FAMILY PARADIGMS AND HUMAN EMOTIONS By Lori A. Hoisington A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Family Studies 2011 ABSTRACT FAMILY PARADIGMS AND HUMAN EMOTIONS By Lori A. Hoisington The primary objective of this research was to explore the relationship between family paradigms and human emotions. The research tested the relationship between closed paradigm vs. random paradigm and positive affect vs. negative affect. As a secondary objective, the research also extended the analysis to include open paradigm and synchronous paradigm. The closed family reflects stability through tradition and focuses on the past. Relationships are cohesive with a strong sense of belonging. The family is group-oriented. Boundaries prohibit information from freely entering or exiting the family. The random family lives for today and values freedom and independence; the individual always comes first. This family often appears chaotic to other paradigms and is discontinuityoriented as it seeks change and new ideas. The family theme supports innovation but not hierarchy. The open family orients to the past, present and future with balance between continuity and change. The family is consequence-oriented with flattened hierarchy and values both the individual and the group. Consensus occurs through communication and boundaries are semipermeable. The synchronous family is a harmonious system that operates on timelessness with no visible hierarchy. Members share consensus without communication through a special way of knowing. This family values individuality but provides stability with rigid system boundaries. The current study occurred at a single-site location throughout four phases of data collection; the first two phases comprised the pilot study and the latter two phases comprised the working study. Participants (N=202) were college students in a Midwest University (primarily 20 – 22 years old). Demographics were collected using two surveys and the research implemented four revised versions of the Relational Paradigmatic Assessment Scale (RPAS) for collection of paradigm and emotions data. The emotions data were coded according to the Circumplex Model to produce measures of valence and arousal for each emotion word. Dialectical logic served as the framework for the study and established a system of opposites (e.g. closed paradigm vs. random paradigm and positive affect vs. negative affect). Analyses were conducted with bivariate correlation (Phase III and Phase III/IV combined), ordinary least squares analysis (Phase IV) and ordinal regression analysis (Phase III/IV combined). Results of the study were interpreted according to Kantor and Lehr’s Distance Regulation Perspective. Findings supported use of the distance regulation model in family paradigms research and further suggested that, in its current state, the model does not adequately consider emotions that accompany change in family structure. Findings addressed the research question, is there a relationship between family paradigms and human emotions? Results showed negative correlation between cohesive paradigms (closed and synchronous) vs. distant paradigms (random and open). In addition, results also indicated that closed-type individuals respond least favorably to alternate paradigms (closed, open or synchronous) and open-type individuals respond most favorably to alternate paradigms (closed, random or synchronous). Moderating effects were revealed for education, relationship and religion. Gender served as a control variable. Results are applicable toward assessments of families undergoing system change. Copyright by LORI A. HOISINGTON 2011 To my husband Carl and Dr. Tom Luster You both should have been here for this day v ACKNOWLEDGMENTS It’s difficult to find words to express the gratitude I feel for the support I received toward completion of my dissertation. When I began this journey nearly five years ago, I never imagined that I would complete the project without all the members of my support system. Two of my key supporters – my husband, Carl and Committee member, Dr. Tom Luster – provided guidance that helped to establish my research and set me on the path to success. Sadly, they are not with us to celebrate its completion. While we may not understand such losses, their influences and ideas continue to live through my research and my professional career. I am deeply grateful to my Committee Chair, Dr. Adrian Blow, for his support, patience and encouragement during the difficult times – to Dr. David Imig for continuous guidance throughout data collection, analysis and interpretation – to Dr. Francisco Villarruel for support, constant guidance and expertise following Dr. Luster’s passing – to Dr. Kevin Berger for helping me to maintain my professional vision and encouraging me to continue during the tough times – and to Dr. Laura Symonds for providing support and encouragement and leading me to the resources I needed for my research. I would also like to thank Dr. Karen Wampler and the faculty and staff of HDFS for continual support and encouragement throughout my research. Additionally, I am grateful to Jason Huang (CSTAT) for his generosity and diligence in assisting me with data analysis. Special acknowledgements to Dr. Jean Davis Schlater, Dr. Verna Lee Hildebrand, the family of Dr. Beatrice Paolucci and the Graduate College at Michigan State University for supporting my research through Dissertation Completion Fellowships. Each project represents collective efforts by individuals with a common vision. I am grateful to each of you for sharing this vision with me. vi TABLE OF CONTENTS LIST OF TABLES .................................................................................................... viii LIST OF FIGURES ................................................................................................... xii CHAPTER 1 INTRODUCTION .................................................................................................... Family Systems ............................................................................................. Closed Paradigm ................................................................................. Random Paradigm .............................................................................. Open Paradigm ................................................................................... Synchronous Paradigm ....................................................................... Assumptions ....................................................................................... Family Systems in Transition: The Role of Emotions .................................. Decision-Making and Emotions ......................................................... Identification of the Problem ........................................................................ Purpose of the Study ..................................................................................... Significance of the Study .............................................................................. Research Hypotheses ..................................................................................... Primary Hypothesis ........................................................................... Secondary Hypotheses ...................................................................... 1 2 4 5 7 8 9 10 10 12 13 14 15 15 15 CHAPTER 2 LITERATURE REVIEW .......................................................................................... Family Process Theory ................................................................................. Subsystems ........................................................................................ Access and Target Dimensions ......................................................... Access Dimensions ............................................................... Target Dimensions ................................................................ Player Parts ....................................................................................... Dialectical Logic ........................................................................................... Distance Regulation Model............................................................................ Human Emotions ........................................................................................... Emotions in Research ................................................................................... 18 19 19 20 20 21 23 25 27 30 32 CHAPTER 3 METHODOLOGY .................................................................................................... Overview ....................................................................................................... Procedure ...................................................................................................... Design ........................................................................................................... Participants .................................................................................................... Pilot Study: Phase I and Phase II ...................................................... Working Study: Phase III and Phase IV ........................................... 34 34 35 35 37 37 38 vii Phase III ................................................................................ Phase IV ................................................................................ Phase III and Phase IV Combined ......................................... Conceptual Definitions .................................................................................. Independent Variable ......................................................................... Dependent Variable ........................................................................... Moderating Variables......................................................................... Control Variables ............................................................................... Inclusion Criteria .......................................................................................... Exclusion Criteria .......................................................................................... Instruments ..................................................................................................... Demographic Survey ......................................................................... Relational Paradigmatic Assessment Scale (RPAS) .......................... Standard RPAS .................................................................................. Revised RPAS.................................................................................... RPAS-1 and RPAS-2: Pilot Phase ......................................... RPAS-3 and RPAS-4: Working Study .................................. Positive and Negative Affect Schedule (PANAS) ............................. Coding ........................................................................................................... Research Hypotheses ..................................................................................... Primary Hypothesis ............................................................................ Secondary Hypotheses ....................................................................... Data Analysis ................................................................................................ Univariate Analyses .......................................................................... Bivariate Analyses ............................................................................ Multivariate Analyses ....................................................................... CHAPTER 4 RESULTS .................................................................................................................. Phase IV Analyses ........................................................................................ Univariate Analyses for Phase IV ...................................................... Bivariate Analyses for Phase IV ........................................................ Correlates for Closed Paradigm, Random Paradigm, Open Paradigm, Synchronous Paradigm and Political Orientation for Phase IV ........................................................................... Multivariate Analyses for Phase IV ................................................... Findings Addressing Research Hypotheses ........................... Findings Related to Primary Hypothesis .................. Findings Related to Secondary Hypotheses .............. Phase III/IV Combined Analyses .................................................................. Univariate Analyses for Phase III/IV Combined ............................... Bivariate Analyses for Phase III/IV Combined ................................. Correlates for Closed Paradigm, Random Paradigm, Open Paradigm, Synchronous Paradigm and Political Orientation . Multivariate Analyses for Phase III/IV Combined ............................ Emotions: Valence and Arousal............................................. viii 38 40 40 41 41 41 41 42 42 42 42 42 43 43 44 45 46 48 49 50 50 50 51 51 53 53 55 56 57 59 59 62 63 64 64 84 85 87 88 89 89 Summary ....................................................................................................... Phase IV Bivariate Correlation ......................................................... Phase IV Ordinary Least Squares Analyses ...................................... Phase III/IV Combined Bivariate Correlation .................................. Phase III/IV Combined Ordinal Regression Analysis ...................... 119 119 119 120 120 CHAPTER 5 DISCUSSION ............................................................................................................ Results ........................................................................................................... Distance Regulation and Emotions ................................................... Addressing the H01 Null Hypothesis: Main Effects.......................... Addressing the H01 Null Hypothesis: Interaction Effects ................. Main Effects for the Control Variable: Gender ................................. Addressing Secondary Hypotheses .................................................... Education ............................................................................... Relation .................................................................................. Religion .................................................................................. Political Orientation ............................................................... Interaction Effects .............................................................................. Applications Toward Paradigmatic Transition ............................................. Closed Paradigm ................................................................................ Random Paradigm .............................................................................. Open Paradigm................................................................................... Synchronous Paradigm ...................................................................... Study Limitations ........................................................................................... Implications for Future Research .................................................................. fMRI as a Research Tool Toward Cognitive Social Science ............. fMRI and Human Emotions ............................................................... Proposed fMRI Research ................................................................... Conclusions ................................................................................................... 