1 EXPLORING RELATIONAL HEALTH AND COMFORT WITH CLOSENESS IN STUDENT COUNSELOR DEVELOPMENT By SARA NASH 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 2012
2 2012 Sara Nash
3 To my community
4 ACKNOWLEDGMENTS I thank my chair, Dr. Sondra Smith Adcock, for her steadfast commitment to me. Her kindness, generosity, and insistence on scholarship have been indispensable, and her contributions to my development are impossible to quantify. I thank Dr. Wayne Griffin for co chairing my committee, setting an example for my career in crisis intervention, and just being an all around lovely pe rson. In addition to his valuable perspective on methodology, Dr. Chris McCarty is the first scholar I really knew, and without his support and friendship, I would not have come this far. Dr. Ana Puig laid the foundation for this dissertation when she intr oduced me to the common factors of therapy seven years ago. She has offered many critical insights on this project, as well as lived by example through her own genuine presence, passionate teaching and creativity. I thank my father, Jim Nash, who taught m e the value of authenticity, exposed me to broad ideas about the human condition, and modeled being real in relationship. I thank my mother, Susan Nash, and my step father, Bill Paine, for being there when I needed refuge, help with a flat tire, musical an d artistic inspiration, or just a good long hug. I thank my siblings Kristen Sink, Justin Nash, and Erin Browning Paine, for loving me through ups and downs, and for tolerating my absence from their lives during my tenure of study. And, while they are bles sedly too young to know what a dissertation is, I thank my niece Anna and my nephews Dylan and Mason for reminding me that enchantment is available whether I have deadlines or not. I thank Marshall Knudson, Ali Martinez, Perry Peace, and Connie Hardstock at the Alachua County Crisis Center for the therapist centered group supervision they have provided. Their investment in that process has helped me learn to trust myself and be
5 more present with people in times of deep suffering and pain. I cannot thank th em thank the late Dr. Pat Korb, who taught me Gestalt therapy and the power of speaking to the inherent, unbreakable wholeness within each person. I thank Sarah Nola n for her careful edits of this manuscript. I thank Meggen Sixbey for her mentorship, leadership, and friendship. I also thank the folks at the Santa Fe College Counseling Center, Kalpana Swamy, Lara Zwilling, Bruce Tucker, and Steve Vutsinas, for being great colleagues and for cheering me to the finish line Lastly I want to thank my friends for making this long journey so interesting and fun. I thank Emi Lenes for her enthusiasm, honesty, creativity and unconditional love. I thank Drake Finn Logan, whos e commitment to living authentically inspires me to live from what is most true in me. I thank Anna Heineman for arriving in Gainesville at just the right time and offering her kitchen table, encouragement, and lots of unrelated but sorely needed conversa tions about art and life. I thank Shengying Zhang for sharing her depth, wisdom, and kindness with me in living rooms and restaurants all over Gainesville. I thank C hris ty McCain for being there, and for introducing me to so many lovely people who now are my community, too. Among them are Lauren Day, Whitney Untiedt, Crystal Goodison Sally Adkins and Kiara Winans. I thank Katie White for early encouragement of my effor ts, both scholarly and painterly, and for renting me their sweet yellow house I thank JJ Buchholz for so many things, but most of all for reminding me to eat and dance.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 LIST OF TERMS ................................ ................................ ................................ ........... 12 ABSTRACT ................................ ................................ ................................ ................... 15 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 17 The Training Debate ................................ ................................ ............................... 17 Established Theory and Research ................................ ................................ .......... 18 New Directions ................................ ................................ ................................ ........ 19 Theoretical Framework ................................ ................................ ........................... 21 Statement of the Problem ................................ ................................ ....................... 25 Need for the Study ................................ ................................ ................................ .. 26 Purpose of the St udy ................................ ................................ .............................. 29 Rationale for the Methodology ................................ ................................ ................ 30 Research Questions ................................ ................................ ............................... 33 Ov erview of Study ................................ ................................ ................................ ... 34 2 REVIEW OF RELATED LITERATURE ................................ ................................ ... 35 An Historical Look at Counselor Development Models ................................ ........... 35 Hogan ................................ ................................ ................................ ............... 37 Stoltenberg ................................ ................................ ................................ ....... 37 Loganbill, Hardy & Delworth ................................ ................................ ............. 38 Stoltenberg, McNeill and Delworth ................................ ................................ ... 40 Research Supporting Counselor Development Models ................................ .......... 41 Liter ature Related to Counselor Development ................................ ........................ 45 Counselor Cognitive Complexity ................................ ................................ ...... 46 Critical Incidents ................................ ................................ ............................... 51 Personal Characteristics Associated with Desirable Counselor Development 54 Theoretical Framework ................................ ................................ ........................... 57 Relational Cultural Theory in Counselor Preparation ................................ ....... 58 Attachment Theory: Implications for Counselor Development .......................... 61 The R ole of Counselor Preparation Programs in Counselor Development ............. 63 Summary and Conclusions ................................ ................................ ..................... 67
7 3 METHODOLOGY ................................ ................................ ................................ ... 69 Overview ................................ ................................ ................................ ................. 69 Sampling Procedures ................................ ................................ .............................. 69 Data Collection Methods ................................ ................................ ......................... 70 Design ................................ ................................ ................................ ..................... 72 Instruments ................................ ................................ ................................ ............. 72 Relational Health Indices ................................ ................................ .................. 72 Adult Attachment Scale ................................ ................................ .................... 76 Supervisee Levels Questionnaire Revised ................................ ..................... 77 Hypotheses ................................ ................................ ................................ ............. 79 Operational Definition of Variables ................................ ................................ ......... 80 Data Analysis ................................ ................................ ................................ .......... 80 Methodological L imitations ................................ ................................ ...................... 83 4 RESULTS ................................ ................................ ................................ ............... 85 Overview ................................ ................................ ................................ ................. 85 Demographic Characteristi cs of Sample ................................ ................................ 86 Regional Distribution of Sample ................................ ................................ ....... 87 Gender and Racial Identification ................................ ................................ ...... 88 Years in a Preparation Program ................................ ................................ ....... 88 Program Emphasis ................................ ................................ ........................... 89 Personal Therapy ................................ ................................ ............................. 89 Professional Counseling Experience ................................ ................................ 90 Supervision Quality ................................ ................................ .......................... 90 Instrumentation ................................ ................................ ................................ ....... 91 Hypotheses Tests ................................ ................................ ................................ ... 92 Hypothesis 1 ................................ ................................ ................................ ..... 93 Hypothesis 2 ................................ ................................ ................................ ..... 94 Hypothesis 3 ................................ ................................ ................................ ..... 95 Hypothesis 4 ................................ ................................ ................................ ..... 95 Summary ................................ ................................ ................................ ................ 96 5 DISCUSSION ................................ ................................ ................................ ....... 105 Overview of Study and Discussion of Findings ................................ ..................... 105 Professional Counseling Experience ................................ .............................. 106 Comfort with Closeness ................................ ................................ .................. 108 Community Relational Health ................................ ................................ ......... 109 Mentor Relat ional Health ................................ ................................ ................ 110 Supervision Quality ................................ ................................ ........................ 111 Implications for Theory ................................ ................................ .......................... 112 Implications for Practice ................................ ................................ ........................ 114 Limitations of the Study and Implications for Research ................................ ........ 11 6
8 Limitations ................................ ................................ ................................ ...... 116 Implications for Research ................................ ................................ ............... 119 Summary ................................ ................................ ................................ .............. 120 APPENDIX A RELATIONAL HEALTH INDICES* ................................ ................................ ....... 122 B COMFORT WITH CLOSENESS ................................ ................................ ........... 126 C SUPERVISEE LEVELS QUESTIONNAIRE REVISED (SLQ R) ........................... 127 D SUPPLEMENTAL AND DEMOGRAPHIC QUESTIONNAIRE .............................. 132 LIST OF REFERENCES ................................ ................................ ............................. 136 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 148
9 LIST OF TABLES Table page 4 1 Frequencies for demographic variables ................................ .............................. 98 4 2 Quality of supervisi on most recently received ................................ ................... 100 4 3 Instrument total scores and subscale scores ................................ .................... 100 4 4 Cronbach alpha reliability coefficients to study scales ................................ ...... 101 4 5 Pearson product moment correlations between independent and dependent variables (one tailed tests) ................................ ................................ ............... 102 4 6 Model summary for hypothesis test 1 ................................ ............................... 103 4 7 Coefficients for hypothesis test 1 ................................ ................................ ...... 103 4 8 Model summary for hypothesis test 2 ................................ ............................... 103 4 9 Coefficients for hypothesis test 2 ................................ ................................ ...... 103 4 10 Model summary for hypothesis test 3 ................................ ............................... 103 4 11 Coefficients for hypothesis test 3 ................................ ................................ ...... 104 4 12 Model summary for hypothesis test 4 ................................ ............................... 104 4 13 Coefficien ts for hypothesis test 4 ................................ ................................ ...... 104
10 LIST OF ABBREVIATION S AAI Adult Attachment Interview AAS Adult Attachment Scale ANCOVA Analysis of Covariance CACREP Council for Accreditation of Counseling and Related Educational Programs CCI Counselor Complexity Index CCM Counselor Complexity Model CCQ Counselor Cognitions Questionnaire CES D Center for Epidemiological Studies Depression Scale IDM Integrated Developmental Model IRB Institutional Review Board LEP Learning Environment Preferenc es MANOVA Multiple Analysis of Variance MPDQ Mutual Psychological Development Questionnaire MSPSS Multidimensional Scale of Perceived Social Support QRI Quality of Relationships Questionnaire RCT Relational Cultural Theory RHI Relational Health Indices RHI C Relational Health Indices Community Subscale RHI M Relational Health Indices Mentor Subscale RHI P Relational Health Indices Peer Subscale SLQ Supervisee Levels Questionnaire SLQ R Supervisee Levels Questionnaire Revised SPSS Statistical Package for th e Social Sciences
11 UCLA University of California, Los Angeles WUSCT Washington University Sentence Completion Test
12 LIST OF TERMS A TTACHMENT THEORY (B OWLBY 1988) A theory of human development that attempts to explain how and why people attach or fail to attach to the significant others experiences with their caregivers shape their expectations of how others will connect with them and attend to their survival and care. According to the theory, these early experiences shape the way people relate as adults, and while these patterns can be changed through reparative relationships, they initially tend to be persistent, particularly in times of stress. Attachment theory has been offered as one way to explain why the therapy relationship is so important, and to investigate outcomes from the counseling, supervisory, and client perspectives. A UTHENTICITY acquisition of knowledge and understanding of self and other (Comstock, 2002). A UTONOMY D EPENDENCY is a developmental construct that reflects the degree of independence or dependence a supervisee is displaying, and to move fluidly throughout several domains of counselor competency. C OMFORT WITH CLOSENESS is a construct der ived from attachment theory and used in the Close subscale of the Adult Attachment Scale (AAS ; Collins & Read, 1990) to assess the extent to whic h a person feels he or she is comfortable with intimacy in human relationships. C OMMUNITY RELATIONAL HEALTH program as a whole (Lian g et al., 2002). For the purposes of this study, community is defined as all the members of the C ONNECTION is defined as any engagement between two or more people ler & Striver, 1997, p. 26). C OUNSELOR D EVELOPMENT refers to the lifelong process by which people learn and practice the art and science of the counseling profession, which includes personal domains such as self awareness, self reflection, and openness to new experiences as well as professional domains such as demonstrable knowledge, skills, and clinical judgment.
13 E MPOWERMENT / ZEST refers to an energetic and vital feeling one gets from participating in growth enhancing connections (Miller, 1986). The ind ividual feels inspired, strong, and more eager to take action (Liang et al., 2002). M ENTOR R ELATIONAL HEALTH describes a relationship the participant has, usually with an older adult, who listens, shares experiences, and offers guidance on an aspect of t ( Liang et al., 2002 ). For the purposes of this study, mentor will be defined as pro g ram. M OTIVATION and fluct uates throughout the training and supervision process. M UTUAL ENGAGEMENT / EMPATHY happens when both parties in a relationship feel a sense of involvement with, commitment to, and emotional attunement to the relationship (Liang et al., 2002). For this to occur, both parties must engage in the authentic expression of their thoughts and feelings and be willing to be impacted by the connection (Miller & Stiver, 1997). N OVICE C OUNSELOR for the purposes of this study, refers to a counselor who is unlicensed a nd under mandated supervision as either a graduate intern or a provisionally state licensed practitioner P EER RELATIONAL HEALTH refers to someone with whom the participant shares interests, feels attached to, trusts, respects, and can rely upon for suppo rt in difficult times (Liang et al., 2002). For the purposes of this study, peer is defined as a classmate within the R ELATIONAL C ULTURAL T HEORY (RCT) is a theory of human development that holds that optimal de velopment is characterized by an increasing capacity to engage in healthy, growth enhancing ways with the important development as a process of individuation and autonomy, RCT holds that developme nt happens in and through relationship and that people grow towards deep, healthy connections rather than away from significant others over the lifespan ( Comstock & Qin, 2005) R ELATIONAL HEALTH / QUALITY is a construct derived from RCT and is the notion t hat healthy relationships involve mutual engagement, empathy, connection, and authenticity that leaves both parties feeling a greater sense of empowerment, zest for life, and a desire for more relationships (Comstock, 2002)
14 S ELF O THER A WARENESS refers to clients are experiencing while also being aware of their own feelings and reactions during counseling sessions. S TUDENT COUNSELOR for the purposes of this study, refers to a person in a master o r doctoral level Counselor Education preparation program. S TUDENT COUNSELOR DEVELOPMENTAL LEVEL is a construct taken from traditional theories of counselor development (Hogan, 1964; Loganbill, Hardy, and Delworth, 1982; Stoltenberg, 1981) that holds th at counselors progress through normative and qualitatively different stages of development on their way to becoming optimally effective or master practitioners. T HERAPEUTIC RELATIONSHIP has been described as the strong emotional bond or therapist clien t connection component of the therapeutic alliance, which broadly refers to a safe, trusting therapeutic interaction that 1979).
15 Abstract of Dissertation Presented to the Graduate School of the U niversity of Florida in Partial Fulfillment of the Req uirements for the Degree of Doctor of Philosophy EXPLORING RELATIONAL HEALTH AND COMFORT WITH CLOSENESS IN STUDENT COUNSELOR DEVELOPMENT By Sara Nash August 2012 Chair: Sondra Smith Adcock Cochair: Wayne Griffin Major: Mental Health Counseling Counselor development has been conceptualized as a gradual progression from the cognitive, technical, and relational rigidity of novices to the optimally effective cognitions, interventions, an d therapeutic alliances of master practitioners (Hogan, 1964; Loganbill, Hardy, & Delworth, 1982; Skovholt & Ronnestad, 1992a; Stoltenberg, 1981). Within this paradigm, have been viewed as normative, yet to date r elational constructs have not been major consideration s in counselor development research or theory. Given the well established importance of the therapist client relationship in counseling outcomes (Wampold, 2001) and the growing recognition that developm ent is enhanced through quality relationships (Comstock, 2002; Jordan, 2002), the current study explored the contribution of relational variables to student counselor development. Stepwise multiple r egressions on a cross section of master and doctoral stu dents in CACREP accredited clinical mental health counseling programs in 19 states showed a positive trend between self reported overall and domain specific development al level s and prior counseling experience, comfort with relational closeness, and relati onal health
16 in a counselor preparation program. Additionally, self reported domain specific development was negatively related to supervision quality and relational quality with a counseling mentor. Relational Cultural Theory (Comstock, 2002) and attachmen t theory (Bowlby, 1988) provided a framework for interpreting the results of this study. Study limitations included the lack of causal inference possible in a correlational study, use of a convenience sample, weaknesses inherent in self reported observatio ns, and questionable scale psychometrics. Implications for theory and research concentrated on improving counselor development models and measures to include a stronger relational emphasis. Implications for practice focused o n the role of counselor educato rs in creating relational ly healthy preparation programs and ways to integrate relational domains in admissions interviews, training, supervision and ongoing evaluation
17 CHAPTER 1 INTRODUCTION The Training Debate s an undefined technique applied to unspecified problems with unpredictable results. For this we recommend rigorous by extension, issues have been research ed and debate d for several decades (Ahn & Wampold, 2001; Buser, 2008; Loganbill, Hardy, & Delworth, 1982; Skovholt & Ronnestad, 1992a; Smith, 2003; Stoltenberg, 1981; Wampold, 2001; Yalom 2005). As early as 1975, Rogers expressed discontent with technique based training practices, arguing that they missed the humanity of both clients and therapists Over the past two decades, counseling and psychology preparation programs have similarly been criticized for overemphasizing techniqu e based training at the expense of the quality of the therapeutic relationship and the inter perso nal development of trainees (Bergin, 1997; Mahoney, 1986; Winslade, Monk, & Drewrey, 1997). Given the considerable body of research suggesting that the ther apeutic relationship is a critical factor in successful therapy outcomes (Ahn & Wampold, 2001; Assay & Lambert, 2004; Lambert & Bergen, 1994; Lambert & Cattani Thompson, 1996; Wampold, 2001), questions about the role of counseling and psychology programs i n promoting relational skills in student counselors have resurfaced in the literature (Whiston & Coker, 2000). While some disagreement persists about what makes therapy effective, the mounting evidence that the quality of the therapeutic relationship accou nts for considerably more therapeutic success than indivi dual theories or techniques has
18 generated calls for increased research on the relational aspects of student counselor development (Assay & Lambert, 2004 ; Wampold, 2001 ; Whiston & Coker, 2000 ) Furthe r, there is a call for research investigating how preparation programs attend to the relational development of their students (Lambert & Ogles, 1997; Torres Rivera, Phan, Maddux, Wilbur, & Garrett, 2001). Established Theory and Research To date, consensus opinions about to foster therapeutic relationships have been largely informed by theories of counselor development and supervision (Hogan, 1964; Loganbill et al., 1982; Stoltenberg, 1981) and qualitative research (Skovholt & Ronnestad, 1992b, 2003). Overall, counseling students and new professionals have been characterized in the literature as simplistic and rigid thinkers who are externally focused, personally incongruent, and preoccupied with those aspects of counseling mos t related to the specific ingredients of therapy, such as easily implemented interventions, fool proof techniques, and reductionist applications of theory that often miss the nuances and importance of the therapeutic relationship (Hogan, 1964; Loganbill et al., 1982; Stoltenberg, 1981 ; Skovholt & Ronnestad, 2003). Student counselors in particular are said to exhibit behaviors and personal attributes associated with negative therapeutic alliances, such as intellectual and emotional rigidity, interpersonal di stancing, tension and anxiety, and over structuring of sessions (Ackerman & Hilsenroth, 2001). Taken together, theories of counselor deve lopment propose that these restricted attributes and behaviors form a normative developmental stage and that only afte r graduation and completion of supervision requirements do counselors become more relaxed, reflective, complex, and relationally effective practitioners ( Stoltenberg, 1981 ; Skovholt & Ronnestad, 2003 )
19 New Directions Despite the theoretical and research b ased assertions in the literature recent research findings have raised questions about the generalizability of these consensus depictions of student and novice counselors (Furr & Carroll, 2003; Goodyear, Wertheimer, Cypers, & Rosemond, 2003; Howard, Inman & Altman, 2006; Lennie, 2007; McAuliffe, 2002 ) Dunkle and Friedlander ( 1996) found that interns at a university counseling center who scored high on comfort with relational closeness, as measured by the Close subscale of the Adult Attachment Scale (AAS ; Colins & Read, 1990), were rated significantly higher by clients on the quality of the therapeutic alliance. In the same study, the authors found that self reported quality of social support iance ratings. These findings suggest that both comfort with relational closeness and perceived quality of social support might be age nor professional experiences were uniquely predictive of ratings of the therapeutic alliance, drawing into question the consensus assumption that new counselors are necessarily more relationally limited than experienced therapists (Dunkle & Friedlander, 1996). Results from studies on reactions to training experiences that encourage authentic interpersonal engagement have lent support to the likelihood that at least some students are developmentally open to and enthusiastic about learning opportunities that involve deep engagement with issues of the therapist client relationship, such as risk taking self exploration, and personal vulnerability (Anderson & Price, 2001; Furr & Carroll, 2003; McAuliffe, 2002; Lennie, 2007 ). Further anxiety does not appear to be as pre valent a feature of the novice experience as Skovholt and
20 Ronnestad (2003) have suggested (Chaplin & Ellis, 2002). In fact, a few studies (Lichtenberg, 1997; Stein & Lambert, 1995; Whiston & Coker, 2000) have found no significant difference in effectivenes s between novice helpers and experienced professionals, throwing doubt on the traditional, linear models of counselor development by suggesting that the key qualities of effective helpers may be as present at early stages of development as later ones. Last ly, researchers who have investigated cognitive complexity in novice counselors reported that some counseling students demonstrate cognitive processes associated with more sophisticated client conceptualization, less rigidity, and more effective in session behaviors ( Borders, 1989; Fong et al., 1997; Welfare & Borders, 2010 ). T hese studies of counselor development raise questions about the extent to which current research and theories have captured ps during a graduate preparation program and in the first years of post graduate practice. Some researchers (Whiston & Coker, 2000) have even argued that current training methods do not seem to impact counselor effectiveness, and that teaching methods nee d reconsideration and revision in line with what is now known about the importance of the therapeutic relationship and the person of the therapist who participates in those relationships. Likewise, counselor preparation programs have been criticized for ov er emphasizing techniques and skills training and under emphasizing the relational development of student practitioners (Bergin, 1997; Whiston & Coker, 2000). Yet to date, the relational environments and the academic cultures in which students learn counse ling have not been a major consideration in counselor development research or theory. Extant theories of counselor development have not investigated relational
21 factors that may be linked to novice development, including the perceived quality of relationshi ps within a preparation program or comfort with closeness in relationships. Furthermore, because the current academic environments of preparation programs may exert pressures and influences at odds with the complex, sensitive, and often ambiguous relationship dynamics that arise in real therapy settings (Skovholt & Ronnestad, 1992 b ), the relational quality within preparation programs and when assessing student counselor development (Whiston & Coker, 2000). Theoretical Framework Theories of counselor development are attempts to understand the normative experiences and changing needs of counselors as they progress through the required education and supervision e xperiences and eventually occupy non supervised professional roles. Though differences in developmental models exist, the prominent theories of counselor development (Hogan, 1964; Loganbill et al., 1982; Stoltenberg, 1981 ; Skovholt & Ronnestad, 1992) have proposed that counselors move through several stages of change in fairly linear fashion, from the naive rigidity, cognitive simplicity, and dependency of the novice to the eventual personal and professional autonomy, relational competence, and personal/pro fessional integration of the master therapist. This proposed linear, stage wise progression bears much in common with traditional Western theories of human development that cast the individual as the central character on a relatively predictable journey fr om dependency to individuation and personal autonomy (Chickering, 1969; Comstock, 2002; Erickson, 1963; Miller, 1976; Worthington, 1987 ).
