THE IMPACT OF RELIGIOUS CAREGIVER TRAINING ON PARAPROFESSIONAL PARTICIPANTS BY Martin Jay waalkes A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1992 /(J} ‘7'.) /\ ABSTRACT THE IMPACT OF RELIGIOUS CAREGIVER TRAINING ON PARAPROFESSIONAL PARTICIPANTS BY Martin Jay Waalkes Sixty one male and 146 female participants were solicited from 20 churches conducting Stephen Ministries (SM) Christian caregiving training to identify personal change among participants. The sample included 29 previously trained SM graduates to identify changes due to ongoing caregiving participation, and a control group of 44 Bible Study (BS) participants. Testing conducted at the beginning and end of training included the Gordon Personal Profile (GPP), the Structural Analysis of Social Behavior-- short form (SASB), the Helping Beliefs Inventory (RBI), and the Adjective Check List (ACL). Multiple repeated measures analysis of variance procedures with socioeconomic status, gender, and age covariates were used to test hypotheses. Unhelpful beliefs for assisting others with problems, as measured by the RBI, were less strongly endorsed by previously trained SM participants at the initial testing. By follow-up testing, new SM participants developed these same beliefs while BS participants remained unchanged. The AOL, GPP, and SASB showed no evidence of change due to not 9‘. covari, either training or caregiving experiences. Post-hoc analysis suggested previous research with analogous religious caregiving training may have over-interpreted personal change attributed to training by failing to include adequate control groups, neglecting sufficient follow-up, or not giving adequate consideration to the effect of covariates. DED I CATI ON For Ellen, You help make all my dreams happen. iv was a' Series satxs: away w - careg. as we; the f; for tr. ACKNOWLEDGEMENTS This was my grandfather's idea. Martin Wallace Waalkes was an established octogenarian when he attended Stephen Series Leadership training. As a family we watched with satisfaction the way his qualities, which we had all appreciated for decades, found formal expression in teaching caregiving to others. I think it made a difference for him as well, in some sense providing personal confirmation of the findings that if one is at least active, no matter what the activity is, some changes happen. Thank you, Grandpa, for the idea. Several people participated in the development of the idea. John Stapert, Ph.D., Gary Voss, M.Div. director of training at Stephen Ministries, and John Powell, Ph.D. of my guidance committee all helped me formulate a sense of what changes I should expect to find. Norman Abeles, Ph.D., who was also the chair of my master's effort has provided excellent guidance for my career and development as a psychologist in addition to serving as the chairperson of this effort. Ray Frankmann, Ph.D. provided many helpful comments on data analyses as well as helpful suggestion on the whole draft in Dr. Abeles' absence. Robert Caldwell, Ph.D. provided helpful technical assistance in developing drafts into accurate documents in both the proposal and final forms. Dr. Caldwell also interrupted his sabbatical to assist in the final defense. Gershen Kaufman, Ph.D. the final member of my committee also deserves appreciation for his participation. Thank-you all! A special thank-you is extended to John McCormick, a friend who, without accepting compensation, rescued my data from a cryptic ASCII doom and wrote the scale-scoring programs with some computer ”Black Magic”. Short of tedious and likely error-filled hand re-entry of thousands of scores, the data would have never seen analyses. Ellen waalkes. for whom this is dedicated, often worked harder on this than I did. Her efforts at keeping the house running, providing finances, coding data, and watching the details of registration and loans often went unacknowledged, but made it possible. A final thank-you to all my participants and group leaders. The testing was an extra task in an already busy curriculum. I hope these results feel confirming of your experience. Vb .at the TABLE OF CONTENTS Page LIST OF TABLES .......... .............................. ix LIST OF FIGURES ....................................... xi INTRODUCTION ...................... . .................. 1 Trends in mental health care .................... l The growing trend towards church provided human services ........... ........... ........... 2 The professional / paraprofessional trend ....... . 6 Paraprofessional identity and training ........... 7 Paraprofessional effectiveness ................... 8 The movement within churches for increased lay involvement in pastoral care, including mental health ministries .............................. 15 The variety of available lay counseling models and training programs ...... ......... . ..... ..... 