123 124 126 128 134 136 139 139 140 140 141 144 145 148 148 149 149 151 152 152 153 154 157 APPENDICES Appendix A: Conceptual and Operational Definitions ................................. Appendix B: Instrument Review – Relational Paradigmatic Assessment Scale (RPAS) ...................................................... Appendix C: Instrument Review – Positive and Negative Affect Schedule (PANAS).................................................................. Appendix D: Demographic Survey #1 .......................................................... Appendix E: Demographic Survey #2 .......................................................... Appendix F: RPAS-1 – Phase I .................................................................... Appendix G: RPAS-2 – Phase II ................................................................... Appendix H: RPAS-3 – Phase III ................................................................. Appendix I: RPAS-4 – Phase IV – Sample Question #1 ............................. Appendix J: Positive and Negative Affect Schedule (PANAS) .................. Appendix K: Revised RPAS Calculations .................................................... ix 160 164 168 172 175 179 183 189 196 206 208 Appendix L: Emotions Survey for Categorizing Phase III/IV Combined Emotions .................................................................................. Appendix M: Guidelines for Categorizing Emotions for Phase III/IV Combined ................................................................................ Appendix N: Guidelines for Categorizing Emotions for Phase IV (PANAS) ............................................................................... Appendix O: The Human Brain .................................................................. BIBLIOGRAPHY 212 220 229 232 .............................................................................................. 242 x LIST OF TABLES Table 1: Mechanisms and Submechanisms for Access Dimensions ....................... 22 Table 2: Mechanisms for Target Dimensions ......................................................... 23 Table 3: Emotions Words Included in Phase III of Development .......................... 39 Table 4: Descriptive Statistics for Categorical Data in Phase IV (PANAS) (N=59) ....................................................................................................... 58 Table 5: Descriptive Statistics for Continuous Data in Phase IV (PANAS) (N=59) ....................................................................................................... 59 Table 6: Correlation Between Closed, Random, Open and Synchronous Paradigm Scores and Political Orientation for Phase IV (PANAS) Using Pearson’s (N=59) ............................................................................ 61 Table 7: Exploratory Ordinary Least Squares (OLS) Regression of the likelihood of Closed Positive Valence, Random Positive Valence, Closed Negative Valence and Random Negative Valence with Closed and Random Paradigm Predictors Included in the Model and Gender Controlled (N=59) ..................................................................................... 64 Table 8: Exploratory Ordinary Least Squares (OLS) Regression Model of the Likelihood of Random Positive Valence with Closed Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled (N=59) .................................................................................... 66 Table 9: Exploratory Ordinary Least Squares (OLS) Regression Model of the Likelihood of Open Positive Valence, Synchronous Positive Valence, Open Negative Valence, Synchronous Negative Valence, with Closed and Random Paradigm Predictors Included in the Model and Gender Controlled (N=59) .................................................................................... 68 Table 10: Exploratory Ordinary Least Squares (OLS) Regression Model of the Likelihood of Open Positive Valence with Random Paradigm Predictor and Moderating Variables Education, Relationship, Religion, and Political Orientation Included in the Model with Gender Controlled (N=59) .................................................................................... 70 xi Table 11: Exploratory Ordinary Least Squares (OLS) Regression Model of the Likelihood of Synchronous Positive Valence with Random Paradigm Predictor and Moderating Variables Education, Relationship, Religion and Political Orientation Included in the Model with Gender Controlled (N=59) .................................................................................... 72 Table 12: Exploratory Ordinary Least Squares Regression (OLS) of the Likelihood of Closed Positive Valence, Random Positive Valence, Open Positive Valence, Synchronous Positive Valence, Closed Negative Valence, Random Negative Valence, Open Negative Valence and Synchronous Negative Valence with Random Paradigm, Open Paradigm and Synchronous Paradigm Predictors Included in the Model and Gender Controlled. (N=59) .............................................. 74 Table 13: Exploratory Ordinary Least Squares (OLS) Regression Model of the Likelihood of Random Positive Valence with Synchronous Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled. (N=59) .................................................................................. 77 Table 14: Exploratory Ordinary Least Squares (OLS) Regression Model of the Likelihood of Synchronous Positive Valence with Synchronous Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled. (N=59) ................................................................................... 78 Table 15: Exploratory Ordinary Least Squares (OLS) Regression of the Likelihood of Closed Positive Valence, Random Positive Valence, Open Positive Valence, Synchronous Positive Valence, Closed Negative Valence, Random Negative Valence, Open Negative Valence and Synchronous Negative Valence with Closed Paradigm, Open Paradigm and Synchronous Paradigm Predictors Included in the Model and Gender Controlled. (N=59) .................................................... 81 Table 16: Exploratory Ordinary Least Squares (OLS) Regression Model of the Likelihood of Open Positive Valence with Open Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled. (N=59) ................................................................................... 83 Table 17: Descriptive Statistics for Categorical Data in Phase III/IV Combined ..... 86 Table 18: Descriptive Statistics for Interval Data in Phase III/IV Combined (N=202) ..................................................................................................... 87 xii Table 19: Correlation Between Closed, Random, Open and Synchronous Paradigm Scores and Political Orientation for Phase III/IV Combined Using Pearson’s (N=202) ......................................................................... 88 Table 20: Exploratory Ordinal Regression of the Likelihood of Random Valence, Closed Valence, Random Arousal and Closed Arousal Responses with Closed Paradigm and Random Paradigm Predictors Included in the Model and Gender Controlled. (N=202) .................................................. 90 Table 21: Exploratory Ordinal Regression Model of the Likelihood of Random Valence with Random Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled.. (N=202) .......................................... 92 Table 22: Exploratory Ordinal Regression of the Likelihood of Open Valence, Synchronous Valence, Open Arousal and Synchronous Arousal Responses with Closed Paradigm and Random Paradigm Predictors Included in the Models and Gender Controlled. (N=202) ....................... 93 Table 23: Exploratory Ordinal Regression Model of the Likelihood of Open Valence with Closed Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled. (N=202) ....................................... 96 Table 24: Exploratory Ordinal Regression Model of the Likelihood of Synchronous Valence with Closed Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled. (N=202) .... 97 Table 25: Exploratory Multiple Regression of the Likelihood of Closed Valence, Random Valence, Open Valence, Synchronous Valence, Closed Arousal, Random Arousal, Open Arousal and Synchronous Arousal Responses with Random Paradigm, Open Paradigm and Synchronous Paradigm Predictors Included in the Models and Gender Controlled. (N=202) .................................................................................................... 98 Table 26: Exploratory Ordinal Regression Model of the Likelihood of Random Valence with Open Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled. (N=202) ........................................... 103 Table 27: Exploratory Ordinal Regression Model of the Likelihood of Open Valence with Open Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled. (N=202) ........................................... 105 xiii Table 28: Exploratory Ordinal Regression Model of the Likelihood of Synchronous Valence with Synchronous Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled. (N=202) .... 106 Table 29: Exploratory Ordinal Regression Model of the Likelihood of Random Arousal with Random Paradigm Predictor and Moderating Variables Education, Relation, Religion and Political Orientation Included in the Model with Gender Controlled. (N=202) ........................................... 107 Table 30: Exploratory Ordinal Regression Model of the Likelihood of Closed Valence, Random Valence, Open Valence, Synchronous Valence, Closed Arousal, Random Arousal, Open Arousal and Synchronous Arousal Responses with Closed Paradigm, Open Paradigm and Synchronous Paradigm Predictors Included in the Models and Gender Controlled. (N=202) .................................................................... 109 Table 31: Summary of Findings from Ordinary Least Squares (OLS) Regression Analysis of Phase IV Data ....................................................................... 113 Table 32: Summary of Findings from Ordinal Regression Analysis of Phase III/IV Combined Data ......................................................................................... 116 Table 33: Conceptual and Operational Definitions .................................................. 161 xiv LIST OF FIGURES Figure 1: Figure 2: Constantine’s Dialectic (Quadruplex) Model (Based on Constantine’s Model of Unified Family Process Theory) ........................ 4 Conceptual Model for Phase III and Phase IV Depicting Inclusion of Question 5 .......................................................................................... 37 Figure 3: Conceptual Model for Phase III and Phase IV Combined ..................... 47 Figure 4: Conceptual Model for Phase IV ............................................................. 48 Figure 5: Interaction Effect of Random Paradigm and Relationship on Open Positive Affect with Gender Controlled ....................................... 71 Interaction Effect of Random Paradigm and Education on Synchronous Positive Affect with Gender Controlled ........................... 73 Interaction Effect of Random Paradigm and Open Paradigm on Random Positive Affect with Gender Controlled .................................. 76 Interaction Effect of Random Paradigm and Open Paradigm on Positive Random Emotions .................................................................... 80 Figure 6: Figure 7: Figure 8: Figure 9: Interaction Effect of Closed Paradigm and Random Paradigm on Open Valence with Gender Controlled .................................................. 95 Figure 10: Interaction Effect of Random Paradigm and Open Paradigm on Closed Arousal with Gender Controlled ................................................ 101 Figure 11: Interaction Effect of Open Paradigm and Synchronous Paradigm on Open Valence with Gender Controlled .................................................. 102 Figure 12: Interaction Effect of Open Paradigm and Religion on Random Valence with Gender Controlled ............................................................ 104 Figure 13: Interaction Effect of Closed Paradigm and Open Paradigm on Synchronous Arousal with Gender Controlled ...................................... 112 Figure 14: Main Effects for Phase IV (PANAS) Analysis ...................................... 132 Figure 15: Main Effects for Phase III/Phase IV (Combined) Analysis ................... 133 Figure 16: Main Effects for Gender in Phase IV Analyses ..................................... 138 xv Figure 17: Main Effects for Gender in Phase III/Phase IV Combined Analyses .... 138 Figure 18: Main Effects for Moderating Variables in Phase IV Analyses .............. 142 Figure 19: Main Effects for Moderating Variables in Phase III/Phase IV Combined Analyses .............................................................................. 143 Figure 20: Magnetic Resonance Image of the Brain Depicting Cerebrum Brainstem and Cerebellum ..................................................................... 234 Figure 21: Magnetic Resonance Image Showing the Cerebrum, Brainstem and Cerebellum ............................................................................................. 236 Figure 22: Magnetic Resonance Image Showing the Temporal Lobe .................... 236 xvi CHAPTER 1 INTRODUCTION Anthropologist Gregory Bateson, who was well known for bringing cybernetics to anthropology, first applied General Systems Theory to family systems during the 1940s (Ingold, 2000; Marcus, 1985; Nuckolls, 1995). His teachings suggested that family systems are comprised of interconnected parts that have bidirectional relationships with other family systems and the surrounding environment. Bateson introduced a new way for social scientists to view interpersonal relationships within the family, an approach that later served as a cornerstone for family therapy (Krause, 2007). Many of Bateson’s ideas emanated from his observations of the naven ritual practiced by the Iatmul people while in New Guinea (Krause, 2007). In this context Bateson developed an interpersonal theory of emotion with emphasis on “social construction of emotion categories” that relied upon “dialectical conflict and its dynamic principles” (Nuckolls, 1995, pp. 370-371). Bateson described emotions among males in the tribe as being centered on individualistic pride painted with competition and flamboyance while he saw Iatmul women reflecting a quiet and cooperative affect (Krause, 2007). Bateson described this mutually reinforcing system of opposites as schismogenesis (Nuckolls, 1995, p. 372). The dynamics that accompany this cultural structure enable the opposing systems to constrain each other when either system approaches excessive differentiation, thus preserving the cultural homeostasis. Bateson concluded from his study of the Iatmul people that emotions and the ethos they represented were central loci for expressing homeostatic balance (Nuckolls, 1995, p. 375). 