|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 DEVELOPMENT OF THE CONCEPTUALIZATION OF GROUP DYNAMICS INVENTORY (CGDI) By KEVIN A. TATE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Kevin A. Tate
3 To my amazing, beautiful wife Tyson a nd my miraculous daughter, Mercy
4 ACKNOWLEDGMENTS I want to thank WIlliam Conwill for his willingness to challenge me to understand the philosophical side of my work Ana Puig for her consistent willingness to help me w ith any part of my research, David Miller for making statistics seem simple and elegant, and Edil Torres Rivera for showing me that I can trust mysel f as a counselor and counselor educator. I would also like to thank my mom, my dad, and my brother Wesley like to thank my wife Tyson for supporting me emotionally, intellectu ally, spiritually and financially throughout this entire process. Without her I could not have done this and would not have wanted to do it without her.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGU RES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 10 Purpose of the Study ................................ ................................ .............................. 13 Definiti on of Terms ................................ ................................ ................................ .. 15 Group Counselor Case Conceptualization ................................ ....................... 15 Group Dynamics ................................ ................................ ............................... 15 Therapeutic Fa ctors ................................ ................................ ......................... 17 Level of Experience ................................ ................................ .......................... 17 Signific ance of the Study ................................ ................................ ........................ 17 2 REVIEW OF THE LITERATURE ................................ ................................ ............ 19 Therapeutic Factors ................................ ................... 19 Group Cohesiveness ................................ ................................ ........................ 21 Universality ................................ ................................ ................................ ....... 22 Altruism ................................ ................................ ................................ ............ 23 Synergy of the Three Dynamics ................................ ................................ ....... 24 Case Conceptualization in Group Counseling ................................ ........................ 27 Content and Organization of Group Counselors Case Conceptualization ........ 29 Factors That Influence Group Based Case Conceptualization ......................... 33 Group Based Case Conceptualization and Therapeutic Factors ............................ 35 A Framework for Researching Group Based Conceptualization ............................. 38 Assessing Group Based Mental Mod els ................................ ................................ 43 Summary ................................ ................................ ................................ ................ 46 3 METHODOLOGY ................................ ................................ ................................ ... 48 Research Quest ions and Hypotheses ................................ ................................ ..... 48 Development of the CGDI ................................ ................................ ....................... 49 Group Dynamics Inventory (GDI) ................................ ................................ ..... 50 Constructing Items for the CGDI ................................ ................................ ...... 52 Demographic Questionnaire ................................ ................................ ............. 53 Studying the CGDI ................................ ................................ ........................... 53 Data Analysis ................................ ................................ ................................ .......... 54
6 4 DATA ANALYSIS ................................ ................................ ................................ .... 55 Descriptive Statistics ................................ ................................ ............................... 55 Studyi ng the CGDI ................................ ................................ ................................ .. 57 Question 1 CGDI Validity ................................ ................................ ............... 57 Question 2 CGDI Reliability ................................ ................................ ........... 60 Demographics and the CGDI ................................ ................................ .................. 61 Qu estion 3 Counselor Race/Ethnicity ................................ ............................ 62 Question 4 Group Member Race/Ethnicity ................................ .................... 63 Question 5 Counselor Level of Experience ................................ ................... 64 Question 6 Group Counselor Gender ................................ ............................ 64 Summary ................................ ................................ ................................ ................ 65 5 DISCUSSION ................................ ................................ ................................ ......... 69 Summary of Findings ................................ ................................ .............................. 69 Validity and Reliability of the CGDI ................................ ................................ ... 69 Demographic Analysis ................................ ................................ ...................... 7 1 Interpre tation of Results ................................ ................................ .......................... 73 Next Steps in Developing the CGDI/GDI Method ................................ ................... 75 Limitations of this Study ................................ ................................ .......................... 77 Future Directions for the CGDI and the Mirror Instrument Method ......................... 77 APPENDIX A Items from the GDI (Phan et al., 2004) ................................ ................................ ... 79 B CGDI Items and Directions ................................ ................................ ..................... 81 C Additional Q uestions ................................ ................................ ............................... 83 D Email Invitation ................................ ................................ ................................ ....... 85 E Informed Consent ................................ ................................ ................................ ... 86 LIST OF RE FERENCES ................................ ................................ ............................... 88 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 93
7 LIST OF TABLES Table page 4 1 Group member race/ethnicity (as a percentage of group) ................................ .. 66 4 2 Counselor level of experience ................................ ................................ ............ 66 4 3 LISREL suggested changes to three factor model ................................ ............. 66 4 4 Relationship between experience and CGDI constructs ................................ ..... 67
8 LIST OF FIGURES Figure page 4 1 Factor l oadings for the three factor CGDI model ................................ ................ 68
9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DEVELOPMENT OF THE CONCEPTUALIZATION OF GROUP DYNAMICS INVENTORY (CGDI) By Kevin A. Tate May 2011 Chair: Dr. Edil Torres Rivera Major: Mental Health Counseling Case conceptualization is of central importance to the practice of group counseling. Although there has been research conducted on this topic there is not research that specifically addresses this topic in terms of widely accepted theor ies of dynamics in group counseling. Further, the resear ch conducted on this topic has not provided a consistent, parsimonious method for investigating group based case conceptualization. As such, this study focused on developing an assessment that will address major theories of group counseling, and will provi de a method to measure group based case conceptualizations in concise and consistent manner.
10 CHAPTER 1 INTRODUCTION The field of counseling is unique in that those who practice as counselors use themselves as the central tool of their professional activities In other words, counselors circumstances with the goal of using these conceptualizations to effectively work with their clients. This also holds true for group counselors Group counselors use their mental abilities to figure out what is happening in group counseling setting s in order to facilitate a therapeuti c process and a healing environment for their clients The centrality of case conceptualization in group counseling positions this skill as an extremely viable area for research. The Association for Specialists in Group Work (ASGW) recognizes th e importan ce of case conceptualization ability in their Professional Standards for the Training of Group Workers (ASGW, 2000) These standards state that grou p counselors should d emonstrate skill in observing and identifying group process, observing the personal characteristics of individual members in a group, developing hypotheses about the behavior of group members, and employing contextual factors ( e.g., family of origin, neighborhood of residence, organizational membership, cultural membership) in interpreta tion [i.e. conceptualization] of individual and group data (ASGW, 2000, B.2.a. d.) ASGW (1999) extends this discussion of contextual factors as it related to group own cultural awareness and the cultural backgrounds of their clients Group ethnic, cultural heritage, gender, socioeconomic status (SES), sexual orientation, practice negative emotional reactions toward Indigenous Peoples, African Americans, Asian
11 Americans, Hispanics, Latinos/Latinas, gays, lesbians, bisexuals, or transgendered p ersons and persons with physical, mental/emotional, and/or learning disabilities that II.A.1.) These factors are assumed influence the way in which group counselors concep tualize their groups and the dynamics within their groups. Many group counselor training models identify the development of case conceptualization ability as an important focus in the training effective group counselors (Bemak & Chung, 2004; Delucia Waa ck, 2002; Granello & Underfer Babalis, 2004; Okech & Rubel, 2007; Orr & Hulse Killacky, 2006; Rubel & Okech, 2006). All of these models are based on the assumption that case conceptualization in group work should be a key focus of effective training. Mor eover they specify factors that are uni que to case conceptualization in group counseling For example, Delucia Waack (2002) u ses a session] in terms of group process, progress toward member goals, critical incidents, 346). Other factors such as group development (Granello & Underfer Babalis, 2004), the multiple levels of interacti on and communication in groups (Okech & Rubel, 2007; Rubel & Okech, 2006), and unique multicultural influences on group process (Bemak & Chung, 2004; Okech & Rubel, 2007) are addressed by these training models as issues to consider when training group coun selors in forming effective case conceptualizations. In essence, these models suggest that there is a need to provide training that specifically addresses group based case conceptualization ability.
12 There is also a significant amount of research about cas e conceptualization in the context of group counseling. This research has focused on the content and Stockton & Morran, 1995; Kivlighan, Martin, Stahl & Salahuddin, 2007; Kivl ighan & Quigley, 1991; McPherson and Walton, 1970) and the factors that influence these conceptualizations (Christensen & Kline, 2000; Okech & Kline, 2006; Rubel & Kline, in terms of cognitive organization and complexity (Kivlighan, Martin, Stahl & Salahuddin, 2007; Kivlighan & Quigley, 1991; McPherson and Walton, 1970), professional experience feelings of competence (Okech & Kline, 2006), thematic content (Browne, 2005; Hines, Stockton & Morran, 1995), and emotional intensity of the group (Browne, 2005). This breadth of knowledge about case conceptualizations in the group context illuminates several important factors in terms of creating a more comprehensive understandin g of the content, organization, and factors that influence this process. Finally, there has been a call in the group counseling literature to be more conscious and purposeful about multicultural issues in relation to both clients and counselors ( ASGW, 19 99 2007; Phan & Torres Rive ra, 2004) This literature highlights the importance of process and outcomes of group counseling practice. These authors describe the reasons why multicultural competency is an indispensible part of competent group counseling.
13 Although the professional organizations (ASGW, 2000) and current training models (Bemak & Chung, 2004; Delucia Waack, 2002; Granello & Underfer Babalis, 2004; Okech & Rubel, 2007; Orr & Hul se Killacky, 2006; Rubel & Okech, 2006) in group counseling have emphasized the important of case conceptualization in terms effective professional practice, the research in this area has not provided a concise, consistent method for conducting research. In particular, the research has not been linked to specific methods for training effective group counselors. Additionally, there are no valid or reliable methods for assessing the effectiveness of practicing professionals in this field in terms of group c ounselor case conceptualization ability. And finally, although there has been a call for multicultural research in group counseling ( ASGW, 1999 ; Bemak & Chung, 2004; Delucia Waack, 2002; Granello & Underfer Babalis, 2004; Okech & Rubel, 2007; Orr & Hulse Killacky, 2006; Rubel & Okech, 2006). there has been no research in the field of group counseling that investigates the effect racial and/or ethnic identity may affect the outcome and process of group counseling. Given this gap and the lack of a method to assess case conceptualization ability, there is a need to create a method of assessment that fills this void. Purpose of the Study The purpose of this study is to adapt a current instrument (Group Dynamics Inventory GDI: Phan & Torres Rivera, 2004) in o rder to create a valid and reliable instrument (Conceptualization of Group Dynamics Inventory CGDI) for measuring accepted theoretical framework of group counseling dyna mics (Yalom, 2005) in order to create an instrument (CGDI) that will allow researchers, group counselor educators, and
14 onceptualization of itself. The CGDI is the accuracy of their conceptualization skills and the relationship between their own case conceptualizations and the way in w hich the group conceptualizes itself. This conceptualization and the conceptualization of the counselor is a significant factor in predicting therapeutic outcomes (Martin, Gars ke & Davis, 2000; Pan & Lin 2004) Additionally, this study investigated the relationship between group counselor responses on the CGDI in order to be sure it is an instrument that is not confounded by multicultural and professional training factors. The following questions were the questions addressed in this study. 1. Will the CDGI have the same three factor structure as the GDI, including the apparent validity based relationship between these three factors and the three factors of interest in this study (group cohesiveness, altruism and universality)? Hypothesis: The CGDI will have the same three factor structure of the GDI 2. Will the CDGI meet an acceptable level of reliability? Hypothesis: The CGDI will meet an acceptable level of reliability 3. and their responses on the CGDI Hypothesis: There will not be a significant relationship between group coun and their responses on the CGDI. 4. Will there be a significant relationship between the level of the three dynamics measured by the CGDI and the race/ethnicity of the group members? Hypothesis : There will not be a significant relations hip between the level of the three dynamics measured by the CGDI and the race/ethnicity of the group members
15 5. Will there be a significant relationship between group counselor level of experience and the three dynamics being measured by the CGDI? Hypothesis : There will not be a significant relationship between the level of the three dynamics measured by the CGDI and the grou p counselor level of experience. 6. Will there be a significant relationship between group counselor gender and the three dynamics being me asured by the CGDI? Hypothesis: There will be a significant relationship between the level of the three Definition of Terms Group Counselor Case Conceptualization As used in this study, group counselor case conceptualization is defined as the in a group, specifically in terms of theoretical constructs used in the group counseling profession (Yalom, 2005). The construct of mental models (Hinsz, 1995) will be developed and described in Chapter 2 of this dissertation. Group Dynamics Three specific factors will be highlighted in terms of discussing case conceptualizations. These factors are group dynamics that Yal om (2005) posits as Cole, 2005, p 25 ) The dynamics defined here are group cohesiveness, universality, and altruism ( Yalom 2005 ) Group cohesiveness Group cohesiveness is considered to be a precondition to effective group counseling ( Yalom 2005). This dynamic is an amalgamation of forces that make the group members feel a sense of ownership and membership in a
16 (Yalom, 20 05, p55). Universality In counseling groups, group members may enter with a heightened feeling of uniqueness in terms of their life experiences. Universality bers disclose concerns similar to their own, clients report feeling more in touch with the world and describe the process provides a context in which members feel more open to others identification with their own experience, and an openness to sharing more of their own experiences. Altruism Altruism is a dynamic in which group members feel positive about not only being helped, but also about having the opportunity to help others in the group. As the receiving giving reciprocal sequence, but also on profiting from something intrinsic ighlights another uniquely beneficial aspect of group therapy insofar as this dynamic can only take place in an environment in which clients are allowed and enabled to receive and provide healing in the same setting. An important caveat to add about these three dynamics is that they are not mutually exclusive in the group counseling process. Each dynamic overlaps with each other and with other group dynamics (Yalom, 2005). These dynamics interact in a synergistic way to provide a therapeutic context for group counseling.