17 Reasons participants volunteer for lay and mental health service ........ .............. 26 Personality changes from training and experience as counselors ................ ..... .. 22 What has counselor training been observed to change? ..................................... 35 Empathy changes from training ............... 36 Locus of control changes ..... ........ ....... 37 Interpersonal functioning changes ........... 37 Anxiety and confidence changes .............. 38 Summary of personality changes as a result of training ........................ 40 What qualities are essential to the therapeutic relationship? ............ .......... 41 Empathy, warmth, and genuineness ............ 41 Interpersonal style and behavior ............ 43 Cognitive style ............................. 45 Interests and abilities ..................... 48 Attitudes, beliefs, and values .............. 48 Gifts of the Spirit ......................... 50 Summary on therapist characteristics ........ 51 Hypotheses ................................. ..... . 52 METHOD ......... ....................... . ............... 57 Participants ................ .......... ........... 57 Procedure ........................................ 59 Training courses ................................. 61 Materials ........................................ 61 Instruments ...................................... 61 The Adjective Checklist ..................... 61 The Gordon Personal Profile ................. 62 The Structural Analysis of Social Behavior .. 64 vi Viv-n!" D 1'" “and... The Helping Beliefs Inventory ............... 66 The Two-Factor Index of Social Position ..... 66 RESULTS ..... . ......................................... 68 Dropouts and predispositions ........ ............. 68 The hypotheses of change ......................... 79 Using the traditional method for analysis of change ............... ...... .. ....... . .......... 74 DISCUSSION ............................................ 77 Summary of findings .............................. 77 Implications for the role of Stephen Ministries training .. ......................... ........... 85 Implications for other research .................. 98 Summary .......................................... 93 NOTES 000000 0.0.0 0000000000000000000000000000000000000 94 Note 1: Use of the word "church" ................. 94 Note 2. Use of the word ”laity” .................. 94 APPENDIX A ............................................ 95 Churches from which the sample was drawn ......... 95 APPENDIX B ............... . ......... . ..... . ..... ... 98 Group descriptive statistics ...... . .............. 98 APPENDIXC 0.0.0.0... ...... ...... OOOOOOOOOOOOOOOOOOOOOO 10“ Comparison of participant scores on the ACL and GPP with available norms ... ......... ....... 190 APPENDIXDOOOOOOOO 00...... ..... 00.000.00.000. ..... .106 Outline of the Stephen Ministries training program and course content ............ ........ . 186 ”PENDIX E O O ........... O .0 O O O. O I .0 O. I. OOOOOOOOOOOOOOOO 110 Letter of reference from Stephen Ministries ...... 118 APPENDIXFO...............OOOOOOOOOOOOOOOO ........ 0.0.112 Letter to trainers describing terms of part1c1pation O I O O O O O O O. O O I O O... O O. ......... .0 O O 112 APPENDIXGO ..... .....OOOOOOOOOOOOO 000000 00000000. 00000 116 Code correspondence form ... ..... ...... ......... .. 116 APPENDIXHOO ..... 0......OOOOOOOOOOOOOOOOOOOOOOOO ..... .117 Consent cover letter for research participants ... 117 APPENDIXI OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO119 Cover letter to trainers providing instructions for administration of instruments .............. 119 ”PENDIXJO.......OOIOOOOOOOOOIOOO-OOOOO......O...0.... 123 Narrative description of results provided to participants ................................ 123 vii 1‘: 1.395532). Der Des- APPENDIX 5 Repea 5A3 Demographic information collection form .......... 128 APPENDIX L ............................................ 129 Description of the Adjective Checklist ........... 129 APPENDIX M ....... .................................. 138 Description of the Gordon Personal Profile ....... 138 APPENDIX N .0. ........ .OOOOOOOOOOOOOOOOOO 000000000000 132 Description of the Structural Analysis of Social Behavior ................................ 132 APPENDIX 0 ........................................... 137 Description of the Relping Beliefs Inventory ..... 137 APPENDIX P .. ......................................... 139 Means and standard deviations for subjects completing both pre and post- -test ... ........... 139 APPENDIX 0 ............................... ' ......... .... 148 Table 16 ......................................... 148 APPENDIXR 00.0.0... OOOOOOOOOOOOOOOOOOOO O... OOOOOOOOO 150 Participant predispositions and characteristics of non- persisting participants ................. 