1 Prior to the 1960s, family therapists and researchers focused primarily on pathological families in their attempt to derive better understanding about how families functioned (Bateson, Jackson, Haley & Weakland, 1956; Haley, 1959; Handel, 1967; Vogel & Bell, 1960; Wynne, Ryckoff, Day & Hirsch, 1958). While this approach generated valuable information toward understanding families, application of the information was limited by its narrow focus on behavioral characteristics of a single family member (Kantor & Lehr, 1975). These findings did not significantly increase knowledge about normal family function from a whole-family perspective. Kantor and Lehr addressed this shortfall in their pioneering work, Inside the Family (1975). Based on information Kantor and Lehr collected within the family’s natural setting, they described family life as a goal-seeking system, or process, that revolves around actual and metaphorical space. “How does a family set up and maintain its territory? How does it regulate distance among its own members?” (Kantor & Lehr, 1975, p. 7). Kantor and Lehr borrowed three key concepts from general systems theory—systems, feedback control and strategies—to explain family process. They described family systems as “…organizationally complex, open, adaptive, and information processing systems” (1975, p. 10). They further suggested that family systems maintain feedback control through feedback loops that determine the outcome of family communication and interaction (1975, p. 12). They also explained that family systems form strategies—rrecurring patterns of interaction—to help regulate and shape relationships among family members (1975, p. 15). Family Systems Family System typology describes family worldviews, or different ways of knowing. One system is neither more valid nor more desirable than another; it is just different from the others. 2 Kantor and Lehr described three types of family systems - closed, random and open (1975). Constantine expanded on Kantor and Lehr’s theory to include the synchronous system (1986). Constantine’s theoretical contribution produced a quadruplex model that is compatible with other four-fold typologies such as Leary’s Interpersonal Circumplex Model (1957), Olson, Sprenkle and Russell’s Circumplex Model of Marital and Family Systems (1979) and Beavers’ Centripetal/ Centrifugal Model (1981). The quadruplex model also enabled integration of the four family types (closed, random, open and synchronous) with Kantor and Lehr’s player parts (mover, follower, opposer and bystander) (1975) allowing for application of the theory in family intervention and industry. Constantine (1986) expanded on Bateson’s earlier ideas about the dynamic principles of dialectical conflict and argued that Family Paradigms reflects dialectical logic, a philosophy that considers alternative ways of viewing the world. To illustrate, Constantine proposed that the closed system is the Thesis. The random system represents the opposite way of thinking, or the Antithesis. The open family system is a combination, or Synthesis of the closed and random systems and the synchronous system (Constantine’s claim to fame) is the Antisynthesis. By proposing alternate ways to view the world, Family Paradigms theory suggests that family systems can shift to alternate types of systems during times of severe stress or crisis. 3 Thesis (Group/Continuity) Closed Synthesis Antisynthesis (Group/Change) (Individual/Continuity) Open Synchronous Antithesis (Individual-Change) Random Figure 1 Constantine’s Dialectic (Quadruplex) Model (Based on Constantine’s Model of Unified Family Process Theory). Adapted from “Family Paradigms, Interpersonal Relationships & Family Systems” by D. R. Imig, 2005, p. 10). Adapted with permission Closed Paradigm The closed family reflects stability through tradition and members value time-tested ideas. This family commonly focuses on the past. Family members are cohesive and maintain very private relationships with loyalty and a strong sense of belonging. The closed family frequently operates under authoritative rule with the father generally in charge. The family is group-oriented. If conflict arises between the family group and an individual member, the group always comes first. Family secrets are quite common in this structure as the relational boundaries are structured and prohibit information from freely entering or exiting the family system (Constantine, 1986, 1993; Imig, 2005; Kantor & Lehr, 1975). 4 The enabled closed family is the picture portrayed by the “Leave it to Beaver” era. This is portrayed as a two-parent family with children, a three-bedroom home and a white picket fence. This structure provides stability for family members with a clear chain-of-command and most of the family’s needs and goals are met (Constantine, 1986; Imig, 2005). Once a member is in, he or she is always in. Family members share appropriate meaning and emotions with each other through a balanced feedback mechanism. This balanced process of sharing is called coupling (Constantine, 1986, p. 196). In addition to stability, the closed system is also highly efficient. Since one person can make instant decisions and delegate tasks, the closed system is quite efficient at carrying out plans (Constantine, 1986; Imig, 2005; Kantor & Lehr, 1975). The military and police operate on the closed model (Imig, 2005). The disabled closed family is inflexible and over-involved. Members of this family express high levels of frustration. This sometimes occurs when children go through adolescence and seek independence (Constantine, 1993). In times of crisis, the natural tendency of families is to do more of the same (Imig, 2005). Under severe stress, the closed family tries to enforce mechanisms that have worked in the past. The family responds by tightening its boundaries and increasing authority until the leadership becomes tyrannical. This only exaggerates the problem and leaves the family so tightly bound that family members become enmeshed (Constantine, 1986; Imig, 2005). Families in this state tend to avoid getting help from the outside. Lack of contact with the outside world only exacerbates the problem. Intervention for the rigidly enmeshed closed family often involves restructuring the family to counter the hierarchical leadership (Constantine, 1986; Imig, 2005). Random Paradigm The antithesis of the closed family is the random family. This family values freedom and independence. If conflict occurs between the needs of the family and the needs of an individual 5 member, then the individual always comes first (Constantine, 1986, p. 105). The random family often appears chaotic and unorganized to other types of families. It is discontinuity-oriented as it seeks change and new ideas. The family theme supports curiosity and innovation but the family does not support hierarchy (Constantine, 1986, 1993; Imig, 2000a; Kantor & Lehr, 1975). Parents in this type of family often treat the children as little adults and allow them to contribute toward family decision-making. Competition is the norm in random families as it sparks innovation. Children in the random family are frequently the “smartest kids on the block”. Family members live “for today” as the family remains focused on the present (Constantine, 1986; Imig, 2005). The enabled random family offers members plenty of freedom with few rules. This is often a very intellectual, high-energy family with a variety of social attachments. The family system encourages freedom in thinking with loose structure that respects individual needs (Constantine, 1986; Imig, 2005; Kantor & Lehr, 1975). This is the most flexible kind of system. In the business world, many high tech companies as well as art agencies and graphic design companies operate as random structures. These are organizations that encourage creativity and innovation. The disabled random family is chaotic and distant. Family members lack cohesion and disengage. If somebody makes a decision, nobody follows it. This family is crisis-oriented (Imig, 2005). System boundaries change from flexible to disorderly as regulation fails. This allows anything to pass through. Members begin to rebel against their own individual freedom as they try to “rescue” the family from others (Kantor & Lehr, 1975, p. 176). Family members continually react to events with no closure. Intervention for the disabled random family often involves moving the family system toward reengagement through creative activities (Imig, 2005, p. 113). 6 Open Paradigm The open family system orients to the past, present and future. This family structure is typically consequence-oriented with a healthy balance between continuity and change. The open system is ideal for many individuals because it stresses practical consensus through open communication. This system encourages multiple perspectives in an environment of flattened hierarchy. The open family values both the individual and the group and encourages participation toward collective goals (Constantine, 1986; Imig, 2005; Kantor & Lehr, 1975). The enabled open family reaches consensus through communication. This is probably the most obvious characteristic of this system; family members talk a lot. The open family values diversity and different perspectives and respects the voice of each individual. Family boundaries are semi-permeable. The Open system allows information to pass through system boundaries for a while in order to allow access to new ideas, and then it closes the boundaries until the system has a chance to process the information. In this way, the open system maintains flexibility and allows change through new ideas, while at the same time, prevents chaos and system overload (Constantine, 1986, Imig, 2005; Kantor & Lehr, 1975). In the disabled open system family members feel exhausted from information overload. Their efforts to resolve issues by gathering more and more information result in ambiguity and confusion (Imig, 2005). Members of the disabled open system sometimes withdraw emotionally. Other times, they simply talk about consensus but no longer share in any type of functional interaction. Intervention for the disabled open system includes purposeful disengagement with more focus on individuality to deter attention from meta-talk about communication. Families in open disablement benefit from focusing on individual mastery rather than collective goals (Imig, 2005, p. 113). 7 Synchronous Paradigm The synchronous family is the least common type. This is a harmonious system that reflects natural alignment and deeply held beliefs. The context of this system is timeless; it does not operate in the past, present or future (Constantine, 1986, Imig, 2005). Members share consensus without communication through a special way of knowing. The synchronous system promotes learning through observation and listening. This system reflects no visible hierarchy, but the context of the synchronous system contains structure and patterns. Members understand common concepts but do not explicitly teach them to each other; they learn by being a part of the environmental context that contains the structure (Constantine, 1986; Imig, 2005). Religious organizations sometimes reflect synchronous structure. Enabled synchronous systems radiate harmony and understanding. Family members share a sense of oneness with feelings of unity. In the synchronous family, members maintain cohesion through unspoken shared understandings. These families value individuality and, at the same time, provide stability with rigid system boundaries (Constantine, 1986; Imig, 2005). Disabled synchronous families appear cult-like and devalue individual differences. These families are rigidly disengaged. The rigidity appears as “invariant repetition of interaction” with no homeostatic process to account for the repetition (Imig, 2005, p. 113). Synchronous systems that are disabled lose their coincidence capabilities and appear lifeless. They experience multiple failures as their previous ways of handling challenges no longer work. The most common intervention for disabled synchronous families involves reengagement of family members. Constantine suggests using one-way mirrors for this intervention. Selected family members initiate “open” communication and discussion while other family members observe (1986, p. 350). This method converts implicit discussions into explicit discussions and encourages more open communication. 8 Assumptions Family paradigm theory describes three important assumptions that relate to family structure. The first of these assumptions states that pure paradigms are extremely uncommon. Family structure is usually a blend of different family types (Constantine, 1986; Imig, 2005). While a family may operate under one primary, dominant structure it still contains elements of other less-dominant structures. The second assumption addresses the validity of different types of structure. This assumption states that one structural arrangement is neither more valid nor more desirable than another. Different families can successfully operate under different structures, and what works for one family might not work for another (Constantine, 1986; Imig, 2005). This is not to say that family systems never fall into a state of disablement, families can begin to function in a way that is not congruent with their paradigm for any number of reasons. The third assumption addresses this idea and recognizes the existence of both enabled and disabled forms of family structure (Constantine, 1986; Imig, 2005). Enablement and disablement of family systems are context-dependent. For example, a random family can function just fine until one of its members develops a serious illness and then becomes chaotic when the situation calls for cooperation from other family members. The family becomes too distant and disengages. Likewise, the closed family can become too close, or enmeshed in times of crisis. Neither condition is better or worse than the other, but both are dysfunctional and disabled for that particular family system. While it is outside the scope of this dissertation to describe detailed characteristics for each disabled structure, Constantine did a remarkable job of illustrating such characteristics in Family Paradigms (1986) and included sound suggestions for intervention. 9 Kantor and Lehr described a fourth assumption for family paradigm theory that relates to distance regulation of family members (1975, p. 159). Different family types maintain different levels of interpersonal distance or cohesiveness. Constantine added the synchronous paradigm to the distance-regulation model in Family Paradigms (1986, p. 194). Closed and synchronous families are typically more cohesive and group-oriented, while random and open families are usually more distant and individual-oriented. The distinguishing characteristic between closed and synchronous families and random and open families is their tendency toward continuity vs. change. Closed and synchronous family systems value continuity and are more structured and connected, while random and open families place more value on adaptability and are more flexible and separate (Constantine, 1986; Imig, 2005). Family Systems in Transition: The Role of Emotions One of the primary challenges for the family system undergoing severe stress brought about by illness, job loss, divorce, etc. is effective management of emotions among its members. At a time when the very structure of the system is at question and roles are often diffused, members of the family system experience an array of emotions. While some research suggests that anger, anxiousness, alienation and depression detract from an individual’s thinking ability, and reduce their ability to take in information (Goleman, 2002), other research suggests that emotions provide implicit or explicit knowledge for the individual that promotes rational decision-making (Bechara & Demasio, 2004). Although the two views present conflicting theory, they both point to the importance of emotions during rational decision-making. Decision-Making and Emotions A basic tenet of Human Ecology Theory presupposes that decision-making in a family system is a rational process (Bubolz & Sontag, 1993). During decision-making, family members 10 first recognize the need for change, then they consider viable options, and finally, they identify the most logical choice based upon the needs of the family and available resources (Paolucci, Hall & Axinn, 1977). While Human Ecology Theory is generally accepted as a valid theory with popular support, the underlying logic that explains decision-making according to this theory is at question based on the information presented above, and also revealed through research that shows a map of human brain activity that occurs during processing of emotions (Posner, et al., 2009). The idea of addressing emotions during stressful times within the family system is not new. For example, the importance of addressing emotions during therapy is well documented in the literature (Constantine, 1986; Franks, Gardner & Wampler, 2008; Griffin 1993, 2003; Smith et al. 1990). As stated by Constantine, “failure to take into account the emotional investment…an individual has in something can hinder the therapist or lead to unexpected outcomes” (1986, p. 76). Other researchers describe the importance of emotions in interactions among married couples, “…negative affect is the most reliable predictor of current and future marital satisfaction as well as future marital dissolution” (Franks, Gardner & Wampler, 2008, p. 111). See Also Gottman 1979; Griffin 1993, 2003; Rausch et al. 1974; Sanford, 2007 and Smith et al. 1990. One of the challenges in addressing human emotions in the context of family systems is identifying measurable constructs for emotions. The research associated with this study underwent several changes in instruments in order to address this problem. After initial pilot phases and preliminary data analysis were conducted, the investigator identified positive affect vs. negative affect as the most suitable constructs for measuring the relationship between family paradigms and human emotions. 