22 This study explores the contribution of previously unexamined relational variables to student counselor development, and use s Relational Cultural Theory (RCT ; Comstock, 2002) as a guiding theoretical framework. Recently, recognition has grown for the importance of relationships and cultural contexts in the development of the individual (Porter, 2002). Since its inception as Relational Theory approximately twenty years ago, RCT has been gaining attention as an alternative theoretical framework for understanding human development and the counseling endeavor (Comstock, 2002; Comstock Frey, Beesley, & Newman, 2005; Comstock, Hammer, Strentzsch, Cannon, Parsons, & Salazar, 2008; Liang, Tracy, Taylor, Williams, & Jordan, 2002). RCT posits that people develop towards relational closeness and complexity over the life span rather than towards individuation and autonomy (Comstock, 2 002). In RCT, optimal development is characterized by the ability to have quality relationships that are mutually empathic and lead to greater connectedness, empowerment, and authenticity (Jordan, 2002). Further, according to RCT, d evelopment is restricted when relationships and cultural contexts do not provide opportunities for growth enhancing connections (Walker, 2005). Within RCT, conflicts and disconnections are seen as a normative part of relationships, and naming barriers to and moving back into conn ection are considered sources of healing, empowerment, and transformation (Comstock et. al, 2003). Alternatively, chronic disconnections, isolation, and alienation from relationships are associated with reduced closeness and fewer opportunities to develop and grow in positive ways. The fundamental premise of RCT is that the development of the individual cannot be examined or understood without also considering the health of a
23 oppo rtunities for development (Miller, Jordan, Kaplan, Stiver, & Surrey, 1991; Miller & Stiver, 1997) Given the importance of relationship in the counseling process, a particularly relevant concept of RCT is mutual empathy (Comstock, 2002). According to RCT, mutual empathy is a critical component of high quality growth fostering relationships directions; in mutual empathy, both the client and counselor are impacted by the exchange and are able to authentically acknowledge this impact (Comstock, Duffey, & St. George, 2003). As a result, both parties emerge from the exchange with the five good t hings (Miller, 1991) : ( 1) increased sense of vitality and zest for life ; ( 2) empowerment and the ability to take action ; ( 3) increased clarity about oneself, the other, and the relationship ; ( 4) greater sense of self worth ; and ( 5) deepened desire for more relationships. In order for mutual empathy to occur, both parties in the exchange must be vulnerable, emotionally responsive, and open to change qualities that require a considerable amount of interpersonal risk but are associated with more growth enhanci ng relationships (Comstock et al., 2008; Jordan, 1992; Miller et al., 1991). The current consensus of counselor development models seems to be that novice counselors are either developmentally unable or unwilling to open themselves to mutually empathic enc ounters, preferring the seeming security of theories, techniques, and the notion that answers and fixes are easy to learn and apply. From the perspective of RCT, context and process factors, including the responsiveness and health of other
24 significant rela tionships and institutional environments, must be considered to better understand these relational disconnections (Jordan, 1997). Given that, from the perspective of RCT development happens with in and through other important lif e contexts, RCT offers several potentially valuable contributions to the study of counselor development. First, it provides a relational framework for revisiting established measures of student counselor development (Comstock et al., 2008). Second, rather than attributing relational limitations to a universal stage through which all counselors eventually pass, RCT offers therapeutic relationships both in the training years and throughout the career span (Qin & Comstock, 2005). Third, by further articulating the growth fostering qualities of relationships, RCT may help counselors, supervisors, and educators to be more aware of their own contributions to relationships (e .g., supervisory relationships) that have the potential to foster or hinder development. Originally, RCT was proposed as an alternative developmental theory to understand the developmental processes of women and other members of marginalized groups whos e experiences of living and growing did not fit the Caucasian and male oriented descriptions of change over time. In the past decade, however, RCT has been used in research studies of mixed gender populations (Adams, 2010; Frey et al., 2005) and as a guidi ng framework for developmental courses for counselors in preparation (Comstock, 2004). Over the past two decades, RCT has been recognized as applicable to both female and male developmental processes ( Bergman, 1991; Bergman & Surrey, 1994; Comstock & Qin, 2005; Dooley & Fedele, 2004; Jordan, 2002; Mirkin & Geib,
25 1995 ). As an alternative to traditional developmental theories, RCT is a potentially meaningful way to explore how relational factors may influence student counselor development. Statement of the P roblem Relational quality between client and therapist has been shown to be a critical component of successful therapy, even more important than theoretical orientation or use of specific techniques or interventions (Ahn & Wampold, 2001; Wampold, 2001). Y et predominant theories of counselor development (Loganbill et al., 1982; Skovholt & Ronnestad, 1992; Skovholt & Ronnestad, 2003; Stoltenberg, 1981) depict novice counselors as relationally limited, becoming more relationally skillful only later in the car eer cycle. While some researchers have found evidence for different developmental levels at different points in counselor preparation (Borders, 1989; Fong et al., 1997; McNeill, Stoltenberg, & Romans, 1992; Welfare & Borders, 2010), these findings typicall y tap constructs within the individual (e.g., cognitive development ) and have neglected the possible contribution of relational variables to counselor development such as the quality of the preparation environment. While stage wise counselor development t heories and research studies have contributed important insights to current knowledge of novice counselor development, more research is needed to understand the contribution of the relational domain to well established conceptualizations of counselor devel opment. To date, the theoretical and reported relational limitations of novice counselors have been interpreted as developmentally normative. Within this paradigm, questions of how counseling students experience relationship s such as their capacity to be close to others and the quality of relationships within their preparation programs, have not been
26 explicitly considered in research on developmental outcomes. Without taking a studied look at relational factors in student counselor development, preparatio n programs risk training and graduating new professionals who may be limited in important relational aspects of the counseling process While current research has indicated that counseling students and new professionals (i.e., novice counselors) seem to ha ve relational limitations (Ackerman & Hilsenroth, 2001 ; Skovholt & Ronnestad, 2003 ), it is unclear whether these are developmentally transient, stem from fixed, intrinsic counselor characteristics, or reflect the relational quality of the preparatory envir onment. However, without research in this area, the conversation remains mainly theoretical and anecdotal. If preparation programs fail to consider their own role in training counselors for the therapeutic relationship a variable that has been overwhelming ly tied to successful client outcomes then opportunities to adjust their training approaches to emphasize relational factors and potentially show better developmental outcomes are lost, and both student and client developmental outcomes may suffer. Additio nally, if counseling students lack a fundamental comfort with closeness in relationships when they enter training programs, they may not be ready to take advantage of the relationally rich developmental opportunities that are available to them. F rom th e pe rspective of RCT, educators, supervisors and students of counseling indeed, all the members of the counselor education community may either create or restrict developmentally significant opportunities for mutually empathic, authentic and growth enhancing engagement with their students. Need for the Study The Council for Accreditation of Counseling and Related Educational Programs Standards (CACREP, 2009) ha s stated that admission to graduate preparation
27 relevant interpersonal relationships in individual and small Furthermore, CACREP Standards (2009) dictate that counselors must be aware of the characteristics and behaviors that lead to successf ul therapy outcomes. Given that success, it follows that student and professional counselors, counselor educators, and supervisors need to be aware of their capacity to for m, model, and facilitate high quality, developmentally enriching relationships. Currently, the consensus opinion is that novice counselors are developmentally limited in their relational abilities. While many theories of counselor development share this v iew, there is a lack of quantitative data on relational factors in counselor development or in the broader training context factors that may somehow be influencing these limitations. In fact, the relational dimension of counselor development has been large ly neglected in studies about student counselor developmental levels during a preparation program. An implicit assumption within traditional theories of counselor development has been that the training environment provides for the developmental needs of st udents. From this perspective, student counselor growth depends largely upon forces within the individual that inspire her or him to grow in response to the optimal environment provided by the program, educators, and supervisors. Yet some have suggested th at the training environment may not be optimal and have called for more extensive research and validation of novice counselor development and more attention to the relationship in counselor preparation (Hansen, 2010; Whiston & Coker, 2000).
28 Given the impo rtan t association between quality of the therapeutic relationship and positive client outcomes, it is important to expand existing knowledge about student counselor development that may be tied to the counselor role in establishing a quality therapeutic re lationship. Based on the RCT framework, factors such as the quality of relationships students report experiencing in their preparation programs and their own reported comfort with closeness in relationship may promote both counselor development and clinica l results. Current developmental literature leaves many questions unanswered about what student counselors are capable of relationally and the contribution of preparation programs to the relational aspects of counselor development. By not addressing these important issues researchers in the field of counseling risk the following: (1) perpetuating the self limiting assumption that novice counselors are, by nature, limited in their relational abilities; and (2) not gaining more insight into the important rol e that relationships among all parties involved in preparation programs (i.e., student counselors, clients, supervisors, educators, and the larger academic and community contexts ) play in the development of counselors Claims about the relational restrict ions of student counselors have not been extensively validated. Further, much of counselor development literature and research has been primarily theoretical, qualitative, and/or cross sectional, making it unwise to widely generalize findings. Reflecting t his concern, Howard et al. (2006, p. 89) have Although theoretical contributions offer a framework from which to understand potentially relevant issues in counselor development, researchers must be cautious in over relying on anecdotal and narrati ve information and seek empirical support for these While valuable, extant quantitative studies of counselor development have
29 been based on traditional theories of human development; thus, the findings have only taken the field so far in under variable capacity to form and sustain effective therapeutic relationships in the educational years and throughout the career cycle. With a richer understanding of the variability within and relation ships among these factors, educators, supervisors and researchers can begin to directly address the relational capacities and limitations of counselors and promote the development of relationally competent counselors earlier in the career cycle. Purpose o f the Study The purpose of this study is to add to the existing research and knowledge about the development of novice counselors, with specific attention to relational variables that may help contribute a relational emphasis to existing counselor developm ent models. The present study had a long term goal of stimulating research interest about the relational capacities of novice counselors and whether it might be possible for preparation programs to enhance relational aspects of developmental outcomes in no vice counselors Another long and generate dialogue about how preparation programs may promote or restrict student c ounselor development and preparation for the therapeutic relationship. Over time, increased attention to relationship and mutuality in the training process may foster relationships and contexts with qualities similar to those of effective therapeutic rela tionships. Additionally, support for the role of relational variables in student counselor development may help preparation programs better identify the factors that are contributing to student counselors who are struggling relationally and/or better
30 asses s these dimensions during the admissions process. The present study was intended to begin the process that could lead to the attainment of these long term goals. More immediately, the goal of the present study was to examine the contribution of previously absent relational constructs to student counselor development. As such, t he present study investigate d the extent to which self reported student counselor developmental level could be attributed to self reported comfort with relational closeness and /or per ceived relational quality with a counseling peer and mentor, and with the counselor preparation program as a whole An additional purpose of this study was to investigate how a number of individual factors (i.e., the amount personal therapy, previous profe ssional counseling experience supervision quality, a common versus specific factors training program orientation, gender, age, and race ) might influence self reported student developmental level, comfort with relational closeness, and relational quality w ith in a preparation program. Rationale for the Methodology This study explored the whether self reported relational health in a preparation program and self reported comfort with closeness had a statistically significant relationship to self reported stud ent counselor development level. Further, this study investigated the possible contribution of time in personal therapy, prior professional counseling experience, quality of supervision, a common versus specific factors training program orientation (as ass essed by students) and the demographic variables of gender, race, and age to student counselor developmental level. The study utilized quantitative data gathering and analyses and established and validated survey instruments to measure the variables of int erest, which included perceived relational health in a preparation program, comfort with relational closeness, student
31 developmental level, amount of personal therapy, prior professional counseling experience supervision quality, a common versus specific factors program orientation, race, age and gender. Theorists and researchers have hypothesized that novice counselors develop in a relatively predictable sequence that has been described as a general movement from dependency and relational limitation to a utonomy, relational skill, and personal/professional congruence (Stoltenberg, McNeill, & Delworth, 1998). The theoretical underpinnings of these studies have placed the trajectory of development within the individual and have given far less attention to th e relational aspects of development such as the relational quality of preparation programs and the contribution of other potentially development fostering constructs (e.g. comfort with closeness, time in personal therapy, etc.) to the developmental process With research on the effective ingredients of psychotherapy highlighting the importance of relationship to outcomes, it was considered probable that relational variables may also play a role in counselor student development. Relational health in a prepar ation program, comfort with relational closeness (an attachment construct), and additional counselor characteristics and experiences were tested for the statistical strength of their relationship to student counselor developmental level. Except for researc h on cognitive complexity, much of the research on novice counselor development has been qualitative, anecdotal, and/or cross sectional. The contributions and limitations of these studies will be discussed in Chapter 2. Assessments of counselor development level have been based either on theories of moral development, cognitive complexity, or linear, stage wise assumptions of
32 development. Stage wise theories of development have been the most prominent over the past several decades, particularly the Integrat ed Developmental Model (IDM ; Stoltenberg, McNeil, & Delworth, 1998). Further, the developmental constructs of the IDM have been translated into a measure with modest reliability and validity in the literatur e (i.e., Supervisee Levels Questionnaire; McNeill et al., 1992 ) and used in several research studies ( Lovell, 1999; Deal, Bennett, Mohr, & Hwang, 2011 ). Therefore, in the present study, this self report measure of novice developmental level was used as the outcome variable to capture a traditional, widel y accepted conceptualization of counselor development. As stated earlier, reported counselor comfort with relational closeness appears tied to client ratings of the therapeutic alliance in at least one outcome study (Dunkle & Friedlander, 1996). Thus, it was thought meaningful to determine whether there are connections between student counselor comfort with relational closeness and self reported developmental level. Therefore, in addition to asking participants to report on relational health and answer a developmental assessment, the present study used Close, an Adult Attachment subscale assessing comfort with closeness in relationships (AAS ; Collins & Read, 1990) to measure this construct in participants. The research questions presented below represent an exploration of the utility of RCT for explaining student counselor developmental level. Additionally, relationships among comfort with closeness, supervision quality, time in preparation program, amount of personal therapy, previous professional couns eling experience, a common versus specific factors program emphasis, gender, race, age and counselor developmental level were explored.
33 Research Questions The questions that follow represent an itemization of the more general question regarding how perceiv ed relational quality of a preparation program and comfort with relational closeness are related to a traditional assessment of counselor development level. The conceptual framework for the independent variables chosen for study was derived from Relational Cultural Theory (RCT) development is driven in large part by and through the quality of their relationships. authenticity, and empowerment; development i s hindered when people suffer chronic disconnections, feel isolated or alienated, or are unable to be authentic in important contexts, such as at work, in school, and/or with family or friends. RCT is particularly attuned to the role of power and status di fferentials in shaping the relational health within dyads and broader institutional cultures. If results of this study found that relational variables could predict novice developmental level, these findings would lend support to the idea that current cou nselor development theory should broaden to include deeper consideration of these variables. This finding could in turn, affect the preparation of counselors. The following questions guided data gathering and analyses. Is there a significant relationship between Developmental L evel (as measured by the total score of Supervisee Levels Questionnaire Revised ) and Relational H ealth (as measured by the total score of the Relational Health Indices ), Comfort with C loseness (as measured by the Adult Attachment subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age ? Is there a significant relationship between stude nt Self Other Awareness developmental level (as measured by the Supervisee Levels Questionnaire Revised) and Peer Relational Health, Mentor Relational Health, and Community Relational Health (as measured by the Relational Health Indices), Comfo rt with C loseness (as measured by the Adult Attachment subscale Close ),
34 Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age? Is there a significant relationship between student Motivation developmental level (as measured by the Supervisee Levels Questionnaire Revised) and Peer Relational Health, Mentor Relational Health, and Community Relational Health (as measured by the Relat ional Health Indices), Comfort with C loseness (as measured by the Adult Attachment subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age? Is there a significant relationship between student Dependency Autonomy developmental level (as measured by the Supervisee Levels Questionnaire Revised) and Peer Relational Health, Mentor Relational Health, and Community Relati onal Health (as measured by the Relational Health Indices), Comfort with C loseness (as measured by the Adult Attachment subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus speci fic factors training emphasis), race, gender, or age? Overview of Study Chapter 1 presents an introduction to the present study based on the theoretical framework. Chapter 2 includes a review of relevant literature. The research methodology is described in Chapter 3. Chapter 4 describes the results of the study. Chapter 5 includes a summary, discussion of the results and implications, as well as recommendations for future research.
35 CHAPTER 2 REVIEW OF RELATED LI TERATURE This chapter takes an historical l ook at models of counselor development and related research. Relational Cultural Theory is reviewed at greater depth, with particular attention to its relevance for this study. The additional relational variable, comfort with closeness, is discussed in the context of the counseling relationship as well as the attachment theory from which it was derived. Chapter 2 concludes with a discussion of contextual issues in counselor education, including the possible impact of accreditation and standardization on the lack of relational emphasis in counselor preparation and the implications of this for novice counselor development. An Historical Look at Counselor Development Models Models and theories of counselor development attempt to explain the process by which no vices learn skills and gain the intra and interpersonal experiences necessary to become effective and eventually master practitioners. Counselor development has been an important area of inquiry since Hogan (1964) first proposed his influential stage mode l of how counselors change and grow over time. Over the decades, theorists have and theories, and added their own observations and research conclusions to form what appears to be general consensus in the literature about novice development. Broadly, the major theories of counselor development (Hogan, 1964; Stoltenberg, 1981; Loganbill et al, 1982; Skovholt & Ronnestad, 1992 a ) have depicted novices as limited relational beings who struggle through a long period of dependency and external focus during training, only arriving at complex thinking, strong clinical skills, self awareness, and solid therapeutic relational capacities later in the career cycle. These assertions
36 have be en substantiated by a handful of qualitative studies reviewed in this chapter; quantitative studies assessing novice development particularly relational development have been difficult to execute and have yielded mixed findings With growing evidence t o support the role of the therapeutic relationship in successful outcomes (Wampold, 2001), some (Buser, 2008; Hansen, 2010; Whiston & Coker, 2000) have called for further investigation into the relational development of novices, more research on training p ractices that enhance interpersonal skills, and renewed attention to the environments that prepare new counselors for the counseling profession. To date, most theories of counselor development (Hogan, 1964; Loganbill, Hardy & Delworth, 1982; Stoltenberg, 1981; Stoltenberg et al., 1998) have been founded upon the epistemological assumptions of western human developmental theories and cognitive developmental theory. The early western human developmental theorists (Chickering, 1969; Erickson, 1963) believed t hat people move through a predictable sequence of stages, with a general trend from dependence on parent style relationships to independent, individuated, and autonomous functioning the world (Comstock & Qin, 2005). During the same era that these stage wis e, individuation based theories of human development were being advanced, psychologists and counselors began to construct the first comprehensive theories of counselor development, fueled in part by the need to understand and inform the practices of counse lor education and supervision. These theories of counselor development shared similar assumptions with the prevailing models of human development (Comstock & Qin, 2005), namely that development occurs in a linear stage wise manner, with optimal development characterized by independence from perceived experts such as professors and
37 supervisors, professional and personal integration within the therapist, and eventual mastery that can then be passed down to supervisees at lower developmental levels. Hogan Hog an (1964) was one of the first theorists to propose distinct stages of counselor development, upon which subsequent theorists have built (Worthington, 1987). Hogan (1964) proposed four levels of development, the first in which the novice counselor is neuro tic, unaware, dependent, insecure, and highly motivated to learn counseling; the second in which the novice struggles with a dependency autonomy conflict alternating between over confidence and ambivalence; the third in which the counselor displays conditi onal dependency on the supervisor, marked by increased confidence, self awareness, and insight; and the fourth and final stage of the master counselor or psychologist, characterized by personal autonomy, insight, and self assurance. Hogan (1964) believed t hat ideal supervision/preparation environments provide the impetus for had themselves progressed through the developmental stages and are capable of optimally assess ing an d guid ing their supervisees through the different levels of change and growth. Stoltenberg popular counselor development models (Tryon, 1996), the Counselor Complexity Model (CCM) which was later expanded and renamed the Integrated Developmental Model (IDM ; Stoltenberg & Delworth, 1987) to incorporate theoretical contributions from Loganbill et al. (1982). In the CCM, Stoltenberg (1981) also theorized four levels of development. L evel 1 novices or supervisees are highly dependent on supervisors and
38 display cognitive rigidity marked by dualistic thinking. Supervisees at Level 2 struggle with dependency autonomy conflicts with their supervisors. At Level 3, supervisees move into a mo re egalitarian relationship with supervisors until finally, at Level 4, supervisees become master therapists and only receive supervision when they seek it out. Similar to Hogan (1964), Stoltenberg conveyed novices as insecure, anxious, and dependent on au thority for direction. According to the theory, developmental progression means the emergence of a personal style and authentic presence that includes risk taking and greater self awareness and insight. According to Stoltenberg, the most developed or maste r therapists possess an integrated personal and professional identity, autonomy, and a deep understanding of their strengths and as cognitively rigid and relationally de pendent, and emphasizes the role of supervisors in facilitating movement through the stages. Both of these theorists assumed that the supervisor adequately attend s to the relational needs of novice counselors and the remaining limitations are a normative d evelopmental stage. Absent from both of these theories, however, is consideration of the broader relational environment in which novices are trained, the possibility that supervisors may not be developmentally advanced, or that there may be qualities withi n the individual counselor that promote or limit the capacity for optimal growth and development. Loganbill, Hardy & De l worth Loganbill et al. (1982) outlined a detailed theory of counselor development on which they based their theory of supervision. The a uthors acknowledged the influence of three prominent western developmental theorists in their work, that of Erickson (1968), Chickering (1969), and Mahler (1979). Loganbill et al. (1982) saw similarities
39 between the counselor development process and Mahle separation individuation process in children and adolescents (Stoltenberg et al., 1998). includes a linear progression through distinct developme ntal stages. In Stage One, Stagnation, the novice counselor is characterized by navet, unawareness of his/her personal issues and reactions to clients, rigidly categorical thinking, narrow minded problem solving, dependency on a supervisor, and either lo w self concept or artificially high self and the counselor is not working through the issues presented by this stage. In Stage Two, Confusion, the ng to instability, disorganization, a desperate search for rebalance, ambivalence about the therapy process, and erratic emotional shifts. In this stage the novice still displays black and white thinking, but is gradually learning that the supervisor does not have magic answers. Only in the Third Stage, Integration, does the novice counselor come to accept the complexity of professional and personal issues and begin to tolerate the ambiguity and ongoing nature of challenge in the counseling process. In thi s stage, the counselor displays a mature worldview with signs of integrating the lessons from the previous two stages. Views of self and supervisor are more realistic, with acceptance of the strengths and limitations of both parties. In the Integration sta ge, the counselor is thought to know him or herself and be committed to a process of ongoing introspection and self reflection. According to this model, counselors need to resolve eight critical issues throughout the three stages before mastering counselin g, including competency, emotional awareness, autonomy, theoretical identity, respect for differences, direction
40 and purpose, personal motivation, and ethics. The authors added that movement through the stages occurs at different rates for different people Stoltenberg, McNeill and Delworth Stoltenberg et al. (1992) amended the CCM to include theoretical tenants of Loganbill et al. (1982) and renamed it the Integrated Developmental Model (IDM). A key addition of the IDM was the recognition that counselors m ay display different levels of development across different domains or with different types of clients. Built on major assumptions of established developmental theories including cognitive developmental theory, western theories of human development, and pr ior models of counselor development, the IDM proposes more complex dimensions of counselor development than previous models. In the IDM, Stoltenberg et al. (1992) theorized four levels of development (Levels 1 through 3 and Level 3 Integrated) and three m ajor developmental structures across which counselors progress over time: Self Other Awareness, Motivation, and Dependency Autonomy. According to this model, novice counselors gradually gain competence and confidence in nine areas of counseling activities (e.g. intervention, skill competence, assessment techniques, client conceptualization, multicultural counseling, interpersonal assessment, theoretical orientation, ethics, and treatment planning, ) as they progress through the four developmental levels ( Ru ssel Chapin & Chapin, 2012; Tryon, 1996) and experience quantitative and qualitative changes in the areas of awareness, motivation, and autonomy. In the IDM theory, Level 1 novice counselors are characterized by dependency on supervisors, high anxiety, and high motivation to learn counseling, but are low in the capacity to understand the complexity of the therapy process. In Level 2, counselors are
41 hypothesized to struggle with a dependency autonomy conflict with their supervisors and other perceived author ities. Their focus gradually shifts from their own fears and appraisal of their skills, abilities, and limitations, Level 2 counselors have fluctuations in motivation and feel confusion or emotional turmoil. At Level 3, counselors function fairly autonomously, display high confidence, are strongly motivated and possess a realistic understanding of their strengths and weaknesses. Also at this stage, counselors display self awareness and reflection and high empathy with clients. Although they may not have achieved the highest development al level in autonomy, motivation, and awareness in all eight domains of therapy, they have made significant gains in many of these areas. In the final level, Level 3 Integrated, the counselors have reached Level 3 development in many structures and shows integration across the structures (McNeill et al., 1992). The IDM represents a widely influential model of the process by which novices beco me expert practitioners. Changes occur in stages across several domains. Progress in major areas of competence may vary slightly depending on experiences, but general trends are proposed. Similar to previous theories, the novice counselor begins at a cogni tively rigid and relationally limited, dependent place and emerges a flexible, complex thinker with relational autonomy, strong clinical skills, self awareness, and a sense of empowerment. Research Supporting Counselor Development Models The developmental progressions portrayed above have received some research support. In a cross sectional design, Stoltenberg et al. (1987) surveyed doctoral students ( N=91 ) from six counseling and two clinical programs across the United States.