17 Therapeutic Factors The concept of therapeutic factors overlaps with the concept of group dynamics in this study. Each of the dynamics described above (group cohesiveness, universality, and altruism) are in fact therapeutic factors. In other words, these dynamics are factors that contribute a therapeutic group counseling context (Yalom, 2005). Although these dynamics are therapeutic factors, there are other group dynamics that do not contribute to creating a positive therapeutic grou p setting. In the context of this study, it is only important for the reader to remember that the three group dynamics being discussed are also therapeutic factors. Level of Experience For the purposes of this study, level of experience will be defined o n three levels. First, this term refers to the number of years that a clinician has been practicing as a group counselor. Secondly, it will be measured as the number of group counseling focused academic courses a counselor has taken. Last, it will be mea sured by the number of group counseling focused continuing education hours a counselor has taken. Significance of the Study This study addresses a topic that is central to the field of group counseling. Although ASGW (2000; 1999) highlights group couns ability as an indispensable factor in effective group counseling, there is no clear or concise way of assessing these skills. Further, the research that has been conducted on this topic has looked at very general aspects of g roup based case conceptualization rather than focusing on the factors that are known to contribute to a therapeutic group counseling environment. This study will provide a parsimonious method of researching es based on therapeutic dynamics that
18 are posited by a widely accepted theory of group counseling (Yalom, 2005). The instrument that will be created will provide a method for further research, as well as assessment tool) f or group counseling trainees and practitioners (in combination with the GDI) to measure their level of case conceptualization skill in regard to any particular group with which they are working. Not only can this instrument be used to study general patter ns in relation to case conceptualization ability acquisition, development, and maintenance, it may also be used as a way to begin understanding the specific ways in which issues of diversity (ASGW, 1999) affect the therapeutic process in group counseling. In total, this provide a powerful research tool for those interested in discovering more about the acquisition, development, and maintenance of group based case concep tualization abilities, and (in tandem with the GDI) it will provide a context specific assessment tool for group counselor educators and practitioners.
19 CHAPTER 2 REVIEW OF THE LITERA TURE The previous chapter provided a brief overview of the importa nce of case conceptualization in the group counseling profession, a discussion of the need for and purpose of this study, and the research questions and hypotheses that this study addressed. This chapter will provide deeper discussion of the literature th at provides a foundation for the issues being addressed by this dissertation. It will conclude with a final summary and argument for the importance of developing an instrument to measure group counselor case conceptualization ability. Group Dynamics & Yal Therapeutic Factors The scientific study of groups is believed to have been founded by Kurt Lewin social processes have on members of a group. More specifically, thes e processes are social interaction, create patterned interrelationships among its members, bind (Forsyth, 1999, p 11) These group dynamics are therefore considered to be a key construct to be used (Cole, 2005, p 25). T he Theory and Practice of Group Psychotherapy (2005) counseling is the therapeutic factors that play a key role in the process and outcome of process that occurs through an intricate in
20 an environment where therapeutic change can happen and act as therapeutic forces to produce change in group members (Yalom, 2005, p. 1). This definition alludes to the fact that these therapeutic factors are a specific type of group dynamics. These therapeutic factors can create a therapeutic group counseling context and produce therapeutic change. In this study these factors will refer to specific, therapeutic types of group dynamics in order to both build the broader field of group dynamics in general. Yalom (2005) describes eleven therapeutic factors: the instillation of hope, universality, imparting information, altruism, the corrective recapitulation o f the family group, development of social techniques, imitative behavior, interpersonal learning, group cohesiveness, catharsis, and existential factors. All of these represent a unique group dynamic which ultimately lead to therapeutic change in group co unseling. Although each of these factors are considered to be unique in important ways, they are Given this tension between uniqueness and interdependence, it is importan t to understand these group dynamics in a way that addresses both their unique qualities and their synergistic nature. As such, this study will also address each of these. Although each of these eleven therapeutic factors are considered to be important i n creating therapeutic change in a group counseling context, three of these group dynamics are uniquely important in that they are both therapeutic factors and group co hesiveness, universality, and altruism. Each of these group dynamics act as an agent of change in group counseling, and constitute conditions for therapeutic
21 change in the group counseling context. These three dynamics will be discussed in terms their im portance in the therapeutic process and the way in which they contribute to conditions for therapeutic change. In addition to describing the unique aspects of each of these dynamics, their synergistic nature will also be discussed. Group C ohesiveness counseling processes and outcomes has shown that the therapeutic relationship accounts for a very l arge percentage of outcomes for clients. In a meta analysis of the literature Lambert (1992) found that the therapeutic relationship in individual counseling and therapy accounted for 30% of the outcome for clients. While it is helpful to compare this re search with the concept of cohesiveness in group counseling, there are some limitations to this comparison. Group cohesiveness is a more complex construct, members, and to defined in many different ways in the literature, Yalom (2005) defines group cohesion as the attraction that group members feel for a particular group. This dynamic plays a foundationa l role in group counseling. It serves as a therapeutic factor in that it allows group members to feel that they are a part of an accepting group that is concerned for their wellbeing. Group cohesiveness also serves as a condition for change in that it se rves as the therapeutic relationship in which change can take place (Yalom, 2005). 55).
22 Research on group cohesion supports the idea that this group cohesion is both a therapeutic factor and a condition for therapeutic change. MacKenzie ( 1983) developed a questionnaire that was intended to assess the climate of group. The Group Clima te Questionnaire Short Form (GCQ S : MacKenzie, 1983) includes a subscale that measures the construct engagement (2005) concept of group cohesion. A factor analysis of this questionnaire showed that engage ment included both support (group cohesion as a precondition for change) and work (group cohesion as a therapeutic factor) dimensions of the group. This instrument provides evidence that group cohesion serve in both of these capacities. Kivlighan and L illy (1997) i nvestigated the effect that therapeutic factors had on outcomes for clients in group counseling using the GCQ S ( MacKenzie, 1983) The engagement subscale of this instrument was investigated in relationship to client outcomes of group counsel ing. The construct of engagement is conceptually related to and thought to encompass the concept of group cohesion ( MacKenzie, 1983) The researchers found that group member engagement was a significant predictor of therapeutic outcomes for these clients Universality that their experiences are not completely unique. In other words, universality describes the dynamic wherein group members discover that there are others wh o have experiences, thoughts and feelings that are similar to their own. Yalom (2005) (p. 6). This dynamic is highly important in terms of experiences or feelings that a re taboo in our culture such as issues of sexuality, suicide, and feelings of inferiority. The
23 dynamic of universality constitutes a therapeutic factor because it allows group members to understand that they are not alone in their life experiences, and he lps them to begin normalize what they may be dealing with. Universality also serves as a precondition for change in that, by building a foundation of common experiences, thoughts and feelings, it provides a context for group members to see the group as a place where their own experiences will be understood and accepted by the group (Yalom, 2005). Research has supported the therapeutic nature of universality. Kivlighan and Mullison (1988) conducted a study to determine the role that therapeutic factors pl ayed in relationship to group developmental stages. The findings showed that group members perceived universality as significantly more important in early stages of the group. This finding highlights both the fact that group members perceived universalit y universality also seems to serve as precondition for therapeutic change via other dynamics in the group counseling context. Altruism A unique feature of gro up counseling can be found in the way in which group members are able to both give and receive healing. Altruism refers to the process giving reciprocal sequence, but also on profiting from something intrinsic to the act of members may feel that they do not have anything helpful to offer to others, and provides a method of empowering clien ts to feel that they have the resources to heal both themselves and the other people in the group. There is also the obvious therapeutic
24 effect this dynamic has for providing multiple sources of change and healing (other than the client therapist dyad) fo r all members of the group. Altruism also serves as a unique precondition for therapeutic change as it sets the context for multiple group members to enact the power of the entire group to engage in the healing and change process in both a giving and rece iving fashion (Yalom, 2005). Altruism has also been shown to be a significant therapeutic factor in group counseling. Waldo, Kerne and Kerne (2007) studied the importance of therapeutic factors between two different types of groups. The results showe d that the importance of altruism did not vary across these two different types of groups. These findings suggest that altruism is a stable factor that contributes to therapeutic outcomes. Although these findings support the idea that altruism is a thera peutic factor, there is a lack of evidence to support the theoretical notion that altruism is also a precondition for therapeutic change. Even so, the interrelation of altruism with universality and group cohesion suggests that it may also be a preconditi on for change. Synergy of the T hree D ynamics While group cohesion, universality, and altruism are unique in important ways, they also overlap and act synergistically to create therapeutic change and to serve as a basis for conditions of healing in the group counseling context (Yalom, 2005). Although the complex, synergistic interaction of these dynamics does not lend itself to an all encompassing description, this author will attempt to outline the ways in which these dynamic needed for all other therapeutic factors to function, it is best begin the discussion with the way in which group cohesion interacts with universality and altruism.
25 Group cohesion interacts with universality and altruism to produce therapeutic c cohesion, they also generate that cohesion, creating durable relationships perhaps for s not explicitly posit how cohesion is generated in terms of universality and altruism, it is implied that group cohesion is generated in part when group members begin to accept each other based on the feeling that they are not alone in the way in which th ey from Yalom (2005) describing the therapeutic power of group cohesion suc cinctly shows how group cohesion is connect ed to universality and altruism. A cceptance and understanding among members [universality] may carry greater power and meaning than acceptance by the therapist. Other group members, after all, do not have to care or understand. They are not paid ] (p 63 ). In discussing the dynamic of altruism, Yalom (2005) also states that, in his experience, group members often state that they improve based on the fact of group members ply having been present and allowing their fellow members to grow as a result of a dynamic and overlapping relationship between altruism and group cohesion as therape utic factors. Group cohesion seems to be connected to and influenced by the process of feeling understood by others and understanding others in the same way (universality), and by actively and freely contributing to the healing of others in the group (alt ruism).