11 Positive affect vs. negative affect serves an interesting role as a system of opposites in the proposed research. As previously described, dialectical logic is a philosophy that considers alternative choices and supports the idea of opposites. For the proposed study, dialectical logic provides the framework for examining the relationship between closed paradigm, random paradigm, open paradigm and synchronous paradigm and positive affect vs. negative affect. This philosophy further explains the process by which individuals or systems transition from one state to another. For example, when a family experiences severe illness or other stress, individual members often respond by accepting new roles (breadwinner, caretaker, etc). A family that operates in a traditional manner with the father acting as breadwinner might transition into a more random-type system with the mother taking over the breadwinner role if the father becomes incapacitated. Much like individuals can transition between different roles in the family, family systems can also evolve through different paradigms, or worldviews when they experience change. Identification of the Problem Stress and transition are common phenomena among family systems as environmental conditions change and family members advance through developmental stages in the life cycle. This is generally understood and accepted as routine – it’s life. However, successful transition of family systems into alternate states – or paradigms – depends largely upon the ability of family members to effectively manage stress associated with the situations and achieve a satisfactory resolution. One of the challenges in this process is framing the situation in a way that family members understand. When presented through the lens of dialectical logic, family paradigm theory provides such a model. Family paradigm theory offers clear descriptions of both enabled and disabled family system types and suggests specific action for family members to take toward 12 reinstating homeostasis. However, in its present state, family paradigm theory does not address human emotions for family systems undergoing transition. This research addressed that shortfall and explores the relationship between family systems and human emotions. Specifically, the primary objective of this study is to determine whether there is a relationship between closed paradigm vs. random paradigm and positive affect vs. negative affect. In addition to this focus, the current research considers relationships that include open paradigm and synchronous paradigm. Purpose of the Study The purpose of this survey study is to test the theory of dialectical logic that relates family paradigmatic orientation to human emotions while controlling for gender of participants at a Midwest University. The independent variable, family paradigmatic orientation, is defined as the overarching worldview of the relationship based on structure, behavior, and image. The dependent variable, human emotions, is defined as positive vs. negative affect, with subtypes of valence and arousal included in Phase III/Phase IV combined analyses (as described in the methods section). Control variables for the study include age, the number of years since birth, and self-reported gender of participants. Modifying variables are defined as follows: four-year education and above is completion of four or more years of education at the college or university level; relationship status is the current status of the participant in terms of being legally married to another individual of the opposite gender or living with a Significant Other; religious group is a group defined by reference to religious beliefs or lack of religious beliefs and political orientation is the thinking that characterizes a group or nation. 13 Significance of the Study The current study focuses on the research question is there a relationship between family paradigms and human emotions? More specifically, the study explores the relationship between closed paradigm vs. random paradigm and positive affect vs. negative affect. Information gained through this research will enhance family paradigm theory by defining the emotions that accompany various paradigmatic transitions. Furthermore, the dissemination of this information to family members who are undergoing transitions should empower them to recognize and appropriate adequate attention to the emotions that occur during transitional times. This awareness of, and consideration for emotions during family system transitions will empower the family system to more successfully transition into an alternate enabled state. Results from this study will be disseminated among the general public as well as professionals. This strategy will maximize the use of research results toward more effective communication and conflict resolution among family members in various contexts. In order to fully appreciate the value of this research toward conflict resolution, it is necessary to expand the focus of opposites - or dichotomy - beyond the boundary of the family and apply this concept to worldviews in general. Other sociologists have supported a general dichotomy in worldviews and argue for its significance in social science and cognitive science. For example, George Lakoff offers a convincing argument based upon morality in politics. Lakoff describes our nation’s political structure in terms of two opposing worldviews founded on morality: the Strict Father model (Conservative) and the opposing Nuturing Parent model (Liberal) (2002, p 33). Lakoff applies conceptual metaphors when suggesting, “the models show how moral reasoning in politics is ultimately based on models of the family” (2002, p. 17). To exemplify this, Lakoff paraphrases columnist William Raspberry (Houston Chronicle, section A, 14 p. 30, February 4, 1995) in his description of the Conservative’s view of the Liberal: “the government is an overindulgent, impractical mother and her citizens are her children, she has no self-discipline; she is indulging her children irresponsibly…this is not merely politics, it is a story with a moral” (Lakoff, 2002, p 6). At the very root of these two opposing worldviews is morality defined as self-discipline and self-reliance within the strict father model vs. love, empathy and nurturance within the nurturing parent model (2002, p 33). The current research argues that interpersonal conflict in any context can result from opposing worldviews such as those presented in family paradigm theory, and furthermore, that individuals who are well versed about these opposing views will be better equipped to work through, or prevent interpersonal conflict in other contexts. The significance of developing methodology for the proposed research lies in its ability to reveal new information that will enable social scientists and neuroscientists to question current limitations and explore new ideas. This information will represent a step toward explaining legitimate differences between individuals of opposing worldviews. Research Hypotheses Primary Hypothesis The H01 null hypothesis for this study states, among participants who complete the revised RPAS, there is no relationship between the participants’ relational paradigmatic orientation and the positive emotions vs. negative emotions they express in response to each paradigm statement. Secondary Hypotheses In order to better understand how moderator variables and control variables affect the relationship between family paradigms and human emotions, several additional hypotheses will 15 be tested in this study. Hypotheses H02 through H05 examine relationships between education, relation, religion and political orientation while controlling for age and gender. The conceptual model used to test hypotheses for this study is illustrated in Figure 2 on page 37. This model addresses the following null hypotheses: H01 Among participants who complete the revised RPAS, there is no relationship between participants’ relational paradigmatic orientation and the emotions they express in response to each paradigm statement H02 Among participants who complete the revised RPAS, the participants’ level of education does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement H03 Among participants who complete the revised RPAS, the participants’ relationship status does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement H04 Among participants who complete the revised RPAS, the participants’ religious status does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement H05 Among participants who complete the revised RPAS, the participants’ political orientation does not affect the relationship between their relational 16 paradigmatic orientation and the emotions they express in response to each paradigm statement 17 CHAPTER 2 LITERATURE REVIEW The proposed research is based upon Family Paradigm theory and structured around dialectical logic. The literature review begins with a history and description of family paradigm theory followed by an explanation of dialectical logic and its potential to support family typology as outlined by family paradigm theory. The review continues by describing how the established system of opposites enables family paradigm theory to enter into research. Finally, the literature review discusses human emotions and suggests future applications of this information toward a greater understanding about the role of emotions in family transition. Family paradigm theory evolved through the efforts of several individuals. Bateson provided a reasonable foundation for the theory with his application of general systems theory toward family systems (Ingold, 2000; Marcus, 1985; Nuckolls, 1995) and introduced the idea of dialectics and homeostasis among families (Krause, 2007; Nuckolls, 1995). Kantor and Lehr (1975) further developed the theory by identifying three distinct family systems – closed, random and open – based upon their observations of families in their natural environments (1975). Constantine (1986) later introduced the synchronous paradigm to Kantor and Lehr’s model, a contribution that produced a quadruplex model for family paradigm theory making it compatible with other four-fold typologies such as Leary’s Interpersonal Circumplex Model (1957); Olson, Sprenkle and Russell’s Circumplex Model of Marital and Family Systems (1979) and Beavers’ Centripetal/Centrifugal Model (1981). The quadruplex model also enabled integration of the four family types (closed, random, open and synchronous) with Kantor and Lehr’s player parts (mover, follower, opposer and bystander) (1975) allowing for application of the theory in family intervention and industry. Imig (2005) further 18 developed family paradigms through his contribution and application of the Relational Paradigm Assessment Scale (RPAS) (2000b). This contribution provided an essential instrument that enabled measurement of paradigms, and moved family paradigm theory deeper into the research arena. Family Process Theory Based upon their observations of families in their homes, Kantor and Lehr identified five fundamental components in the family process theory: subsystems, access dimensions, target dimensions, structures and player parts. These components serve as the foundation for understanding structure and function in family systems and empower the theory with the ability to explain everyday family process. Subsystems Kantor and Lehr introduced the idea of subsystems in family process to explain different types of interaction that occur in family systems. They recognized the personal subsystem that belonged to each individual member, the interpersonal subsystem that two or more family members shared and the family-unit subsystem that included all family members (1975, p. 23). Kantor and Lehr observed that family members identified boundaries to define their relational system as a separate entity from the surrounding environment. Boundaries determined how much and what type of information passed into and out of the family system. Family process theory applies the theme of metaphorical space to define the spatial boundary around each subsystem. Overlapping boundaries define the interface where members form strategies to achieve common goals. Identifying the correct boundary interface is significant because it enables therapists to make “manifest a system’s latent or covert aims” (Kantor & Lehr, 1975, p. 33). The intention and meaning of the strategy depends on the interface where it occurs. 