42 The purpose of the study wa this study, developmental level was operationalized by experience and level of education. Researchers reported differences in needs as a function of education and counseling experience, lending empirical support to the theoretical claim in the IDM that novices need different inputs from supervisors as they develop (Stoltenberg, McNeill, & Crethar, 1994). Skovholt and Ronnestad (1992 a ) conducted a seminal qualitative study that has lent credibility to the progression depicted in the counselor development models previously reviewed including those of Stoltenberg (1981), Loganbill et al (1982), and Stoltenberg et al. (1998). The researchers interviewed 100 counselors at different points in their career to discover stages and themes of counselor development. Using semi structured interviews in a cross sectional research design, the researchers elicited information on b oth professional and personal development over the counselor career cycle. Participants were divided into five groups of twenty according to their level of training and clinical experience. The sample was 96% White, 50% male and 50% female. All participan ts lived in Minnesota. The mean age was 42.4 years with a range from 24 to 71 years. While professionals in the sample had been trained at a wide variety of institutions, the students in the two counselor in preparation groups were all enrolled in one of two Minnesota universities. Researchers analyzed data from their 100 interviews for themes. Then, they conducted follow up interviews with 60 of the informants to validate their selection of themes. Ten years later, they revisited their findings, reconstr ucted meanings, and
43 presented their reflections again (Ronnestad & Skovholt, 2003). Researchers concluded that counselor development unfolded in stages (later renamed phases ) with early development characterized by prescriptive thinking, fear of negative evaluation, over attachment to theories and techniques, and over reliance on perceived experts. As counselors got further away from education programs, however, their personal development and use of self in the therapeutic relationship became paramount, an d they reported greater personal/professional integration and authenticity (Ronnestad & Skovholt, 2003). Participants in the later stages of their careers also consistently reported greater appreciation for the power of therapeutic relationship in their ou tcomes, and deeper ease in working with the personal and interpersonal dimensions of their profession. These results suggest a gradual movement from lower to higher personal development and interpersonal skill. This research seems to suggest that most coun selors develop and use their capacity for therapeutic relationships only several years after the conclusion of graduate education and post graduate supervision, a conclusion that, while influential, has not been extensively validated in subsequent literatu re. By focusing on changes across the lifespan, Skovholt and Ronnestad (1992 a ) provided a closer look at the longer term themes in counselor development, an area of education pr ograms (Goodyear, Wertheimer, Cypers, & Rosemond, 2003). Their study is often the cited as evidence of limitations in novice development and eventual broad generalizations While the overall sample size was robust, the subgroups of
44 counselors in level and doctoral level students) included 20 participants per subgroup still robust for a qualitative study but small for forming the foundation of a wide s weeping theory. All study participants hailed from two training programs in the same state, and thus may not have reflected the regional or national diversity of counseling students and/or counselor education programs. It is also unclear whether the counse lors in preparation were describing essential and universal aspects of the new counselor learning process, or if they were merely reflecting the orientation and the culture of their educational environment and the particular emphases of the training they r eceived. Further, because this study was cross sectional, there is no way to know if the participants in the earlier stages of development would actually progress to the advanced stages reported by seasoned professionals. Another concern with this study i s the lack of additional empirical validation for the anxiety reported by the counseling student informants. As Goodyear et al. (2003) noted, in two other studies (Chaplin & Ellis, 2002; Ellis, Krengel, & Beck, 2002) on trainee anxiety, only ten percent of counselors in preparation reported even moderate anxiety in situations that were considered highly anxiety provoking, such as showing a practice tape to a supervisor for the first time or working with their first client. Thus, Skovholt 992 a ) conclusion that anxiety is a major feature of the counselor education experience that causes students to gravitate towards the concrete and rigid aspects of therapy practice would benefit from further research. A finding from the study conducted by S kovholt and Ronnestad (1992 a ) that has received relatively little attention is that novice participants seemed to fundamentally possess either an open attitude towards the developmental journey, including an
45 openness towards learning, personal growth, and eagerness to work through developmental challenges, or a closed attitude of rigidity and restriction in response to the complexities of counseling, the latter of which led to stagnant development. This closed orientation to their own development may impact the degree and pacing at which they progress towards While Skovholt a a ) seminal study made a valuable contribution and offered support for the general developmental progression in preceding counselor development models, more research is needed to understand the possible role played by the culture of the preparation impact of the program was mentioned only cursorily in the research, as well as the forthcoming discussion of related lite rature demonstrates, some novice counselors seem to demonstrate developmental characteristics of therapists in the later stages of this model. Novice counselor development (as well as counselor development over the career span) may involve more complex var iables than those for which this model accounts. Literature Related to Counselor Development In addition to the comprehensive models of counselor development reviewed above, researchers have attempted to understand changes in specific domains of develop ment. Counselor cognitive complexity has been one of the few areas of counselor development to be researched quantitatively and has yielded preliminary findings about how students may or may not change over the course of a preparation
46 program and how cogni tive complexity may be related to client outcomes. Research on critical incidents and personal characteristics of novices associated with positive therapeutic outcomes have been largely qualitative in nature, but also offer insights into relevant aspects o f novice development. Counselor Cognitive Complexity While the first theories of counselor development (Hogan, 1964; Stoltenberg, 1981) contained some assumptions of cognitive developmental theory, the development of cognitive complexity has recently emerg ed as a specific lens through which to understand counselor development (Fong & Borders, 1997; Granello, 2002; Welfare & Borders, 2010). Cognitive developmental theories describe and explain how people acquire skills and competencies over time. According t o this approach, desirable counselor development reflects movement from less to more complex cognitive thought processes as novices accomplish coursework, gain clinical experience, and engage in productive supervision contexts. Early cognitive theorists ( Harvey, Hunt, & Schroeder, 1961; Kelly, 1955; Loevinger, 1976) proposed that novice counselors form mental constructs to organize what they have learned and observed. Counselor cognitive complexity represents a twofold capacity lity to differentiate between a variety of constructs and client characteristics in increasingly sophisticated ways, and second, the ability to integrate these characteristics, which may be discordant or paradoxical, into an organized, therapeutically rele vant whole (Crocket, 1965). Both the CCM (Stoltenberg, 1981) and IDM (Stoltenberg et al., 1987) present typologies for more and less cognitively complex counselors. These typologies describe complexity as a movement away from rigid dualistic thinking about clients and issues to complex cognitions that
47 reflect a relativistic worldview, increased empathy, and better therapeutic relationships (Skovolt & Ronnestad, 1992 a ). The relationship between counselor cognitive complexity and effectiveness has received e mpirical support (Welfare & Borders, 2010). Higher levels of cognitive complexity have been positively correlated with important aspects of practice such as objectivity in sessions (Borders, 1989), acceptance of and exploration with clients (Goldberg, 1974 ), complexity, verbal skills, and confidence (Fong, Borders, Ethington, & Pitts, 1997), tolerance of ambiguity (Holloway & Wampold, 1986), avoidance of stereotypes (Spengler & Strohmer, 1994), the formation of complex clinical hypotheses (Holloway & Wollea t, 1980) and more complex case conceptualizations (Ladany, Marotta, & Muse Burke, 2001). Based on these findings, Welfare and Borders (2010) able to formulate a complet e understanding of the client and use effective techniques in In an early study, Holloway and Wolleat (1980) used a paragraph completion method with trained rater scoring to illuminate the complexity of counselor thoughts about sources of information and their relationship to clinical hypotheses. In another study, Holloway and Wampold (1986) conducted a meta analysis of studies about cognitive complexity and the ability to perform counseling tasks. They found that high co nceptual complexity was positively associated with counseling ability and the capacity to work well in low structure (non rigid) environments. Fong, Borders, Ethington, and Pitts (1997) traced the longitudinal development of counselor trainees as they mov ed through their graduate preparation program. Using a
48 sentence completion test and trained raters, they measured ego development and student cognitions five times during a 2 year training period. They found no significant ognitive complexity over the 2 years. Contradictorily, Duys and Hedstrom (2000) tested the impact of a 1 semester counseling skills course and post levels of cognitive complexity towards their peers in the course. The researchers rep orted increases in complexity and concluded that participation in the counseling skills course may positively impact cognitive complexity. However, the study had a serious limitation in that it assessed cognitions about a peer, not about a client. As a res ult, the findings have limited generalizability to counselors in preparation who work with clients in an actual counseling context. conducted a cross sectional analysis of novice cogn itive development at three points in a training program. Perry (1970) proposed that cognitive development is the result of increasingly complex and relativistic epistemological assumptions. Granello (2002) administered the Learning Environment Preferences (LEP; Moore, 1989), a sentence level students at 13 college and university counseling programs. The researcher analyzed data with multivariate analysis of variance (MANOVA) to determine if time in program was related to cognitive development. Findings revealed that time in program and scores on the man service field, or grade point average. Students at the beginning of a program tended to reflect thought second stage, a multiplistic position that
49 recognizes that even the experts do not have the answers to the problems of the field. Towards the end of their program, participants reflected the view that while there may not be a right answer, the best answers can be supported by data, which is considered a slightly more advanced cognitive position. None of the par ticipants in this study demonstrated relativistic thought, similar to the findings in an earlier study (Simpson, In an attempt to determine the impact of years of experience on cou nselor cognitive complexity, Granello (2010) administered the LEP to a sample of 122 professional counselors from one mid western state. The researcher found a positive relationship between years of counseling experience and cognitive complexity as measure developmentally advanced relativistic thinking, a finding consistent with Skovholt and s (1992) finding that more experienced professionals reported greater cognitive complexity, a finding that has been consistent in the research on factors impacting this domai n. Welfare and Borders (2010) used the Counselor Cognitions Questionnaire (CCQ ; Welfare, 2006) and the Washington University Sentence Completion Test (WUSCT ; Loevinger & Wessler, 1970) to measure domain specific complexity, which potentially represented an advance in this area of inquiry. The researchers recruited 120 master level counseling students and post graduate or doctoral level counselor practitioners from CACREP institutions in 5 widely dispersed states. Participants were asked to write
50 in depth co nceptualizations of clients, evaluate client characteristics and group issues and behaviors into themes. Trained raters then scored participants for cognitive complexity based on the balance of positive and negative assessments as well as evidence for cons idering behavioral, emotional, contextual, cognitive, and spiritual dimensions of the case. Additionally, participants who included the therapeutic relationship in the case were awarded points for complexity. Further, a widely used and validated measure of ego development, the WUSCT, was administered to assess development. The researchers reported that the measure of domain specific complexity was better able to about clients than the general measures. They also reported that general measures of complexity did not predict domain specific complexity, and that counseling experience was a predictor for domain specific but not general complexity. The authors concluded that counseling experience does seem to needed to understand the particular training and clinical experiences that promote cognitive complexity. Based on the assumption t hat cognitive complexity is critical to effective practice, the above studies represent attempts to understand the training experiences that counselor cognitions, and a s ystem of ratings to evaluate changes in cognitive complexity over time. Higher levels of cognitive complexity have been linked with increased counseling performance, and in general reflect a movement from simple and rigid thought processes to more sophisti cated cognitions about client problems,
51 personalities, solutions, and the ongoing complex dimension of the therapeutic relationship itself (Welfare, 2007). Studies on cognitive complexity of counselors in training have provided support for the claim that sophisticated ways about their professional roles and responsibilities. However, these studies do not include relational constructs in their designs. It is yet unknown the extent to which cognitive complexity may or may not be related to the capacity to form and sustain effective therapeutic relationships, an area of counseling directly linked to successful outcomes. More research is needed to understand the variables that improve nov Critical Incidents Critical incidents, or intra and interpersonal incidents that involve distress, paradox, and challenge, are one way to examine the significant developmental expe riences of counselors in preparation (Skovholt & McCarthy, 1988). Furr and Carroll (2003) conducted a qualitative study in which they examined the impact of critical incidents on counselor development. Critical incidents were defined as both positive and n egative experiences considered by a graduate student as significant to his/her development as a counselor and may include experiences within and outside of a formal preparation a CACREP counseling program. The sample was 84% female, 81% White, and 14% African American, with the remaining 5% ethnic and racial minorities. Students were given time in class to fill out the 2 page package th at elicited the nature of the event and its signifi cance for their development.
52 Data were analyzed using a five part phenomenological method of coding meaning units and classifying these meaning units into categories. In their responses, students listed fieldwork experiences as the events within the progr am that impacted them the most. External interpersonal events were listed more frequently than events within the program as affecting counselor development, and experiential classroom learning seemed to have more impact on development than cognitive learni ng, skill acquisition and theory. Personal therapy was also frequently listed as significant. Students appreciated the opportunities for self disclosures in class and for learning about complex relationship dynamics in classes and personal therapy. While this in preparation, the results suggest that students realize that personal critical events have an impact on their development as therapist s Based on the results reported in this study, it does s eem that at least some of these students exhibited an interest in more advanced development and integration of personal and professional dimensions of counselor issues o n professional development, as well as research into faculty and student ideas on how to foster the most developmentally facilitative classroom experiences. McAuliffe (2002) conducted a qualitative study of 15 senior students in an undergraduate counselor education program, asking how trainees changed during the course of the program and what aspects of their program influenced these changes. Data were collected first via focus groups and second through in depth one on one interviews. The researcher employ ed a four part coding and analysis procedure of individual analysis, group coding and analysis, reanalysis, and triangulation. Three
53 major categories of change emerged. Students reported: 1.) Increased reflection (considering multiple perspectives before a ction, increased self awareness of defenses and projections, awareness of cultural and familial influences), 2.) Greater autonomy and interdependence (ability to act independently, the development of healthy boundaries, and needing less control of counseli ng sessions), and 3.) Valuing dialogue (valuing differences, being less judgmental, feeling more comfortable with ambiguity) While the sample of this study was small, these themes of change seem to indicate more advanced developmental processes than those constructed by Skovholt and Ronnestad (1992 b level counselors in training because the sample was in an undergraduate program; still, this study raises important questions about whether advanced development as characterized by openness, self awareness, self reflectivity, and enhanced relational connection is necessarily a function of age or time spent in the field. One limitation of this study was that the researchers did not ask the students what was unhelp ful in their program. Thus, students highlighted only those changes they perceived to be positive. Some students however, this study was not designed to capture that inf ormation and may therefore be biased toward positive experiences. Howard, Inman, and Altman (2006) conducted a qualitative study of critical incidents of nine students in a master level counseling program at a northeastern university. For fifteen weeks, p articipants made weekly journal entries of incidents they experienced during the semester that they judged as critical for their professional development as counselors. Researchers used a discovery oriented approach to
54 analyze the data, first for themes an d then for categories which included professional identity, personal reactions, competency, supervision experiences, and philosophy of counseling (Howard et al., 2006). The researchers reported that supervisees described negative experiences with supervis ors as often as they described positive ones, suggesting that supervision can be a source of dissatisfaction as well as a positive growth experience. Overall, their results suggested the importance of the supervisory relationship to counselor development. Additionally, the authors suggested that their findings indicated that novices possessed greater insight, ability to conceptualize clients, and capacity to reflect on therapeutic process than was suggested by Loganbill et al. (1982) and Stoltenberg (1981 ). These studies on critical incidents seem to support the role of the relational domain in counselor preparation. Indeed, after reviewing over fifty critical incidents, Cormier (1988) concluded that the relationship between a novice and supervisor or ment or appeared to be as important as the client counselor relationship to the quantitatively, as the developmental impact of significant relationships has historically been difficult t o research in the quantitative domain. Personal Characteristics Associated with Desirable Counselor Development While some researchers have found links between cognitive complexity and time in training or experience in the field (Duys & Hedstrom 2000; Gr anello, 2010; Welfare & Borders, 2010), research on the therapeutic alliance has provided mixed findings on the with clients. At the heart of the debate are questio ns about the extent to which the ability
55 or characteristics or the result of training and experience. Some have suggested (Mallinckrodt & Nelson, 1991; Whiston & Coker, 2000) that skills and knowledge may have less to do with quality therapeutic relationships than personal qualities of the therapists. This hypothesis has been pursued in the literature investigating factors that positively or negatively impact the therapeutic alliance. Because the therapy relationship draws on the person of the counselor, researchers have sought to understand the personal characteristics of the therapist that are associated with positive therapy outcomes (Ackerman & Hilsenroth, 200 3; Orlinksy, Grave, & Parks, 1994; Dunkle & Friedlander, 1996). For example, in a comprehensive review of studies on the personal qualities of therapists that negatively influence the alliance, Ackerman and Hilsenroth (2001) summarized these negative qual ities to include rigidity, personal distancing, being critical, uncertain, tense, and emotionally absent. Further, certain techniques such as over structuring sessions, making inappropriate self disclosures, using silence inappropriately, and over attribut ing a transference motivation to clients were negatively associated with the therapy alliance. In a meta analytic study on outcomes, Orlinsky, Grave and Parks (1994) identified therapist credibility, skill, empathy, understanding, engagement, and ability positively impacting treatment outcomes. From their research, Hovarth and Symonds (1991) concluded that client ratings of ess. Dunkle and Friedlander (1996) built on this finding and designed a study on the contribution of er hostility towards
56 self, and higher social support were all related to higher client ratings of the therapeutic alliance, but that counseling experience or age did not predict client ratings of the alliance (Dunkle & Friedlander, 1996). Their findings le nd support for the idea that personal characteristics of counselors may have a strong role in successful outcomes, perhaps even stronger than experience or age. Lastly, in literature on the characteristics associated with clinical expertise, reflection has comfort with complexity and tolerance for ambiguity (Jennings, Goh, Skovholt, Hanson, & Banjeree Stevens, 2003). Jennings et al. (2003) have argued that awareness of the complex and often ambiguous nature of therapy is a critical foundational piece for optimal professional development and performance. Notably, the characteristics and behaviors negatively associated with the therapeutic alliance seem to have much in common with the qualities and behaviors attributed to nov ice counselors, yet studies on the contribution of experience to the therapeutic alliance have not supported this conclusion (Dunkle & Friedlander, 1996; Hovarth & Symonds, 1991). Herman (1993) has argued that although professional standards typically emph asize therapist knowledge and experience in discussions of therapist competency, the research on the effective ingredients of therapy indicates that therapist personal characteristics particularly the ability to sustain a quality therapeutic bond play a si gnificant role in outcomes. The author expressed concern with the lack of emphasis on these personal characteristics in the literature and training standards, and suggested that findings about the importance of personal characteristics are threatening to a field that believes training and experience creates effective practitioners.
57 In sum, the personal characteristics of therapists and the role of the person of the therapist in promoting quality therapy relationships have received some attention and resear ch support in the literature. However, this area of inquiry continues to be a source of debate and there are many important but yet unanswered questions about how much the personal characteristics of the therapist promote successful outcomes, and what role if any, training programs play in facilitating personal qualities associated with positive therapy outcomes. Theoretical Framework This study used the theoretical framework of Relational Cultural Theory to guide research questions, analyses, and conclus ions (RCT ; Comstock et al. 2008). RCT enhancing ways with others. The present study was concerned with the contribution of relational factors to traditional conceptualizations of novice counselor development. As yet, a comprehensive perspective on relational issues in novice counselor development has not been offered. Instead, theo ries and research studies have been largely predicated on the assumption that development progresses within the individual counselor, who is thought to become more cognitively complex, relationally skilled, and professionally autonomous in the years follow ing mandatory training and supervision. Because RCT considers the role of the environment and significant others in development, it provides an alternative framework for investigating and interpreting relationship factors in novice counselor development.
58 R elational Cultural Theory in Counselor Preparation Relational Cultural Theory (RCT) proposes that growth and development happens in and through relationship in a process that is life long. Optimal development is characterized by the opportunity to have inc reasingly rich relational engagement throughout the lifespan, a view that has been supported by research that supportive relationships enhance wellbeing, development, and resilience (Spencer, 2000; Hartling & Ly, 2000). On the other hand, classic developme ntal theories propose that adult development involves separation, independence, and autonomy (Chickering, 1969; Erikson, 1963). Unlike RCT, traditional theories of development do not consider relationship and connection as a key factor in development Much of the extant counselor development theory and research is predicated on traditional developmental constructs. From the traditional perspective, the novice counselor is viewed as gradually separating from a supervisor much in the same way an adolescent se parates from his or her parents. The relational limitations of novices are seen as normative, as an inauthenticity, and difficulty attending to the therapeutic relationship are considered a normal part of the individuation process (Stoltenberg et al., 1998; Skovholt & Ronnestad, 2003). ess as well as other developmental challenges may be strongly related to the relational quality of their preparation programs because, according to RCT, relational quality drives development. From this framework, adult counselors in preparation who have di fficulty with forming quality relationship s do not necessarily represent the symptoms of fundamental, normative limitation. Rather, the relationships that novices have with important others in
59 their preparation programs may have a significant impact on the ir willingness to take the risks necessary to establish and maintain high quality therapeutic relationships. Hartling and Sparks (2002) acknowledge that the principles of RCT present a challenge to the dominant values of most institutional settings. These values reflect a broad cultural socialization towards competition, independence and self sufficiency (Jordan, 1999), and presume that optimal development is the result of separation, individuation and autonomy rather than the result of deep, genuine conn ection in relationship (Putnam, 2000). Hartling and Sparks (2002) suggest that institutions vary in the extent to which they promote cultures of connection or disconnection. Cultures that promote relational disconnection may contain overt dominance submiss ion dynamics, while others may be so stressful and demanding that there is little space for genuine connection. Still other settings may present themselves as connected but lack the authenticity and comfort with conflict necessary for real relational healt h. To date, it is unknown how counselors in preparation experience the relational quality of their preparation programs. It is also unknown how comfortable they feel taking the risks suggested to be so necessary for authentic engagement in relationship (B irrell & Freyd, 2006). Rather, counselor development models assume that the supervisor is advanced enough to provide for the relational needs of the novice (Hogan, 1964; Stoltenberg, 1981). From the perspective of RCT, relational quality with supervisors, peers, professors, mentors, and the larger community play s an important role in novice counselor development. Given the current evidence that the therapeutic relationship is critical to successful outcomes, and the consensus in the developmental literatur e that new counselors have
60 significant relational limitations, RCT offers a potential advance ment in understanding perceptions of the relational quality of their counselor prep aration programs have received little attention; instead, internal constructs such as cognitive complexity or self efficacy have been emphasized. Relational themes have appeared in developmental literature only in the context of the assumption that supervi sors attend to the relational aspects of novice development in skillful and appropriate ways; previously reviewed research indicated this may not be an accurate assumption. While students learn counseling within a network of relationships and relational d ynamics, the settings in which students are prepared to be counselors has not been researched as a potentially relevant variable in the process or outcome of novice counselor development. Several researchers have called for greater attention to how prepara tion programs prepare novices for the therapeutic relationship (Assay & Lambert, 2005; Hansen, 2010; Torres Rivera et al., 1998; Whiston & Coker, 2000). Others have suggested that the current emphasis on knowledge and skills does not get to the heart of th e therapeutic endeavor or prepare skillful practitioners (Herman, 1993; Rogers, 1973; Schaef, 1993). Still others point to research that the personal characteristics of therapists impact successful outcomes (Ackerman & Hilsenroth, 2003). All of these argum ents point to the importance of the relational dimension of counselor preparation and the practice of professional counseling, but the models of counselor development continue to emphasize domains of competence that are not specifically relational in natur e.