26 In the same way that these three dynamics interact to produce therapeutic change, they also interact to produce conditions for this type of change to take place. In only a potent therapeutic force in its own right. It is a precondition for other therapeutic factors therapeutic factor and a condition for therapeutic change, the interplay of universality and altruism highlighted above also applies to this creation of therapeutic context. An environment in which members feel understood and accepted by others in terms of having similar experiences (universality), and feel that they are part of group that both receive and give healing in (altruism) allows for the other eight therapeutic factors to be enacted. Empirical support for the interrelationship of these three dynamics has been provided by MacNair Semands and Lese (2000) They cond ucted a study to investigate problems. These researchers measured the therapeutic factors near the beginning and the end of the group sessions. The results showed that per ception of the levels of altruism, universality and group cohesion all increased as a function of time in the group. While inconclusive in terms of understanding the exact relationship among these dynamics, these findings suggest that these dynamics may b e related in some way. In Pan and Lin (2004) i therapeutic factors and their perception of group counselo finding in this study showed that there was a positive relationship between the
2 7 (group cohesion, universality and altruism). These three dynami cs correlated positively with leader behaviors above all the other therapeutic factors. In other words, when group members had very positive perceptions of group counselor behaviors, the factors of group cohesion, universality and altruism were rated as i mportant dynamics in the group in terms of therapeutic change. This finding shows a direct connection between the way in which group counselors facilitate their groups and the dynamic interaction of group cohesion, universality and altruism. The research cohesion, universality and altruism are important in terms of facilitating therapeutic change in the group counseling context. The research has also revealed that these dynamics are dynamically r elated to one another within counseling groups. Further, one study has shown that these three dynamics are significantly related to the way in which group counselors facilitate their groups. This finding in particular shows not only the importance of the on the ways in which these dynamics are experienced. To better understand group counselor skill and effectiveness, it is important to discuss what unique skills are needed. Of particular i mportance is understanding the way in which group counselors conceptualize what is happening with their client groups. Case Conceptualization in Group Counseling One of the unique features of group counseling is found in the notion that those involved in this profession use themselves as the central tool for carrying out their work. More specifically, group counselors use themselves to conceptualize what is happening in a group at any particular moment in time. The Association for Specialists in Group
28 Wo rk (ASGW) highlights the importance of conceptualization in group work in their Professional Standards for the Training of Group Worke rs (ASGW, 2000) These standards state that grou p counselors should d emonstrate skill in observing and identifying grou p process, observing the personal characteristics of individual members in a group, developing hypotheses about the behavior of group members, and employing contextual factors ( e.g., family of origin, neighborhood of residence, organizational membership, cultural membership) in interpretation of individual and group data (ASGW, 2000, B.2.a. d.) ASGW (1999) exten ds this discussion of contextual factors as it related to group own cultural awareness and the cultural backgrounds of their clients ASGW asserts that g roup counselors should be aware of the influence of and i nfluence on their clinical practice as group workers (ASGW, 1999, I.A.2.), and should Peoples, African Americans, Asian Americans, Hispanics, Latinos/Latinas, gays, lesbians, bi sexuals, or transgendered persons and persons with physical, knowledge about these populations (ASGW, 1999, II.A.1.) These factors are assumed to have an influence on the way in which group counselors conceptualize their groups and the dynamics that occur within these groups. These professional standards dynamics that occur and de velop within groups and the factors that influence these dynamics.
29 The importance of case conceptualization in group counseling is supported by a significant body of research. In general, this research in separated into two categories the factors that influence group based case conceptualization. Content and Organization of Group Counselors Case Conceptualization This area of research focuses on the specific content ( e.g., cognitions or thoughts) and/or organization ( e.g., cognitive complexity or order of importance) of these conceptualizations. The primary purpose of these studies is to study group based case conceptualization from a perspective that focuses on what formulations group counselors make of their groups and the structu re of these formulations. McPherson and Walton (1970) conducted a study in order to determine the way in which experienced group clinicians organized their thoughts as these thoughts relate conceptualizations were assessed by having each of them list ways in which individual group members were alike or different in terms of the constructs they found important in differentiating among members. McPherson and Walton (1970) found that there were t hree bi polar constructs that the clinicians used to conceptualize the clients in the study. These constructs were dominance submissiveness warm/sensitive cold/insensitive and hinders attainment of group goals aids attainment of group goals Based on t he consistency of these constructs across seven clinicians and the similarity of these constructs to ones used in previous literature on this subject, the authors drew the implication that these constructs hold enough validity to be considered worthy of fu rther study.
30 In an effort to further the understanding of the way in which group workers conceptualize their clients, Kivlighan and Quigley (1991) investigated the differences in content, cognitive complexity and organization of conceptualizations of novice and experienced group therapists. These research ers directed the participating group therapists to first rate all possible pairs of group members in terms of their similarity to one another. The participants then used words to describe these differences. It was discovered that experienced and novice g roup therapists differed in their conceptualizations in three ways First, the experienced therapists had a more cognitively complex way of organizing their conceptualizations about group members. Secondly, the experienced therapists used more differenti ated distinctions in their formulations about group members. Lastly, experienced and novice group members used different dimensions to conceptualize the group members. Experienced therapists used dominant/submissive friendly/unfriendly and supporting t herapeutic work/hindering therapeutic work dimensions, whereas novice therapists used dominant/submissive and rate of participation dimensions. Kivlighan and Quigley (1991) state that, given the similarity of the constructs used by expert therapists found in this study with the constructs found in the McPh erson and Walton (1970) study, these constructs seem to be recurring themes in terms of the ways in which group counselors conceptualize their clients. Using the same method of comparing all possible pairs of individuals in a group, Kivlighan, Martin, Stahl and Salahuddin (2007) researched the ways in which novice ognitive organization and complexity changed as a result of completing their first group practicum course. Within the context of this group
31 practicum course, the researchers compared the conceptualizations of an expert therapist (instructor of the course ) with those of novice therapists (students in the course). The study participants acted as both observers and leaders of an actual therapy group, alternating between these roles throughout the semester. There were three findings that emerged. First, th is study replicated the findings of Kivlighan and Quigley (1991) in regard to novice versus expert conceptualizations of group therapists. conceptualizations of the group members increased in complexity. Third, the novic e those of the expert therapist. The authors suggest that these finding are important over time in terms of recognizing the complexity of group members as well as (Kivlighan et al. 2007, p. 185) A significant amount of research has also been conducted on the self talk (cognitions) of group counselors. The initial research on this subject was directed at the as it relates to level of expertise in group work. Hines, Stockton and Morran (1995) report the results of study designed to record novice and expert self recorded thoughts about a staged video segment of a counseling group in an effort to find the differences between the self talk of these two groups. This study found that there were 17 distinct categories of thoughts that interpretation of group process category was the strongest predictor for expertise in group work. Of the 17 categories, it also seems that three o f them are directly related to the conceptualizations formed by
32 group counselors interpretation of members interpretation of group process and interpretation of co leader In a study of the intentions behind interventions of group leaders (Stockton, Morran, & Clark, 2004) the intention category assessing growth also tentions to conceptualize the development the group and/or its members. In an effort to summarize the larger body of research on self talk, Browne (2005) described dissertations that Rex Stockton supervised. Additional findings about self talk (2005) r eview. Self talk is not only affected by level of experience, but is also affected by the emotional intensity of the group being lead. Findings from self talk research have also suggested that the ability to identify critical incidents (a specific type of conceptualization) is related to level of experience (Browne, 2005). In summary, each of these studies addressed the content and organization of this in terms of thematic content, cognitive complexity, and organization of content (Kivlighan et al. 2007; Kivlighan & Quigley, 1991; McPherson & Walton, 1970) A separate body of research on group counselor self talk (cognitions) specifically addressed the content of these conceptualizations (Browne, 2005; Hines et al. 1995; Stockton et al. 2004) These studies highlight a focus on the individual group counselor and the specific content and organization of their group based case conceptualizations. Although this is an important focus for research on this topic, ot hers have focused on factors that
33 Factors That I nfluence Group Based Case Conceptualization As opposed to the study of the content and organization of group based case conceptualizati ons, this area of research has focused on factors that influence these conceptualizations. For example, as opposed to focusing on cognitions of counselors in experie nce of group supervision influences their conceptualizations of their client groups. The general purpose of this stream of research is to understand how Christensen and Kline (2000) conducted a qualitative investigation of group supervision group for group counselo rs in a counselor education program, two themes were discovered that seem to speak directly to the concept of conceptualization in a group setting. The first theme was the multiphasic learning process that the students upervisees used [group] themes, theories, and philosophical assumptions to comprehend their experiences in group supervision and (Christensen & K line, 2000, p. 389) The second theme was multiphasic learning outcomes These outcomes flowed from the multiphasic learning process and represented a deeper understanding of group process in the supervision groups and in the groups that they were cu rrently leading. An interesting relationship was also found in this study in that, as the supervisees began to better understand their experience in the supervision group (increased conceptualization ability), anxiety related to participation in the super vision group decreased. Okech and Kline (2006) explored the way in which group co leaders who were working together for their first time experienced their relationship with each other in a
34 clinical setting. The do minant theme that emerged from this investigation was that co leaders begin their relationship with significant concerns about their competence, as evaluated by both themselves and their co leader. In terms of conceptualization, it was discovered that as conceptualizations of group events, interventions, and intervention goals that differed from those of their co (Okech & Kline, 2006, p. 178) Theses findings suggest that the conceptualizations that are formed by group co leaders are related to the quality of the relationship between them. An important thing to note about t his study is that all the participants had at least a nominal amount of experience as group leaders, which may suggest that competency concerns remain a significant influence on conceptualization in groups across different levels of counselor experience. A recent study by Rubel and Kline (2008) sought to understand the experiences that expert g roup leaders have as they lead groups. The dominant theme that emerged from this investigation was that experiential influence was the dominant factor that contributed to the phenomenological experiences the participants had as group leaders. The partici sub themes of experiential influence The sub theme of leader resources served as a basis from which the participants developed their conceptualizations. The process of developing case conceptualizations is reflected in the sub theme of leadership process This process is related to conceptualization in that: t he experts described the experience of leading groups as a recurring circular process of engaging with the group, percei ving interaction, developing understanding of these perceptions, and deciding how to
35 engage further based on their understandings. (Rubel & Kline, 2008, p. 150 151). In particular, it seems that the notion of understanding in the above statement is direc tly related to the concept of conceptualization. The expert group leaders in this study felt experiential resources. Another finding of this study is that some of the participants felt that their extensive experience had given them that ability to have int uitive understandings (conceptualizations). This finding suggests that a qualitative change in conceptual ability occurs as a result of significant experience as a group leader. concept ualizations were influenced by the learning process in group supervision (Okech & Kline, 2006) the quality of group co leader relationships (Okech & Kline, 2006), experience working with groups, and the actual proc ess of working with current groups (Rubel & Kline, 2008). Each of this studies highlight a focus on understanding what factors play a role in influencing the conceptualizations that group counselors form about their client groups. Group Based Case Concept ualization and Therapeutic Factors Although one research study highlighted above specifically addresses the issue of the way in which group counselors conceptualize the therapeutic factors in a group counseling context (Hines, Stockton & Morran, 1995) there is a paucity of research on this topic. Although there has perceptions of therapeutic factors ( Kivlighan & Lilly 1997; Kivlighan & Mullison, 1988;
36 MacNair Semands & Lese, 2000; MacKenz ie, 1983 ; Waldo, Kerne & Kerne, 2007 ) the way in which these factors are related to counselor effectiveness ( Pan & Lin, 2004 ), and on the way in which group counselors conceptualize their client groups (Browne, 2005; Hines, Stockton & Morran, 1995; Kivlighan & Quigley, 1991; Kivlighan et al. 2007; McP herson & Walton, 1970), none of are actually present and what the quality of these dynamics are in a particular group. Further, none of these studies focused on the way in which multicultural factors influence case conceptualization in a group context. This is problematic given th e direct call for professionals to focus on these dynamics in group counseling (ASGW, 2000), and the call to focus on the contextual issues that a ffect the ways in which group counselors conceptualize their client groups (ASGW, 19 99). Although there is a lack of research on this topic, some of the professional literature has pointed out the need to ese issues in the context of training future group counselors. To create a framework for training group counselors, DeLucia Waack (2002) proposed a method for preparing group workers for supervision sess ions. This framework called for the use of a detailed writing plan to facilitate effective supervision of group workers. One particular section within this writing plan deals specifically with conceptualization of group dynamics in the group context. In the sectio n of the writing toward member goals, critical incidents, therapeutic factors, ef fective interventions, and
37 DeLucia Waack (2002) focus on therapeutic factors highlights the key role that group dynamics should play in forming effective conceptual izations of client groups. Bemak and Chung (2004) provided a framework for teaching multicultural group counseling. Within this framework, the authors suggested that several factors are directly related to grou it is important that group counselor educators are culturally competent and understand cross cultural and socio ecological issues that impact group process and con individual orientations, historical and sociopolitical backgrounds of clients, client awareness. Bemak and Chung (2004) provided several specific guidelines for training culturally competent group workers. Of the 17 guidelines put forth by these authors, 11 that th e authors are placing a large emphasis on the factors that affect the conceptualizations that group workers form about their clients and client groups. This training program implies a direct focus on addressing contextual issues in terms of understanding conceptualizations of the dynamics present in a particular group. Rubel and Okech (2006) addressed c onceptualization of group dynamics within their model for supervising group workers. Specifically, the supervisor directs her or his supervision of conceptualization of group dynamics by focusing on the individual, interpersonal, and group as a whole leve ls of the group sessions. All of these levels
