119. O'Connor, W., and Stachowiak, J. "Patterns of Interaction in Families with High Adjusted, Low Adjusted, and Mentally Retarded Members." Familerrocess, Vol. 10, 1971, pp. 229-241. Odom, Linda; Seeman, J.; and Newbrough, J. R. "A Study of Family Communication Patterns and Personality Integration in Children." Journal 9: Child Psychiatry and Human Devel- opment, Vol. 1, No. 4, 1971, pp. 275-285. Orden, S. R., and Bradburn, N. M. "Dimensions of Marriage Happiness." American Journal 2; Sociology, Vol. 73, 1968, pp. 715-731. ' Paris, Bethel L., and Luckey, Eleanore B. "A Longitudinal Study of Marital Satisfaction." Journal sf Sociology and Social Research, Vol. 50, 1966, pp. 212-213. Parsons, Talcott. The Social System. Glencoe: Free Press, 1951. Parsons, Talcott, and Bales, R., eds. Family, Socialization, and Interaction Process. Glencoe: Free Press, 1955. Pineo, Peter C. "Disenchantment in the Later Years of Marriage." Marriage and Famileriving. Vol. 23, 1961, p. 3-11. Plank, R. "The Family Constellation of a Group of SchiZOphrenic Patients." American Journal pg Orthopsychiatry, Vol. 23, 1953, pp. 817-825. Provence, S., and Lipton, R.C. Infants ls Institutions. New York: International University Press, 1962. Rausch, Harold; Barry, W.; Hertel, R.; and Swain, M. A. Commu- nication, Conflict and Marriage. San Francisco: Jossey- Bass, 1974. Renne, Karen S. "Correlates of Dissatsifaction in Marriage." Journal p: Marriage and the Family, Vol. 32, No. l, 1970, pp. 54-67. Riskin, J., and Faunce, E. "Family Interaction Scales: III, Discussion of Methodology and Substantive Findings." Archives pf General Psychiatry, Vol. 22, 1970, pp. 526-537. Rogers, L. E., and Farace, R. V. "Analysis of Relational Commu- nication in Dyads: New Measurement Procedures." Human Communication Research, Vol. 1, 1975, pp. 222-239. 196 Rollins, B. C., and Feldman, H. "Marital Satisfaction Over the Family Life Cycle." Journal sf_Marriage and the Family, Vol. 32, No. l, 1970, pp. 20-28. Rosen, B., and Andrade, R. "The Psychosocial Origins of Achieve- ment Motivation." Sociometry, Vol. 20, 1959, pp. 185-218. Schaefer, E. S. "Converging Concpetual Models for Maternal Behavior and for Children's Behavior." In Parental Atti- tudes and Child Development, ed. Glidewell, 1961, pp. 124-146. Schaefer, E. S., and Bayley, N. "Maternal Behavior, Child Behavior, and their Intercorrelations from Infancy through Adolescence." Monographs g: the Society for Research ls Child Development, Vol. 28, No. 87, 1963, pp. 1-127. Schulman, R. E.; Shoemaker, D. J.; and Moelis, I. "Laboratory Measurement of Parental Behavior." Journal p£_Consulting Psychology, Vol. 26, 1962, pp. 109-114. Schutz, William. Firo: §_Three-Dimensiona1 Theory g: Inter- personal Behavior. New York: Holt, Rinehart & Winston, 1958. Sears, R. R.; Whiting, J. W. M.; Nowlis, V.; and Sears, P. S. "Some Child-Rearing Antecedents of Aggression and Depend- ency in Young Children." Genetic Psychology Monographs, Vol. 47, 1953, pP. 135-236. Silberpfennig, Judith. "Mother Types Encounterd in Child Guidance Clinics." American Journal sf Orthopsychiatry, Vol. 11, 1941, pp. 475-484. Singer, J. L. "Projected Familial Attitudes as a Function of Socioeconomic Status and Psychopathology." Journal g: Consulting PsychologY. Vol. 18, 1954, pp. 99-104. Slater, P. E. "Parent Behavior and the Personality of the Child." Journal sf Genetic Psychology, Vol. 101, 1962, pp. 53-68. Sperling, M. "The Role of the Mother in Psychosomatic Disorders in Children." Psychosomatic Medicine, Vol. 11, 1949, pp. 377-385. Spiegel, J. P. "The Resolution of Role Conflict Within the Family." In 5 Modern Introduction pg the Family. Ed. N. W. Bell and E. F. Vogel. Glenco: The Free Press, 1968. Spitz, R. "Hospitalism: Genesis of Psychiatric Conditions in Early Childhood." Psychoanalytic Study sf the Child, Vol. 1, 1945. PP. 53-74. 