61 Attachment Theory: Implications for Counselor Development The process of forming and sustaining effective therapeutic and supervisory relationships has been conceptualized and researched through the lens of attachment theory ( Dunkle & Friedlander, 1996 ; Ligiro & Gelso, 2002; Meyer & Pilkonis, 2001 ). Attachment theory (Bowlby, 1988) explains how and why people attach to emotionally significant relationships, and how these attachments inform learning and development in a variety of adult contexts (Pistol e & Fitch, 2008). Bowlby (1969) proposed that the establish bonds with emotionally important others throughout the lifespan (White & Queener, 2003). Bowlby (1988) hypothesi zed that three interconnected structures attachment, care giving, and exploratory behavior affect personality development, patterns of relating, and ways of learning. notion of attachment styles to include work ing models, or fairly stable individual internal maps created in infancy and childhood for how people react to significant others. Pine (2004) described the essential patterns of these working models through three different attachment styles, which are sec ure, avoidant, and insecure ambivalent. According to attachment theory, people with secure attachment styles are comfortable with closeness to others and feel confident that others will be there during stressful times. Those with avoidant attachment are su to closeness in relationships. People with an insecure ambivalent attachment style them. Because these inter nal working models were originally formed to allow the infant to survive in response to the particular relational conditions of his or her early life, the theory posits that, even in adulthood, times of stress tend to activate internal working
62 models. The theory is developmental in nature, meaning that people can have reparative experiences and adjust their internal working models based on subsequent relationships with important others ( Miller, Notaro, & Zimmerman, 2002) Over the years, some have noticed that the counseling and supervisor roles seem to parallel the care creates a secure base from which he r child can gradually venture out into the larger world, counselors and supervisors may provide a secure base for clients and supervisees to explore themselves, experiment with relational intimacy, and try these new experiences outside of the office. Thus, the attachment styles and internal working models of supervisors, counselors, clients, and supervisees may impact the bonding process and outcome (i.e., how close, secure, and deep the relationship can become ) Attachment styles have been related to commu nication, self disclosure, empathy, the processing of information, and psychological development (Mikulincer, 1998; Mikulincer et al., 2005; Vivona, 2000) ; all domains that have relevance in counseling or supervision sessions. Attachment theory has been s upported by research across different cultures and types of emotionally significant relationships, including parental (Cassidy & Shaver, 1999), romantic (Mikulincer & Shaver, 2003), group (Rom & Mikulincer, 2003), and counseling (Dunkle & Friedlander, 1996 ; Ligiro & Gelso, 2002; Meyer, Pilkonis, Proletti, Heape, & Egan, 2001). Because relationship is a significant factor in both counseling and supervision, and attachment theory attempts to explain how pe ople engage in relationships, attachment theory has b een considered a relevant framework
63 for investigating process and outcomes between clients and counselors (Dunkle & Friedlander, 1996) as well as supervisors and supervisees (Bennett & Vitale Saks, 2006; Hill, 1992; Ladany, Friedlander, & Nelson, 2005; Nes wald McCalip, 2001; Pistole & Watkins, 1995; Riggs & Bretz, 2006; White & Queener, 2003). Dunkle and to form close adult relationships (attachments) were positively r elated to the therapeutic alliance. Hill and Corbett (1993) suggested that attachment theory could provide a framework with which to examine therapist competency. Dozier, Cue, and Barnett (1994) administered the Adult Attachment Interview (AAI; George, Ka plan, & Main, 1985) to eighteen clinical case managers who were working with clients with severe psychopathology. The researchers reported that the clinicians with secure attachment styles were more effective, and more likely to respond to the underlying n eeds of their clients and offer interventions that addressed core issues. Clinicians with insecure attachment were more likely to deal with the surface dependency and encourage this dependency from clients. The Role of Counselor Preparation Programs in Co unselor Development According to extant models and theories of counselor development, students are ill prepared for the relational dimension of counseling. Some qualitative research on critical incidents has suggested that students may be more attuned to i ssues of relationship, and open to reflection about complex dynamics, than suggested by these models, but these claims have not been substantiated by quantitative studies. Given the research on the common factors of psychotherapy that indicates relationshi p is a critical component of successful outcomes, some have raised questions about the extent to
64 which counselor preparation programs emphasize relational issues and dynamics in training experiences (Lambert & Ogles, 1997; Torres Rivera, Phan, Maddux, Wilb ur, & Garrett, 2001: Whiston & Coker, 2000). One place to begin to answer this question is by looking to CACREP, the national organization tasked with the standardization and credentialing of master and doctoral level counselor education programs. CACREP was established in the late 1980s after decades of preliminary regulatory efforts (Sweeney, 1992). Since then, CACREP has continued to revise its standards, emphasizing outcomes and accountability as well as the gate keeping function of CACREP programs in the profession (Ziomek Daigle & Christensen, 2010). Over the years, CACREP has both shaped and reflected the prevailing standards of the profession. The CACREP certification indicates that graduates are prepared with the knowledge and skills necessary for effective professional practice (CACREP, 2009). CACREP determines a large degree of the counselor preparation curricula, and thus represents the standard in terms of preparation and programmatic emphasis. Current CAC REP standards outline eight major areas of competency and make curricular recommendations for each of these areas, which include Professional Orientation and Ethical Practice, Social and Cultural Diversity, Human Growth and Development, Career, Helping Rel ationships, Group Work, Assessment, and Research and Program Evaluation. Within these areas of competency, the therapeutic relationship is mentioned in the context of admissions procedures, stating that decisions for admittance must consider otential to form effective interpersonal, culturally appropriate relationships with individuals and small groups. The therapeutic relationship is also briefly mentioned
65 under the heading of Helping Relationships, stating that students must be exposed to st (CACREP Standards, 2009, p. 12). CACREP does not require that counseling studen ts undergo personal therapy as part of their training. Neukrug and Williams (1993) outlined several benefits of personal therapy for counselors, including increased emotional health, self awareness, knowledge of blind spots, and greater respect for clients Wheeler (1991), however, summarized research findings that failed personal therapy. While these studies had serious methodological limitations, they also reflect the d ifficulty the field has had in establishing which experiences increase novice CACREP standards specify the amount of supervision and clinical hours required for graduation. However, according to Bradley and Forini (1999), asi de from requiring supervision to accompany practicum experiences, the nature of these clinical experiences is left up to the individual programs, resulting in little consensus about the relational competencies students must demonstrate. Further, they note several older studies (Haase, DiMattia, & Guttman, 1972; Roffers, Cooper, & Sultanoff, 1988) whose findings suggest that the novices do not necessarily retain skills learned in introductory classes (e.g. Microskills, attending behaviors), and that these sk ills may not carry over to actual clients. Further, the practice of using doctoral students as clinical supervisors raises questions about the assumption that supervisors are necessarily more advanced than their supervisees. Both the CCM (Stoltenberg, 1981 ) and IDM (Stoltenberg et al.,
66 1987) theories of development and supervision place great responsibility on supervisors to provide the kind of environmental supports and challenges that supposedly help counselors in preparation move through the developmenta l levels. As such, supervisors are expected to have access to the four developmental levels themselves in order to adequately assess and provide the optimal environment for supervisees (which may not be the case for many doctoral students). Further, most s tates require counselors to receive only two years of post graduate supervision, while the developmental constructs development. There is also considerable variability in the content and quality of supervision across clinical training sites and within any given community. Given the consensus in the literature that relationship is important to counselor development and client outcomes, the available knowledge on counselor d evelopment would be programs. Within the literature on counselor development, overt discussion of the role of preparation programs in facilitating development has been relati vely absent. Skovholt and Ronnestad (1992 b ) briefly mentioned that the academic atmosphere in graduate school may not be conducive to the personal authenticity and quality relational connections reported by the more experienced practitioners in their study In fact, Whiston and Coker (2000) have proposed restructuring clinical training to reflect current research findings on best practices. Research suggests that counselors who receive formal training are no more effective than paraprofessionals trained wi th basic counseling skills (Berman & Norton, 1985; Christensen & Jacobson, 1994; Durlak,
67 1979; Weisz, Weiss, Alicke, & Klotz, 1987). Furthermore, research studies have found that experienced therapists are not more effective than novices (Stein & Lambert, 1995). However, counselor skillfulness has been found to be positively associated with outcomes (Orlinksy, Gawe, & Parks, 1994), but according to Whiston & Coker (2000), current therapist training methods do not affect skillfulness. The authors argue that counseling training needs to move students beyond introductory counseling skills to master the more complex conceptualization, assessment and intervention skills that research suggests can lead to better outcomes. The authors stress the importance of the therapeutic relationship in outcome literature, and suggest that training should better prepare students to maximize the common factors of therapy shown to have most ab ility to form a deeply empathic working relationship. The authors advocate for and require them to think more critically about the contexts and factors that contribut e to a call for more relational approaches to training. Summary and Conclusions An abundance of research on the effective ingredients of psychotherapy supports the rol e of the therapeutic relationship in successful outcomes. Predominant theories of counselor development depict novices as relationally limited as compared to their more experienced counterparts. While the literature on counselor development contains a weal th of theories, to date most research on counselor development has been qualitative and has resulted in mixed findings. Quantitative studies have primarily investigated development through the lens of cognitive complexity, an important domain
68 but one that leaves unanswered questions regarding the relational dimension of novice development. The IDM offers one framework for looking at novice development that includes aspects of cognitive complexity but also includes an assessment of intra and inter personal domains. Currently, no quantitative studies have examined the role of relational closeness may relate to their developmental progress in a training program. Neither have studi es included an investigation of the contextual issues that may be preparatory environments. Instead, most theories of development have placed the developmental journe y within the individual, assuming that the supervisor adequately facilitates novice development, the program inputs are fixed, and that development is a fairly linear process that unfolds similarly for most individuals. While this line of inquiry has revea led important findings, it has also neglected the possible roles played by peers, mentors, and the community in a preparation program as well as how comfortable novices may be with close relationships. Given that therapeutic outcomes significantly rely on clients, it was thought necessary to address gaps in current knowledge by exploring the possible role played by relational variables in novice development along with traditional underst andings and assessments. The results of this study contribute to the literature on which counselor educators, supervisors, and researchers can base appropriate developmental interventions, decisions about curricula, and further research studies.
69 CHAPTER 3 METHODOLOGY Overview Given the need for this study, the following chapter outlines the methods used to investigate the contribution of relationship factors and counselor characteristics to student counselor developmental level. This study employed the a vailable and validated measures that most closely operationalized the constructs of interest in this research design. Sampling Procedures For the purposes of this study, the population was defined as all master level and doctoral level student counselors e nrolled in Community Counseling, Clinical Mental Health Counseling, or Mental Health Counseling programs at CACREP colleges and universities in the United States. (These three program track titles were virtually synonymous, and at the time of this study, w ere in the process of changing to the Clinical Mental Health Counseling name.) Eligible programs in each state were randomly selected from the total list of programs available on the CACREP website (www.cacrep.org), and department chairs and program coordi nators were contacted via email and asked to forward an email request for counseling student participation to their student listservs. Because one measure in this study, the SLQ R (McNeil et al., 1998) assumed the presence of a supervisory relationship, on ly students who were currently or previously under supervision for a practicum or internship were eligible to participate In an attempt to control for possible differences in curricula and academic culture between program tracks, only students who were en rolled in mental health related
70 tracks were invited to part icipate. Students enrolled in dual track s were permitted to participate as long as one of track s included a mental health program. Given the difficulty obtaining a random sample for an online surve y, convenience sampling was used. Upon completion of informed consent, all counseling students who met criteria for participation in the study were included until the target number of participants was reached. In order to strive for a geographically repre sentative sample of student counselors, students from randomly selected CACREP colleges and universities per state were sampled. Although the original plan was to contact one school per state, the response rate was insufficient. Thus, four schools per stat e were randomly selected and contacted, except for instances where states did not have an eligible program or had less than four CACREP programs. In the latter case, all eligible programs were contacted in the state. Two additional reminders were sent to e ach school. Limitations of the methodology and generalizability of the results are discussed in Chapter 5. This study was conducted in accordance with the guidelines and protocol of the Institutional Review Board (IRB) at the University of Florida. After o btaining IRB approval, a list of department chairs and program coordinators at CACREP programs in throughout United States were contacted by electronic mail with a letter that described the purpose of the study and invited eligible students to participate. Contacts at each school were asked to forward the request to their students. Three reminders emails were sent after the initial invitation, spaced one week apart. Data Collection Methods P otential p articipants who accessed the online survey were asked to read and agree to the informed consent statement before starting the survey instrument. Those
71 to begin the online survey. Those who did not agree to the terms of the info rmed consent were not able to complete the survey. The use of an online survey offers several benefits, including the elimination of paper and the costs of postage, mail out, and manual data entry (Dillman, 2000). This method also lessens the time required for implementation (Murray & Fisher, 2002). A limitation of online surveys is a potential lack of computer access among some populations, but it is unlikely that graduate students in this population of interest are unable to access computer and Internet r esources. The online survey in this study included an introduction, directions, an informed consent, the Relational Health Indices (Liang et al, 2002), the Close subscale of the Adult Attachment Scale (AAS ; Collins & Read, 1990), the Supervisee Levels Ques tionnaire Revised (McNeill et al., 1998), and a demographic questionnaire. Data were collected on participant demographics, quality of supervision, previous experiences with counseling, the ability to develop close relationships as measured by the Close s ubscale of the Adult Attachment Scale (AAS; Collins & Read, 1990), perceived relational health of a stude as measured by the Relational Health Indices (RHI; Liang et al., 2002), and student counselor developmental level as me asured by Supervisee Levels Questionnaire Revised (SLQ R ; McNeil, Stoltenberg, & Delworth, 1992). Data were stored on a secure online database at the internet based survey company Survey Monkey (www.surveymonkey.com). When data collection was complete, res ponses were transmitted to the researcher via secure technology purchased with the survey package.
72 Data were analyzed using descriptive statistics, correlations and multiple regression analyses. Design This study used a cross sectional, correlational, mu ltivariate survey design. Twelve variables (i.e., Comfort with C loseness, Total Relational Health, P eer Relational Health Mentor Relational Health Community R elational H ealth, Personal Therapy, Professional Counseling Experience Supervision Quality, Pr ogram Emphasis age, gender, and race ) were assessed for their relationships to the dependent variable (i.e., scores on the total scale of SLQ R and the three SLQ R subscales : Self Other Awareness, Motivation, and Dependency Autonomy ). Gender was converte d to a point bi serial format to permit correlational analyses. Additionally, post data collection, responses from the series of questions rating various supervision types were re calculated as a mean score for each respondent. This recalculation permitte d use in the multiple regression analyses. Bivariate analyses were conducted for all variables. Multivariate analyses were used to explore the contribution of every combination of predictor variables. All relevant two way interactions among the predictor and dependent variables were investigated. Relational Cultural Theory Attachment Theory, and other theories of counselor development provide the framework for interpretation, discussion of results, methodological considerations, and suggestions for futur e research. Instruments Relational Health Indices Liang et al. (2002) developed the Relational Health Indices (RHI) from key item self
73 report instrument comprised of statements abo peer, mentor, and community. The RHI can be scored either by measuring relational quality within the three relationship domains (peer, mentor, and community) or by measuring the three subscale dimension scores of eng agement, authenticity, and empowerment (Frey, Beesley, & Newman, 2005; Liang et al., 2002). The fourth dimension of relational quality, the capacity to deal with conflict, is not assessed by the RHI. The instrument consists of statements about attitudes an d feelings towards specific relationships (e.g., I feel as though I know myself better because of my mentor and This community provides me with emotional support ) Respondents rate the statements on a 5 point Likert scale that ranges from 0 = Never to 4 =Always (Appendix C). Higher scores on the RHI reflect higher reports of relational health. Given that differences in power dynamics might exist in these relationships, Liang et al. (2002) believed the instrument should assess each of these relationshi ps separately. A mentor relationship, for example, potentially involves power issues not present in a community or peer relationship. A community relationship might involve feelings of alienation and disconnection not present with a peer. A higher score on the This study considers peer as a classmate, mentor as a professor within the counseling program, and community rogram. The RHI describes a mentor as an adult who is not a parent, guardian, peer, or romantic listen, share her or his own experiences, and guide you through some area
74 through respect, affection and/or common interests, someone you can depend on for The RHI was originally applicable to mixed gender populations ( Bergman, 1991; Bergman & Surrey, 1994; Dooley & Fedele, 2004; Jordan, 2002; Mirkin & Geib, 1995 ). Internal consistency was established using the subscale composite Cronbach alpha coefficients for peer ( .85 ; n = 448), mentor ( .86 ; n = 303), and community ( .90 ; n = 445) (Liang et al., 2002). Convergent validity was determined by comparison with three previously validated instru ments, including the Mutual Psychological Development Questionnaire (MPDQ; Genero, Miller, Surrey, & Baldwin, 1992), the Quality of Relationships Questionnaire (QRI; Pierce, Sarason, Sarason, Solky Butzel, & Nagle, 1997), and the Friend Support subscale of the Multidimensional Scale of Perceived Social Support (MSPSS ; Zimet, Dahlem, Zimet, & Farley, 1988). The MPDQ assesses perceptions of mutuality in close dyads such as friendships and mentor relationships. The QRI assesses levels of depth, support, and c onflict in dyadic relationships. The Friend Support subscale of the MSPSS measures perceived social support from friends and was used to establish convergent validity on the RHI Peer scale. Convergent validity was established by the correlation of the MPDQ and the RHI Peer scale ( r =.69), the RHI P scale and the Support ( r = .61) and Depth of Relationship ( r = .64) scales of the QRI, and the RHI P and Friend Support subscale of the MSPSS ( r = .50). Convergent validity for the RHI M was established by the co rrelation of the RHI M to the MPDQ ( r = .68) and the RHI M and QRI Support scale ( r = .58) and the Depth
75 of Relationship scale ( r = .51). Unfortunately, there were no instruments designed to assess community relationships that could serve as a comparison m easure. Concurrent validity of the RHI was examined by comparison with The Rosenberg Self Esteem Scale (Rosenberg, 1965), the UCLA Loneliness Scale (Russell, Peplau, & Cutrona, 1980), the Center for Epidemiological Studies Depression Scale (CES D, Radloff 1977), and Perceived Stress Scale (Cohen, Kamarck, & Mermelstein, 1983). Loneliness, as measured by the UCLA Loneliness Scale, was negatively correlated with the RHI P, RHI M, and RHI C ( r = .35, .14, .47 respectively ). RHI peer and community relati onships were weakly related to self esteem, and depression and perceived stress were moderately related to relational health in community relationships (negative correlations of .39 and .32 respectively) (Liang et al., 2002). In their original validation study, Liang et al. (2002) reported that a confirmatory factor analysis supported the factors of engagement, authenticity, and empowerment as strongly related but conceptually distinct. However, in a reanalysis of the RHI, Frey, Beesley, and Newman (2005) raised concerns about the multidimensionality of these characteristics. Frey et al. (2005) administered the RHI to 247 women and 135 men during intake at a Midwestern college counseling center and investigated the structure and psychometric properties of the RHI. The results supported the use of this instrument with both genders, as well as the relative independence of the relationship domains. The results did not support independence of the growth fostering characteristics of empowerment, authenticity, an d engagement. Thus, this study used the scores for relational health across peer, mentor, and community domains but not the subscales for growth fostering characteristics. Frey et al. (2005) reported a one
76 dimensional structure for peer and mentor domains, but found a two component structure for the community domain that they suggested represented connection with or alienation from the community. The current study followed the recommendation to measure the overall quality of relationships for each type of r elationship and will produce a composite score of relational health and three subscale scores (peer, mentor, and community relational health). Adult Attachment Scale The Adult Attachment Scale (AAS; Collins & Read, 1990) is an 18 item measure of adult atta chment patterns. The measure consists of three subscales of six statements each: (a) Depend, or the extent to which a person trusts others in stressful times ; (b) Anxiety, or how much anxiety a person experiences or the fear of being abandoned in close rel ationships ; and (c) Close, or the extent to which a person is comfortable with intimacy in close relationships. For the purposes of this study, only the subscale Close was used for its previously demonstrated ability to predict client and therapist rating s of the quality of the therapeutic alliance ( Dunkle & Friedlander, 1996). Each item is rated with a 5 point Likert scale ranging from 1 ( not at all characteristic of me ) to 5 ( very characteristic of me). Subscale scores range from 6 to 30. A higher scor e on th e Close subscale reflects greater comfort with relational intimacy. This measure has been used with groups of professionals including nurses, as well as with novice therapists in st personal characteristics to the working alliance. In a sample of 101 college students over a 2 month period, Collins and Read (1990) reported test retest reliability rates of .68, .71, and .52 and internal consistency rates of .69, .75, and .72 for the Depend, Anxiety, and Close subscales respectively