38 are seen to be a necessary component of a complete conceptualization of what may be happening in a group. Okech and Rubel (2007) suggest that the Supervision of Group Work Model may also be used to integrate issues of diversity in the supervision includes knowledge of the impact of dive rsity upon individuals, subgroups, and overall counselors to better conceptualize the dynamics at work in a particular group, as well take into account factors that are influence t hese dynamics and their conceptualizations thereof. An important thing to note is the intentional focus on multicultural issues in two of the articles (Bemak & Chung, 2004; Okech & Rubel, 2007). These authors point to the direct connection between the mul ticultural factors associated with both the client and the group counselor, and the dynamics that are present within a counseling group. This issue is also asserted in term of the birth language of clients in a group counseling ) also highlights the important role that racial and cultural understanding of the dynamics pre sent in a counseling group, but also the effect that multicultural issues may be having on this understanding A Framework for Researching Group Based Conceptualization Given the importance of the therapeutic factors in group counseling (Yalom, 2005), and the consequent professional responsibility to form accurate conceptualizations of these dynamics ( ASGW, 2000) and the factors that influence them (ASGW, 199 9), a coherent framework for research, training, and practice is needed in
39 order to promote competen t practice in the group counseling field in terms of conceptualizing group dynamics. conceptualization in a group context p rovides such a coherent model. This framework highlights the specific nature of group ba sed conceptualization, and posits a phenomenological perspective for understanding this type of conceptualization. Hinsz (1995) uses the concept of mental models to address group based (Hinsz, 1995, p. 201). In other words, this framewor k describes mental models as the phenomenological experience individuals have of a group they are a member of. insz (1995) also highlights several specific traits of groups that underlie the way in which these mental models are to be understood. These traits include that metal models are (a) dynamic (they change over time with respect to changes in the individual and the system in question), (b) specific to a particular system, (c) gained through experience, and (d) idiosyncratic to the individual group member. Although this framework is applied to several contexts ( e.g., juries or work groups), it has specific i mplications for group counseling. and group based case conceptualizations, this paper will discuss this model in terms of the information covered so far in this chapter. In terms of group cohesiveness, universality and altruism, the framework of mental models positions these dynamics as a property of the group. Group members form a subjective understanding and mental
40 representation of these dynamics. Specifically, these dynamics represent a reality of the group that each member forms a subjective perception about. This also holds for the group counselor in that the counselor also forms a mental representation of these dynamics in relation to the reality of the group in q uestion. The importance of this mode of viewing group based case conceptualization is supported by research findings. Martin, Garske and Davis ( 2000 ) therapeutic alliance is more reliable that the therapi relationship between the conceptualization of group members and the conceptualizations of the group counselor given that the perspective of t he group member may be more relaible. MacKenzie ( 1983) also reported that when both group members and group therapists are given the GCQ S ( MacKenzie, 1983) the perspectives differ significantly. Again, this shows the importance of understanding group c ased case conceptualization in terms of the unique perspective of the group members and the group counselor. The mental models framework (Hinsz, 1995) provides a very useful lens for understanding these multiple perspectives and their relationship to ther apeutic change. Another important aspect of the mental models framework is that mental models are posited to be dynamic in that they change over time and in respect to changes within the group (Hinsz, 1995). This dynamic nature of group based conceptu alizations has specific implications for the way in which this topic should be studies. In other words, given that these mental models change in a dynamic fashion over time, it follows that research on these conceptualizations should be designed to be sen sitive to these
41 conceptualizations change dynamically in relation to factors that occur within the group context. Both Pan and Lin (2004) and MacNair Semands and Lese (2000) found that therapeutic factors change as result of time. This suggests that as group dynamics change over the life of a group, the mental models of these dynamics also change. This also echoes what many authors in this field have called for in terms of research on group counseling that seeks to understand how groups develop over time in regard therapeutic factors ( Kivlighan & Mullison, 1998; MacNair Semands, 2000; Marmarosh, Holts & Schottenbauer, 2005; Nitza, 2005; Yalom, 2005). This specifically applies to group based case conceptualization in that it is important to understand the way in which group based case conceptualizations develop and change, especially in regard to dynamics ther ein (Martin, Garske & Davi, 2000) The specifically unique nature of each group also has implication for group based h the specific nature of group counseling and the unique traits and experiences that the group members bring to the group counseling context. In terms of the specific nature of group counseling, it is important to group counselors to focus on specific fac tors (group cohesion, universality and altruism) that facilitate therapeutic change (Yalom, 2005; Delucia Waack, 2002). It also important for group counselors to focus on the unique traits and experiences that both themselves and the group members bring t o the group counseling context ( e.g., race, ethnicity, and gender) ( ASGW, 2000; ASGW, 1999; Bemak & Chung, 2004)
42 Because mental models are specific to a particular group, these specific attributes are a key component of effective group based case conce ptualizations. The fact that mental models are gained through experience also has clear implications for conceptualization in group counseling. A significant amount of research has addressed the way in which professional experience influences the case conceptualizations of group counselors (Hines, Stockton & Morran, 1995 ; Kivlighan et al. 2007; Kivlighan & Quigley 1991 ; M cPherson & Walton, 1970; Rubel & Kline, 2008). This body of research provides definitive evidence that mental models do in fact change in terms of experience with a certain type of group (group counseling). This is also relevant to group members in that each member or collection of members may have varying levels of experience in group counseling. Again, this relates to the relationship between group counselor conceptualizations of the group dynamics and the group mics in that they may differ based on the level of experience with the group counseling context. The idiosyncratic nature of mental models provides a final implication for group yncratic to that individual because individuals develop their mental models through their particular viewing the conceptualization of group dynamics from a group as a whole perspective versus an individual experience perspective. Rubel and Okech (2006) highlight this tension in their model for training group workers. This perspective on teaching group based case conceptualization highlights the importance of conceptualizing group dynamics at the individual, interpersonal, and group as a whole level. The mental
43 models framework allows for this tension by positing a phenomenological ex perience of these dynamics that are unique to individuals and subgroups within the group, and as as a whole. This is important given that mental models of groups are dynamic, specifi c to the particular group, and gained through experience. All of these coalesce in several phenomenologically unique conceptualizations of any particular group and the dynamics therein (Hinsz, 1995). The use of mental models (Hinsz, 1995) to address grou p based case conceptualizations provides a comprehensive, holistic view of the way in which both group counselors and group members conceptualize group cohesion, universality and altruism (Yalom, 2005). In addition, this framework describes specific trait s of counseling groups that allow for an understanding and analysis of the factors that influence these conceptualizations. While research on group therapeutic factors and group based case conceptualizations has provided a wealth of knowledge, there has b een a lack of a coherent framework to integrate these constructs. Given this succinct and parsimonious method for addressing this integration, one question remains how can researchers and practitioners study and measure group based case conceptualizatio ns using the mental models framework? Assessing Group Based Mental Models In order to use the framework of mental models (Hinsz, 1995) to measure group based case conceptualization, two factors need to be addressed. First, the phenomenological experience of the group members must be assessed (group
44 measurement that retains fidelity to this framework must include an assessment procedure that addresses that fact that mental models are dynamic, specific to a particular group, gained through expereince, and idiosyncratic to the individual (Hinsz, 1995). Finally, this assessment procedure must address th e three therapeutic factors of group cohesion, universality and altruism given that these group dynamics are key to both therapeutic change and the creation of a therapeutic environment in the group counseling context. As such, this section will highlight currently used assessments that measure the mental models of group members in relation these three group dynamics, and describe the need for an instrument that assesses the mental of models of group counselors as they directly relate to the mental models of the group members. There are several assessments that have been created to assess the mental one in particular that is interesting in terms of developing a measur ement consistent the Group Dynamics Inventory (GDI: Phan, Torres Rivera, Volker & Garrett, 2004) The Group Dynamics Inventory. The Group Dynamics Inventory (GDI: Phan, Torres Rivera, Volker & Garrett, 2004) was developed to assess the mental m odels of group members in relation to group cohesion, universality, and altruism. This instrument drew items from a wide range of group counseling measures, and in definitions of group cohesion, universality and altruism. A total of 60 items were selected, and after review by an expert panel, 20 were retained as being representative of one of these three dynamics. A piloting of the instru ment was conducted, and an
45 exploratory factor analysis revealed that there were in fact three constructs represented in the GDI, each of which seemed to be one of the three therapeutic factors that were targeted in its development. This assessment provide s a method of assessing mental models of group members in terms of their perception of group cohesion, universality and altruism. Additionally, the form of this measure allows for the multiple administration of the instrument across the duration of a grou p to determine changes that occur in these mental models. The GDI (Phan et al. 2004) seems to be very well suited to be used as one half of the assessment procedure for assessing group counselors mental models of group dynamics as they relate to group me particular, it is uniquely suited for this task because it is brief enough to be used as method of multiple measures over the life of a group (addressing the dynamics nature of mental models), its design allows i t to assess both the individual and group as a whole nature of group based mental models (addressing the idiosyncratic nature of mental models), and is designed to be used for the assessment of different types of groups (addressing the group specific natu re of mental models) (Hinsz, 1995). Although it does not directly address the fact that mental models are gained through experience (Hinsz, 1995), this aspect will be addressed by the proposed instrument described in the next chapter. Given that the GDI (Phan et al. 2004) is equipped to address the mental (2005) factors of group cohesion, universality and altruism, the next step is creating an assessment procedure that assesses the mental models (phenomenological experience
46 of the group cohesion, universality and altruism) of group counselors in relation to the mental models of group members in terms of these three factors. Summary Group dynamics are a central force in group counseling. In particular, three specific group dynamics are both therapeutic factors and conditions for therapeutic change group cohesion, universality and altruism (Yalom, 2005). These dynamics are also cen tral in terms of what must be considered by group counselors when conceptualizing their client groups. These dynamics should be addressed in terms of factors that may be influ (ASGW, 1999). Researchers in this field have addressed the content, organization, and research in terms of specif conceptualize these particular dynamics and the factors that may be influencing these conceptualizations of group cohesion, universality and altruism. The mental models framework (Hinsz, 1995) provides a useful basis for creating an assessment process mental models of these three dynamics. Further the GDI (Phan et al. 2004) is an assessment that specifically measures g roup cohesion, universality and altruism, and is well suited to be used as one half of an assessment process that will employ the mental models framework (Hinsz, 1995) in the service of creating a method for understanding the relationship between mental mo dels of group members and the mental models of group counselors. The purpose of this study was thus to create such an assessment
47 and subject it to initial validation and reliability tests, as well as study issues of demographics that may confound response s to the CGDI
48 CHAPTER 3 METHODOLOGY The rationale and literature basis for the importance of understanding group presented in Chapters 1 and 2. The professional literature sho wed that each of these group dynamics act as both therapeutic factors and as conditions for therapeutic change. Additionally, the literature supports the need for an assessment method that ely conceptualize these dynamics in a given counseling group. This chapter will outline the methodology for the development and the initial studying of an instrument that will measure said conceptualization the Conceptualization of Group Dynamics Invento ry (CGDI). Research Questions and Hypotheses The first two research questions are focused on the development of the CGDI and the establishment of validity and reliability thereof. 1. Will the CDGI have the same three factor structure as the GDI, including the apparent validity based relationship between these three factors and the three factors of interest in this study (group cohesiveness, altruism and universality)? Hypothesis: The CGDI will have the same three factor structure of the GDI (based on accept able factor loadings and goodness of fit measures found in a confirmatory factor analysis). 2. Will the CDGI meet an acceptable level of reliability? Hypothesis: The CGDI will meet acceptable levels of reliability (based on study). The last four questions dealt specifically with the investigating the relationship between 3. and their re sponses on the CGDI
49 Hypothesis: There will not be a significant relationship between group multiple regression analysis of the relationship between each counselor reported race/ethnicit y, and each of the three constructs measured by the CGDI) 4. Will there be a significant relationship between the level of the three dynamics measured by the CGDI and the race/ethnicity of the group members? Hypothesis: There will not be a significant relatio nship between the level of the three dynamics measured by the CGDI and the race/ethnicity of the group members (based on a multiple regression analysis of the relationship between each counselor reported race/ethnicity of their group members, and each of t he three constructs measured by the CGDI). 5. Will there be a significant relationship between group counselor level of experience and the three dynamics being measured by the CGDI? Hypothesis : There will not be a significant relationship between the level of the three dynamics measured by the CGDI and the group counselor level of experience (based on a multiple regression analysis of the relationship between measures of counselor reported level of experience, and each of the three constructs measured by the C GDI). 6. Will there be a significant relationship between group counselor gender and the three dynamics being measured by the CGDI? Hypothesis: There will not be a significant relationship between the level of the three dynamics measured by the CGDI and the g gender (based on a individual sample t test) Development of the CGDI In order to assess the conceptualization ability of group counselors in terms of group cohesion, universality and altruism, this section will describe the development of the Conceptualization of Group Dynamics Inventory (CGDI). This inventory was developed by adapting the Group Dynamics Inventory (GDI; Phan et al. 2004). As such, the GDI will be described in order to highlight the theoretical basis and assessment methodo logy that the CGDI will be drawn from.