19 Access and Target Dimensions Family interaction takes place within a social field, that contains access and target dimensions. These dimensions represent common resources and goals shared by family members. “Members of families gain access to target [dimensions] through the way in which they and their families regulate [access dimensions]” (Kantor & Lehr, 1975, p. 37). Access dimensions include the physical aspects of “family members’ quest for experience” and target dimensions describe the conceptual aspects…” (p. 36). Access Dimensions Kantor and Lehr (1975) identified three access dimensions in their framework: space, time and energy. They further identified one regulating mechanism and two operational mechanisms for each of these to help explain how the dimensions regulate family systems (see Table 1). Constantine introduced a fourth access dimension, material, in Family Paradigms (1986, p. 145). However, he did not identify mechanisms for this dimension in his theoretical framework. Imig (2005) later suggested the mechanisms of utilization, availability and suitability for material dimension. Since submechanisms have not yet been identified for material, this dimension remains a work in progress. The first access dimension, time, reflects the meshing of individual rhythms in a family process. Family members structure activities around time as they exist in everyday life. This element contains three mechanisms: synchronizing, orienting, and clocking that help to explain the function of time within the family process (Kantor & Lehr, 1975, pp. 78-89). The second access dimension, energy, refers to quantitative and qualitative attributes of the family system. Family members carry energy in high or low quantities and with positive or negative charges (Constantine, 1986; Imig, 2005; Kantor & Lehr, 1975, pp. 90-102). In a family system, energy 20 represents the “co-constructed strategies and rules” (Imig, 2005, p. 51). This dimension also contains three mechanisms: mobilizing, investing and fueling (Imig, 2005; Kantor & Lehr, 1975, pp. 90-102). The third dimension, space, includes both interior and exterior space and also refers to system boundaries. Interaction within family systems occurs primarily within the interior space (Constantine, 1986; Imig, 2005; Kantor & Lehr, 1975, pp. 66-77). The mechanisms for space: bounding, linking and centering, prohibit certain ideas and language for some family systems and allow new ideas to flow freely for others (Imig, 2005, p. 104). Constantine introduced a fourth access dimension, material, as a relational element that reflects the family’s attitude toward acquiring and consuming goods (Constantine, 1986, p. 152; Imig, 2005). For some families, material goods are obstacles to personal freedom. For others, material items are valued artifacts from the past. The meaning of material for each family system depends upon the family’s worldview. Target Dimensions Target dimensions refer to family goals. These elements comprise the informational dimension in family systems (Constantine, 1985, p. 146). Kantor and Lehr introduced three target dimensions in their family process theory: power, affect and meaning. They described one regulating mechanism and two operational mechanisms for each element. Constantine later referred to the power dimension as control and introduced content as a fourth target dimension (see Table 2) (Constantine, 1986, pp. 145,155). Content describes immediate and literal interpretation, as opposed to more finite meaning, during family interaction (p. 145). This dimension allows individuals to make sense out of the situation at hand. 21 Table 1 Mechanisms and Submechanisms for Access Dimensions Time Energy Space Synchronizing** (Monitoring, PrioritySetting, Programming, Coordinating, Reminding*) Fueling (Surveying, Tapping, Charging, Storing, Requisitioning*) Orienting (past, present, future, nontemporal, integrating*) Bounding (Mapping*, Screening, Routing, Patrolling Linking (Bridging, Buffering, Blocking Out, Channeling, Recognizing*) Material Utilization** ** Availability Regulating Mechanism ( ) Investing (Reconnoitering, Attaching, Committing, Detaching, Accounting*) Clocking (Sequencing, Frequency-Setting, Duration-Setting, Pacing, Scheduling*) Mobilizing** (Gauging, Budgeting*, Mustering, Transforming, Distributing) Centering** (Locating*, Gathering, Designing, Arranging, Spreading) Suitability Submechanisms In family systems the target dimension control often resides at the center of conflict. Control describes “the ability of the family to accomplish and achieve what it wants in a manner consistent with its paradigmatic design” (Imig, 2005, p. 60). The mechanisms of mastery, efficiency and efficacy provide guidelines for the family system as its members determine how to get things done. The second target dimension, affect, describes the patterns families engage in to provide members with an affirmative sense of warmth, closeness and engagedness (Constantine, 1986, p. 162, Imig, 2005, p. 63). Affect includes the mechanisms of reciprocity, belonging and expression. Family members display affect in different ways depending on the structure of their system. For example, in some family systems members express affection in a private manner with close physical contact. In other systems members express affection through more playful, public interaction (Constantine, 1986, Imig, 2005, Kantor & Lehr, 1975.) 22 Kantor and Lehr defined meaning as “some kind of philosophical framework that provides us with explanations of reality and helps us define our identity” (1975, p. 37). They describe the primary target of the family’s meaning system as “purposeful identity” (1975, p. 37). Meaning includes the mechanisms of purposefulness, connectedness and continuity. While some family systems derive meaning through a common set of shared values, others derive meaning by recognizing individual perspectives and “not” sharing. Meaning is closely related to content, the fourth target dimension Constantine introduced in family paradigm theory (1986, pp. 154-155). Constantine identified content as a separate dimension from meaning to distinguish between immediate, literal interpretations of situations and more continuous, value-laden interpretations. Imig further explains, “what is being sought is a rendering of reality (content). How that reality is interpreted (opportunity or problem) is the function of meaning” (2005, p. 71). The way individuals interpret various situations depends to a large degree on the image and structure of their family systems. Table 2 Mechanisms for Target Dimensions Control Affect Meaning Capability** Belonging** Purposefulness** Efficiency Reciprocity Connectedness Efficacy Expression Continuity Content ** Reality** Modulating Mechanism Relativity Representativeness Player Parts The behavioral component of each family system describes the roles family members assume during goal-seeking activity within the context of the family’s paradigm. Kantor and Lehr (1975) described four primary roles, or player parts that constitute the entire behavioral 23 range of social interaction for family systems: mover, follower, challenger and bystander. Constantine supported this player-part typology with family paradigm theory (1986). The mover in a family system represents a collective action that becomes the central focus of the family. This role most often belongs to one or more family members but may represent an individual or family commitment or illness. The mover is whatever initiates action within the family (Constantine, 1986, p. 124; Imig, 2005, p. 74; Kantor & Lehr, 1975, pp. 183-184). The follower supports and agrees with an existing action. This position requires strong interpersonal skills. An effective follower can have as much impact on family decisions as the mover. Moverfollower interaction occurs most often in the closed family structure. These dynamics add stability to the system. While the closed family system thrives on this stability, other family systems seek change and new ideas. The opposer in a family brings new ideas to the system. The individual who assumes this player part often stops or interferes with existing activity and enables the family to change directions or entertain new ideas (Constantine, 1986, p. 125-126; Imig, 2005, p. 78; Kantor & Lehr, 1975, pp. 184-185). Imig (2005) refers to the opposer as the challenger because of negative connotations that are frequently associated with the term “opposer” (2005, p. 78). For the purpose of this research, the term opposer and the term challenger are interchangeable. However, the term opposer will be used to describe this player part. The opposer is most often valued in the random family and least often valued in the closed family as this individual may be viewed as a threat to the stability of the closed system (Imig, 2005, p. 79). The bystander in a family system has a much less active role. This individual does not typically speak much but comments on the family process in a non-biased manner. The bystander is essentially a non-participant. However, although the bystander does not initiate or follow action, this player part is essential to the family system. The bystander is in the best 24 position to see and understand the family process (Constantine, 1986, pp. 126-127; Imig, 2005, pp. 79-80; Kantor & Lehr, 175, pp. 188-198). Individuals who possess the skills to interchange between all four player parts hold the best positions within a family system. These individuals stand to gain the most through effective interpersonal communication (Imig, 2005, p. 74). Families comprised of individuals who possess these skills achieve a high level of enablement and realize the most goals. Dialectical Logic Much like individuals can transition between different player parts, family systems can evolve through different paradigms. Dialectical logic is a philosophy that considers alternative ideas to the ones believed to be true (Imig, 2005, p. 9). In some circumstances, when none of the alternatives seem logical, a completely new idea is constructed, or synthesized, that makes more sense than any of the others. Constantine paralleled the idea of dialectical logic with a process he identified as morphogenetic sequence to explain how some family systems evolve through different paradigms over time (Constantine, 1986, p. 174). Without any order or structure a group tends to fall into a predictable pattern of stability and change over its life cycle (Constantine, 1986, p. 171). A group with no structure or leadership has a natural tendency to identify a leader. As time elapses, members become discontent and rebel against the leader’s authority. They focus more on individual feelings than group tasks. After a while, members begin to look to each other for direction as they establish new goals and norms. At some point the group completes its work and dissolves. The members leave the group with a sense of “integrated identification” and “separate but common memories” that define the group that now exists only in the minds of its members (Constantine, 1986, p. 170). 25 Constantine applied morphogenetic sequence to a family system’s orientation to continuity vs. change over time. He suggested that the natural sequence of family systems favors the early emergence of closed–type systems in families followed by random-type systems and then open-type systems, ending with synchronous-type systems (Constantine, 1986, p. 174). Constantine described the closed paradigm as a group-oriented, continuity-oriented system that seeks to continue the past into the future (1986, p. 97). When the needs of individual members conflict with the needs of the group, then the group always comes first. According to Constantine, the closed structure is the least complex. Families with young children frequently function in this structure to provide direction and leadership for the children. As the children grow into adolescence and begin to rebel, the family frequently falls into a more random structure that supports independence. This structure is primarily individual-oriented and discontinuity-oriented maximizing change from the past in a radical focus on the present” (1986, p. 97). When the needs of the random family conflict with the needs of the individual, then the needs of the individual always come first. Once the children reach adulthood, the conflict typically subsides and the family system falls into more of a consensual, egalitarian, open structure. Constantine described the open structure as predominantly consequence-oriented with a tendency to integrate the past, present and future into a blend of continuity and discontinuity that maximizes effectiveness (1986, p. 97). In this paradigm, the individual and the group are both equally important; the needs of one do not overpower the needs of the other. After many years of renegotiating the family’s guiding principles, the family system sometimes achieves a state of “unspoken closeness” characterized with “high levels of agreement” (1986, p. 174). Constantine introduced this as the synchronous paradigm and 26 described it as a coincidence-oriented paradigm characterized by complete agreement. Since everyone is of one state of mind, the issue of group vs. individual does not exist (1986, p. 97). Constantine suggested that members within a synchronous paradigm connect through coincidence rather than organized behavior (1986, p. 97). He further described “timeless, atemporal quality” about the synchronous system that is “neither continuity – nor discontinuityoriented” (1986, p. 97). This application of morphogenetic sequence illustrates dialectical logic in family paradigm theory. The family system changes between paradigms in response to the changing needs of its members. However, before the family actually transitions, it must first consider the new paradigm as a viable option, or in dialectic terms, an alternative. Constantine’s application of dialectical logic identifies the closed paradigm as the thesis, the random paradigm as its opposite, or antithesis, the open paradigm as a combination of the closed and random, or the synthesis and the synchronous paradigm as not closed, random or open: the antisynthesis (see Figure 1 on page 4) (Constantine, 1986, p. 97). The significance of this dialectical quality about family paradigm theory lies in its compatibility with other “systems of opposites” such as that described by Kantor and Lehr’s (1975) distance regulation perspective. Distance Regulation Model Kantor and Lehr (1975) identified five fundamental components in family process theory: subsystems, access dimensions, target dimensions, structures and player parts. These components are the fundamental building blocks for the theory enabling it to go beyond standard description and achieve a more dynamic position within the family system. When considered through the lens of dialectical logic, these components of family process help to explain distance regulation within family systems through everyday family process. 27 Two basic elements of family systems that influence interpersonal dynamics are system feedback and relational orientation (Imig, 2005, p. 9). While system feedback mechanisms in a family system regulate change vs. continuity for the family, relational orientation determines the family’s focus on individual vs. group (p. 9). Interestingly, the theoretical image formed by these components produces a system of opposites that mirrors dialectical logic (see Figure 1 on page 4). This relationship is especially useful in the current research because it provides a framework for interpreting emotions data. Family systems typically reflect one of four possible combinations of system feedback and relational orientation: • Closed paradigm – cohesive continuity • Random paradigm – distant change • Open paradigm – cohesive change • Synchronous paradigm – distant continuity Kantor and Lehr’s distance regulation perspective suggests that subsystems, access dimensions, target dimensions, structures and player parts all work together to maintain the characteristic conditions described above (e.g. cohesive continuity among members in the closed paradigm and distant change among members in the random paradigm) (1975, pp. 221-224). To illustrate, members of the closed family commonly claim areas of the home as private space and expect other family members to request permission to enter into those areas. Additionally, boundaries in the closed paradigm often prevent new ideas from entering into family conversation and limit conversation among family members to “approved” topics. Furthermore, in the closed family the needs of the family typically come before the needs of the individual (Constantine, 1986, Imig, 2005, Kantor & Lehr, 1975). 