77 The researchers claimed extensive construct validity for the AAS, reporting that participants with greater comfort with closeness reported higher self esteem, higher trust in others, and higher belief i n love relationships (Collins & Read, 1990). Reported reliability scores from other studies using the AAS have been moderate (Sperling Foelsch, & Grace, 1996). Sperling et al. (1 996) also reported convergent validity between this measure and their Attachm ent Style Inventory (Sperling & Berman, 1991). Supervisee Levels Questionnaire Revised The Supervisee Levels Questionnaire Revised (SLQ R ; McNeill et al., 1992) is a 30 item self report instrument that was designed to assess supervisee developmental le vel along a continuum of the three domains (self other awareness, autonomy dependency, and motivation) of the IDM supervision model. The original instrument, the Supervisee Levels Questionnaire (SLQ), was developed by McNeill, Stoltenberg, & Pierce (1985) and has received support in the literature (Borders, 1990 ; McNeill et al., 1985). In a pre post short term longitudinal study, Borders (1990) administered the SLQ to students before and after their practicum to collect self reported changes on the SLQ dom ains of self awareness, autonomy, and skills/theory acquisition. Borders (1990) analyzed data with analyses of covariance (ANCOVAs) and reported that the sample ( N = 44) made significant mean gains on the three dimensions of development measured by the SL Q as well as on the composite score. However, Borders (1990) also reported that some students in the sample showed decreases in pre post test scores, and that this finding warranted further research. McNeill et al. (1985) administered the SQL to a sample ( N = 91 ) from eight counseling and counseling psychology programs in the Eastern, Midwestern, Southern and Western regions of the United States. Using analyses of variance ( ANOVAs ) to
78 analyze the data, the authors reported expected differences between Le vel 1 and Level 2 supervisees on the Self awareness and Autonomy scales, and difference s between Level 2 and Level 3 supervisees on the Autonomy and Theory/Skills Acquisition scales. Further, they reported significant differences across all three scales f or trainees from Level 1 to Level 3. In 1992, McNeill et al. revised the SLQ to better reflect the three domains of the IDM (self other awareness, dependency autonomy, and motivation) and administered the revised questionnaire (SLQ R) to a sample ( N = 104 ) of doctoral students in the beginning, intermediate, and advanced stages of their training. Participants were drawn from eight counseling and clinical preparation programs across the United States. Researchers analyzed data with multivariate analyses of variance ( MANOVA s) ANOVA s and one tailed t tests, and found no significant differences between beginning and intermediate supervisees. Advanced supervisees scored significantly higher on all three scales than both beginning and intermediate supervisees. McNeill et al. (1992) reported Cronbach alpha reliability coefficients for the three subscales (Self Other Awareness, Motivation, and Dependency Autonomy Conflict) at .83, .74, and .64 respectively, with an overall reliability coefficient of .88. To asses s the construct validity, the researchers examined differences between subscales and total scores of the beginning, intermediate, and advanced participants. They reported Pearson correlation coefficients for subscale scores that indicated significant relat ionships for Self and Other Awareness and Dependency Autonomy, r = .53 p < .001; for Self and Other Awareness and Motivation, r = .58 p < .001; and Motivation and Dependency Autonomy, r = .43 p < .001. While the associations are significant the
79 authors stated that they did not think these intercorrelations measured the same attributes. To search for differences between the groups, the researchers used a MANOVA and reported that the three trainee groups differed on a linear combination of SLQ R subscale scores, F (6,198) = 2.45 p < .026. An ANOVA indicated that the total SLQ R scores of the groups differed significantly, F (2,102) = 7.37 p < .001. The researchers conducted one way planned contrasts to test their hypotheses and further delineate the direct ion of the obtained effects. They reported that at a .05 alpha level, significant differences were found in mean subscale and total SLQ R scores between beginning and advanced trainee groups and intermediate and advanced groups. However, they found no sign ificant differences between the beginning and intermediate groups. To measure effect size, the authors used the product moment correlation (Rosnow & Rosenthal, 1988) and reported correlations for significant effects that fell in the "medium" range as defin ed by Cohen (1977). Hypotheses The following null hypotheses were evaluated in this study. HO1 : T here is no relationship between Developmental L evel (as measured by the total score of SLQ R) and Relational H ealth (as measured by the t otal score of the RHI), Comfort with C loseness (as measured by the AAS subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. HO2 : T here is no relationship between student Self Other Awareness developmental level (as measured by the SLQ R) and Peer Relational Health, Mentor Relational Health, and Community Relational Health (as measured by the RHI), Comfort with C loseness (as measured by the AAS subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. HO3 : T here is no relationship b etween student Motivation developmental level (as measured by the SLQ R) and Peer Relational Health,
80 Mentor Relational Health, and Community Relational Health (as measured by the RHI), Comfort with C loseness (as measured by the AAS subscale Clo se ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. HO4 : T here is no relationship between student Dependency Autonomy de v elopmental level (as measured by the SLQ R) and Peer Relational Health, Mentor Relational Health, and Community Relational Health (as measured by the RHI), Comfort with C loseness (as measured by the AAS subscale Close ), Personal T herapy Professional Co unseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. Operational Definition of Variables Scores on the RHI operationally define d mutual engagement, authenticity, and emp owerment/zest. Additionally, RHI subscale scores operationally define d Peer Relational Health, Mentor Relational Health, and Community Relational Health Developmental level were operationally defined by a total score on the SLQ R which is comprised of fiv e components: microskills, process, difficult client behaviors, cultural competence, and awareness of values (Larson et al., 1992). SLQ R subscale scores defined the domain specific developmental levels (i.e., Self Other Awareness, Motivation, and Dependen cy Autonomy developmental levels). Scores on the AAS subscale Close operationally define d student counselors comfort with closeness in relationships. Data on amount of personal therapy, prior professional counseling experience, supervision quality, prepar ation program orientation (i.e. a common versus specific factors training emphasis), race, gender, and age were self reported at the end of the questionnaire. Data Analysis The present study proposed to determine if relational quality (as measured by the t otal scale and subscales of the RHI) and comfort with closeness (as measured by the
81 AAS subscale Close ) were significantly related to developmental level (as measured by the total scale and subscales of the SLQ R). This study also inves personal therapy, supervision quality, program orientation (i.e. common or specific factors), and demographic variables. The data obtained for this study were analyzed using the Statistical Package for the Social Sciences (SPSS). The data were collected online at www.SurveyMonkey.com and stored via an Internet database, which the researcher downloaded and transferred to SPSS upon completion of data collection. To test the prop osed hypotheses, descriptive statistics, correlation, and multiple regression analyses were used to explore the interrelationships between all independent variables and counselor student developmental level. Specifically, step wise multiple regression anal yses were conducted to determine if gender, race, age, amount of personal therapy, supervision quality, comfort with closeness, specific versus common factors program orientation, previous professional counseling experience, and the total and subscales of the RHI (relational quality with a peer, mentor, and community) predicted the total scale and subscales of the SLQ R. In the use of multiple regression, Cohen (1992) based desired sample size on effect size at power = .80 and a given level of probability. For a multiple regression study with eleven predictor variables, a medium effect size, power = .80, and probability = .05, the target sample size was estimated at 120 individuals. Prior to commencing the analysis, the data was assessed to ensure the assu mptions of regression had been met. The assumptions of multiple regression are as follows: ( 1) the independent variables (IV) are fixed and the same values on the IVs
82 would have to be used if the study were to be replicated; ( 2) the IV s are measured withou t error; ( 3) the relationship between the IVs and dependent variable (DV) is linear; ( 4) the mean of the residuals for each observation on the DV over replications is zero; ( 5) errors associated with any single observation on the DV are independent or not correlated with errors associated with any other observation on the DV; ( 6) the errors are not correlated with the IV; ( 7) the variance of the residuals across all values of the IV is consistent or homoscedasticity of the variance of the residuals; and ( 8) the errors are normally distributed. Assumptions 1, 2, and 4 are research design issues. Assumptions 3, 5, and 6 address linearity and assumptions 7 and 8 address homoscedasticity and normality. Normality was evaluated using skewness, kurtosis, and Kolmo gorov Smirnov statistics and linearity was assessed through inspection of bivariate scatterplots. Homoscedasticity can be assessed through interpreti examining residual scatterplots. Examination of the residual scatterplot s provided a test of each of these three assumptions (Tabachnick & Fidell, 1996). Little is gained by adding variables to a regression analysis measuring the same construct, and multi collinearity can cause real problems with the analysis. Stevens (2009) pointed out three reasons why multi collinearity can cause problems including : (a) multi collinearity limits the size of the R since the IV s are going after much the same variability in the DV (b) multi collinearity can cause difficulty because effects a re confounded when there is overlapping information, and (c) multi collinearity tends to increase the variances of the regression coefficients resulting in unstable prediction equations. The simplest method of diagnosing multi collinearity is to check for high inter correlations between the
83 predictor variables. A second method is to inspect the variance inflation factor ( VIF ), an indicator of the relationship between predictors (Stevens, 2009). Stevens also noted VIF values greater than 10 are generally cause for concern. The data for all regression analyses were assessed so that multi collinearity would not present a problem in the analysis. Methodological Limitations Certain limitations were anticipated in the present study. Several were expected fr om the use of self report surveys to collect developmental data. The self report scales for comfort with closeness, relational quality, and developmental level represented have been partially influenced by have been different from perceptions of others, such as peers, mentors, and professors Given sampling procedures in which participants were solicited through program listservs, the response rate will be unknown, making any generalizations from this study cautious and preliminary at best. No causal interpretations of results were made, and implications for practice based on t he results were phrased and read with tentativeness, describing possible alternative explanations for significant findings. Research that manipulates training in relational aspects of therapy is needed for causal conclusions. This correlational study may serve as a springboard for future causal research. It was expected that random selection of participants was not to occur due to the desirability of achieving an adequate sample size for an online survey administration within a reasonable period of time. As such, results were susceptible to a bias reflecting differences between counseling students willing to complete a research survey on this
84 area and those who were not. Such differences possibly affect ed responses to survey items. These and other poss ible limitations were considered likely in the present study.
85 CHAPTER 4 RESULTS Overview This chapter presents the results of the web based survey administered to master and doctoral counseling students enrolled in the CACREP Clinical Mental Health progra m track (also known as Community Counseling or Mental Health Counseling tracks) at graduate institutions across the United States. The study investigated the assessed developmental level and self reported rela tional health, comfort with closeness, previous experience with professional counseling and personal therapy, and other demographic characteristics. Th is chapter is arranged in the following order: (a) a summary responses to demographic an d supplemental questions; (b) data on the psychometric properties of the instruments used in the study; and ( c ) findings of the hypothesis testing. The following null hypotheses were evaluated in this study: HO1 : There is no relationship between student Developmental L evel (as measured by the total score of Supervisee Levels Questionnaire Revised ) and Relational H ealth (as measured by the total score of the Relational Health Indices ), Comfort with C loseness (as measured by the Adult Attachment subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. HO2 : There is no relationship between O ther Awareness developmental level (as measured by the Supervisee Levels Questionnaire Revised ) and Peer Relational Health, Mentor Relational Health, and Community Relational Health (as measured by the Relational Health Indices), Comfort with C loseness (as measured by the Adult Attachment subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. HO3 : There is no relationship be tween developmental level (as measured by the Supervisee Levels Questionnaire Revised) and Peer Relational Health, Mentor Relational Health, and Community
86 Relational Health (as measured by the Relational Health Indices), Comf ort with C loseness (as measured by the Adult Attachment subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. HO4 : There is no relationship between student Dependency Autonomy developmental level ( as measured by the Supervisee Levels Questionnaire Revised) and Peer Relational Health, Mentor Relational Health, and Community Relational Healt h (as measured by the Relational Health Indices), Comfort with C loseness (as measured by the Adult Attachment subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factor s training emphasis), race, gender, or age. Demographic Characteristics of Sample A total of 250 students began the survey for the study, indicating their consent to participate. Of those, 204 indicated that they were currently enrolled in an eligible grad uate training program. The next inclusion criteria that students were currently taking or had completed at least one practicum or internship reduced the sample size to 146 participants. Between the screening criteria and students who gave consent but chose not to participate during the first three screening questions, 142 participants began the first instrument in the survey. In total, nearly all data was missing for 108 responses, so these observations were omitted from the analysis As with most surveys, not all 142 participants who began the survey completed the questionnaire. The objective was to keep as much data as possible with the acknowledgement that data were missing for some participants. No attempt was made to input data for missing observations Most missing data entailed a large number of unanswered items (early terminated surveys). Therefore, while 142 people gave consent, met eligibility criteria, and began the survey, 119 questionnaires were deemed useable for the multiple regression analyse s For
87 reported demographic information and frequency tables, the total participants and percentages will vary based on the number of responses for each question. P articipants were recruited from CACREP masters and doctoral level counseling programs throu ghout the United States. It was determined through the CACREP website (www.cacrep.org) that 47 states had eligible programs (excluding Hawaii, Alaska, and Rhode Island). Up to five eligible counseling programs in every state (or ranging from one to four pr ograms where states did not have five eligible programs) were contacted via email through department chairs and program coordinators. These contact persons were requested to forward the invitation to participate, which included the Internet link to the inf ormed consent and questionnaire, to their graduate student listservs. Contact persons received three additional reminder emails over a period of five weeks. It is not known how many times the contact persons forwarded the invitations to their students. Re gional Distribution of Sample Given the anonymous nature of this survey and the evaluative questions about mentors and supervisors, participants were not asked to report their individual school, only the state in which they were attending school. Of the pr ograms in all 47 states contacted, participants from 19 states completed the questionnaire. The percentage of participants from each state is as follows: Colorado ( n = 8, 6.6%), Florida ( n = 19, 15.7%), Georgia ( n = 7, 5.8%), Illinois ( n = 3, 2.5%), Indian a ( n = 2, 1.7%), Louisiana ( n = 13, 10.7%), Minnesota ( n = 3, 2.5%), Mississippi ( n = 1, .8%), New Hampshire ( n = 1, .8%), North Carolina ( n = 19, 15.7%), Ohio ( n = 3, 2.5%), Oregon ( n = 7, 5.8%), Pennsylvania ( n = 3, 2.5%), South Carolina ( n = 6, 5.0%), T ennessee ( n = 3, 2.5%),
88 Texas ( n = 4, 3.3%), Washington ( n = 1, .8%), Wisconsin ( n = 12, 9.9%) and Wyoming ( n = 6, 5.0%). Gender and Racial Identification Table 4 1 provides a summary of the data collected for this question. The sample included females ( n = 103, 85.1%) and males ( n = 18, 14.9%). Participants varied in years of age categories as follows: 18 22 years ( n = 2, 1.7%), 23 25 years ( n = 32, 26.4%), 26 30 years ( n = 41, 33.9%), 31 35 years ( n = 19, 15.7%), 36 40 ( n = 2, 1.7%), 41 45 years ( n = 9, 7.4%), 46 50 years ( n = 9, 7.4%), over 50 years ( n = 6, 5.0%), and Not Indicated ( n = 1, .8%). When asked about their racial/ ethnic identification, students indicated they were White or Caucasian ( n = 95, 78.5%), Black or African American ( n = 11, 9.1%); A sian ( n = 2, 1.7%); Native American ( n = 1.8%); Latino/Hispanic ( n = 6, 5.0), Other ( n = 3, 2.5%), Multiracial ( n = 2, 1.7%) and Not Indicated ( n = 1, .8%). Years in a Preparation Program Participants were asked to estimate the number of years they had c ompleted in a graduate counselor education program. This information is summarized in Table 4 1. Due to a programming error, this question was added after the survey had been fielded. Thus, the number of respon ses for this question was 84, while the remain ing demographic items had approximately 120 observations per question. Participants indicated their years of education as follows: less than one year ( n = 7, 8.3%), one year ( n = 2, 2.4%); one and a half years ( n = 21, 25%); two years ( n = 5, 6.0%); two an d a half years ( n = 20, 23.8%); three years ( n = 6, 7.1%); three and a half years ( n = 6, 7.1%); four years ( n = 4, 4.8%); four and a half years ( n = 4, 4.8%); five years ( n = 2, 2.4%); five an d a half years ( n = 1, 1.2%); six years ( n = 2, 2.4%); seven ye ars or more ( n = 4 4.8%).
89 Program Emphasis Students were asked to rate their counseling program (including classes, curriculum, professors, and within department supervisors) on a continuum, where 1 represented the greatest program emphasis on the specif ic factors of therapy, 4 meant about equal emphasis on specific and common factors, and 7 represented the greatest emphasis on common factors. Table 4 1 illustrates the program emphasis reported by students. Only 3 respondents rated their program as the gr eatest emphasis on specific factors (2.4%), and just 2 students indicated that their program mostly focused on specific factors (1.6%). Another 11 students indicated that their program placed a little more emphasis on specific factors than common factors ( 8.7%). Equal emphasis on specific and common factors was selected by 32 students (25.4%), an additional 33 students rated their program as a little more emphasis on common factors (26.2%), and the highest number of students indicated that their program emp hasized mostly the common factors ( n = 36, 28.6%). An additional 9 students rated their program as placing the greatest emphasis on the common factors of therapy (7.1%). Personal Therapy Participants were asked to indicate whether they had voluntarily att ended personal counseling or therapy since the age of 18. Details of the se results can be found in Table 4 1. Of those who responded, 100 (82.0%) indicated that they had received some voluntarily personal counseling, while 22 (18.0%) reported no voluntary counseling. Students who indicated that they had rece ived counseling were then asked to estimate how many sessions they had attended since the age of 18. Participants selected 1 5 sessions ( n = 12, 12.0%), 6 10 sessions ( n = 18, 18%), 11 15 sessions ( n = 8, 8.0%), 16 20 sessions ( n = 12, 12.0%), 21 20 sessions ( n = 14, 14.0%), and over 30 sessio ns
90 ( n = 36, 36.0%). Using a five point Likert response scale of Not at all Helpful, A Little Helpful, Somewhat Helpful, Very Helpful, and Extremely Helpful, studen ts were asked to rate the helpfulness of the counseling they had received. The majority of students ( n = 73, 73.0%) rated their counseling as Very Helpful or Extremely Helpful. An additional 19 (19.0%) students indicated counseling had been Somewhat Helpf ul, 7 students ( 7.0 % ) reported A Little Helpful, and just one student (1.0%) selected Not at all Helpful. Professional Counseling Experience The survey asked students to indicate the number of years worked as a professional counselor prior to entering the ir current counseling program. Table 4 1 provides a summary of the data collected for this question The majority of respondents ( n = 87, 71.9%) had no prior professional counseling experience, 20.7 % ( n = 25) had between one and three years, four to six y ears ( n = 4, 3.3%), seven to ten years ( n = 2, 1.7%), and over ten years ( n = 3, 2.5 %). Supervision Quality Survey respondents were also asked to rate the quality of the most recent types of supervision they received during their current or most recent cl inical practicum or internship. Table 4 2 depicts students supervision most recently received. Quality of supervision was defined as the usefulness of the supervision to their clinical work, personal developm ent, and professional development The majority of respondents rated their various types of supervision as Good, Very Good, or Excellent as follows: 1) group supervision from a faculty member ( n = 96, 89.7%), group supervision from a doctoral student ( n = 38, 82.6%), group supervision from a clinical site supervisor ( n = 63, 80.8%), individual supervision from a faculty member ( n = 34, 75.6%), individual supervision from a
91 doctoral student ( n = 79, 88.8%), and individual supervision from a clinical site sup ervisor ( n = 89, 84.8 %). Percentages of students rating their supervision quality as Fair or Poor ranged from 10.8% to 24.4%. Instrumentation Three instruments were used to collect data for the study: the Relational Health Indices (RHI ; Liang et al. 200 2), the Close subscale of the Adult Attachment Scale (AAS ; Collins & Read, 1990), and the Supervisee Levels Questionnaire Revised (SLQ R ; McNeill, Stoltenberg, & Roman, 1992). The means and standard deviations for the total scales and sub scales are reporte d in Table 4 3 Additionally, t he reliability correlations of the study instruments are depicted in Table 4 4. Liang et al. (2002) reported internal consistency for the RHI as ranging from .69 to .90 for subscales and the composite. In this study, the RHI composite yielded a high Cronbach alpha of .9 5 Reliability coefficients for the subscales Mentor, Peer, and Community were .93, .9 2 and .92 respectively. Individual items for the RHI were not modified in this study, but respondents were asked to answ er the items in terms of a peer, mentor, and community within their counselor preparation program. The reported test and retest alpha for the AAS subscale Close was .69 and .68. The study alpha for the AAS subscale Close was .7 4 slightly higher than previ ously reported. The reliability coefficients for the SLQ R composite study alpha was .89, and subscales Self Other Awareness, Motivation, and Dependency Autonomy Conflict were .90, .8 1 and .41 respectively. Only the study alpha for the Dependency Autonom y subscale was not as internally consistent or reliable as it had been with other populations (reported .64 ).