50 Group Dynamics Inventory (GDI) The GDI (Phan et al. 2004) was developed in the effort to assess the presence and level of group dynamics in group counseling. This instrument is based theoretically theory of group counseling and psychotherapy. Specifically, this assessment was designed to measure group cohesiveness, universality and altruism because the research these authors reviewed supports the importance of these dynamics (as well as the literat Yalom ranks altruism, cohesiveness, and universality as the top three therapeutic factors in a working group. Thus, in the development of the Group Dynamics Inventory these three factors Phan et al. 2004, p. 2 35). Potential items for the GDI were drawn from several existing group counseling instruments ( Frank Saracini et al. 1998 ; Hurley & Brooks, 1987 ; Kivlighan, Multon, & Brossart, 1996; Lese & MacNair Semands, 2000; Marcus, 1998 ; Mullen et al. 1998; Phan, 2001; Wilbur et al. ., 1997a, 1997b [all as cited in Phan et al. 2004] ). In particular, sort instrument given that some items in this assessment had a direct theoretical connection with the three fac tors being addressed by the GDI. In total, 60 items were selected to be considered for inclusion in the GDI. In order to establish content validity, these 60 items were then evaluated by a panel of six counsellor educators that were deemed to be experts in the field of group counselling. Twenty of these items were retained by keeping statements that were agreed upon by at least five of the experts as fitting for one of three constructs being measured. Each of the twenty items is a statement about group memb behaviours in regard to the group they are a part of. The responses to the GDI are given via a Likert scale that ranges from 1 (strongly disagree) to 4 (strongly agree). This
51 response system is constructed such that higher scores on each co nstruct specific item corresponds to a higher level of that dynamic in a group. The GDI (Phan et al. 2004) was piloted by administering the assessment to two hundred counseling students. This sample of counseling students consisted of 183 females, 17 ma les, eight Asian Americans, seven Latina/os, two self identified biracial (Asian and African American, and Filipina and Caucasian), and 183 White students. The 54, with a mean age of 31. In terms of their academic career, t he respondents were in their final stage of their training as supervision course. The GDI was administered to the students during one of the sessions of this group supervi sion course. In order to investigate the factor structure of the GDI (Phan et al. 2004) an exploratory factor analysis (EFA) was conducted on the data gathered from the pilot study. Three factors emerged from this analysis. Factor 1 accounted for 49.6% of the variance, and 64.4% was accounted for in a three factor solution. Factor 1 (labelled cohesiveness) included 10 of the items. Items in this category had factor loadings ranging from .70 to .83. These items addressed issues related to feeling a part o f the group and being accepted by the group members ( e.g., helping others and making an impact in their therapeutic process ( e.g., loadings ranging from .59 to .84. Factor 3 (labelled universality) consisted of five items
52 struggle in life ( e.g., ( e.g., loadings ranging from .77 to .87. The int ernal consistency of all items on the GDI was half reliability coefficient was calculated to be .87 and .92. The three factor solution of the EFA supports the initial ta rget constructs of the GDI. Further, the large amount of variation accounted for by Factor 1 (cohesiveness) is consistent with the empirical (Perrone & Sedlacek, 2000) and theoretical (Yalom, 2005) literature on group dynamics. The significant correlation among the three factors is also consistent with the literature on this topic (Lese & MacNair Semands, 2000, Yalom, 2005). Given the consistency with which the GDI addresses the concurrent uniqueness and relatedness of group cohesion, universality and altru ism, it provides an ideal basis conceptualizing these three dynamics. Constructing Items for the CGDI In order to adapt the GDI (Phan et al. 2004) in the effort to create an instrument group cohesion, universality and altruism, it was necessary to re write the items from the GDI ( APPENDIX A ) to reflect the attitudes of counselors rather tha n the attitudes of group members. In this vein, each of the twenty items on the GDI was re written in this fashion ( APPENDIX B ). For example, an item that addresses the construct of cohesiveness on the GDI ( e.g., I felt a
53 sense of belongingness to the gro on the CGDI to reflect the perception of the group counsellor ( I perceive that the group members feel a sense of belongingness to the group and that the group accepts ). Demographic Question naire responses to the CGDI, during the study a demographic questionnaire was administered in t andem with the CGDI (APPENDIX C ). This questionnaire gather ed data about the parti and gender race/ethnicity and amount of relevant experience Studying the CGDI Once the f inal items were re written, a study was conducted to determine the factor structure of the CGDI. The CGDI was sent out i n electronic format to respondents via email. Participant were considered eligible for the study if they were currently leading a group ( e.g., counseling group or group supervision) at the time they recvied the invitation email for the study. This restrict ion was put in place due to the intended use of the CGDI as an instrument to be used in applied clinical settings, and as such the CGDI needed tp be investigated by using a sample of participants that are currently working as a group counsellor or leader. As specified by Gable and Wolf (1993), the sample size should result in a 10:1 ratio of items to participants ( e.g., for 20 CGDI items there should be at least 200 participants in the study). Subjects for the study were recruited via targeting CACREP accredited counseling programs, non profit community counseling agencies, and university counseling centers that conducted or supervised group counseling. Recruiting also was
54 done via the Associati on for Specialists in Group Work (ASGW) and CESNET counselor education listserve. The electronic package mailed to these participants included a invitation email including a brief description of the study ( APPENDIX D ) an informed consent ( APPENDIX E ), an screening question and demographics questionnaire (APPENDIX C ) and the electronic form of the CGDI ( APPENDIX B ). A follow up email was sent to maximize response rates. This email reminder was repeated at two weeks and four weeks after the initial emailing for those that have not yet responded. Data Analysis Once a sufficient number of participants have responded to the study, the data from responses to the CGDI will be subjected to confirmatory factor analysis (CFA), as well as a within demographic questionnaire will then be used to study the rela tionships between the chapter will detail this analysis.
55 CHAPTER 4 DATA ANALYSIS ite ms, as well as the demographic questionnaire that was completed in tandem with the CGDI. Descriptive statistics will be reported, each of the research questions asked in the study will be addressed, and the results will be interpreted. This chapter will co nclude with a brief summary of the results and statistical conclusions about the research questions. Descriptive Statistics This section will provide a discussion of details of how potential respondents were contacted, response patterns to the electronic version of the CGDI and demographics questionnaire (sent as one instrument to potential participants), and a description of those who actually responded to the survey. The survey was broken into three pages (not including the informed consent) a screeni ng question to determine eligibility for the study, the CDGI items, and a demographics questionnaire. Each of these sections was on a separate webpage. Potential respondents were contacted in two ways. The first method was through direct emails to CACREP a ccredited counselor education program (42 total), non profit community counseling centers (39 total), and university counseling centers that listed group counseling as services on their website (51 total). There were a total of 132 direct emails sent to po tential respondents The second method used to recruit potential respondents was through a counselor education listserve (CESNET) and the listserve for the Association for Specialists in Group Work. It is not know the number of people on these listserves. As such it is not possible to generate an accurate response rate.
56 Of the 286 respondents who agreed to the informed consent, 261 indicated that they were currently facilitating a qualifying group and 25 said that they were not facilitating such a group (s ee APPENDIX C qualification statement). Of the 261 who passed the screening question, 223 responded to the CGDI items. There were no missing responses on this section, as the survey was set up such that all items had to be responded to in order for the dat a to be recorded and to be permitted to continue the survey. Of the 223 who responded to the CGDI items, 219 individuals responded to the demographics questionnaire. As with the CGDI items, this page was set up such that all items had to be responded to in order for the data to be recorded and to be permitted to continue the survey. One exception to the number of responses to the demographics survey was the question that asked respondents how many years experience they had running groups (see APPENDIX C It em 2 for the actual question). Because of a problem with the Survey Monkey survey delivery platform, this question did not begin collecting data until several days after the survey was first open to respondents. As a result, only 142 respondents provided d ata for this question. Of those who responded to this survey (specifically to the demographics section), they identified as the following 157 as male, 61 as female, 176 as White, 14 as African American/Black, 8 as Asian American/Asian, 11 as Latino/His panic, and 9 as Other. The respondents that indicated Other identified as Bi Racial, Caucasian/Latina, Greek/Jewish, Jewish, Mixed (Japanese/Latino), multi ethnic, multi ethnic Japanese American, n/a, and Norwegian. Respondents also reported the racial/e thnic proportional makeup of the groups they used as a reference point for responding to the CGDI items. In other words,
57 respondents reported what the percentage of each of the following groups comprised their group Hispanic/Latino, African American/Blac k, Asian American/Asian, Native American or Alaska Native, Native Hawaiian or Other Pacific Islander, and White. The descriptive statistics for this question can be found in Table 4 1. The results from the three measures of group counselor level of experi ence (years of experience facilitating groups, number of group focused academic courses, and number of group focused continuing education hours) can be found in Table 4 2. Studying the CGDI The first two research questions are focused on the development of the CGDI and the establishment of validity and reliability thereof. Question 1 CGDI Validity Before discussing the construct validity of the CGDI as measured by a CFA, a note should be made about the content validity. During the initial study on the G DI (Phan et al. 2004), the content validity of the items was verified via a panel of group counseling experts. The items were rewritten for the CGDI, with the only change being the point of view of the question changing from the group member ( e.g., I felt a sense to the group counselor ( I perceive that the group members feel a sense of belongingness to the group and ). The theoretical basis of each items w as not altered, and as such the content validity from the GDI carries over to the CGDI. The main validity issue being addressed in this study was the construct validity fo the CGDI. The question is as follows: 1. Will the CDGI have the same three factor stru cture as the GDI, including the apparent validity based relationship between these three factors and the three factors of interest in this study (group cohesiveness, altruism and universality)?
58 Hypothesis: The CGDI will have the same three factor structure of the GDI (based on acceptable factor loadings and goodness of fit measures found in a confirmatory factor analysis). The CDGI item s (see APPENDIX B ) consisted of twenty likert scale items intended to measure the three dynamics of interest in this research study altruism (items 1 5), group cohesiveness (items 6 15), and universality (items 16 20) A confirmatory factor analysis (referred to as CFA he reafter) was used to study the construct validity of the CGDI because there is pre conceived theoretical assumption that the CGDI measures these three constructs. The initial study on the GDI (Phan et al. 2004) suggests that these three dynamics were meas ured by the GDI. Given that the CGDI items were directly adapted from the GDI, it is assumed that the CGDI measures the same three constructs. The CFA will study the factor structure of the CGDI using this theoretical assumption this assumption. Using the LISREL software application ( J oreskog & S o rbom, 1993) to conduct a CFA, two models were studied a one factor model and a three factor model. These two models were compared to determine of the assumed three factor model (i.e. altruism, group cohesiveness and universality) is a better model than a model that assumes there is only one common factor. This comparison approach was used on the basis of (2004) assertion that goodness of fit studies using so as effective as comparing competing models when goodness of difference in misspecification between two 337 340). Ultimately, this approach suggests that meaningful results can be reached by
59 (p.340). Although this is the approach used to s tudy goodness of fit for all if these models, guidelines for acceptable cut off values will be used a secondary method of assessing goodness of fit for these two models. (1999) suggest the following cut off values a Root Mean Square Error of Approximation (RMSEA) < 0. 06 a Non Normed Fit Index (NNFI) > 0. 95 a Comparative Fit Index (CFI) > 0. 95 and a Standardized Root Mean Square Residual (SRMR) < .0 8. This secondary approach was used because there are several com peting approaches to assessing goodness of fit in the literature ( Marsh, Hau & Wen, 2004), and as such it is necessary to consider multiple methods of assessing the construct validity of the CGDI. The first model studied was a one factor model, which held that all of the items on the CDGI were measuring one common factor. This model was chosen because of the literature that suggests group cohesiveness is a factor that may be a ubiquitous dynamic in counseling groups (Martin, Garske & Davis, 2000). This mod el had a RMSEA of 0.12, a NNFI of 0.80, a CFI of 0.82, and a SRMR of .097. All if these fell well below guidelines for acceptable cut off values. The next step was to compare this model to the three factor model in order to use Marsh, Hau, and (2004) approach of comparing competing models in order to assess goodness of fit. The three factor model was based on the theoretical assumption that this CGDI measured altruism, group cohesiveness, and universality. This three facto r model had a RMSEA of 0. 078 a NNFI of 0.8 8 a CFI of 0. 90 and a SRMR of .0 8 The RMSEA, NNFI and CFI all fell below cut off values, while the SRMR was
60 just within their guidelines. Using (2004) approach of comparing competing models, this model was better fit than the one factor model. Because this model seems to be a significantly better fit, a deeper analysis of the factor loadings of this model was conducted. The model diagram illustrating the factor lo adings of the three factor model generated by the LISREL program can be seen in Figure 4 1. These loadings indicate off value for loadings > .40. The loadings for items 2 (.25) and 3 (.18) on the al truism construct indicate that these items may not be measuring this construct. The same is also true for item 14 (.33) in relation to the group cohesiveness construct, and for item 18 (.37) in regard to the universality construct. LISREL also provides suggestions for reorganization of the model to increase the goodness of fit. These suggestions are made in terms of adding paths from the three latent constructs to other items in the model, and in terms of allowing for covariation between some of the item s in the model. Those suggestions are illustrated in Table 4 3. Although each of these changes would make a positive change to the loadings and reduce error variance between items, none of these changes fit with the theoretical assumptions the CFA is based on in the first place. As such, a change was not made given the problematic issues around using a CFA in an exploratory fashion ( Klein, 2005). Question 2 CGDI Reliability The second question addressed the reliability of the CGDI. This research questio n is as follows: 2. Will the CDGI meet an acceptable level of reliability?