197 Stabenau, J.; Turpin, J.; Werner, M.; and Pollin, W. "A Compara- tive Study of Families of Schizophrenics, Delinquents, and Normals." Psychiatry, Vol. 28, 1965, pp. 45-59. Stuckert, R. P. "Role Perception and Marital Satisfaction - A Configurational Approach." Marriage and Family Living, Vol. 25, 1963, pp. 415-419. Sullivan, Harry S. The Interpersonal Theory pf Psychiatry. Eds. Perry and Gawell. New York: W. W. Norton and Company, 1953. Swensen, Clifford. Introduction pg Interpersonal Relations. Glenview: Scott, Foresman and Company, 1973. Talbot, Mira. "School Phobia: A Workshop: I: Panic in School Phobia." American Journal 2: Orthopsychiatry, Vol. 27, 1957, pp. 286-295. Taylor, A. B. "Role Perception, Empathy, and Marriage Adjustment." Sociological and Social Research, Vol. 52, 1967, pp. 22-34. Taylor, D. "The Development of Interpersonal Relationships: Social Penetration Processes." Journal 21: Social Psychology, Vol. 75, 1968, pp. 79-90. Terman, Lewis, M. Psychological Factors ip_Marital Happiness. New York: McGraw-Hill, 1938. Thibaut, John, and Kelley, Harold. The Social Psychologyg§ Groups. New York: John Wiley & Sons, 1959. Thurwald, Richard. Economics is Primitive Communities. London: Oxford University Press, 1932. Tietze, Trude. "A Study of Mothers of Schizophrenic Patients." Psychiatry, Vol. 12, 1949, PP. 55-65. Turner, James L.; Foa, E.; and Foa, U. "Interpersonal Rein- forcers: Classification, Interrelationship and Some Differential Properties." Journal g£_Personallty.spg Social Psychology, Vol. 19, No. 2, 1971, pp. 170-177. Turner, Jonathon H. The Structure gr Sociological Theory. Homewood: Dorsey Press, 1974. Villard, Kenneth L., and Whipple, Leland J. Beginnings is Relational Communication. New York: John Wiley & Son, 1976. Wahl, C. W. "Some Antecedent Factors in the Family Histories of 392 Schizophrenics." American Journal 9: Psychiatry, Vol. 110, 1954, pp. 668-676. 198 . "Some Antecedent Factors in the Family Histories of 568 Male Schizophrenics of the United States Navy." American Journal gngsychiatry, Vol. 113, 1956, pp. 201-210. Watson, G. A. "A Comparison of the Effects of Lax Versus Strict Home Training." Journal 9: Social Psychology, Vol. 5, 1934, pp. 102-105. Watzlawick, P.; Beavin, A. B.; and Jackson, D.D. Pragmatics 9: Human Communication. Palo Alto: Norton and Company, 1967. Weinstein, Laura. "Social Schemata of Emotionally Disturbed Boys." Journal of Abnormal Psychology, Vol.70, No. 6, 1965, pp. 457- 461. Weinstein, W. A., and Deutschberger, Paul. "Tasks, Bargains and Identities in Social Interaction." Social Forces, April 1963, pp. 451-456. Westley, William A., and Epstein, Nathan B. "Family Structure and Emotional Health: A Case Study Approach." Marriage and Family Living, Vol. 22, 1960, pp. 25-27. ' Wilson, M. J. "Grandmother, Mother, and Daughter in Cases of School Phobia." Smith College Studies Ln Social Work, Vol. 25, 1955, pp. 56- 57. Winter, W. and Ferreira, A. "Interaction Process Analysis of Family Decision Making." Family Process, Vol. 6, 1967, pp. 155-172. Winter, W.; Ferreira, A.; and Olson, J. "Hostility Themes in the Family TAT. " Journal of Projective Techniques and Personality Assessment, Vol.30, 1966, pp. 270- 274. Winterbottom, M. "The Relation of Need for Achievement in Learn- ing Experiences in Independence and Mastery." In J. W. Atkinson (Ed. ) Motives Ln Fantasy, Action and Society. New York: Van Nostrand, 1958, 453-478. Worthy, W.; Gary, A.; and Kahn, G. "Self- Disclosure as an Exchange Process." Journal of Personality and Social Psychology, Vol. 13,1966, pp. 59-63. Wynne, L. C.; Ryckoff, I.M.; Day, J.; and Hirsch, S. I. "Pseudo- mutuality in the Family Relations of Schizophrenics." Psychiatry, Vol. 21, 1958, pp. 205-220. 199 ‘Wynne, L. C., and Singer, M. T. "Thought Disorder and Family Relations of Schizophrenics: II, Classification of Forms of Thinking." Archives 9; General Psychiatry, Vol. 9, 1963a, pp. 199-206. . "Thought Disorder and Family Relations of SchiZOphrenics: I, A Research Strategy." Archives g£_General Psychiatry, Yarrow, L. J. "Maternal Deprivation: Toward an Empirical and Conceptual Re-evaluation." Psychological Bulletin, vol. 58, 1961, pp. 459-490. Young, M. "Marital Counseling with Affection Deprived Spouses." Social Casework, Vol. 47, 1966, pp. 571-574. MICHIGAN sms UNIV. LIBRARIES 11111111111 "I1111111111WWIWMIIUWI 31293100135924