92 Hypotheses Tests The dependent variable for this study was counselor development, as measured by the composite score and three subscale scores of the SLQ R: Self Other Awareness, Motivation, and Dependency Autonomy Conflict. A total of 12 independent variables were tested in the this study, with a maximum of 11 variables in a model: the composite score of the RHI, Peer Relational Health Mentor Relational Health, Community Relational Health Comfort with Closeness (from the AAS subscale Close ) amount of personal therapy, supervision quality, a common or specific factors orientation in preparation program, years of prior professional counseling e xperience age, race, and gender An additional independent variable, years in a preparation program, was originally proposed and approved. Due to a programming error, this variable was omitted from the initially fielded questionnaire. Although eventually added to the instrumen t, this question yielded only 84 responses before data collection concluded. Including this variable in the regression analyses limited the data that could be include d in the analyses from 119 to 84 Therefore, the decision was made to omit this variable from the regression analysis and explore other variables of interest. The variables previous professional counseling experience and age were significantly correlated to years of graduate education and explored as other possible indica tors of educational attainment. Correlations between independent and dependent variables are displayed in Table 4 5. Four null hypotheses were tested using the statistical procedure of step wise multiple regression Step wise multiple regression analyse s w ere chosen due to the exploratory nature of this study. An alpha level of p < .05 was established to interpret the results of the statistical analyses. The assumptions of multiple regressions were
93 assessed prior to beginning the analysis to address the hy potheses proposed for this study. Due to the nature of the descriptive variables used in the analysis, some of the variables were not normally distributed and were skewed in distribution. The subscales and total scale scores of the instruments used in thi s study were mostly normally distributed and were not skewed. The residual plots for the regression s were centered around a horizontal line indicating the data was acceptable for multiple regression s Most importantly, the VIF and Tolerance indicated multi col l inearity was not problematic for this analysis. The VIF was less than Stevens (1992) criteria for having to address collinearity (i.e., VIF > 10) and Tolerance was 1.0 or less. Both measures indicated collinearity was not a problem in this analysis. The results of the hypotheses tests are reported below. Hypothesis 1 Hypothesis 1 stated that t here is no relationship between Developmental L evel (as measured by the total score of SLQ R) and Relational H ealth (as measured by the tota l score of the RHI), Comfort with C loseness (as measured by the AAS subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age Tables 4 6 and 4 7 illustrate the regression model summary for the SLQ R total mean scores and the regression c oefficient model for the first hypothesis test A step wise m ultiple regression procedure was used to identify the best model. The total SLQ R score was determined by calculating the mean scores for individual responses on the SLQ R Results indicated Professional Counseling Experience and Comfort with Closeness were significant predictors ( R = .46, R 2 = .21, R 2 adj = 20 F (1,
94 116) = 11.2 9 p = <.0 01 ). The two step model account ed for 21.3% of the variance with Professional Counseling and Comfort with Closeness accounting for 13.6% and 7.7% of the variance, respectively, indicating that f actors other than these two variables accounted for the re maining 78.7% of the variance. This null hypothesis was rejected since at least two variables, Professional Counseling and Comfort with Closeness, were found to be significant to the model. Hypothesis 2 Hypothesis 2 stated that t here is no relationship be tween student Self Other Awareness developmental level (as measured by the SLQ R) and Peer Relational Health, Mentor Relational Health, and Community Relational Health (as measured by the RHI), Comfort with C loseness (as measured by the AAS sub scale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. Tables 4 8 and 4 9 illustrate the results of the regression analyses for the second hypothesis. A stepwise multiple regression procedure was used to identify the best model. Results of the three step model indicated that P rofessional C ounseling Experience, Comfort with Closeness, and Community Relational Health were signi ficant predictors of Self Other Awareness ( R = .47 R 2 = .22 R 2 adj = .20 F (1, 115 ) = 6.03 p = <.001 ). T he regression model accounted for 22.2 % of the variance indicating that 77.8 % of the variance was due to factors other than these three variables Th e three significant variables in this model, P rofessional C ounseling Experience, Comfort with Closeness, and Community Relational Health, accounted for 10.6%, 7.5%, and 4.1% of the variance of scores on Self Other Awareness, respectively. T his null hypothe sis was
95 rejected for P rofessional C ounseling Experience, Comfort with Closeness, and Community Relational Health. Hypothesis 3 Hypothesis 3 stated that t here is no relationship between student Motivation developmental level (as measured by th e SLQ R) and Peer Relational Health, Mentor Relational Health, and Community Relational Health (as measured by the RHI), Comfort with C loseness (as measured by the AAS subscale Close ), Personal Therapy Professional Counseling Experience, Supervision Q uali ty, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. Tables 4 10 and 4 11 illustrate the results of the regression analyses for hypothesis 3 A stepwise multiple regression procedure was used to identify t he best model. Results of the three step model indicated that Professional Counseling Ex perience, Community R elational H ealth, and Supervision Q uality were predictive of Motivation ( R = 39 R 2 = 15 R 2 adj = 13 F (1, 115) = 4.48 p = <.0 01 ). The three st ep model account ed for 15.3 % of the variance indicating that 84.7 % of the variance was due to factors other than these three variables Professional Counseling Ex perience, Community R elational H ealth, and Supervision Q uality accounted for 8.3%, 3.7%, and 3.3% of the variance on Motivation level, respectively. As a result, the null hypothesis was rejected for these three variables Hypothesis 4 Hypothesis 4 stated that t here is no relationship between student Dependency Autonomy developmental l evel (as measured by the SLQ R) and Peer Relational Health, Mentor Relational Health and Community Relational Health (as measured by the RHI), Comfort with C loseness (as measured by the AAS subscale
96 Close ), Personal Therapy Professional Counseling Experi ence, Supervision Q uality, Program Emphasis (i.e. a common versus specific factors training emphasis), race, gender, or age. Tables 4 12 and 4 13 illustrate the results of the regression analysis for hypothesis 4 A step wise multiple regression procedure was used to identify the best model The findings indicated that Professional Counseling E xperience, Comfort with Closeness and Mentor Relational Health were predict ors of Dependency Autonomy ( R = 46 R 2 = 21 R 2 adj = 19 F (1, 115) = 4.95 p =<.0 01 ). T he three step model accounted for 20.7 % of the variance with Professional Counseling E xperience, Comfort with Closeness and Mentor Relational Health accounting for 13.7%, 3.6%, and 3.4% of the variance, respectively. Thus, the null hypothesis was rejecte d for Professional C ounseling E x perience, Comfort with Closeness, and Mentor Relational Health. Summary Despite the limitations of the study, there was evidence of a statistical ly significant relationship between the independent variables and the outcome variables for the CACREP Clinical Mental Health counseling students participating in this study. S ignificant relationship s were found in each of the four hypotheses tested, though the strength of the relationships varied. Specifically, Professional Counse ling Experience, Comfort with Closeness, Community Relational Health, Mentor Relational Health, and self reported developmental level Overall the results of this study provide prelimin ary evidence for the potential use of relational health and comfort with relational closeness as a framework for understanding the role of relationship factors in student counselor development
97 Chapter 5 will provide detailed analysis of each hypothesis as well as theoretical and practical implications of findings. Chapter 5 will also include explication of study limitations and recommendations for future research
98 Table 4 1. Frequencies for demographic variables Variable Frequency Percent Race White or Caucasian Black or African American Asian Native American Latino/Hispanic Other Multiracial Not available 95 11 2 1 6 3 2 1 78.5 9.1 1.7 .8 5.0 2.5 1.7 .8 Age 18 22 years 23 25 years 26 30 years 31 35 years 36 40 years 41 45 years 46 50 years Over 50 years Not available 2 32 41 19 2 9 9 6 1 1.7 26.4 33.9 15.7 1.7 7.4 7.4 5.0 .8 Years of education Less than 1 year 1 year 1 yea rs 2 years 2 years 3 years 3 years 4 years 4 years 5 years 5 years 6 years 6 years Seven years or more 7 2 21 5 20 6 6 4 4 2 1 2 2 2 8.3 2.4 25.0 6.0 23.8 7.1 7.1 4.8 4.8 2.4 1.2 2.4 2.4 2.4 Professional c ounseling e xperience 0 years 1 3 years 4 6 years 7 10 years Over 10 years 87 25 4 2 3 71.9 20.7 3.3 1.7 2.5
99 Table 4 1. Continued Variable Frequency Percent Personal t herapy Did not recei ve 1 5 sessions 6 10 sessions 11 15 sessions 16 20 sessions 21 30 sessions Over 30 sessions 22 12 18 8 12 14 36 5.5 9.8 14.8 6.6 9.8 11.5 29.5 Helpfulness of counseling received Not at all helpful A littl e helpful Somewhat helpful Very helpful Extremely helpful 1 7 19 33 40 1.0 7.0 19.0 33.0 40.0 Program e mphasis Greatest on specific factors Mostly on specific factors A little more on specific factors Equal emp hasis on both A little more on common factors Mostly on common factors Greatest on common factors 3 1 11 32 33 36 9 2.4 1.6 8.7 25.4 26.2 28.6 7.1 Differences in sample size indicate exclusion of observations with missing variables.
100 Table 4 2. Quality of supervision most recently r eceived Poor Fair Good Very Good Excellent N/A N % N % N % N % N % N % Group Supervision from Faculty Member 5 4.1 6 4.9 21 17.2 34 27.9 41 33.6 15 12.3 Group Supervision from Doctoral Student 4 3.3 4 3.3 12 9.8 13 10.7 13 10.7 76 62.3 Group Supervision from Site Supervisor 2 1.6 13 10.7 11 9.0 26 21.3 26 21.3 44 36.1 Individual Supervision from Doctoral Student 5 4.1 5 4.1 11 9.0 24 19.7 44 36.1 33 27.0 Individual Supervision from Faculty Member 7 5.7 4 3.3 7 5.7 13 10.7 14 11.5 77 63.1 Individual Supervision from Site Supervisor 4 3.3 12 9.8 14 11.5 23 18.9 52 42.6 17 13.9 Table 4 3 Instrument total scores and subscale scores Mean Standard Deviation Range Overall Relational Health (RHI) 3. 692 .553 1.70 4.70 Peer Relational Health (RHI) 3.895 .688 1.18 5.00 Mentor Relational Health (RHI) 3.867 .763 1.82 5.00 Community Relational Health (RHI) 3.364 .650 1.64 4.71 Comfort with Closeness ( Close subscale, AAS) 3.676 .679 1.17 5.00 Overall D evelopmental Level (SLQ R) 5.082 .625 3.06 6.45 Self Other Awareness (SLQ R) 5.236 .823 2.50 7.00 Motivation (SLQ R) 5.241 .762 2.38 6.88 Dependency Autonomy (SLQ R) 4.565 .593 3.09 6.00
101 Table 4 4 Cr onbach al pha r eliability coefficients to study s ca les No of items Study Alpha Reported Alpha Relational Health Indices 37 .949 .69 .90 Adult Attachment subscale Close 6 .738 .68 .69 Supervisee Levels Questionnaire Revised 30 .892 .88
102 Table 4 5 Pearson product moment correlations between independen t and dependent variables ( one tailed tests) n = 119 Variables 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 1. SLQR Tot 1.00 2. SLQR SO .930** 1.00 3. SLQR MO .903** .779** 1.00 4. SLQR DA .719** .5 25** .520 1.00 5. RHI Tot .107 .127 .150 .147 1.00 6. RHI M .047 .060 .117 .203* .805** 1.00 7. RHI P .061 .033 .083 .021 .739** .371** 1.00 8. RHI C .141 .195* .150 .105 .837** .494** .481** 1.00 9. Close .302** .296* .209* .216* .212* .152 .214* .151 1.00 10. SFCM .008 .030 .019 .037 .029 .013 .080 .009 .058 1.00 11. Pers Coun .160* .139 .178* .084 .112 .025 .076 .158* .017 .197* 1.00 12. Prof Coun .369** .326** .288 .370** .171* .146 .119 .142 .071 .065 .244* 1.00 13. SupQual .069 .019 .149 .136 .285* .296* .156* .220* .065 .095 .023 .046 1.00 14. Gend .016 .002 .034 .026 .064 .035 .129 .006 .109 .144 .036 .036 .012. 1.00 15. Race .013 .030 .081 .101 .058 .032 .012 .159* .149 .200* .072. .177* .241* .022 1.00 16. Age .208* .165* .220* .210* .166* .145 .207* .053 .075 .122 .362** .272* .029 .201* .013 1.00 p<.05 (one tailed), ** p<.001 (one tailed).
103 T able 4 6. Model sum mary for hypothesis test 1 R R 2 R 2 adj 2 F chg df P Model 461 213 199 077 11.285 116 <.001 Table 4 7. Coefficients for h ypothesis test 1 Model B T P Bivariate r Partial r Professional counseling experience .250 349 4.231 <.001 .369 .366 Comfort with closeness .242 .277 3.359 .00 1 .302 .298 Table 4 8. Model summary for hypothesis test 2 R R 2 R 2 adj 2 F chg df P Model 471 222 202 041 6.032 115 <.001 Table 4 9. Coefficients for h ypothesis test 2 Model B T P Bivariate r Partial r Professional counseling experience 298 .339 4.058 <.001 .326 .354 Comfort with closeness .257 .241 2.880 .005 .296 .259 Community relational health .235 .207 2.456 .016 .195 .223 Table 4 10. Model summary for hypothesis test 3 R R 2 R 2 adj 2 F chg df P Model 392 153 131 033 4.47 7 115 <.001
104 Table 4 11 Co efficients for h ypothesis test 3 Model B T P Bivariate r Partial r Professional counseling experience .272 .313 3.614 <.001 .288 .319 Community relational health .264 .236 2.655 .009 .150 .240 Supervision qualit y .167 .186 2.116 .037 .149 .194 Table 4 12. Model summary for hypothesis test 4 R R 2 R 2 adj 2 F chg df P Model 455 207 186 034 4.94 115 .028 Table 4 13. Coefficients for h ypothesis test 4 Model B T P Bivariate r Partial r Profession al counseling experience .211 .326 3.871 <.001 .370 .340 Comfort with closeness .174 .222 2.629 .010 .216 .238 Mentor relational health .132 .189 2.224 .028 .185 .203
105 CHAPTER 5 DISCUSSION C ounselor development has been the focus of numerous theor ies and studies (Hogan, 1964; Loganbill et al., 1982; Skovholt & Ronnestad, 1992a; Stoltenberg, 1981; Stoltenberg & McNeill, 2010). To date, counselor development has been typically conceptualized as a series of normative developmental stages, with the st udent or novice stages featuring counselor behaviors commonly associated with poor therapeutic alliances and less effective outcomes (Ackerman & Hilsenroth, 2001 ; Herman, 1993 ). At the same time, concerns have bee n raised that counselor education programs may not attend optimally to student development by neglecting interpersonal issues germane to therapeutic relationships ( Hansen 2010; Whiston & Coker, 2000 ). Relational Cultural Theory (RCT) and attachment theory offer novel frameworks from which to inte rpret traditional theories and findings related to counselor development. This study explored programs and their comfort with relational closeness as possible influen ces in self reported developmental level. Other variables including supervision quality, amount of personal therapy, prior professional counseling experience, a common or specific factors program orientation, and demographics were also explored for their c orrelation with self assessed student counselor development. This chapter contains a discussion of the study findings and implications to theory and practice, and concludes with limitations and recommendations for further research. Overview of Study and D iscussion of Findings This study used a convenience sample to collect data from counseling graduate student volunteers. A total of 144 eligible programs in 47 states were approached with a
106 request to forward the invitation to their counseling students. Re spondents were mostly Caucasian and female. Comparison demographics for the population of graduate counseling students in Clinical Mental Health and related program tracks in the United States were not available According to the CACREP Director of Researc h and Information Services, CACREP does not collect information related to student age nor demographic data either by specific program or as a whole (T. Kimbel, personal communication, April 13, 2012). CACREP does collect general demographic information on CACREP students by institution, but analysis of this dataset was beyond the scope and resources of this study. CACREP institutions are comprised of students in private, public, or faith based colleges and universities with traditional, online, or hybrid p rograms seeking MA, MEd, or EdS degrees in addiction counseling, career counseling, clinical mental health counseling, marriage, couple, and family counseling, school counseling, student affairs and college counseling, or doctoral degrees in Counselor Educ ation and Supervision ( http://www.cacrep.org ). There are a total of 599 CACREP accredited institutions, with 240 of those programs listing degrees in Clinical Mental Health or related counseling program tracks Profe ssional Counseling Experience reported counselor development and domain specific areas of (1) awareness and understanding of self and others; (2) motivation to learn counseling; and (3) the ability to become autonomous. The finding that students with more experience rated themselves more highly on a developmental measure was relatively expected and consistent with findings in the li terature that suggest small to moderate gains in basic counseling skills (Baker & Daniels, 1989;
107 Baker, Daniels, & Greenley,1990), cognitive complexity (Welfare & Borders, 2010 ; Duys & Hedstrom, 2000; Little, Packman, Smaby, & Maddux, 2005), self appraisal of counseling skills (Fong et al., 1997), self efficacy (Adams, 2010) and self reported level of counselor development (Ronnestad & Skovholt, 2003). While these areas are widely considered in the literature as important domains of counselor development, m any of these studies suffer from major methodological limitations (Buser, 2008). Further, with the exception of cognitive complexity and basic counseling skills, both of which can be assessed by external raters, the remaining constructs have been measured through self report. Additionally, few studies of counselor development have linked these constructs directly to therapeutic outcomes, and many calls have been made for rigorous research that investigates the impact of various types of training on client o utcomes (Bennett Levy, 2006; Buser, 2008; Christensen & Jacobson, 1994; Henry, Strupp, Butler, Schacht, & Binder, 1993; Stein & Lambert, 1995). Because there is still a dearth of evidence linking counselor education, experience, or self reported constructs to therapeutic outcomes, the present finding must be interpreted cautiously Given that t his findin g was consistent with findings in the literature the impact of previous experience with professional counseling on self reported student counselor develop ment was expected. Additionally, because developmental level was self reported, this result does not mean that students who rated themselves more highly possess greater understanding of themselves and their clients, are less ambivalent at learning counseli ng, or execute effective interventions independent from their supervisors; rather, as they progress through education and professional experiences,
108 they at least perceive themselves as increasingly accomplished in these domains over time. Comfort with Clos eness with being close in their relationships is a self reported construct derived from attachment theory (Bowlby, 1988). Results from one study (Dunkle & Friedlander, 1996) showed that counseling interns who rated themselves hi ghly on comfort with closeness received higher ratings from their clients on the therapeutic alliance. Previous counselor development theory and research has assumed that counselors eventually develop relational expertise (Ronnestad & Skovholt, 2003) but h as neglected to investigate whether some counselors possess more innate capacity to relate effectively than others. Thus, being close in relationships was included as a variable in an attempt to see how much of what has been traditionally considered develo pmental (subject to significant change over time) may be related to a relatively stable personal propensity for closeness in relationships that can be formed in early childhood experiences. Findings showed that students who rated themselves highly on comf ort with closeness also rated themselves more highly on their overall development, awareness of self and others, and their autonomy from a supervisor. It is possible that students who are more comfortable with relational intimacy feel more competent and co nfident with clients and thus would view their overall development more favorably. Similarly, students who reported more comfort with closeness also reported greater self awareness and capacity to understand their clients. From a relational cultural perspe ctive, students with closer relationships would also be more practiced at
109 understanding and sharing their own feelings and empathizing with others (Comstock, 2002). In terms of attachment theory (Bowlby, 1988), this finding lends support for the idea that some developmental constructs may in fact be more fixed and stable than previously framed in counselor development theory (Skovholt & Ronnestad, 1992b). The current study also found a link between student ratings of their autonomy from supervisors and the ir comfort being close to others. While the contribution of this variable was relatively small, this finding suggests that comfort with closeness may be connected to a greater sense of independent decision making with clients, a factor that has been consid ered a desirable marker of development in the literature (Stoltenberg & McNeill, 2010). Furthermore, comfort with closeness in relationships was not significantly correlated with either years of previous education or prior counseling experience, supporti ng previous theory and research that comfort being close to others is relatively intrinsic and stable over time, including across cultures (Bowlby, 1988). Comfort with closeness in relationships was also significantly correlated with community relational h ealth and mentor relational health. This finding is reason to explore whether students who are more comfortable with relational closeness could be better equipped to form meaningful, growth enhancing relationships in their preparation programs. Community R elational Health college students in three previous studies (Adams, 2010; Frey et al., 2005; Liang et al., 2002). Adams (2010) was the first to explore the relational health of counseling students. The researcher adapted the RHI to collect data about a peer, mentor, and
110 classroom in which counseling students had been exposed to gender issues. The present study was the first to survey counseling students about the relational heal th of their preparation programs as a whole. Findings revealed that higher reports of community relational health were important in the regression models for both self other awareness and motivation to learn counseling. These findings suggest that an enhan ced sense of connection in a counselor preparation program may impact therapeutic skillfulness. This finding is in line with the RCT tene t s that development is enhanced wh en environments provide for enriching relationships that empower people and convey authenticity and empathy (Comstock, 2002; Jordan, 2002; Walker, 2005). Research and theory also suggests that students may feel more motivated to learn and grow when they fe el connected to their community (Comstock et al., 2008; Skovholt & Ronnestad, 1992b). Mentor Relational Health autono my, meaning that higher reports of relational health with me ntor s were correlated with self reports of greater dependency on supervisors. The effect of this variable in the regression model was small; however, the finding is interesting nonetheless. The traditional counselor development paradigm holds that developm ent is a trajectory from dependency to autonomy (Stoltenberg & McNeill, 2010). In this paradigm, dependency on a supervisor represents something to outgrow. From an RCT perspective, the very concept of dependency is reconsidered (Comstock, 2010) in light o f a greater emphasis on interconnection and interdependency. Thus, this finding could be interpreted that a stronger, higher quality relationship with a mentor in a counselor preparation program
111 leads students to feel more connected to and trusting of the guidance and input of their mentor. From this theoretical lens, slightly higher dependency on supervisors during training would not necessarily mean that students with closer mentor relationships are less able to think critically about their clients and as sume ownership for the direction of their clinical decisions. However, another possible interpretation is that more supportive mentor relationships may promote dependency in students and supervisees, or that students who are less secure in themselves may r ely on the perceived authorities rather than making their own clinical decisions. Notably, the effect of prior counseling experience had a strong positive influence on ratings of autonomy, meaning that this cross section of students reported feeling more a utonomous along with reports of more professional experience. Also, supervision quality and mentor relational quality had a high positive correlation, indicating that they likely tap similar constructs. Supervision Quality Self reported quality of supervis ion received was negatively related to self reported motivation to learn counseling. That is, higher reports of supervision quality were correlated with slightly lower reported levels of motivation. This variable only accounted for a small portion of the v ariance on reported motivation. One interpretation of this finding is that, according to the authors of the scale, (SLQ R; McNeill et al., 1992), motivational level is thought to fluctuate throughout the trai ning and supervision experience as part of norma tive development. In other words, lower motivation is not necessarily a negative indicator of development, and motivation is theorized to rise and fall throughout training and supervision as students encounter difficult clients, recognize their own limitat ions, and periodically doubt their abilities. It is thus possible that
112 students who reported higher supervision quality felt more challenged by their supervisors. Such challenges may have influenced slightly lower reports of motivation to learn the complex practice of counseling. However, this finding can also be interpreted more negatively It is possible that students gave their supervisors high ratings for being supportive rather than for challenging them to grow personally and professionally. Most prog rams use doctoral student supervisors to provide master students with required supervision, and it is unknown the extent to which doctoral student supervisors are willing or able to challenge their supervisees. Thus, it could also be that perceptions of hi gh supervision quality were associated with lower levels of motivation because supervisors are not providing their supervisees with enough challenges or motivators. Implications for Theory To date, counselor development theory has not included a relationa l emphasis. Instead, theories propose that counselors progress through normative stages that are relatively independent from the relational context of their training environment or their own capacity for maintaining close relationships in their personal li ves. Within this paradigm, an untested but common assumption is that preparation programs provide optimally for the relational needs of students (McNeill et al., 1992), and that preparation programs are not implicit in novice developmental and relational limitations. The results of the current study provide a closer examination of relational quality within preparation programs as a correlate to self assessments of development. Findings suggest that there are relational components to how counseling student perceive their development. When asked about the health of relationships with a mentor, peer, and counselor education community as a whole, the strongest effect was
113 St udents who felt they were more connected to and authentic within the overall context of their programs also reported higher developmental levels. Although counselor development theory has rested on Western assumptions that value individuation and autonomy as developmental markers, these findings lend initial support for the inclusion of a relational health perspective in counselor development theory. Additionally, results from this study found that comfort with closeness had a strong impact on self reporte d overall developmental level, as well as self other awareness and autonomy from a supervisor. Given that attachment theory, from which in less developmental terms, this finding is striking. Rather, an attachment framework would suggest that counseling students bring with them a stable blueprint for relating that impacts their self appraisal of important developmental markers, and is not impacted by years of education or amount of professional counseling experience. Comfort with relational closeness may impact effectiveness in therapeutic relationships (Dunkle & Friedlander, 1996), and this study as well as attachment theory suggests that comfort with closeness may not be a domain easily affected by counselor training and experience. Over the past decade, constructivism has been emerging as a guiding theory for counselor education. Constructivism proposes that reality is a product of meanings created in social environments and that all truths can and should be actively examined and are open to reformulation ( Nelson & Neufeldt, 1998). Constructivism in counselor education challenges educators and students alike to participate in ways in which
114 students are also teachers and teachers learn from students. A context of critical thought, ongoing reflection, and greater power sharing amongst the traditional experts (i.e., professors and supervisors) and students is also thought to be a more optimal environment for fostering multic ultural competencies and diversity training. These environments where students are encouraged to question their assumptions and accept a multiplicity of perspectives also are likely to involve greater attention to the key aspects of relational cultural the ory authenticity, connection, and mutual empowerment. Implications for Practice This study has several practical implications. For counselor educators and those tasked with program development, these findings suggest that the student perceptions of relati onal quality within counseling programs are linked to self assessed student developmental level. Educators in particular occupy important positions within counseling programs, and thus need to consider the ways in which they shape the relational tone of th eir classrooms. Educators also occupy mentoring roles, modeling to various degrees the growth enhancing or disconnecting features of relationship for future counselors, supervisors and educators. Because educators, supervisors, and students comprise the co unselor education community, results of this study suggest that all participants within the preparation environment need to attend to relational dimensions. Unfortunately, the culture of the academy may not be optimal for fostering the growth enhancing, re ciprocal relationships championed by RCT. Hartling and Sparks (2002) discuss the challenges of doing counseling from an RCT perspective in workplaces that foster competition, hie rarchy, and self sufficiency. Yet i f students were
115 trained in environments th at maximized RCT values, they could then bring these values into their practices and perpetuate them in counselor education and supervision as well At the graduate level, it is likely that counseling students, supervisors, educators have already internal ized many of these dominant western values (Nelson, Gizara, Hope, Phelps, Steward, & Weitzman, 2006). To begin to reconstruct the cultural paradigm, counselor education faculty and administrators could explore the importance of creating learning atmosphere s that foster optimum relational health and critically examine the barriers to building relationally healthy academic environments They would need to assess their relationships with each other as well as considering their potential impact on students. Th en, broader conversations could begin between faculty and students, as well as within the literature on counselor education theory, practice, and research. In the short term, student progress evaluations could expand to include peer, mentor, and supervisor y relational feedback. Students and colleagues could also provide instructors and supervisors with feedback on their facilitation of empathic, mutually empowering relationships. Relational health with all members of the counselor education community could be integrated as part of the curriculum in student and instructor/supervisor assessments, classroom activities, and program evaluation. Further, RCT promotes multicultural awareness and oppression consciousness by acknowledging the sociocultural and conte xtual challenges faced by marginalized peoples in a society with traditional western values (Comstock et al., 2008). The theory role socialization, power, dominance, marg inalization, and subordination affect the
116 Thus, a focus on relational health in preparation programs may contribute to student skill acquisition and personal developme nt while also promoting multicultural ly competent practice. assessed comfort being close in their relationships and perceptions of their developmental level. Practically, prepar ation programs may want to consider attending to this dimension during application procedures and candidate interviews as well as throughout the training years Students who enter counseling programs uncomfortable with close relationships may struggle to form close therapeutic bonds with their clients not just in the initial learning phase, but throughout their careers They may also have difficulty fostering close relationships with their peers and mentors, which could in turn detract from the relational health of the preparation community. Limitations of the Study and Implications for Research Limitations This study had several limitations First, although participants spanned a large geographic area, sampling was not random. Because this study utilized a convenience sample, the response rate and population characteristics were unknown. It is possible that student counselors in other states and regions of the United States may have reported different data on the constructs of interest T herefore, it wou ld be beneficial to replicate the study in other areas to assess the reliability of the findings. Also, student counselors in this study volunteered to participate Co unselors who volunteer may have significantly differed from those who chose to not volu nteer. Additionally, data collection was conducted in the month prior to Although graduate student counselors are busy throughout the year, the pressures of final exams and the
117 proximity to major holidays may have impacted the response and completion rates as well the students who chose to participate While difficult, random sampling would Results of the current study were also limited by ins trumentation. While t he SLQ R (McNeill et al., 1992) has been used a fair amount in counseling research, the authors recently reported that the measure does not adequately tap the domain specific constructs it was designed to assess (Skovholt & McNeill, 20 10). It would be beneficial to broaden the scope of expert opinion on the SLQ R and consider changes to improve its utility. A new validated measure of counselor development that assessed relational domains would be beneficial to this area of study. Likewi se, the AAS subscale Close (AAS; Collins & Read, 19 90) had less than desirable reliability. Future scale development would benefit from a refinement of this subscale to yield more reliable data. The RHI (Liang et al., 2001) has been validated with college freshman and senior females (Liang et al., 2002) and a mixed gender sample of clients at a university the first study to query graduate counseling students about relati onal health within their preparation programs as a whole. Adams (2010) amended the RHI to assess relational health in a counselor education classroom in which students were exposed to gender issues. Future use of the RHI with counseling students will provi de researchers a richer understanding of the role of relational health in counselor development, counselor preparation programs, and therapeutic outcomes.