61 Hypothesis: The CGDI will meet an acceptable level of reliability (at least a 0.7 on a a study (Cronbach, 1951) for each of the three proposed factors that the CGDI was designed to measure. This approach was used because a Cronbachs alpha study operates on the assumption that there is only one construct that is being measured by the items in question (Cronbach, 1951). To meet acceptable levels of reliability for an exploratory coefficient ( Nunnally, 1978 ). The first construct that was measured by the CGDI was al truism. Items 1 5 on the CGDI (see APPENDIX B ) purportedly measured this 0.7 acceptable level. The second construct measured by items 6 15 on the CGDI (see APPENDIX B ) w 0.78, meeting the acceptable level of reliability for this study. The third construct measured by items 16 items was 0.72, also meeting the acceptable level of reliability for this study. Given the low alpha for the altruism construct (0.64), a further analysis was done based on the factor loadings found in the CFA from research question 1. When removing the two items that did not meet off value (items 2 and 3), the new off for acceptable reliability. Demographics and the CGDI The last four questions will deal specifically with the investigating the relationship
62 Question 3 Counselor Race/Ethnicity The first question about demographics addressed the relationship between counselor race/ethnicity and responses to the CGDI. The quest ion is as follows: 3. and their responses on the CGDI Hypothesis: There will not be a significant relationship between group e CGDI (where significance is set at p < 0.01) For the race/ethnicity category question, respondents were given the options of Hispanic/Latino, African American/Black, Asian American/Asian, Native American or Alaska Native, Native Hawaian or Other Pacific Islander, White, and Other (with an option for giving an open ended response for the Other category).This question was answered based on a multiple regression analysis of the relationship between counselor reported race/ethnicity, and each of the three con structs measured by the CGDI. There was not a significant relationship found between the altruism construct and counselor race/ethnicity, where the whole model ANOVA F=2.03, and p=.11. There was also no significant relationship found between group cohesion and counselor race/ethnicity, where the whole model ANOVA F=1.42, and p=.24. There was a significant relationship found between universality and counselor race/ethnicity, where the whole model ANOVA F=3.50, and p=.01. Further investigation of this shows t hat the significance seems to be between counselors that self identified as African American and the universality construct measured by the CGDI, where t=3.31 and p=.001. For every unit that universality increases for whites (the constant in the ANOVA mode l was White counselors because of their majority status), this study shows that universality increases 2.11 units for African Americans. Even when the item for this construct that
63 did not meet acceptable factor loadings (item 18) was excluded from the anal ysis, the ANOVA remained significant at F=3.72 and p=.01 with the African American/Black relationship remaining significant where t=2.78 and p=.01 Question 4 Group Member Race/Ethnicity The second question about demographics addressed the relationship between group member race/ethnicity and responses to the CGDI. The question is as follows: 4. Will there be a significant relationship between the level of the 3 dynamics measured by the CGDI and the race/ethnicity of the group members? Hypothesis: There will not be a significant relationship between the level of the 3 dynamics measured by the CGDI and the race/ethnicity of the group members (where significance is set at p < 0.01). For the group member race/ethnicity question, respondents were asked to estimat e the percentage of the following categories that made up the group they used as a reference point for responding to the CGDI items Hispanic/Latino, African American/Black, Asian American/Asian, Native American or Alaska Native, Native Hawaian or Other P acific Islander, and White. This question formatted such that all of the percentages had to total exactly 100. This research question was answered based on a multiple regression analysis of the relationship between group member reported race/ethnicity, and each of the three constructs measured by the CGDI. There was not a significant relationship found between the altruism construct and group member race/ethnicity, where the whole model ANOVA F=.24, and p=.87. There was also not a significant relationship f ound between the group cohesion construct and group member race/ethnicity, where the whole model ANOVA F=.26, and p=.86. Finally, there no significant relationship found between the universality construct and group member race/ethnicity, where the whole mo del ANOVA F=2.16, and p=.10.
64 Question 5 Counselor Level of Experience The third question about demographics addressed the relationship between 5. Will there be a signifi cant relationship between group counselor level of experience and the 3 dynamics being measured by the CGDI? Hypothesis: There will not be a significant relationship between the level of the 3 dynamics measured by the CGDI and the group counselor level of experience (where significance is set at p < 0.01). There were three questions that assessed group counselor level of experience ( see APPENDIX C ). The three categories were the number of years experience in facilitating groups, the number of group focused program of study, and number of group focused continuing education hours taken by each respondent. Based on the analysis of the correlation among these two groups of variables (the three measures of experience and the three constructs measured by the CGDI), there were no significant correlations found between any of the three measures of group counse lor experience, or the three constructs measured by the CGDI (see Table 4 4). Question 6 Group Counselor Gender The last demographics question was in relation to self reported gender of the group counselors. The question was: 6. Will there be a significant relationship between group counselor gender and the 3 dynamics being measured by the CGDI? Hypothesis: There will be a significan t relationship between the level of (where significance is set at p < 0.01). This was analyzed by doing a t test to determine if there was significant difference between each gender category (female and male) on each of the three constructs
65 measured by the CGDI. There was a significant relationship found between gender and the altr uism construct where t=2.26 and p=.01. There was no relationship found between gender and group cohesiveness (where t=3.37 and p=.76) or universality (where t=.98 and p=.021). A further analysis of the altruism construct showed that when removing the two i off value (items 2 and 3), there was no significant relationship found between gender and the altruism construct where t=2.42 and p=.09. Summary There were mixed results found in this statistical analys is. The altruism construct did not meet acceptable levels of reliability. Although the three factor model that the CGDI was built on had better goodness of fit values than the one factor model ( Marsh, Hau & Wen 2004), its goodness of fit values fell below some accepted standards for this type of analysis ( Hu & Bentler, 1999 ). There were also some issues with low loading values for some items on the CGDI and reliability concerns for the altruism construct. Additionally, there were some significant relations hips found between some demographic factors and the CGDI constructs. Overall, the three factor model is good fit, but has some flaws that may warrant the need to adjust the CGDI for future use. The next chapter will highlight a detailed discussion of the r esults found in this chapter, and implication of these results will be also be discussed.
66 Table 4 1. Group member race/ethnicity (as a percentage of group) N Minimum Maximum Mean Std. Deviation Hispanic/Latino 144 0 100 13.83 16.198 African American/Black 153 0 100 24.18 23.350 Asian American/Asian 106 0 50 7.23 9.517 White 213 0 100 70.81 24.687 Table 4 2. Counselor l evel of e xperience N Minimum Maximum Mean Std. Deviation Years Experience 142 0 42 9.87 9.397 Classes 218 0 20 1.94 1.896 Continuing Ed Hours 218 0 1000 34.52 109.209 Table 4 3 LISREL suggested changes to three factor model Suggestion to add path: from to New Estimate Item 6 Altruism 0.54 Item 12 Universality 0.38 Item 13 Universality 0.47 Item 15 Universality 0.52 Suggestions to allow for covariance: between and New Estimate Item 3 Item 2 .11 Item 10 Item 9 .05 Item 13 Item 10 .05 Item 14 Item 10 .07 Item 14 Item 13 .09 Item 15 Item 10 .05 Item 15 Item 12 .07 Item 15 Item 13 .09 Item 16 Item 14 09 Item 18 Item 17 09 Item 19 Item 16 .10 Item 20 Item 17 .10
67 Table 4 4 Relationship b etween e xperience and CGDI c onstructs Years Experience Classes Continuing Ed Altruism Pearson Correlation .033 .033 .098 Sig. (2 t ailed) .698 .628 .148 N 142 218 218 Cohesion Pearson Correlation .046 .025 .039 Sig. (2 tailed) .585 .718 .563 N 142 218 218 Universality Pearson Correlation .193 .013 .087 Sig. (2 tailed) .021 .844 .203 N 142 218 218 Correlation is significant at the 0.01 level (2 tailed).
68 Figure 4 1. Factor Loadings for the three factor CGDI model
69 CHAPTER 5 DISCUSSION This chapter will provide a discussion of the results found in this dissertation study. It will begin with a summary and an interpretation of these findings, a limitations section will be included to temper any conclusions made from this research and a discussion about the future of this assessment method will conclude th is chapter Summary of Findings The research questions asked in this study were broken down into two sections validity and reliability study of the CGDI, and a study of the relationship between the three constructs on the CGDI and the demographic info rmation that was collected. Each of these sections will be summarized separately. Validity and R eliability of the CGDI There were mixed results in the construct validity study of the CGDI. The original research question and hypothesis about validity were: 1. Will the CDGI have the same three factor structure as the GDI, including the apparent validity based relationship between these three factors and the three factors of interest in this study (group cohesiveness, altruism and universality)? Hypothesis: The C GDI will have the same three factor structure of the GDI (based on acceptable factor loadings and goodness of fit measures found in a confirmatory factor analysis). Conducting this type of analysis required comparing this three factor model to another mode l that would be considered a plausible alternative (primary consideration), as well On the primary consideration, the three factor model was a much better fit than the model used as a comparison (one factor model). On the secondary consideration, the three factor model did not meet the cut off values. Because the CGDI three factor model
70 passed the primary test, a further analysis of the model was done to attempt to unde rstand the mixed results. When looking at the ways each of the twenty items fit into the model, it was found that the majority of them (sixteen) were a good measure of the constructs they were designed to measure. The re maining four seemed to be poor measu res of their respective constructs. This deeper analysis, while not providing conclusive proof, provided more support that the three factor model of the CGDI was a valid measure of altruism, cohesiveness, and universality. This conclusion is tempered with the understanding that further study on this instrument should be done taking into account these findings, mainly due to the imperfect results on both the factor structure and factor loadings. The reliability study of the CGDI also had some mixed results. The original research question and hypothesis about validity were: 2. Will the CDGI meet an acceptable level of reliability? Hypothesis: The CGDI will meet an acceptable level of reliability (at least Using this method of analysis, the items that were designed to measure the group cohesion and universality constructs were found to be reliable measures of these two constructs. Alternatively, the items that were designed to measure the altruism construct were not found to be reliable indicators of this construct. A further analysis was done to attempt to understand these results in light of the validity study described earlier. After removing the two items that did not seem to be good predictors of the altruism construct (ite ms 2 and 3), the reliability of the remaining items intended to measure altruism (items 1, 4 and 5) were found to be a reliable measure of the altruism construct.
71 Demographic A nalysis The next three questions addressed by this study were focused on analy zing the relationship between the three constructs being measured by the CGDI and the demographic information collected. The first question about demographics addressed the relationship between counselor race/ethnicity and responses to the CGDI. The questi on is as follows: 3. and their responses on the CGDI Hypothesis: There will not be a significant relationship between group the CGDI (where significance is set at p < 0.01) A significant relationship was not found between the altruism or group cohesion found between the universality constru further analysis, it seems that the relationship is between group counselors who reported African American/Black as their identity and their responses to the items measuring the universality construct. This rel ationship remained significant even when an item (item 18) that did not predict the universality construct well, as found in the CFA, was excluded from the analysis. The second question about demographics addressed the relationship between group member r ace/ethnicity and responses to the CGDI. The question is as follows: 4. Will there be a significant relationship between the level of the 3 dynamics measured by the CGDI and the race/ethnicity of the group members? Hypothesis: There will not be a significant relationship between the level of the 3 dynamics measured by the CGDI and the race/ethnicity of the group members (where significance is set at p < 0.01).