28 Let’s assume for a moment that an adolescent child in a closed-type family walks into his father’s office without knocking and interrupts a business call by demanding $200 to help purchase a motorcycle. In doing so, the adolescent uses profane language that is typically not tolerated in the household. This interaction sets off a reaction that elicits anger and disappointment for the father. Additionally, the father feels the need to regain control of the adolescent in order to restore homeostasis within the family system. In this example, conflict at the interface of interpersonal vs. personal subsystems occurs because the adolescent is using prohibited language in demanding money for personal goals that may not be compatible with family goals. The conflicts that are apparent in this example reflect distance vs. cohesion (the adolescent desires to place personal goals – spending money on self ahead of family goals – saving money for family) and continuity vs. change (adolescent introduces new language into the family system that is outside acceptable boundaries and violates the father’s private space). There are several other conflicts that my be extracted from this example but the dynamics are essentially the same; the feedback loop in this interaction elicits responses based upon the subsystems, access dimensions, target dimensions, player parts and structure of the family unit in an attempt to restore homeostasis. While the distance regulation model provides a useful mechanism for recognizing sources of conflict in the family system, this model does not necessarily consider the emotions that accompany the conflict. In family paradigm theory, family crisis or severe illness sometimes acts as a perturbation that pushes the family system into a different paradigm. A primary goal of the current research is to gain a better understanding about emotions that accompany conflict within family systems. Specifically, the research aims to understand positive vs. negative emotions that occur among family members in response to alternative paradigms. This 29 information should be applicable toward assessing families in crisis and making recommendations toward successful paradigmatic transition. Human Emotions What are emotions? Past scholars described emotions in terms of the observable body reactions that characterized them: the rapid heartbeat, goose bumps, increased respiration and dry mouth that occurred during emotional responses. James, for example argued, “If we fancy some strong emotion and then try to abstract from our consciousness of it all the feelings of its characteristic body symptoms, we find that we have nothing left behind, no ‘mind stuff’ out of which the emotion can be constituted” (1884,193). This prompted later scholars like LeDoux (1996) and Damasio (1994) to examine the constituent parts of emotions that James described. The latter scholars recognized unconscious bodily reactions to emotions as well as the associated feelings that accompany emotions. They also acknowledged the more cognitive, conscious input of the cerebral cortex that contributes toward the body’s response. However, during their exploration of emotions the researchers also recognized that the basic primitive fear response did not require cognitive processing of information or overt feelings of the body. Another researcher, Ohman (1999) conducted emotions research about the same time and demonstrated that the fear response does not require consciousness. The famous snake experiment used participants who feared snakes to show an unconscious reaction of fear. The individuals were presented with slides of snakes in rapid succession such that they could not consciously process the images, yet the individuals still experienced elevated skin conductance responses (Franks, 2006, p. 53). In reaction to these findings, LeDoux (1996) and Damasio (1994) introduced new ideas that changed the way modern sociologists view emotions. These researchers acknowledged not 30 only the significance of the body’s overt response toward emotions but also the importance of cortical reasoning in offsetting emotional responses (Franks, 2006, p. 53). The scholars placed emotions in the unconscious, as well as the conscious realm. What is the significance of this information in relation to family paradigms and human emotions? The response to this question is best considered in terms of decision-making. As previously stated, some noteworthy scholars in the field believed that decision-making was a rational process based primarily upon available resources and goals (Bubolz & Sontag, 1993; Paolucci, Hall & Axonn, 1977). This idea is significant to the history of Human Ecology theory as the idea of rational decision-making is one of the basic premises of the theory and discounts the value of emotions in the decision-making process (Bubolz & Sontag, 1993). In more recent times, scholars have questioned whether emotions subconsciously influenced decision-making, and may have actually provided benefit during the decision-making process. In their neural theory of economic decision, Bechara and Damasio reported that emotions moderate the interaction between environmental factors and decision-making, thus enabling “fast and advantageous decisions” (2004, p. 336). The distinction of whether emotions play a significant role in the decision-making process is relevant to the current research. In its present form, Family Paradigm theory does not provide adequate focus on emotions during interpersonal interaction. This is an important consideration for families undergoing change because the primary goal in these situations is to reduce stress and restore homeostasis within the family system. Without ample consideration for emotions, recommended paradigmatic changes for families experiencing stress may elicit unexpected negative emotions and leave the family in a more disabled state. 31 Emotions in Research This study previously identified the usefulness of dialectical logic and distance regulation as optimal frameworks for understanding family paradigms. The next logical step toward understanding the relationship between family paradigms and human emotions is to identify measurable dimensions of emotions. Posner et al. (2009) offered a useful approach toward this goal in their study focused on emotions and the Circumplex Model of Affect. Based on findings from their fMRI study, the researchers reported that “valence” and “arousal” represented two distinct, measurable dimensions of emotions with separate neural networks in the brain. This finding provided a framework for combining data from Phase III and Phase IV in the current research. Historically, neuroscientists and social scientists did not have access to technology that provided information about brain activity during neural processing of emotions. This changed with the introduction of functional Magnetic Resonance Imaging (fMRI) and Positron Emission Tomography (PET). These imaging devices now enable scientists to monitor and record brain activity in response to specific stimuli, including stimuli that are known to evoke certain emotions. This technology opened the door for integrated research among neuroscientists and behavioral scientists; it is now possible to study brain activity that occurs in conjunction with social behavior. However, in spite of this opportunity, researchers have been slow to join efforts toward this goal. In response to this observation, the current study proposes future research aimed toward combining efforts among social scientists and neuroscientists into common research. While it is outside of the scope of this dissertation to conduct the proposed research and report the findings, 32 the groundwork for future research is established (see Implications for Future Research for additional explanation). 33 CHAPTER 3 METHODOLOGY Overview This study includes methodology that evolved over the course of three years and occurred in four phases. The first two phases comprised a pilot study that served as a guide for revisions of the instruments and study design. The primary research question remained the same during all phases of the study, is there a relationship between closed paradigm vs. random paradigm and positive affect vs. negative affect? In answer to this question, all phases included collection of data related to demographics, family paradigms and emotions. The distinguishing factor between phases occurred in the format of the emotions instrument; the first and second phases used an open-ended instrument (see Appendix F and Appendix G), the third phase used a multiple-choice instrument (see Appendix H) and the fourth phase used the Positive and Negative Affect Schedule (PANAS) (see Appendix I) to collect emotions data. Investigators reviewed the data following each phase of the study and implemented appropriate changes. The data that elicited the most concern were the emotions data. The open-ended responses in Phase I and Phase II revealed a general inability among participants to articulate their emotions. Investigators predicted that participants would articulate emotions more clearly if they selected from a list of 30 emotions words rather than responding to open-ended questions (see Figure 3 on page 47). This format was implemented for Phase III. However, additional review of data following this phase revealed continued lack of clarity among participants in recording their emotions. Investigators implemented the PANAS in response to this finding and the PANAS was used to collect emotions data during Phase IV. 34 Following data collection, researchers explored several options for coding and analyzing data. Review of recent literature (Posner, et al., 2009) revealed potential usefulness for the Circumplex Model of Affect (Russell, 1980) to serve as a foundation for recoding and combining emotions data from Phase III and Phase IV of the study, thereby increasing the power of the study. After careful consideration, Investigators opted to convert the emotions data according to the Circumplex Model of Affect and performed analysis using the transformed data (see Appendix L for description of data transformation). In addition, because the PANAS is a validated instrument, a separate analysis was performed using emotions data collected during Phase IV in order to evaluate the usefulness of the instrument for future research focused on family paradigms. Procedure Researchers contacted students through the instructor of a 400-level college course at a Midwest University and invited them to participate in the study. The course focused on interpersonal relationships within the family. Participation in the study was offered as an alternative option for required participation in the course. After explaining the study to the participants, researchers requested their consent to participate. Individuals who agreed to participate completed the revised Relational Paradigmatic Assessment Scale (RPAS-1, RPAS-2, RPAS-3, or RPAS-4) and the demographic survey. Design Development of methodology for the proposed research was survey-based and conducted using an exploratory design with a revised Relational Paradigmatic Assessment Scale (RPAS-1, RPAS-2, RPAS-3 or RPAS-4) and a demographic survey. The study included a total of 257 35 participants: 40 in the pilot study (Phase I and Phase II) and 217 in the working study (Phase III and Phase IV). Of the 257 participants, 253 completed the surveys at a single-site location. The major dependent variable for this research was emotion and the major independent variable was paradigmatic orientation as described by Family Paradigm Theory. The theory defines four family paradigms: closed, random, open and synchronous (see Appendices F, G, H, and I). Moderator variables for the study included: education, relationship, religion and political orientation and control variables included age and gender. Participants in all four phases completed a revised RPAS and provided demographic information including gender, age, level of education, marital status, and political orientation. In addition, participants in Phase III and Phase IV also provided information related to significant relationship with others, religious affiliation, home country, native language and government organization of home country. With the exception of religious affiliation and significant relationship with others, all variables in the last group described were excluded from final analyses because the data were insufficient to produce significant findings. However, these data were coded in case they can contribute toward future research conducted in response to this study. The changes in study design during development of the methodology improved the measurability of emotions. Conceptual models for the development of the four-phase model are depicted in Figure 2 and Figure 3. 36 Input Variable Family Paradigm Statements from RPAS-3 and RPAS4 (Self-Reported Mixed Open/Closed Responses plus PANAS)– (See Emotions list in Table 3) • Closed • Random • Open • Synchronous Participants with Closed Paradigm Participants with Random Paradigm Outcome Variable Emotions Words Participants with Open Paradigm Participants with Synchronous Paradigm Controls: age and gender Figure 2 Conceptual Model for Phase III and Phase IV Combined Participants Pilot Study: Phase I and Phase II The pilot study collected information about emotions using open-ended questions. This part of the study was comprised of 40 participants at a Midwest University. Of these students, 36 were enrolled in a 400-level college course on Interpersonal Relations of the Family at the University. The researcher or study representative explained the study to the participants and requested consent for participation. All 40 participants (100%) completed the study. Demographics for participants included 10 males (25%) and 30 females (75%). The students ranged in age from 19 years to 48 years with a mean age of 23.5 years. 37 The marital status of participants included 30 single, never married (75.0%), three single, living with a significant other (7.5%), five married (12.5%), one divorced (2.5%), and one missing data (2.5%). The highest level of education reported by the participants included six sophomore (15.0%), 17 junior (42.5%), 11 senior (27.5%), three graduate/professional degree (7.5%) and three other (7.5%). The political orientation for this group included four moderately liberal (10%), 33 neither liberal nor conservative (82.5%), two moderately conservative (5%) and one extremely conservative (2.5%). Working Study: Phase III and Phase IV The third and fourth phases of the study comprised the primary data collection phases that served as the foundation for data analysis and interpretation of results and the discussion. Phase III During the third phase of the study participants provided emotional responses on the RPAS-3 (see Appendix H) from a list of 30 emotions words (see Table 3). This part of the study was comprised of 143 participants at a Midwest University who were students in a 400-level course on Interpersonal Relations of the Family. The researcher, or study representative, explained the study to participants and requested consent for participation. Initially, 158 participants provided informed consent and completed the study. However, 15 participants (9.5%) were removed from the study prior to data analysis because their emotion responses included multiple occurrences of missing data. This resulted in 143 participants for Phase III of the study. Demographics for participants in this group included 60 males (42.0%) and 83 females (58.0%). The students ranged in age from 20 years to 51 years with a mean age of 22.3 years. The marital status for participants in Phase III included 118 single, never married (82.5%), 17 38 single, living with a significant other (11.9%), six married (4.2%), one divorced (.7%), and one missing data (.7%). The highest level of education reported by the participants included 93 attended college (65.0.