118 This study contained a single non validated item about student perceptions of a specific or common f actors emphasis in their training programs. The item included a brief definition of the terms followed by a continuum ranging from the most emphasis on specific factors to the most emphasis on common factors. Most students located their programs on the com mon factors end of the continuum, which seems inconsistent with a number of critiques of counselor training programs ( Bergin, 1997; Mahoney, 1986; Miller, 1989; Nelson & Neufelt,1998; Whiston & Coker, 2000; Winslade et al., 1997) A thorough analysis of th e content of preparation programs from multiple perspectives would increase the reliability and validity of this item. Because the common factors are so important to therapeutic outcomes, a better understanding of the extent to which programs train their s tudents for the common factors would be helpful. education. This partial data was omitted from the final analyses to allow for an adequate sample size in the regression models. In preliminary analyses that included this data, years of education accounted for the largest proportion of variance in all four counseling experience emerged a s the strongest varia ble in four regression models. Social desirability may have also contributed to student c ounselor responses. Counseling students may have overestimate d t heir skills and development on their self assessment s of these constructs. Some students may have also rated their comfort with closeness more positively than other s would have, given that comfort with closeness is desirable in the field of counseling. Because t his study was based on self report, it is
119 unknown how similar student cou assessments would be to those offered by counseling professionals, educators and supervisors Implications for Research Future research examining counselor student development may be beneficial in several areas. Linking participant response s to specific preparation programs could reveal relationships between programs and developmental outcomes. The current study was correlational only. While research has indicated that supportive relationships promote human development and resilience (Spence r, 2000; Hartling & Ly, 2000; Skovholt & Ronnestad, 2003), more research is needed to determine whether the relational health of specific programs has an impact on counselor development. In this research design, it was not possible to compare student ratin gs of relational health in one program to those of another. Future studies could explore the relational health of several programs in depth and compare developmental variables with program specific perceptions of relational health. It would be also helpful to explore correlations between programs. Additionally, programs could be designed and piloted with relational health as a guiding framework, and student development and client outcomes could be compared with those of traditional educational models. To date, the literature on counselor development has reflected the view that novice ( Comstock & Qin, 2005 ; Stoltenberg, 1981; Loganbill et al., 1982; Stoltenberg & McNeill, 2010; Skovholt & Ronnestad, 1992a) However, the findings of this study point to the need for further research into the more stable construct, comfort with relational closeness, as a possible indicator of counselor development in the preparation years
120 and beyond. The present study found that students who rated their programs highly on relational health also rated themselves highly on comfort with relational closeness. Thus, studies that explored this link would help researchers better understand how each experience quality relationships with their clients and within their preparation programs Most notably, the present area of study would benefit from the creation of a new counselor development measure that incorporates relational aspects of development. Assessment of relational health, comfort with closeness in relationships, and dev elopmental level could be improved by gathering multiple sources of data, including evaluative feedback from peers, educators, supervisors, and even clients. New studies are needed to explore possible causal relationships between relational aspects of coun selor development and therapeutic outcomes. Summary The results of this exploratory study found significant positive relationships assessed of developmental level and their self reports of: (1) years of prior professional counseling experience; (2) comfort with closeness in relationships; and (3) relational health within a counselor preparation community. Additionally, small negative effects for self assessed development were found for self reported relational quality with a mentor and supervision quality. Taken together, these findings provide preliminary support for the presence of relational features in student counselor development. Given the role of relationship quality to therapeutic outcomes, as well as the growing t heoretical and practical appreciation for the importance of quality social connections in human development, both counselor development theory and counselor education practice would benefit from an ongoing
121 dialogue about how preparation programs do or do n ot attend to relational dimensions. Future research should focus on establishing new measures of counselor development that include relational constructs, as well as the contribution of counselor relational health, comfort with closeness, and counselor dev elopment to specific therapeutic outcomes.
122 APPENDIX A RELATIONAL HEALTH IN DICES (RHI ; Liang, Tracy, Williams Taylor, Jordan, & Miller, 2002) The following questions pertain to your relationships with "mentors" (other than your parents or whoever raised you) to who you go for support and guidance. A mentor is not a peer or romantic partner. For the purposes of this study, please choose a professor or supervisor within your counselor education department (not an off site supervisor) whom you view as a men tor, or to whom you feel the strongest sense of connection. By mentor we mean someone who often is older than you, has more experience than you, and is willing to listen, share her or his own experiences, and guide you through some area of your life (e.g., academic, social, athletic, religious). If you have more than one mentor, please answer the following questions regarding the mentor who is most important to you. If you have more than one mentor, please answer the following questions regarding the me ntor who is most important to you. RHI Mentor For each statement below, please indicate the number that best applies to your relationship with this mentor. 1. I can be genuinely myself with my mentor. 2. I believe my mentor values me as a whole person (e.g., professionally/academically and personally). 3. My mentor's commitment to and involvement in our relationship exceeds that required by his/her social/professional role. 4. My mentor shares stories about his/her own experiences with me in a way that enhances my life. 5. I feel as though I know myself better because of my mentor. 6. My mentor gives me emotional support and encouragement
123 7. I try to emulate the values of my mentor (such as social, academic, religious, physical/athletic). 8. I feel uplifted and energized by interactions with my mentor. 9. My mentor tries hard to understand my feelings and goals (academic, personal, or whatever is relevant). 10. My relationship with my mentor inspires me to seek other relationships like this one. 12. I feel comfortable expressing my deepest concerns to my mentor. Always RHI Peer The following questions pertain to your friendships with peers in your Counselor Education department (excluding family members or a romantic partner). A close friend is someone whom you feel attached to through respect, affection and/o r common interests, someone you can depend on for support and who depends on you. Please answer the next questions regarding just ONE of your closest friends in your Counselor Education Department If you do not have a close friend in the department, pleas e select the peer with whom you have the closest friendship. (Please do not select a family member or romantic partner). Next to each statement below, please indicate the number that best applies to your relationship with a close friend. 1. Even when I have difficult things to say, I can be honest and real with my friend. 2. After a conversation with my friend, I feel uplifted. 3. The more time I spend with my friend, the closer I feel to him/her 4. I feel understood by my friend.
124 5. It is important to us to make our friendship grow. 6. I can talk to my friend about our disagreements without feeling judged. 7. My friendship inspires me to seek other friendships like this one. 8. I am uncomfortable sharing my deepest feelings and thoughts with my friend. 9. I have a greater sense of self worth through my relationshi p with my friend. 10. I feel positively changed by my friend. 11. I can tell my friend when he/she has hurt my feelings. 12. My friendship causes me to grow in important ways. RHI Community The following questions pertain to your classroom community For the purposes of this study, your community is your Counselor Education department as a whole, including your cohort, professors, and other members of your department with whom you come into contact. Next to each statement below, please indicate the n umber that best applies to your relationship with or involvement in this community. 1. I feel a sense of belonging to this community. 2. I feel better about myself after my interactions with this community. 3. If members of this community know something is bot hering me, they ask me about it. 4. Members of this community are not free to just be themselves.
125 5. I feel understood by memb ers of this community. 6. I feel mobilized to personal action after meetings within this community. 7. There are parts of myse lf I feel I must hide from this community. 8. It seems as if people in this community really like me as a person. 9. There is a lot of backbiting and gossiping in this community 10. Members of this community are very competitive with each other. 11. I have a greater sense of self worth through my connection with this community. 12. My connections with this community are so inspiring that they motivate me to pursue relationships with othe r people outside this community. 13. This community has shaped my identity in many ways. 14. This community provides me with e motional support. *With minor adaptations to specify that, for the purposes of this study, respondents should select a peer and mentor in their counselor preparation program, and the communi
126 APPENDIX B COMFORT WITH CLOSENE SS (AAS, Collins & Read, 1990) For each of the following items, please indicate to what extent the statement is characteristic of you. Thank you. 1.) I find it r elatively easy to get close to others. Not at all Very characteristic characteristic 1 2 3 4 5 2.) I do not often worry about someone getting too close to me. Not at all Very characteris tic characteristic 1 2 3 4 5 3.) I am somewhat uncomfortable being close to others. Not at all Very characteristic characteristic 1 2 3 4 5 4.) I am nervous when anyone gets too close Not at all Very characteristic characteristic 1 2 3 4 5 5.) I am comfortable having others depend on me. Not at all Very characteristic characteristic 1 2 3 4 5 6.) Often, love partners want me to be more intimate than I feel comfortable being. Not at all Very characteristic characteristic 1 2 3 4 5
127 APPENDIX C SUPERVISEE LEVELS QU ESTIONNAIRE REVISED (SLQ R) (Mc Neill, Stoltenberg, & Roman, 1992) The following instrument is designed to study the behavior of counselors/therapists in training. The gaining of skills as a counselor/therapist is a learning process, and therefore it is necessary to continuously gather new information. Your total honesty will be appreciated. In terms of your current behavior, please answer the items below according to the following scale. For questions that refer to a supervisor, please refer to the supervisor within your department ra ther than the supervisor appointed to you at your practicum or internship site. 1: Never 2: Rarely 3: Sometimes 4: Half of the Time 5: Often 6: Most of the Time 7: Always 1) I feel genuinely relaxed and comfortable in my counseling/therapy sessions. Never Always 1 2 3 4 5 6 7 2) I am able to critique counseling tapes and gain insights with minimal help from my supervisor. Never Always 1 2 3 4 5 6 7 3) I am able to be spontaneous in counseling/therapy, yet my behavior is relevant. Never Always 1 2 3 4 5 6 7 4) I lack self confidence in establishing counselin g relationships with diverse client types. Never Always 1 2 3 4 5 6 7
128 5) I am able to apply a consistent personalized rationale of human behavior in working with my clients. Never Always 1 2 3 4 5 6 7 6 ) my ability to handle the unexpected. Never Always 1 2 3 4 5 6 7 7 ) The overall quality of my work fluctuates; on some days I do well, on other days, I do poorly Never Always 1 2 3 4 5 6 7 8) I depend upon my supervisor consid erably in figuring out how to deal with my clients. Never Always 1 2 3 4 5 6 7 9) I feel comfortable in confronting my clients. Never Always 1 2 3 4 5 6 7 10) Much of the time in counseling/therapy, I find myself thinking about my next response, instead of fitting my intervention into the overall picture. Never Always 1 2 3 4 5 6 7 11 ) My motivation fluctuates from day to day. Never Always 1 2 3 4 5 6 7 12) At times, I wish my supervisor could be in the counseling/therapy session to lend a hand. Ne ver Always 1 2 3 4 5 6 7
129 13 ) During my counseling/therapy sessions, I find it difficult to concentrate because of my concern with my own performance. Never Always 1 2 3 4 5 6 7 14 ) Although at times, I really want advice/feedback from my supervisor, at other times I really want to do things my own way. Never Always 1 2 3 4 5 6 7 Never Always 1 2 3 4 5 6 7 16) It is important that my supervisor allows me to make my own mistakes. Never Always 1 2 3 4 5 6 7 17) Given my current state of professional development, I believe I know when I need Never Always 1 2 3 4 5 6 7 18 ) Sometimes I question how suited I am to be a counselor/therapist. Never Always 1 2 3 4 5 6 7 19 ) Regarding counseling/therapy, I view my supervisor as a teacher/mentor. Never Always 1 2 3 4 5 6 7 20 ) Sometimes I feel that counseling/therapy is so complex, I will never be able to learn it all. Never Always 1 2 3 4 5 6 7
130 21 ) I believe I know my strengths and weaknesses as a counselor/therapist sufficiently well to understand my professional potential and limitations. Never Alwa ys 1 2 3 4 5 6 7 22 ) Regarding counseling/therapy, I view my supervisor as a peer/colleague. Never Always 1 2 3 4 5 6 7 23 ) I think I know myself well and am abl e to integrate that into my therapeutic style. Never Always 1 2 3 4 5 6 7 24 evaluate alternatives. Never Always 1 2 3 4 5 6 7 25 ) At my current level of professional development, my confidence in my abilities is y. Never Always 1 2 3 4 5 6 7 26 on problem resolution. Never Always 1 2 3 4 5 6 7 27) I am able to adequately assess my interpersonal impact on clients and use that knowledge therapeutically. Never Always 1 2 3 4 5 6 7 28 ) I am adeq therapeutically. Never Always 1 2 3 4 5 6 7
131 29 ) I believe I exhibit a consistent professional objectivity, and an abil ity to work within my role as a counselor/therapist without undue over involvement with my clients. Never Always 1 2 3 4 5 6 7 30 ) I believe I exhibit a consistent professional objectivity, an d an ability to work within my role as a counselor/therapist without excessive distance from my clients. Never Always 1 2 3 4 5 6 7
132 APPENDIX D SUPPLEMENTAL AND DEM OGRAPHIC QUESTIONNAI RE Please read the following section before proceeding to the next item. common factors orientation or use of p articular interventions (Wampold, 2001). Some common factors include the quality of the counselor and support. s pecific factors outcomes. Specific factors differ from counselor to counselor and include the agnosis, and theory driven interventions (Wampold, 2001). 1. Please indicate the extent to which your counseling program AS A WHOLE emphasizes the common factors and specific factors in your training. Please exclude experiences at your practicum and/or inter nship sites. 1 indicates a much greater emphasis on specific factors than common factors. 4 indicates an equal degree of emphasis on both specific and common factors. 7 indicates a much greater emphasis on common factors than specific factors. Specific F actors Equal Amount Common Factors 1 2 3 4 5 6 7 2. Are you currently under clinical supervision? If no, please answer the following items based on the last semeste r you received supervision. Yes No 3. What type(s) of supervision do you currently receive? Select all that apply: Group supervision from a faculty member in my department Group supervision from a doctoral student in my department Group supervision at my Practicum/Internship site
133 Individual supervision from a doctoral student in my department Individual supervision from a faculty member in my department Individual supervision at my Practicum/Internship site 4. Please rate the quality of the supervision you receive. For the purposes of this study, supervision quality includes your assessment of the usefulness of the supervision to your clinical work and to your personal and professional development. Group supervision from a faculty member in my departme nt: 1: Poor 2: Fair 3: Good 4: Very Good 5: Excellent Group supervision from a doctoral student in my department: 1: Poor 2: Fair 3: Good 4: Very Good 5: Excellent Group supervision at my Practicum/Internship site: 1: Poor 2: Fair 3: Good 4: Very Good 5: Excellent Individual supervision from a faculty member in my department: 1: Poor 2: Fair 3: Good 4: Very Good 5: Excellent
134 Individual supervision from a doctoral student in my department: 1: Poor 2: Fair 3: Good 4: Very Good 5: Excellent Individual supervision at my Practicum/Internship site: 1: Poor 2: Fair 3: Good 4: Very Good 5: Excellent 5. Since the age of 18, have you voluntarily seen a counselor(s) for your own personal counseling? 1. Ye s 2. No 3. Not available 6. If yes, approximately how many counseling sessions have you attended? Please include all voluntary counseling sessions since the age of 18 in your estimate. 1 5 sessions 5 10 sessions 10 15 sessions 16 20 sessions 20 30 sessions Over 30 sessions 7. Overall, how helpful was the counseling that you received? 1: Not at all helpful 2: A little helpful 3: Somewhat helpful 4: Very helpful 5: Extremely helpful
135 8. What is your gender? 1: Female 2: Male 3: Transgendered 4: Not availab le 9. What is your race (select all that apply)? White/Caucasian African American/Black Asian Native American Latino/Hispanic Other:_____________ 10. Please indicate your age in years. 18 21 22 25 26 30 31 35 36 40 41 45 46 50 Ove r 50 11. Please indicate the number of years you have worked as a professional counselor prior to entering your current counseling program 0 1 3 4 6 7 10 Over 10 years 12. Please indicate the state in which you attend graduate school: Drop down menu of all 50 states with abbreviations
136 LIST OF REFERENCES Ackerman, S. J., & Hilsenroth, M. J. (2003). A review of therapist characteristics and techniques positively impacting the therapeutic alliance. Clinical Psychology Review 23 1 33. Adams, H. C. (2010). Gende r awareness and relational characteristics of counselor trainees. (Doctoral dissertation). Retrieved from Udini. (3446939) Ahn, H., & Wampold, B. E. (2001). Where oh where are the specific ingredients? A meta analysis of component studies in counseling and psychotherapy. Journal of Counseling Psychology, 48 (3), 251 257. Anderson, R., & Price, G. (2001). Experiential groups in counselor education: Student a ttitudes and instructor participation. Counselor Education and Supervision 41 (2), 111 119. Assay, T. P ., & Lambert, M. J. (2004). The empirical case for the common factors in therapy: Quantitative findings. In M. A. Hubble, B. L. Duncan, & S. D. Miller (Eds.), The heart and soul of change: What works in therapy ( pp. 23 49). Washington, DC: American Psyc hological Association. Baker, S. B., & Daniels, T. G. (1989). Integrating research on the microcounseling program: A meta analysis. Journal of Counseling Psychology, 36, 213 222. Baker, S. B., Daniels, T. G., & Greeley, A. T. (1990). Systematic training of graduate level counselors: Narrative and meta analytic reviews of three major programs. Counseling Psychologist, 18 355 421. Bennet Levy, J. (2006). Therapist skills: A cognitive model of their acquisition and refinement. Behavioral and Cognitive Psychot herapy, 34, 1 22. Bennett, S., & Vitale Saks, L. (2006). A conceptual application of attachment theory and research to the social work student field instructor supervisory relationship. Journal of Social Work Education, 42 669 682. Bergin, A. E. (1997). N eglect of the therapist and the human dimensions of change: A commentary. Clinical Psychology: Science and Practice, 4, 83 89. Bergman, S. J. (1991). Men's psychological development: A relational perspective Work in Progress, No. 48 Wellesley, MA: Stone Center Working Paper Series. Bergman. S. J., & Surrey. J. L. (1994). Couples in therapy: A relational approach. Work in Progress, No. 66 Wellesley, MA: Stone Center Working Paper Series. Berman, J. S., & Norton, N. C. (1985). Does professional training ma ke a therapist more effective? Psychological Bulletin, 98 (2), 401 407.
137 Birrell, P. J., & Freyd, J. J. (2006). Betrayal trauma: Relational models of harm and healing. Journal of Trauma Practice, 5 (1), 49 63. Borders, L. D. (1989). A pragmatic agenda for de velopmental supervision research. Counselor Education and Supervision, 79, 16 24. Borders, L. D. (1990). Developmental changes during supervisees' first practicum. The Clinical Supervisor, 8 (2), 157 167. Bordin, E. S. (1979). The generalizability of the ps ychoa nalytic concept of the working alliance. Psychotherapy: Theory, Research, and Practice, 16 (3), 252 260. Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York NY : Basic Books. Bowlby, J. (1988). A secure base: Parent chil d attachment a nd healthy human development. New York NY : Basic Books. Bradley, C. & Fiorini, J. (1999). Evaluation of counselin g practicum: National study of programs accredited by CACREP. Counselor Education and Supervision, 39, 110 119. Buser, T. (2008). Counselor t raining: Empirical fi ndings and current approaches. Counselor Education & Supervision 48 (2), 86 100. Cassidy, J., & Shaver, P. R. (Eds ) (1999). Handbook o f attachment: Theory, research and clinical applications. New York NY : Guilford. Chapin, J., & El lis, M. V. (2002). Effects of role ind uction workshops on supervisee anxiety. Paper presented at the annual meeting of the American Ps ychological Association, Chicago. Chickering, A. W. (1969). Education and identity. San Francisco CA : Jossey Bass Publish ers. Christensen, A., & Jacobson, N. S. (1994). Who (or w hat) can do psychotherapy: The status and challenge of nonprofessional therapies. Psychological Science, 5 (1), 8 14. Cohen, J. (1977) Statistical power analysis for the behavioral sciences (Rev. ed. ). New York: Academic Press. Cohen, L. (1992). A power primer. Psychological Bulletin 112 (1), 155 159. Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385 396.