72 The analysis showed that there was not a significant relationship between any category of group membe r race/ethnicity and the three constructs measured by the CGDI. experience. The research question was: 5. Will there be a significant relationship between group counselor level of experience and the 3 dynamics being measured by the CGDI? Hypothesis: There will not be a significant relationship between the level of the 3 dynamics measured by the CGDI and the group counselor level of experience (where significance is set at p < 0.01). As with the last question, there was no significant relationship found between the three measures of group counselor experience (years experience as a group counselor, number of group focused academic courses, and number of group focused continui ng education hours) and the three constructs measured by the CGDI. The last demographics question was in relation to self reported gender of the group counselors. The question was: 6. Will there be a significant relationship between group counselor gender an d the 3 dynamics being measured by the CGDI? Hypothesis: There will be a significant relationship between the level of (where significance is set at p < 0.01). There was a significant relationship found between gender and the altruism construct, whereas there was no significant relationship found between gender and the group cohesiveness or the universality construct. Again, when the two items that did not seem t o be good predictors of the altruism construct (items 2 and 3) were removed from the analysis, there was no significant relationship found between the altruism construct and gender.
73 Interpretation of Results The purpose of this dissertation was to conduct an initial validity and reliability study of the CGDI, as well as to better understand the relationship between key demographic variables and the constructs measured by this instrument. The results of this study will be interpreted in terms of how these f indings inform a general understanding of validity and reliability of the CGDI. The fist topic that will be covered is validity. The CGDI was found to have a moderate level of construct validity. This is due in part to the inadequate goodness of fit value s for the resulting CFA data, and in part to the lack of agreement on the best way to interpret CFA models in general. The pre determined three factor model (altruism, group cohesion, and universality) for the CGDI was a significantly better fit that the o ne factor model (group cohesion only) used for comparison. These suggest that although some authors have posited that group cohesion is the dominant dynamic present in counseling (Martin, Garske & Davis, 2000), the CGDI was in fact likely measuring two con structs other than group cohesion (presumably altruism and universality). Although this three factor model was a better fit than the competing model, it did not meet cut off values proposed by some leading authors in the CFA field ( Hu & Bentler, 1999) Mar (2004) disagree with between theoretical models as opposed to absolute values of significance or goodness of fit. Further, 16 of the items that were desig ned to measure each of three constructs in the GDI were good measures, but 4 were not. In particular, two of the items that measured the altruism construct seemed to be problematic in terms of certain demographic factors and reliability for measuring altr uism (this will be covered in more
74 detail in the following sections). The CGDI factor structure and item loadings seem to indicate that this instrument is doing a moderately good job of measuring the constructs it was designed to measure, with the notable exception of the two items (items 2 & 3) that caused consistent problems throughout the data analysis. This suggests that more work should be done to improve the construct validity of the CGDI. Specifically, a content and theoretical analysis should be don e on the items that did not have acceptable loading values for the constructs they were designed to measure (i.e. items 2, 3, 14, and 18). This is particularly true for those items that, when removed, improved the reliability of the altruism construct and nullified the significance between gender and the altruism construct (i.e. items 2 and 3). This will be discussed in more detail later in the sections on reliability and demographic factors. In total, the CGDI seems to be measuring the three construct it w as designed to measure, but will require more refinement to be confident in this claim. When looking at validity in terms of generalizability across demographic categories addressed in his study, the CGDI also showed of moderate level of validity There wa s no significant relationship found between the items on the CGDI and group member race/ethnicity or group counselor level of experience. There was significant relationship found between the universality construct and group counselors who self identified a s African American. This may suggest that the items designed to measure this construct cannot be determined with this data, particularly given that only 14 respondents i dentified as African American. There was also a significant relationship found between counselor gender identification and the altruism construct. This relationship disappeared
75 standard method of determining psychometrics of an instrument such as this one. Once again, the CGDI seems to have a moderate level of validity in terms of generalizability, but there are some key issues that need to be resolved. Further items analysis and study will be need in this area for the CGDI. In terms of reliability, the CGDI once again showed mixed results. Although the items that were designed to measure the universality and group cohesiveness constructs were found to be reliable measures of the se constructs, the items that were designed to measure the altruism construct were not found to be reliable measures. This reliability problem for the altruism items may be related to items 2 & 3, which were also found to not be good measures of this const ruct during the CFA discussed earlier in this section. Although it does seem these two items may be the problem, more analysis is needed to understand the nature of this low reliability. I total, the CGDI showed promise in terms of construct validity, ge neralizability, and reliability. This is tempered by the fact that there were also some problems found in all of these areas. The next section will discuss how to address these issues, as well as further steps needed to fully develop this instrument. Next Steps in Developing the CGDI/GDI Method The first steps needed to further develop the CGDI are related to addressing the psychometric problems revealed in this study. The first step is to do a theoretical content analysis and do new item development. This needed because of the items that loaded poorly during the CFA, particularly the items 2 & 3 that caused consistent problems through the data analysis. If a theoretical decision cannot be made about why these items are not functioning as intended, new item s will need to be developed and
76 added to the instrument. The next step would be to do new construct and generalizability study that included a more racially and ethnically diverse response pool. This is needed to both determine if the issues with measuring counselor race/ethnicity have been addressed, as well as be sure there are not problems with other race/ethnicity categories when there is a more representative sample. Once these issues have been addressed, the CGDI can undergo further psychometric devel opment. One of the key developmental steps for the CGDI will be studying it for external validity. For example, a widely used instrument called the Group Climate Questionnaire Short Form (GCQ S : MacKenzie, 1983) includes a subscale that measures the con struct engagement group cohesion. By giving this instrument in tandem with this CGDI/GDI assessment method, it can be determined if the group cohesion construct as measured by the CGDI/GDI is convergent with the GCQ S subscale of engagement. This type of study is needed to be sure this assessment method is couched within current understandings of the constructs it was designed to measure. The CGDI/GDI method also needs to be studying in ter ms of its validity as a instruments need to be given together simultaneously in a group environment to determine if they have validity to be used in this manner. For exa mple, even if both instruments are shown to be measuring universality, altruism and group cohesiveness from both the group member and group counselor perspectives, when given in tandem this construct validity may not hold up. As such, this type of further study is needed.
77 Limitations of this Study There are several limitations for this study. The first is the self report nature of the data collected. In general this also a limitation for data, but there is particular limitation for this study in relation to the data collected about group member race /ethnicity. This study relied on group counselors to accurately determine and estimate the race ethnicity of their group members. There were also a very low number of non white counselors that responded to this study. While this may reflect the actual perc entage of practicing counselors, it makes it difficult to determine statistical significance related to these populations. If this instrument is to be considered culturally sensitive and appropriate, this limitation should be addressed in future studies of the CGDI. A glaring limitation of this study is the malfunction of the Survey Monkey software were 77 responses missed. Ultimately, this means that no statistical conclu sions can be drawn about this demographic category. This is a significant problem given the importa nce of this demographic factor. This also must be addressed in any future study of the CGDI. Future Directions for the CGDI and the Mirror Instrument Method conceptualization thereof has implications for the field of group counseling. In addition, this method also has implications for the future of outcome research in counseling as a whole. ACA (2005), the central professional organization for counselors, has outlined professional standards that place the client at the center of concern for the field of counseling. The mirror instrument method provides a way to place the client as the
78 centra l focus of process and outcome assessment in counseling. As stated at the beginning of this dissertation, this study was conducted because group counseling (and counseling in general) is conducted by using the self of the counselor as the central facilitat or of therapeutic action. As such the counseling field is caught between two competing goals. One is to provide individualized counseling to each and every client based on their own personal background and circumstance. The other is to answer the call for evidence based mental health practice (Committee on Quality Health Care in America, 2001) and outcome assessment of counselor training methods (CACREP, 2009). Although it is not possible to meet both of these goals in total, the mirror instrument method of concurrently assessing client and counselor conceptualizations of the counseling process and outcomes provides a way to meet both goals in a significant way. The individual perspective of each client and/or group member can be taken into account, as well points can then be compared with outcome assessments to better understand the inner workings of effective counseling. As with any assessment method, trade offs are made and not all issues can be covered. Even so, this method provides a bridge between these two goals of the contemporary counseling field. If the mirror instrument method can, if fact, deliver on its design, other tandem instruments can be developed to suit other forms and foc i of counseling. Development of the CGDI is the first step toward investigating if this mirror instrument approach can deliver for the counseling profession.
79 APPENDIX A ITEMS FROM THE GDI ( PHAN ET AL. 2004) 1. I felt that helping others has given me more self respect. 2. 3. I was forgetting myself and thinking of helping others 4. I was giving parts of myself to others 5. I felt that I was helping others and having an important impact in their lives 6. I felt a sense of belongingness to the group and that the group accepted me 7. I felt like keeping in touch with other people 8. I felt that after revealing embarrassing things about myself, I was still accepted by the group 9. I have the feeling that I am no lon ger alone 10. I feel that I belong to a group of people who understand and accept me 11. same boat) 12. I am seeing that I was just as well off as other people 13. I learned that oth 14. I learned that others had parents and backgrounds as unhappy or mixed up as I 15. I learned that I am not very different from other people and that the group gave eling. 16. I recognized that life is at times unfair and unjust 17. 18. I recognized that no matter how close I get to other people, they still must face life alone 19. I learned that by facing the basic issues of their life and death, I am more able to live my life more honestly and be less caught up in trivialities
80 20. I learned that I must take ultimate responsibility for the way I live my life no matter how much support and guidance I get from others
81 APPENDIX B CGDI ITEMS AND DIRECTIONS For the following items, please think about ONE group that you are currently facilitating, leading, and/or running. It can be a counseling group or a counseling supervision group. Use this group as a reference point for answering all of the following items. Please rate all of these items as they relat e to the group described above. 1. I perceive that the group members feel that helping others has given them more self respect. 2. needs 3. I perceive that the group members are forgetting themselves and thinking of others in the group 4. I perceive that the group members are givin g parts of themselves to each other 5. I perceive that the group members feel they are helping each other and having 6. I perceive that the group members feel a sense of belongingness to the group and that the group accepts them 7. I perceive the group members feel like keeping in touch with each other 8. I perceive the group members feel that after revealing embarrassing things about themselves, they are still accepted by the group 9. I perceive that the group members feel they are no longer alone 10. I perceive that the group members feel they belong to a group of people who understand and accept them 11. I perceive that the group members learned that they are not the only ones with e same boat) 12. I perceive that group members see that they are just as well off as other people 13. thoughts that they do
82 14. I perceive that group members learned that others had paren ts and backgrounds as unhappy or mixed up as their own 15. I perceive that group members learned that they are not very different from other 16. I perceive that group members recognized that life is at times unfair and unjust 17. I perceive that group members recognized that ultimately there is no escape from 18. I perceive that group members recognized that no matter how close I get to other people, they still mus t face life alone 19. I perceive that group members learned that by facing the basic issues of their life and death, they are able to live their lives more honestly and be less caught up in trivialities 20. I perceive that group members learned that they must take ultimate responsibility for the way they live their lives no matter how much support and guidance they get from others
83 APPENDIX C ADDITIONAL QUESTIONS Screening Question Are you currently facilitating and/or running a counseling group, psycho educational group or counselor supervision group? A counseling or psycho educational group is defined as a group that is being run by a counselor in effort to help clients in any way. A counselor supervision group is defined as a supervisor who is e ither running group supervision, or teaching a course that involves group supervision within the course. Demographic Questionnaire 1. What is your age in whole years? 2. How many years experience do you have running/facilitating counseling groups, psycho educat ional groups and/or counselor supervision groups? 3. What gender do you identify as? a. Male b. Female c. Other (please specify) 4. What race/ethnicity do you identify as? a. Latino/Hispanic b. African American/Black c. Asian American/Asian d. Native American or Alaska Native e. Native Hawaiian or Other Pacific Islander f. White g. Other (please specify)
84 5. Please estimate the racial/ethnic percentages of the members of the group you used as reference point for responding to this instrument. Please use percentages (total should equal 100) a. Latino/Hispanic b. African American/Black c. Asian American/Asian d. Native American or Alaska Native e. Native Hawaiian or Other Pacific Islander f. White 6. How many classes did you take in your academic counselor training program that was specifically focused on group co unseling? 7. How many continuing education hours have you completed that were specifically focused on group counseling? Please estimate if you do not know the exact amount
85 APPENDIX D EMAIL INVITATION Dear Fellow Group Workers, I am writing this email to in vite your participation in a study being conducted in dynamics. The resulting data from this study will be used to determine the reliability and factor structure of the Conceptualization of Group Dynamics Inventory (CGDI). The criterion for participation in this study is that each person is that each participant should be currently leading or facilitating a working group (i.e. a counseling group, therapy group, psychoeduc ational group and/or a clinical supervision group). If you meet this criterion and are interested in contributing to the understanding and study of group take this asses sment. Please contact me via this email address ( email@example.com ) or by phone at 352 871 3208 if you have any questions or concerns about this study. Thank you for your time in considering participating in this study, a nd look forward to hearing from you. Have a great day! Kevin Tate, M.Ed. Ed.S. Doctoral Candidate University of Florida, College of Education Counselor Education Program
86 APPENDIX E INFORMED CONSENT Protocol Title : Development of Conceptualization of Group Dynamics Inventory (CGDI). Please read this consent document carefully before you decide to participate in this study. Purpose of the research study : The purpose of this study is to develop an instrument lization ability, as well as to understand the relationship between this ability and demographic attributes of the counselor and group members. What you will be asked to do in the study : Your participation consists of filling out the Conceptualization of Group Dynamics Inventory and a short demographics questionnaire. Time required : Approximately 15 minutes Risks and Benefits : The only potential risk is the reporting of demographic information. This is almost completely eliminated by the use of an online s urvey program that allows for truly anonymous responses to these questions. Compensation : There is no compensation for this study. Confidentiality : Upon participation in this study, demographic information will be collected and compared to your responses on the assessments. All of this demographic information and the information shared in this survey will remain confidential and be reported in an an any of the data during reporting nor will individual responses be divulged. The results of the assessments will be recorded and stored on a secure hard drive 3 years. Voluntary particip ation : Participation is completely voluntary and there will be no
87 negative consequences for refusal to participate. Right to withdraw from the study : You have the right to withdraw from the study at anytime without consequence. Whom to contact if you have questions about the study : Kevin Tate, Doctoral Candidate, University of Florida Counselor Education Program, 1206 Norman Hall, POB 117046, Gainesville, FL 32611 7046, 352 871 3208. Whom to contact about your rights as a research participant in the study : UFIRB Office, Box 112250, University of Florida, Gainesville, FL 32611 2250, 352 392 0433. Agreement voluntarily agree to participate in the study.