%), 21 college, two-year degree (14.7%), 26 college, four-year degree (18.2%), two graduate/professional degree (1.4%) and one missing data (.7%). Table 3 Emotions Words Included in Phase III of Development 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. Exhausted Confused Ecstatic Guilty Suspicious Angry Hysterical Frustrated Sad Confident Embarrassed Happy Mischievous Disgusted Frightened Enraged Ashamed Cautious Smug Depressed Overwhelmed Hopeful Lonely Jealous Bored Love-struck Surprised Anxious Shocked Shy 39 Phase IV During the fourth phase of the study participants provided emotion responses to RPAS-4 using the Positive and Negative Affect Schedule (PANAS) (see Appendix I and Appendix J). This phase was comprised of 59 participants at a Midwest University who were students in a 400level course on Interpersonal Relations of the Family. The researcher, or study representative, explained the study to the participants and requested consent for participation. All 59 participants (100%) completed the study. Demographics for participants included 17 males (28.8%) and 42 females (71.2%). The students ranged in age from 20 years to 50 years with a mean age of 22.4 years. The marital status of participants in Phase IV included 49 single, never married (83.1%), four single, living with a significant other (6.8%), four married (6.8%), one remarried (1.7%), and one missing data (1.7%). The highest level of education reported by the participants included 46 attended college (78.0%), three college, two-year degree (5.1%), nine college, four-year degree (15.3%) and one other (1.7%). Descriptive statistics for this group are recorded in Table 4 on page 58. Phase III and Phase IV Combined The working phase of the study was comprised of 202 participants: 77 males (38.1%) and 125 females (61.9%). Ages of participants in the combined group ranged from 20 to 51 years with a mean age of 22.3 years. The marital status of participants in this group included 167 single never married (82.7%), 21 single, living with a significant other (10.4%), 10 married (5.0%), one divorced (.5%), one remarried (.5%) and two missing data (1.0%). The highest level of education reported by the participants included 139 attended college (68.9%), 24 college, two-year degree (11.9%), 40 35 college, four-year degree (17.3%), two graduate/professional degree (1.0%), one other (.5%) and on missing data (.5%). Descriptive statistics for the working group are recorded in Table 18 on page 87. Conceptual Definitions This study includes several variables that require conceptual definitions in order to fully understand the relationship between family paradigms and human emotions. These definitions are provided in a list that follows. Independent Variable Paradigmatic orientation: Recorded as values assigned to family paradigm statements for closed paradigm, random paradigm, open paradigm and synchronous paradigms on RPAS-1, RPAS-2, RPAS-3 and RPAS-4. Dependent Variable Human emotions: Recorded as emotions words on RPAS-1, RPAS-2, RPAS-3 and RPAS-4. Moderator Variables Education, relationship status, religious affiliation and political orientation: Selfreported through selection of appropriate categories by participants on demographic surveys. Additionally, participants in Phase III and Phase IV of the study indicated their marital status by responding to open-ended question “C2” that described relationship type for others in the same household (see Appendix E). Conceptual and operational definitions for these variables are included in Appendix A. In addition to these data, additional information was collected from participants that described number in household, closeness of significant relationship, native language, home country, and government organization. The latter data were excluded from final 41 data analyses but coded and included in Appendix A in case these data are useful in future research related to this study. Control Variables Age and gender:Age was self-reported by indicating number of years and gender was indicated through selection of appropriate category by participants on demographic surveys. Conceptual and operational definitions for these variables are included in Appendix A. Inclusion Criteria Inclusion criteria for this study were English-speaking college students, 18 years of age or older who were enrolled in a 400-level family relations course at a Midwest University. Additional inclusion criteria were mental competency and the ability to read and write in the English language. Exclusion Criteria Exclusion criteria were students who were less than 18 years of age, or who were mentally incompetent. Instruments Demographic Survey The study used two different demographic surveys to collect information about the participants (see Appendix D and Appendix E). The original demographic survey was used during Phase I and Phase II of the pilot study. This instrument collected information related to gender, age, education, marital status, number in household and political orientation (see Appendix D). Based upon review of preliminary data and additional review of related literature, the demographic survey was revised prior to Phase III of the study. The revised demographic survey was used during Phase III and Phase IV of the working study (see Appendix E). The 42 modified survey collected information related to gender, age, education, student status, number in household, relationship of others in household, significant relationship, home country, native language, religious group, government organization and political orientation. Data extracted from this demographic survey and used for data analyses included gender, age, education, relationship status, religious affiliation and political orientation. Relational Paradigmatic Assessment Scale (RPAS) Standard RPAS The standard RPAS consists of ten questions followed by four statements for each question (Imig, 2000b). Four of the questions on the scale relate to target dimensions, or goal elements (affect, control, content and meaning), and four questions relate to access dimensions, or resource elements (space, material, time and energy). Each question includes four response statements (a,b,c,d) reflecting one statement for each paradigm - closed, random, open and synchronous. Participants respond by assigning a 0-10 value for each statement that indicates how closely the statement describes their relationship or family system. The two remaining questions rate the overall importance of each element in the relationship. Prior to completing the scale, participants are instructed to assign a value of ten to only one statement for each question and then respond to the remaining three statements using a scale from 0-9. Responses on the scale are analyzed to produce quartile scores and cluster scores that reflect the paradigmatic orientation related to each element (see Appendix K). It should be noted that on the original RPAS, participants rate their “current” relationships and (separately) also rate their “ideal” relationships. This information was also collected for the current research, but based upon the massive amount of information collected, Investigators decided to focus on data provided about “current” relationships for this study. 43 The Relational Paradigm Assessment Scale (RPAS) underwent several revisions during development. However, all versions are felt to have construct validity because they were specifically designed to measure family paradigm structure and player parts as described by family paradigm theory (Constantine, 1993a, Hidecker, 2004; Imig, 1993a, 2000a; Imig & Phillips, 1992). One of the earliest versions of RPAS, the Family Regime Assessment Scale (FRAS), is recognized as a useful research instrument within the discipline of family science (Touliatos, Perimutter & Strauss, 1999, pp. 50-51). During development of the next version, the Paradigm Assessment Scale (PAS), participants in a study commented on the appropriateness and the wording of the PAS (Hidecker, 2004; Imig & Phillips, 1992). Hidecker (2004) summarized several reports (Imig, 1993a, 2000a; Imig et al., 1996; Imig & Phillips, 1992; Pate, 1994; Pegorarro, 1999; Villarruel et al., 1995; Ward, 1997) and concluded, “family participants have felt that the FRAS and the R-PAS are representative of family functioning which is a component of content validity” (2004, pp. 22-23). Since RPAS (or variations of this scale) is the only instrument available to measure family paradigms, construct validity has not been established. Variations of RPAS have been used to assess paradigmatic orientation in a variety of situations related to divorce (Pate, 1994); behavior under stress (Imig, 1993a, 2000a); home schooling (Pegorraro, 1999), men’s groups (Imig et al., 1996), family businesses (Imig et al., 1996) and older child adoption (Ward, 1997). Test-retest reliability has not yet been established for RPAS. Revised RPAS The primary instrument used to collect data related to paradigmatic orientation for each relationship in the current study was the Revised Relational Paradigmatic Assessment Scale (RPAS-1, RPAS-2, RPAS-3, or RPAS-4) (see Appendices F, G, H and I) (Imig, 2000b). 44 RPAS-1 and RPAS-2: Pilot phase. For this study, the RPAS was revised to include four questions: two that represented access dimensions (space and material) and two that represented target dimensions (control and affect). Investigators abbreviated the RPAS to ensure that participants could complete the survey in the two-hour time frame allotted for each research session. Researchers felt that a 10-question scale along with the demographic survey would be difficult to complete in this amount of time. After considering the value of each element to the outcome of the study, Investigators decided to include questions related to the elements of space, material, affect and control in the relationship. Prior to phase I of the study, the RPAS underwent further revisions to include open-ended responses that described emotions participants felt in response to each statement as it applied to their relationship. Additionally, Phase I also collected information describing whether each statement applied to the participant’s relationship in a positive (+) way, a negative way (-), neither a positive nor a negative way (0) or both a positive and a negative way (+/-). This resulted in the development of RPAS-1 (see Appendix F). Preliminary data analysis following Phase I of the study revealed difficulty in measuring the emotion responses. Most responses were presented by participants in sentence form resulting in difficulty with coding the information. In response to this, the Investigators for the study revised RPAS-1 to include instructions for participants to list “one or two words” that described how they felt in response to each statement. Additionally, the Investigators also discussed the value of including an intensity value for each emotion word on a scale from 1-5. This variable was also added to RPAS-1 and resulted in RPAS-2 (see Appendix G). Further data analysis following phase II of the study revealed continued difficulty in measuring participant emotion responses. The responses suggested a general inability among 45 participants to articulate the emotions they experienced. Based on this observation, Investigators revised RPAS-2 prior to Phase III with the addition of a list of 30 emotion words for participants to use in describing their emotions (see Table 3 on page 39). RPAS-3 and RPAS-4: Working study. For phase III of the study, researchers introduced RPAS-3 that included the list of 30 emotion words (see Table 3 and Appendix H). In addition, researchers also added question 5 to RPAS-3 in order to determine the overall importance of each element (control, affect, space and material) in the relationship (see Figure 3 on page 47). Phase III produced somewhat more measurable results. However, data analysis following this phase continued to show multiple responses that did not reflect emotions; participants frequently responded with words that were not included on the 30-word emotion list. In order to produce more measurable results, researchers implemented the use of the Positive and Negative Affect Schedule (PANAS) during the fourth phase. The RPAS-4 resulted from combining the RPAS with the PANAS. This instrument recorded family paradigm information as well as information that described positive and negative affect. The data set collected during Phase IV was combined with Phase III data and comprised the Working Study. This data set was categorized according to the Circumplex Model of Affect and was further used to test the hypotheses for the study (see Figure 4 and Appendix M). 46 Q3 Space A. Random B. Closed C. Open D. Synchronous A. Space Q1 Control A. Synchronous B. Closed C. Random D. Open D. Control Q5 B. Material Q4 Material A. Closed B. Random C. Open D. Synchronous C. Affect Q2 Affect A. Open B. Closed C. Random D. Synchronous Figure 3 Conceptual Model for Phase III and Phase IV Depicting Inclusion of Question 5. For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this dissertation. 47 Participants with Closed Paradigm Participants with Random Paradigm RPAS-4 Family Paradigms Statements and PANAS Emotions Words (Positive vs. Negative Affect) Participants with Open Paradigm Participants with Open Paradigm Moderator Variables Control Variables • Gender • Highest Level of Education • Relationship Status • Religious Group • Political Orientation Figure 4 Conceptual Model for Phase IV Positive and Negative Affect Schedule (PANAS) The Positive and Negative Affect Schedule (PANAS) was used to record emotions during Phase IV of the study (see Appendix C and Appendix J). The PANAS is a self-administered mood assessment. Scores have a direct interpretation and indicate the intensity of the emotional response for two affective state dimensions: positive affect and negative affect. Scores are determined by summing responses from each scale: positive affect (PA) items (interested, 48 excited, strong, enthusiastic, proud, alert, inspired, determined, attentive, and active) and negative affect (NA) items (distressed, upset, guilty, scared, hostile, irritable, ashamed, nervous, jittery and afraid). Scoring yields separate scores for PA and NA that range from 10-50. Reliability and validity of PANAS have been widely reported in the literature (Crawford & Henry, 2004; Watson & Clark, 1994; Watson & Vaidya, 2003). Test-retest reliability results indicate coefficient for general ratings are sufficiently high to suggest they indicate participant’s trait affect and internal consistency for this scale is reported at 0.84 - 0.90 (Watson, Clark & Tellegen, 1988). Appendix C provides additional information related to this instrument. Coding In preparation for data analysis, data for the study were coded according to the scales of measurement used during data acquisition. Variables measured at the nominal level were coded with dummy variables that indicated whether the variables were present (0 = no, 1 = yes). Variables measured at this level included gender, live alone, significant relationship, and religious affiliation. Categorical variables included education, marital status and political orientation. These variables were also measured at the nominal level. However, education and marital status were first coded with unique codes for each categorical level and annotated with unique labels and then recoded with dummy variables for analysis. Additionally, based upon the discrimination achieved between different levels of political orientation (respondents selected from five categories ranging from liberal to conservative) these data were analyzed at the interval level. The variable age was also measured at the interval level and coded with the value reported by participants during data acquisition. 