138 Collins, N L., & Read, S.J. (1990). Adult attachment, wo rking models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58, 644 663. Comstock, D. L. & Qin, D. (2005). Diversity and development: Critical contexts that shape ou r lives and relationships (pp. 25 46). D. L. Comstock (Ed.) Belmont, CA: Brooks/Cole. Comstock, D. L., Hammer, T. R., Strentzsch, J., Cannon, K., Parsons, J., & Salazar, G. II. (2008). Relational cultural theory: A fram ework for bridging relational, multi cultural, and social justice competencies. Journal of Counseling and Development, 86, 279 287. Cormier, L. S. (1988). Critical incidents in cou nselor development: Themes and patterns. Journal of Counseling and Development, 67 131 132. Council for Accredi tation of Counseling and Related Educational Programs. (2009). 2009 standards. Retrieved from http://www.cacrep.org/doc/2009%20Standards%20with%20eover.pdf Crockett, W. H. (1965 ). Cognitive complexity and impression formation. In B. A. Maher (Ed.), Progress in experimental personality research (p p. 47 90). New York, NY: Academic Press. Deal, K. H., Bennett, S., Mohr, J., & Hwang, J. (2011). Effects of field instructor training on student competencies and the supervisory alliance. Research on Social Work Practica, 21 (6), 712 726. Dillman, D. A. (2000). Mail and i nternet s urveys: The t ailored d esign m ethod. New York NY : Wiley. Dooley, C., & Fedele. N. M. (2004). Mothers and sons: Rai sing relational boys. In J. V. Jordan, M. Walker, & L. M. Hartling (Eds.), The complexity of connection (pp. 220 249). New York NY : Guilford. Dozier, M., Cue, K., & Barnett, L. (1994). Clinicians as caregivers: The role of attachment organization in tr eatment. Jour nal of Counseling and Clinical Psychology, 62, 793 800. Dunkle, J. H., & Friedlander, M. L. (1996). The contributi on of therapist experience and personal characteristics to the working alliance. Journal of Counseling Psychology, 43 (4), 456 460 Durlak, J. A. (1979). Comparative effectiveness of par aprofessional and professional helpers. Psychological Bulletin, 86, 80 92. Duys, D., & Hedstrom, S. (2000). Basic counselor skills training and counselor cognitive complexity. Counselor Education and Supervision 40, 8 19.
139 Erikson, E. H (1963). Childhood and society. New York: Norton. Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton. Fong, M. L., Borders, L. D, Ethington, C. A., & Pitts, J. H. (1997). Becoming a counselor: A longit udinal study of student cognitive development. Counselor Education and Supervision, 37, 100 114. Frey, L., Beesley, D., & Newman, J. (2005). The relational health indices: Reanalysis of a measure of relational quality. Measurement an d Evaluation in Counsel ing and Development, 38 153 163. Furr, S.R., & Carroll, J.J. (2003). Critical incidents in student counselor development. Journal of Counseling & Development, 81, 483 489. Genero, N. P., Miller, J. B., Surrey, J., & Baldwin, L. M. (1992). Measuring percei ved mutuality in close relationships: Validati on of the Mutual Psychological Development Questionnaire. Journal of Family Psychology, 6, 36 48. George, C Kaplan, N., & Main, M. (1985). Adult a ttachment i nterview. ( Unpublished manuscript ). University of California, Berkeley. Goldberg, A. D. (1974). Conceptual system as a predisposition toward therapeutic communication. Journal of Counseling Psychology, 21, 364 368. Goodyear, R.K., Wertheimer, A., Cypers, S., & Rosemon d, M. (2003). Refining the map of the counselor development journey: Response to Ronnestad and Skovholt. Journal of Career Development, 30 (1), 73 80. Granello, D.H. (2002). Assessing the cognitive development of counseling students: Changes in epistemological assumptions. Counselor Education & Supervision, 41, 279 293. Granello, D. H. (2010). Cognitive complexity among practicing counselors: How thinking changes with experience. Journal of Counseling and Development, 88 (1), 92 100. Gray, L. A., Ladany, N., Walker, J. A., & Ancis, J. R. (2001). experience of counterproductive events in supervision. Journal of Counseling Psychology, 48 (4), 371 383. Hansen, J. T. (in press). Extending the humanisti c vision: Toward a humanities' foundation for the counseling profession. Jour nal of Humanistic Counseling Hartling, L.M., & Ly, J. (2000). Relational references: A selected bibliography of theory, research, and applications. Project Report, No. 7. Wellesley, MA: Stone Center Working Paper Series.
140 Hartling, L., & Sparks, E. (2002). Relational cultural practice: Working in a nonrelational world Work in Progress, No. 97 Wellesley, MA: Stone Center Working Paper Series. Haase, R. F., DiMattia, D. J., & Guttman, M. A. (1972). Tr aining of support personnel in three human relation skil ls: A systematic one year follow up. Counselor Education and Supervision, 11, 194 199. Harvey, O., Hunt, D., Schoerder, H. (1961). Conc eptual Systems and Personality Organization. New York NY : Wiley. Hazan, C., & Shaver, P. R. (1987). Romantic love conc e ptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511 524. Henry, W. P., Strupp, H. H., Butler, S. F., Schacht, T. E., & Binder, J. L. (1993). Effects of training in time limited dynamic psychotherapy: Changes in therapi st behavior. Journal of Consulting and Clinical Psychology, 61, 434 440. Herman, K. C. (1993). Reassessing predictors of therapist competence. Journal of Counseling and Development, 72, 29 32. Hill, C. (1992). An overview of four measures developed to test the Hill Process Model: Therapist intentions, therapist response modes client reactions, and client behaviors. Journal of Counseling & Development 70 (6), 728 739. Hill, C. E. & Corbet, M. M. (1993). A perspective on the history of process and outcome r esearch in counseling psychology. Journal of Counseling Psychology, 40 (1), 3 24. Hogan, R. A. (1964). Issues and approaches in supervision. Psychotherapy: Theory Research and Practice, 1, 139 141. Holloway, E. L., & Wolleat, P. L. (1980). Relationship of counselor conceptual level to clinical hypothesis formation. Journal of Counseling Psychology, 27, 539 545. Holloway, E. L., & Wampold, B. E. (1986). Relatio n between conceptual level and counseling related tasks: A meta analysis. Jou rnal of Counseling Psy chology, 33, 310 319. Horvath, A. O., & Symonds, D. B. (1991). Relationshi p between working alliance and outcome in psychotherapy: A meta analysis. Jou rnal of Counseling Psychology, 38, 139 149. Howard, E. E., Inman, A. G., & Altman, A. N. (2006). C ritical incidents among novice counselor trainees. Counselor Education and Supervision, 46, 88 102.
141 Jennings, L., Goh, M., & Skovholt, T. M, Hanson, M. & Banerjee Stevens, D. (2003). Multip le factors in the development of the expert counselor and therapist Journal of Career Development 30 59 72. Jordan, J. V. (1992). Relational resilience. Work in Progress, No. 57 Wellesley, MA: Stone Center Working Paper Series. Jordan, J. V. (1999). Toward connection and competence. Work in Progress, No. 79 Wellesley, MA: Stone Center Working Paper Series. Jordan, J. V. (2002). Learning at the margin: New models of strength. Work in Progress, No. 98 Wellesley, MA: Stone Center Working Paper Series. Kelly, G. (1955). The psychology of personal constructs. New York, NY: Norton. Ladany, N. (2004). Psychotherapy supervision: What lies beneath. Psychotherapy Research, 14, 1 19. Ladany, N., Friedlander, M. L., & Nelson, M. L. (2005). Critical events in ps ychotherapy supervision: An interpersonal approach. Washington, DC: American Psychological Association. Ladany, N., Marotta, S., & Muse Burke, J. L. (2001). Counselor experience related to complexity of case conceptualization and supervision preference. Co unselor Education and Supervision, 40 (3), 203 219. Lambert, M. J., & Bergin, A. E. (1994). The effectiven ess of psychotherapy. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change ( pp. 143 189). New York NY : Wiley. Lamber t, M. J., & Cattani Thompson, K. (1996). Current findings regarding the effectiveness of counseling: Implications for practice. Journal of Counseling and Development, 74 601 608. Lambert, M. J. & Ogles, B. M. (1997). The effectiveness of psychotherapy su pervision. In C. E. Watkins (Ed.), Handbook of Psychotherapy Supervision (pp. 421 446). New York NY : Wiley. Larson, L. M., Suzuki, L. A., Gillespie, K. N., Potenza, M. T., Bechtel, M A., & Toulouse, A. L. (1992). Development and validation o f the Counsel ing Self Estimate Inventory. Journal of Counseling Psychology, 39, 105 120. Lennie, C. (2007). The role of personal development groups in counsellor training: understanding factors contributing to self awareness in the personal development group. British J ournal of Guidance & Counselling 35 (1), 115 129.
142 Liang, B., Tracy, A., Taylor, C., Williams, L.M., Jordan, J., & Miller, J.B. (2002). The Psychology of Women Quarterly, 26 25 35. Ligiro, D. P. & Gelso, C. J. (2002). Countertransferenc e, attachment, and the working Ps ychotherapy: Theory, Research, Practice, Training, 39, 3 11. Little, C., Packman, J., Smaby, M. H., & Maddux, C. D. (2005). The Skilled Cou nselor Training Model: Skills acquisition, self assessment, and cognitive complexity. Counselor Education and Supervision, 44 189 200. Loevinger, J. (1976). Ego development. San Francisco, CA: Jossey Bass. Loganbill, C., Hardy, E., & Delworth. U. (1982). Supervision: A conceptual model. The Counseling Psychologist, 10 (1), 3 42. Lovell, C. (1999). Supervisee cognitive complexity and the Integrated Developmental Model. The Clinical Supervisor, 18 (1), 191 201. Lovell, C. W. (2002). Development and disequilib ration : Predicting counselor trainee gain and loss scores on the Supervisee Levels Questionnaire. Journal of Adult Development, 9 (3), 235 241. Mahler, M. S. (1979). Separation individuation. The selected papers of Margaret S. Mahler, M.D. (Vol. 2). New Yo rk: Wiley. Mahoney, M. J. (1986). The tyranny of technique. Counseling and Values, 30 (2), 169 174. Mallinckrodt, B. & Nelson, M. L. (1991). Counselor training level and the formation of the psychotherapeutic working alliance. Journal of Counseling Psycholo gy, 38 (2), 133 138. McAuliffe, G. (2002). Student changes, program influences, and adult development in one program of counselor training: An exploratory inductive inquiry. Journal of Adult Development 9 (3 ), 203 214. McNeill, B. W., Stoltenberg, C. D., & Pierce, R. A. (1985 ). Supervisees' perceptions of their development: A test of the Counselor Complexity Model. Journal of Counseling Psychology, 32, 630 633. McNeill, B. W., Stoltenberg, C. D., & Romans, J. S.C. (1992). The Integrated Developmental Model of supervision: Sc ale development and validation procedures. Professional Psychology: Research and Practice, 23, 504 508.
143 Meyer, B., Pilkonis, P. A., Proletti, J. M., Heape, C. L, Egan, M. (2001). Attachment styles and personality disorders as predicator s of symptom course. Journal of Personality Disorders, 15 (5), 371 389. Mikulincer, M. (1998). Adult attachment style and indivi dual differences in functional versus dysfunctional experiences of anger. Journal of Personality and Social Psychology, 74 513 5 24. Mikulincer, M., & Shaver, P. R. (2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. In M. P. Zanna (Ed.), Advances in experimental social psychology: Vol. 35 (pp. 53 152). San Diego, CA: Elsev ier Academic Press. Mikulincer, M., Shaver, P. R., Gillath, O., Nitzberg, R. A. ( 2005). Attachment, caregiving, and altruism: Boosting attachment security increases compassion and helping. Journal of Personality and Social Psychology, 89 (5), 817 839. Mille r, A. L., Notaro, P. C., & Zimmerman, M. A. (2002). Stability and change in internal working models of friendship: Associations with multiple domains of urban adolescent functioning. Journal of Social and Personal Relationships, 19 (2), 233 259. Miller, J. B. (1986). Toward a new psychology of women (2 nd ed.). Boston: Beacon Press. f self. In J. V. Jordan, A. G. Kaplan, J. B. Miller, I. P. Striver, & J. L. Surrey (Eds.), connection: W ritings from the Stone Center (pp. 11 26). New York NY : Guildford. Miller, J. B. & Stiver, I.P. (1997). The heal ing connection: How women form relationships in therapy and in life. Boston: Beacon Press. Miller, M. J. (1989). A few thoughts on the relati onship between counseling techniques and empathy. Journal of Counseling and Development, 67, 350 351. Mirkin, M. P., & Geib, P. (1995). Consciousness of context in relational couples therapy (Work in Progress, Monograph No. 73). Wellesley, MA: Wellesley Co llege. Moore, W. S. (1989). The Learning Environment Prefer ences: Exploring the construct validity of an objective measure of the Perry schem e of intellectual development. Journal of College Student Development, 30 504 514. Moore, W. S. (1990). The Learni ng Environment Preferences: An instrument manual (Rev. edition). Olympia, WA: Center for the Study of Intellectual Development. Murray, D. M., & Fisher, J. D. (2002). The internet: A virtually untapped tool for research. Journal of Technology in Human Ser vices, 19, 5 18.
144 Nelson, M., & Neufeldt, S. (1998). The pedagogy of counseling: A critical examination. Counselor Education & Supervision 38 (2), 70. Nelson, M. L., Gizara, S., Hope, A. C., Phelps, R., Steward, R., & Weitzman, L. (2006). A feminist multic ultural perspective on supervision. Journal of Multicultural Counseling and Development, 34, 105 115. Neswald McCalip, R. (2001). Development of the s ecure counselor: Case examples ussion of attachment theory in co unseling supervision. Counselor Education and Supervision, 41, 18 27. Neukrug, E., & Williams, G. (1993). Counseling c ounselors: A survey of values. Counseling and Values, 38 (1), 51 62. Orlinksy, D. E., Grawe, K., & Parks, B. K. (1994). Process and outcome in psychotherapy Noch einmal. In A. E. Bergen, & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change ( pp. 270 376). New York NY : Wiley. Pierce, G. R., Sarason, I. G., Sarason, B. R., Solky Butzel, J. A., & Nagle, L. C. (1997). Assessing the quality of personal relationships. Journal of Social and Personal Relationships, 14, 337 353. Perry, W. G., Jr. (1970). Forms of intellectual and ethi cal development in the college years. New York: Holt, Rinehart, & Winston. Pistole, M. (1999). Caregi ving in attachment relationships: A pers pective for counselors. Journal of Counseling & Development 77 (4), 437 446. Pistole, M., & Watkins, C. E. Jr. (1995). Attachment t heory, counseling process, and supervision. The Counseling Psychologist, 23, 457 478 Porter, N. (2002). Contextual and developmental frameworks in diagnosing children and adolescents. In M. Ballou & L. Brown, Rethinki ng mental health and disorder: Feminist perspectives ( pp. 262 278) New York NY : The Guilford Press. Putnam, R. D. (2000) Bowling alone: The collapse and revival of American community. New York: Simon & Schuster. Qin, D., & Comstock, D. L. (2005). Diversity and devel opment: Critical contexts that shape our lives and relationships D. L. Comstock (Ed.) Belmont, CA: Brooks/C ole. Radloff, L. S. (1977). The CES D Scale: A self report depression scale for research in the general population. Applied Psychological Measurement, 1, 385 401. Raimy, V. C. (Ed.). (1950). Training in clinical psychology. Upper Saddle River, NJ: Prentice Hall.
145 Riggs, S. A., & Bretz, K. M. (2006). Attachmen t processes in the supervisory relationship: An exploratory investigation. Pro fessional Psychology: Research and Practice, 37, 558 566. Roffers, T., Cooper, B., Sultanoff, S. M. (1988). Can counselor tra inees apply their skills in actual client interviews? Journal of Counseling and Development, 66, 385 388. Rogers, C. R. (1975). Empathic: An unappreciated way of being. The Counseling Psychologist, 5, 2 10. Rom, E., & Mikulincer, M. (2003). Attachment t he ory and group processes: The association between attachment style and group r elated representations, goals, memories, and functioning. Journal of Personality and Social Psychology, 84, 1220 1235. Ronnestad, M. H., & Skovholt, T. M. (2003). The journey of t he counselor and therapist: Research findings and perspectives on professional development. Journal of Career Development, 30 (1) 5 44. Rosenberg, M. (1965). Society and the adolescent self image Princeton, NJ: Princeton University Press. Russell, D., P eplau, L. A., & Cutrona, C. E. (1980). The revised UCLA Loneliness Scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology, 39 (3), 472 480. Russell Chapin, L. A., & Chapin, T. J. (2012). Clinical supervision: Theo ry and practice. Belmont, CA: Brooks/Cole. Schaef A.W. (1999). Living in process: Basic truths for living the path of the soul. New York: Random House. K. (1986). Student and faculty assumptions about the na ture of uncertainty in m edicine and medical education. The Journal of Family Practice, 23, 468 472. Skovholt, T. M., & McCarthy, P. R. (1988). Critical incidents: Catalysts for counselor development. Journal of Counseling and Development, 67, 69 72. Skovho lt, T., & Ronnestad, M. (1992a). Themes in ther apist & counselor development. Journal of Counseling & Development 70 (4), 505 515. Skovholt, T. M., & Ronnestad M. H. (1992b). The evolving professional self: Stages and themes in therapist and counselor deve lopment. Chichester: John Wiley & Sons. Skovholt, T. M. & Ronnestad, M. H. (2003). Struggles of the novice counselor and therapist. Journal of Career Development 30(1), 45 58.
146 Smith, E. W. L. (2003). The person of the therapist. Jefferson, NC: McFarland & Co, Inc. Spencer, R. (2000). A comparison of relational psychologies. Project Report, No. 5. Wellesley, MA: Stone Center Working Paper Series. Spengler, P. M., & Strohmer, D. C. (1994). Clinical jud gmental biases: The moderating roles of cognitive compl exity and counselor client preferences. Journal of Counseling Psychology, 41 (1), 8 17. Sperling, M. B., & Berman, W. H. (1991). An attachment classification of desperate love. Journal of Personality Assessment, 56, 45 55. Sperling, M. B., Foelsch, P., & Grace, C. (1996). Measur ing adult attachment: Are self report instruments congruent? Journal of Personality Assessment, 67, 37 51. Stein, D. M., & Lambert, M. J. (1995). Graduate training in psychotherapy: An integrative review. Professional Psychology: Re search and Practice, 25, 182 196. Steward, R., Morales, P., Bartell, P., Miller, M., & Weeks, D. (1998). The multiculturally responsive versus the multiculturally react ive: A study of perceptions of counselor trainees. Journal of Multicultural Counseling & Development 26 (1), 13 27. Stoltenberg, C. D. (1981). Approaching supervision fro m a developmental perspective: The counselor complexity model. Journal of Counseling Psychology, 28, 59 65. Stoltenberg, C. D., & McNeill, B. W. (2010). IDM supervision: An i ntegrative developmental model for supervising counselors and therapists. New York, NY: Taylor & Francis. Stoltenberg, C D. McNeill, B., & Delworth, U. (1998 ). IDM s upervision. San Francisco, CA: Jossey Bass Inc. Sweeney, T. J. (1992). CACREP: Precurs ors, promises, and prospects Journal of Counseling & Development, 70 667 672. Torres Rivera, E., Phan, L., Maddux, C., Wilbur, M., & Garrett, M. (2001). Process versus content: Integrating personal awareness and counseling skills to meet the multicultura l challenge of the twenty first century. Counselor Education & Supervision 41 ( 1), 28 40. Tryon, G. S. (1996). Supervisee development during the practicum year. Counselor Education and Supervision, 35 (4), 287 294. Vivona, J. M. (2000). Parental attachment styles of late adolescents: Qualities of attachment relationships and consequences for adjustment. Journal of Counseling Psychology, 47 (3), 316 329.
147 Walker, M. (2005). Critical thinking: Challenging developmental myths, stigmas and stereotypes. In D. Co mstock (Ed.), Diversity and development: Critical contexts that shape our lives and relationships (pp. 47 66). Belmont, CA: Thomson, Brooks/Cole. Wampold, B. E. (2001). The great psychotherapy debate: Models, methods, and findings. Mahwah, NJ: Lawrence Erl baum Associates, Inc. Welfare, L. E. (2006). Counselor Cognitions Questionnaire. Copyrighted instrument. University of North Carolina at Greensboro. Weisz, J. R., Weiss, V., Alicke, M. D., & Klotz, M. L. (1987). Effectiveness of psychotherapy for children and adolescents: A meta analysis for clinicians. Journal of Consulting and Clinical Psychology, 55, 542 549. Welfare, L.E. (2007). Counselor cognitive complexity: Instrument development and validation. (Doctoral dissertation). Retrieved from Worldcat. (15 6998507) Welfare. L.E., & Borders, L. D. (2010). Counselor c ognitions: General and domain specific complexity. Counselor Education and Supervision, 49 (3), 162 178. Wheeler, S. (1991). Personal therapy: An essential aspe ct of counselor training, or a dist raction from focusing on the client? Internation al Journal for the Advancement of Counseling, 14, 193 202. Whiston, S.C., & Coker, J.K. (2000). Restructuring clinic al training: Implications from research. Counselor Education & Supervision, 39 (4), 228 254. White, V. E., & Queener, J. (2003). Supervisor and sup ervisee attachments and social provisions related to the supervisory working alliance. Counselor Education and Supervision, 42, 203 218. Winslade, J., Monk, G., & Drewery, W. (1997). Sharpening the cri tical edge: A social constructionist approach in counselor education. In T. Sexton & B. Griffin (Eds.), Constructivist thinking in counseling practice, research, and training (pp. 228 245). New York: Teachers College Press. Yalom, I. D. (2005). Theory and practice of group psychotherapy (5 th ed.). New York: Basic Books, p. 559. Zimet, G. D., Dahlem, N. W., Zimet, S. G., & Farley, G. K (1988). The multidimensional scaled of perceived social support. Journal of Personality Assessment, 52, 30 41. Ziomek Daig le, J. & Christensen, T. M. (2010). An emergent theory of gatekeeping practices in counselor education. Journal of Counseling and Development, 88 (4), 407 415.
148 BIOGRAPHICAL SKETCH Sara Nash received a Bachelor of the Arts degree in political s cience from t he University of Florida. She later attained a Master of Education degree and an Ed ucational Specialist degree in mental health c ounseling from the University of Florida. Her research interest in student counselor development grew out of her experiences in preparation environments both in and out of the academy that provided vastly different opportunities for authentic engagement. expectations of new counselors may be unnecessarily low, and that more students ma y respond to challenges for greater co nnection when they feel safe, supported and can see their mentors, professors, and supervisors modeling the same behaviors She has witnessed counseling students mocking peers for risking vulnerability in classroom an d supervision settings, as well as heard ins tructors and supervisors state that attending to and emotional disclosure was not their responsibility. She wonders often about what would happen to restricted novice development if counseling program cultures promoted and expected more authenticity, engagement, and interpersonal process from students, professors, and supervisors alike. Sara is currently an assistant clinical professor at the University of Florida where she practi ces psychotherapy and helps coordinate crisis and emergency response for the campus. She also sees counselors in private practice, plays music and paints as often as she can. Out of this dissertation study she intends to develop an experiential training to prepare student counselors for the relational challenges of crisis work, and advocate for increased attention to relational constructs in CACREP training programs.