88 LIST OF REFERENCES American Cou nseling Association (ACA) (2005). ACA c ode of ethics Alexandria,VA: Author. Association for Specialists in Group Work (ASGW) (2000) Association for specialists in group work professional standards for the training of group workers The Journal for Spec ialists in Group Work, 25 (4), 327 342. Association for Specialists in Group Work (ASGW) (1999) Principles for diversity competent group workers Journal for Specialists in Group Work, 24, 7 14. Bemak, F., & Chung, R C. Y (2004) Teaching Multicultura l Group Counseling: Perspectives for A New Era The Journal for Specialists in Group Work, 29 (1), 31 41. Browne, F R (2005) Self Talk of Group Counselors: The Research of Rex Stockton The Journal for Specialists in Group Work, 30 (3), 289 297. Burli ngame, G. M., Fuhriman, A. J., Johnson, J. (2004). Process and outcome in group counseling and psychotherapy: A perspective. In J. L. DeLucia Waack, D. A. Gerrity, C. R. Kalodner & M. T. Riva (Eds.), Handbook of Group Counseling and Psychotherapy (pp. 49 6 1). Thousand Oaks, CA: Sage Publications. Christensen, T M., & Kline, W B (2000) A qualitative investigation of the process of group supervision with group counselors The Journal for Specialists in Group Work, 25 (4), 376 393. Cole, M B (2005) Group Dynamics in Occupational Therapy Danvers, MA: SLACK Incorporated Committee on Quality of Health Care in America. (2001). Crossing the quality chasm: A new health system for the 21st century Washington, DC: Institute of Medicine, National Academie s Press. Council for Accreditation of Counseling and Related Educational Programs (CACREP) (2009). Alexandria, VA: Author. DeLucia Waack, J L (2002) A written guide for planning and processing group sessions in anticipation of supervision The Journal for Specialists in Group Work, 27 (4), 341 357. Falvey, J. E. (2001). Clinical judgment in case conceptualization and treatment planning across mental health disciplines. Journal of Counseling and Development 79, 292 303.
89 Falvey, J. E., Bray, T. E., Hebe rt, D. J. (2005). Case conceptualization and treatment planning: Investigation of problem solving and clinical judgment. Journal of Mental Health Counseling 27, 348 372. Forsyth, D R (1999) Group dynamics (3 rd ed.) Belmont, CA: Wadsworth. Frank Sarac ini, J., Wilbur, M., Torres Rivera, E., & Roberts Wilbur, J. (1998, September). GCT: Structure and chaos of group supervision Paper presented at the annual convention of the North Atlantic Association for Counselor Education and Supervision, Wells, ME. Gab le, R. K, & Wolf, M. B. (1993). Instrument development in the affective domain: Measuring attitudes and values in corporate and school settings (2 nd ed.). Norwell, MA: Kluwer Academic. Granello, D H., & Underfer Babalis, J (2004) Supervision of group work: A model to increase supervisee cognitive complexity The Journal for Specialists in Group Work, 29 159 173. Hines, P L., Stockton, R., & Morran, D K (1995) Self talk of group therapists Journal of Counseling Psychology, 42 (2), 242 248. Hinsz, V B (1995) Mental Models of Groups as Social Systems: Considerations of Specification and Assessment Small Group Research, 26 (2), 200 233. Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Convention al criteria versus new alternatives Structural Equation Modeling, 6 1 55. Hurley, J. R., & Brooks, L. J. (1987). International Journal of Group Psychotherapy, 37 441 448. J oreskog K., S o rbom, D., (1993). LISREL 8: Structural Equation Modeling with the SIMPLIS Command Language Scientific Software International, Inc: Lincolnwood, IL. Kivlighan, D M., Lilly, R L (1997) Developmental changes in group climate as they relate to therapeutic gain Group Dynamics: Theory, Research, and Practice 1, 208 221. Kivlighan, D M., Markin, R D., Stahl, J V., & Salahuddin, N M (2007) Changes in the ways that group trainees structure their knowledge of group members with training Group Dynamics: Theory, Research, and Practice, 11 (3), 176 186. Kivlighan, D M., Mullison, D (1988) group counseling: The role of interpersonal style and stage of group development Small Group Research 19, 452 468.
90 Kivlighan, D. M., Multon, K. D., & Brossart, D. F. (1996). Helpful impacts in group counseling: Development of a multidimensional rating system. Group Dynamics: Theory, Research, and Practice, 42, 347 355. Kivlighan, D M., & Quigley, S T (1991) Dimensions used by experienced and novice group therapists to conceptualize group process Journal of Counseling Psychology, 38 (4), 415 423. Klein, R. B. (2005). Principles and Practice of Structural Equation Modeling (2 nd ed.). The Guilford Press: New Y ork. Lambert, M J (1992) The Handbook of Psychology Integration New York, NY: Basic Books Lese, K P., MacNair Semands, R R (2000) The therapeutic factors inventory: Development of a scale Group 24, 303 317. MacKenzie, K R (1983) The clinical application of a group climate measure In R R Dies & K R MacKenzie (Eds.), Advances in group psychotherapy: Integrating research and practice (pp 159 170) Madison, CT: International Universities Press. MacNair Semands, R R., Lese, K P (2 000) Interpersonal problems and the perception of therapeutic factors in group therapy Small Group Research 31, 158 174. Marcus, D. K. (1998). Studying group dynamics with social relations model. Group Dynamics: Theory, Research, and Practice, 2 230 24 0. Marmarosh, C., Holta, A., Schottenbauer, M (2005) Group cohesiveness, grup derived collective self esteem, group derived hope, and the well being of group therapy members Group Dynamics: Theory, Research, and Practice 9, 32 44 Marsh, H. W., Hau, K ., Wen, Z. (2004). In search of golden rules: Comment on hypothesis testing approaches to setting cutoff values for fit indexes and dangers Structural Equation Modeling 11, 320 341. Martin, D J., Garsk e, J P., Davis, M K (2000) Relation of the therapeutic alliance with outcome and other variables: A meta analytic review Journal of Consulting and Clinical Psychology 68, 438 450. McPherson, F M., & Walton, H J (1970) The Dimensions of Psychotherapy Group Interaction: An Analysis of Clinicians' Constructs British Journal of Medical Psychology, 43 281 290. Mullen, B., Driskell, J. E., & Salas, E. (1998). Meta analysis and the study of group dynamics. Group Dynamics: Theory, Research, an d Practice, 2, 213 229.
91 Neufeldt, S. A., Pinterits, E. J., Moleiro, C. M., Lee, T. E., Yang, P. H., Brodie, R. E., Orliss, M. J. (2006). How do graduate student therapists incorporate diversity factors in case conceptualization?. Psychotherapy: Theory, Res earch, Training, Practice 43, 2006. doi: 10.1037/0033 3184.108.40.2064 Nitza, A G (2005) The contributions of Rex Stockton to group development and outcome research and practice TheJjournal for Specialists in Group Work 30, 271 281. Okech, J E A., & K line, W B (2006) Competency Concerns in Group Co Leader Relationships The Journal for Specialists in Group Work, 31 (2), 165 180. Okech, J E A., & Rubel, D (2007) Diversity Competent Group Work Supervision: An Application of the Supervision of Gro up Work Model (SGW) The Journal for Specialists in Group Work, 32 (3), 245 266. Orr, J J., & Hulse Killacky, D (2006) Using voice, meaning, mutual construction of knowledge, and transfer of learning to apply an ecological perspective to proup work train ing The Journal for Specialists in Group Work, 31 (3), 189 200. Pan, P J D., Lin, C W (2004) experiences, and therapeutic factors in group counseling Small Group Research 35, 174 194. Perrone, K. M., & Sedlacek, W. E. (2000). A comparison of group cohesiveness and client satisfaction in homogeneous and heterogeneous groups. The Journal for Specialists in Group Work, 25 243 251. Phan, L. T. (2001). Structured group supervision from a multicultural persp ective Unpublished doctoral dissertation, University of Nevada, Reno. Phan, L. T., & Torres Rivera, E. (2004). The impact of language on dynamics in counseling and psychotherapy groups. In J. L. DeLucia Waack, D. A. Gerrity, C. R. Kalodner & M. T. Riva (E ds.), Handbook of Group Counseling and Psychotherapy (pp. 283 294). Thousand Oaks, CA: Sage Publications. Phan, L T., Torres Rivera, E., Volker, M A., Garrett, M T (2004) Measuring group dynamics: An exploratory trial Canadian Journal of Counseling 38, 234 245. Rubel, D J., & Kline, W B (2008) An Exploratory Study of Expert Group Leadership The Journal for Specialists in Group Work, 33 (2), 138 160. Rubel, D., & Okech, J E A (2006) The supervision of group work model: Adapting the Discrimination Model for supervision of group workers The Journal for Specialists in Group Work, 31 (2), 113 134. Stockton, R., Morran, D K., & Clark, M B (2004) An Investigation of Group Leaders' Intentions Group Dynamics: Theory, Research, and Pra ctice, 8 (3), 196 206.
92 Stoltenberg, C., McNeill, B. W., & Delworth, U. (1998). IDM supervision: An integrated developmental model for supervising counselors and therapists San Francisco: Jossy Bass. Waldo, M., Kerne, P A., Kerne, V V H (2007) Therapeutic factors in guidance versus counseling sessions of domestic violence groups The Journal for Specialists in Group Work 32, 346 361. Wilbur, M. P., Frank Saracini, J., Roberts Wilbur, J., & Torres Rivera, E. (1997a, September ). A chaos model of group supervision process research Paper presented at the meeting of the North Atlantic Regional Association of Counselor Educators and Supervisors, Plattsburgh, NY. Wilbur, M. P., Frank Saracini, J., Roberts Wilbur, J., & Torres Rivera, E. (1997b, Septe mber). A chaos model of group process research. Unpublished manuscript. Yalom, I D (2005) The Theory and Practice of Group Psychotherapy (5th ed.) New York: Basic Books.
93 BIOGRAPHICAL SKETCH Kevin A. Tate received his B.S. in p sychology, M.Ed. and Ed.S. in m ental h ealth c ounseling, and Ph.D. in c ounselor e ducation from the University of Florida. He currently serves as Senior Assistant Director for Career Development at the University of Florida Career Resource Center. He is married to Tyson Tate and has one daughter named Mercy Tate.