49 Research Hypotheses Primary Hypothesis The H01 null hypothesis for this study states, among participants who complete the revised RPAS, there is no relationship between the participants’ relational paradigmatic orientation and the emotions they express in response to each paradigm statement. Secondary Hypotheses In order to better understand how moderator variables and control variables affected relationships between family paradigms and human emotions, several additional hypotheses were tested in this study. Hypotheses H02 through H05 examined relationships between education, relationship status, religious group and political orientation while controlling for gender. The conceptual model used to test hypotheses for this study is illustrated in Figure 4 on page 48. This model addresses the following null hypotheses: H01 Among participants who complete the revised RPAS, there is no relationship between participants’ relational paradigmatic orientation and the emotions they express in response to each paradigm statement H02 Among participants who complete the revised RPAS, the participants’ level of education does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement H03 Among participants who complete the revised RPAS, the participants’ relationship status does not affect the relationship between their relational 50 paradigmatic orientation and the emotions they express in response to each paradigm statement H04 Among participants who complete the revised RPAS, the participants’ religious status does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement H05 Among participants who complete the revised RPAS, the participants’ political orientation does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement Data Analysis Data analysis for the study occurred in four stages. Analyses were conducted using Mac OS × version 16.0 SPSS statistical analysis software (SPSS Inc, Chicago, IL). Univariate Analyses The first stage of analysis consisted of univariate with descriptive statistics. Categorical variables in these analyses included gender, relationship status, education and religious status. Data for each variable were analyzed with frequency tables and then graphically plotted using histograms (for interval data) and bar charts (for categorical data) to assess for shape, frequency distribution, central tendency, and variability. Data were also analyzed for skewness, kurtosis, outliers, gaps and peaks. In addition, interval data were tested for normal distribution using the Shapiro-Wilk normality test. Frequency distributions and percentages for these variables were reviewed. Review of the charts and graphs for gender revealed data that were well suited for further analysis. However, 51 review of marital status, education and religion showed several categories that contained less than three data points. In order to include these variables in further analysis, categories for “marital status’ were collapsed to include “married or living with a significant other” and “single or divorced, not living with a significant other”; categories for “education” were collapsed to include “less than a four-year degree” and “four-year degree and higher” and categories for “religion” were collapsed to include “religious affiliation” and “no religious affiliation” (see Table 4 and Table 17). Descriptive statistics for continuous variables were calculated and reported as mean, standard deviation, minimum value and maximum value. Continuous variables for the study included age, political orientation, closed paradigm, random paradigm, open paradigm and synchronous paradigm (see Table 5 and Table 18). Age values ranged from 20 to 50 (M = 22.38, SD = 4.591), political orientation values ranged from 0 to 4 (M = 1.79, SD = 1.056), closed paradigm values ranged from 0.08 to 0.34 (M = .2033, SD = .0540), random paradigm values ranged from 0.15 to 0.52 (M = .2973, SD = .0662), open paradigm values ranged from 0.11 to 0.42 (M = .2898, SD = .0628) and synchronous paradigm values ranged from 0.09 to 0.31 (M = .2096, SD = .0555). Normality of distribution for continuous variables was assessed using the Shapiro-Wilk test and each dataset was analyzed for kurtosis and skewness. Data for closed paradigm, random paradigm, open paradigm and synchronous paradigm passed the normality test at the 95% confidence interval. However, data for age and political orientation failed the normality test at this confidence interval. Further review of box plots for political orientation revealed a bellshaped curve with normal distribution so this variable was retained for further analysis. 52 Review of box plots for age were analyzed and showed skewness in distribution greater than four. Box and whisker plots were generated in order to identify outliers and these values were checked for accuracy against original data. These data were transformed using natural log, square root and exponent transformations to improve symmetry, but transformation of data was not successful in compensating for the skew. The data indicated that a high number of participants (50 out of 59) reported ages between 20 – 22 years. The resulting bias rendered the data invalid for use as a moderator variable and age was omitted from further Phase IV analysis. Bivariate Analyses The next stage of data analysis consisted of bivariate analyses to test for significant relationships between continuous variables in each phase of the study. During this stage, analyses were performed using Spearman’s correlation to examine relationships between closed paradigm, random paradigm, open paradigm, synchronous paradigm and political orientation. Multivariate Analyses In order to determine whether emotional responses were related to paradigmatic orientation, two different multivariate analyses were conducted. During Phase IV data analyses, ordinary least squares (OLS) regression was used to compare paradigm scores with corresponding levels of emotions (positive affect vs. negative affect) while statistically controlling for gender. In addition, education, relationship status (relation), religious status (religion) and political orientation were included in the model to test for other moderating effects. A probability of p < 0.05 was established as the required value to reject the null hypotheses. Scores for each paradigm reflected the sum of corresponding paradigm complex coefficient values across Q1, Q2, Q3 and Q4 (see appendix K): 53 For Phase III/IV combined analyses, ordinal regression models of emotions were tested separately for valence (pleasure vs. displeasure) and arousal (activated vs. deactivated) using paradigms as primary predictors. Separate ordinal regression analyses were run for each outcome variable: closed valence, random valence, open valence, synchronous valence, closed arousal, random arousal, open arousal and synchronous arousal. Findings were summarized and interpreted in the results section of this dissertation (see Table 31 on page 113 and Table 32 on page 116). 54 CHAPTER 4 RESULTS The primary purpose of this study was to explore the relationship between closed vs. random family paradigms and positive vs. negative human emotions. Following data collection, the study focused on two phases from data collection for analyses: Phase IV included data from the 59 participants who rated their emotions according to the PANAS and Phase III/IV combined included data collected using PANAS combined with data collected using the open-ended emotions survey. The initial plan for analysis was to focus on Phase IV data collected using PANAS because the PANAS is a validated instrument. For this reason, Phase IV analysis is presented first. The decision to combine Phase III and Phase IV was based upon literature related to the Circumplex Model of Affect (Posner et al., 2009; Russell, 1980) that revealed potential benefit in framing emotions according to the Circumplex Model. This process entailed combining Phase III and Phase IV data to provide a larger dataset with more comprehensive analysis. Additionally, as described by Posner et al. (2009) the circumplex model posits that, “all emotions can be described as a linear combination of two underlying, largely independent neurophysiological systems, valence and arousal” (p. 2). Results that follow reflect a two-tier model of analysis – Phase IV analysis followed by Phase III/Phase IV combined analysis - with emotions for Phase IV coded as positive and negative affect (see Appendix N), and emotions for Phase III/Phase IV combined coded as valence and arousal (see Appendix M). 55 Phase IV Analyses These analyses tested the H01 hypothesis, among participants who complete the revised RPAS, there is no relationship between participants’ relational paradigmatic orientation and the emotions they express in response to each paradigm statement. Additional sub-hypotheses included: H02 Among participants who complete the revised RPAS, the participants’ level of education does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement H03 Among participants who complete the revised RPAS, the participants’ relationship status does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement H04 Among participants who complete the revised RPAS, the participants’ religious status does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement H05 Among participants who complete the revised RPAS, the participants’ political orientation does not affect the relationship between their relational paradigmatic orientation and the emotions they express in response to each paradigm statement 56 Univariate Analyses for Phase IV Analysis of data for this study was conducted using Mac OS × version 16.0 SPSS statistical analysis software (SPSS Inc, Chicago, IL). The first stage consisted of univariate analyses with descriptive statistics of Phase IV data. Categorical variables in this analysis included gender, relationship status, education and religious status. Frequency distributions and percentages for these variables were reviewed using bar charts and graphs. As described in methods section, review of the charts and graphs for gender revealed data that were well suited for further analysis. However, review of marital status, education and religion showed several categories that contained less than three data points. In order to include these variables in further analysis, categories for “marital status’ were collapsed to include “married or living with a significant other” and “single or divorced, not living with a significant other”; categories for “education” were collapsed to include “less than a four-year degree” and “four-year degree and higher” and categories for “religion” were collapsed to include “religious affiliation” and “no religious affiliation” (see Table 2). Descriptive statistics for continuous variables were calculated and reported as mean, standard deviation, minimum value and maximum value. Continuous variables for the study included age, political orientation, closed paradigm, random paradigm, open paradigm and synchronous paradigm (see Table 3). Age values ranged from 20 to 50 (M = 22.38, SD = 4.591), political orientation values ranged from 0 to 4 (M = 1.79, SD = 1.056), closed paradigm values ranged from 0.08 to 0.34 (M = .2033, SD = .0540), random paradigm values ranged from 0.15 to 0.52 (M = .2973, SD = .0662), open paradigm values ranged from 0.11 to 0.42 (M = .2898, SD = .0628) and synchronous paradigm values ranged from 0.09 to 0.31 (M = .2096, SD = .0555). 57 Normality of distribution for continuous variables was assessed using the Shapiro-Wilk test and each dataset was analyzed for kurtosis and skewness. Data for closed paradigm, random paradigm, open paradigm and synchronous paradigm passed the normality test at the 95% confidence interval. However, data for age and political orientation failed the normality test at this confidence interval. Further review of box plots for political orientation revealed a bellshaped curve with normal distribution so this variable was retained for further analysis. Table 4 Descriptive Statistics for Categorical Data in Phase IV (PANAS) (N=59) Frequency Percent Valid Percent Category Gender Relationship Status Education Religion Female Male Married or Living with Significant Other Single or Divorced, Not Living with Significant Other Missing Less than Four Year Degree Four Year Degree or Higher Missing Religious Affiliation No Religious Affiliation Missing 58 42 17 9 71.2 28.8 15.3 71.2 28.8 15.5 49 83.1 84.5 1 49 9 1 46 12 1 1.7 83.1 15.3 1.7 78.0 20.3 1.7 84.5 15.5 79.3 20.7 Table 5 Descriptive Statistics for Continuous Data in Phase IV (PANAS) (N=59) Mean Demographics Age Political orientation Paradigms Closed Random Open Synchronous SD Min Max 22.38 1.79 4.591 1.056 20 0 50 4 .2033 .2973 .2898 .2096 .0540 .0662 .0628 .0555 0.08 0.15 0.11 0.09 0.34 0.52 0.42 0.31 Review of box plots for age were analyzed and showed skewness in distribution greater than four. Box and whisker plots were generated in order to identify outliers and these values were checked for accuracy against original data. These data were transformed using natural log, square root and exponent transformations to improve symmetry, but transformation of data was not successful in compensating for the skew. The data indicated that a high number of participants (50 out of 59) reported ages between 20 – 22 years. The resulting bias rendered the data invalid for use as a moderator variable and age was omitted from further Phase IV analysis. Bivariate Analyses for Phase IV Bivariate analyses were conducted to test for significant relationships between continuous variables in Phase IV. Specifically, analyses were performed using Spearman’s correlation to examine relationships between closed paradigm, random paradigm, open paradigm, synchronous paradigm and political orientation. Correlates for Closed Paradigm, Random Paradigm, Open Paradigm, Synchronous Paradigm and Political Orientation in Phase IV Based upon the theory of dialectical logic, the researcher for the current study predicted that bivariate correlation would reveal negative correlation between individuals and groups with 59 opposing worldviews. Specifically, the researcher for the study predicted that bivariate correlation would reveal negative correlation between participants who scored high on closed paradigm compared to participants who scored high on random paradigm. Additionally, based upon Kantor and Lehr’s model of distance regulation (1975), the researcher also predicted that paradigms characterized by cohesiveness among family members (closed paradigm and synchronous paradigm) would reveal negative correlation with paradigms characterized by interpersonal distance between family members (random paradigm and open paradigm). Table 6 presents findings from these analyses. Results of the analyses showed significant negative correlation between closed paradigm and random paradigm. This finding indicates that individuals who reported higher levels of closed paradigmatic orientation also reported proportionally lower levels of random paradigmatic orientation compared to individuals who reported lower levels of closed paradigmatic orientation. In other words, participants who perceived high levels of cohesiveness in their relationship with a focus on group goals perceived lower levels of personal freedom and less sense of individuality compared to individuals who perceived low levels of cohesiveness in their relationship with less focus on group goals. 60 Table 6 Correlation Between Closed, Random, Open and Synchronous Paradigm Scores and Political Orientation for Phase IV (PANAS) Using Pearson’s (N=59) Correlations Closed Pearson’s Closed Rand Correlation 1 -.497*** .000 Coeff. Sig. (2-tailed) Rand Correlation -.497*** 1 Coeff. .000 Sig. (2-tailed) Open Correlation -.164 -.420*** Coeff. .215 .001 Sig. (2-tailed) -.194 Synch Correlation -.233 .140 Coeff. .076 Sig. (2-tailed) .053 Politic Correlation -.050 .693 Coeff. .709 Sig. (2-tailed) *.05