158 Persistence main effects ..... .. ............. 158 Group type main effects .............. . ...... 158 Group x persistence interactions ........... . 151 APPENDIXS OI... 000000000000 OI... ........ O... 0000000000 155 Repeated measures 2 x 3 (group x time) ANOVA SASB pOSt‘hOC main-efoCts ooooooooooooooooooooo 155 APPENDIX T .................... ....... ... ....... ....... 157 Examination of variables with a relationship to the Helping Beliefs Inventory and comparisons to previously measured populations ............. 157 RBI score relationships with other variables.................... ............. . 157 Comparative values of the RBI scores to other known populations ............ ....... 158 REFERENCES .... ................... . ............ ........ 168 viii 11 12 13 14 15 LI ST OF TABLES Table Page 1 Pre and post-test RBI means, standard deviations, and 95 % confidence intervals by group . ......... 71 2 Mean pre and post-test scores for NSM participants on variables with significant paired t-test values .......................................... 75 3 Summary of significant ACL scale changes resulting from lay and CPE education in caregiver technique ............. . ............... 83 Appendix Tables 4 Characteristics of participating groups ........... 98 5 Means and standard deviations of participant ACL raw scores compared with a normative sample ..... 181 6 Comparison of female participant GPP scores with a general adult female sample ................... 184 7 Comparison of male participant GPP score with a sample of male lower lever managers ............. 185 8 Stephen Ministries chapter modules ................ 186 9 Format for typical Stephen Series training 8e881on3 0......O.......OOOOOOOOOOOOOOOOOOOOOOOOO107 l8 Topical resources available in the trainee manual . 188 11 Split-half reliabilities for the Gordon Personal Profile scales .................................. 131 12 Patterns of scores and their clinical manifestations for ”Self" and-”Other” surfaces in the SASB CO0.0.0.0.000...0.00.00.00.00...00.0. 135 13 Aspects of the interpersonal relationship assessed by the SASB short form ................. 136 14 Item content for the Relping Beliefs Inventory .... 137 15 Means and standard deviations for subjects completing both the pre-test and post-test ...... 139 ix 16 17 18 19 Dependant variables with a relationship with age, SES, and gender ....................... ..... 148 variables with significant interactions using 2 x 2 repeated measures (persistence x group type (CSM excluded)) ANOVA with covariates age, SES, and gender for pre-test measures .......... ...... 152 SASB variables and corresponding F-ratios found to have significant main effects for group type in 2 x 3 (Group x Time) repeated measures ANOVA's using age, SES, and gender as covariates ...................................... 156 Means and standard deviations on the RBI for selected samples ........................ ..... ... 158 LIST OF FIGURES Figure Page 1 Mean Relping Beliefs Inventory score at pre and post-test ............. . ....... ............. 73 Appendix Figures 2 Diagram of the axes and quadrant qualities of the "Other" surface of the SASB ........... ..... 133 3 Diagram of the axes and quadrant qualities of the ”Self" surface of the SASB ........... ...... 134 xi T n e t l e t Car Hobbs (1964) has called the increased use of non- traditional mental health solutions the "Third Revolution" of mental heath care. The clinical model, where a professional focuses wise and trained attention on a single client has been deemed not just ineffective, but short sighted as well, since it does not consider the problem of individuals as part of a larger problem of the community. Hobbs (1964) cited Rollingshead and Redlich (1958) and Albee (1959) as influential to his thinking about the need for a revolution in how mental health services are provided. Hobbs noted the growing evidence that community mental health needs will not be met through traditional professional interventions. The Joint Commission on Mental Health (JCMR) echoed this same sentiment in their report of a 5 year study in 1961 (JCMR, 1961). A few years later, Cowen, Gardner, and Zax (1967) suggested that mental health service providers would need to increase their number by 25 percent to 75 percent to respond to the demands of people desiring services. In more recent years, cutbacks in spending on mental health services has again reduced the capacities of mental heath public service agencies to respond to community needs with the zeal they might wish. Siegel (1973) outlined the growing trend in mental health towards utilizing volunteers, like college students, seniors, homemakers, and patients, as agents of mental l «.1 health c to incze populati Cemzunit; response Siegel hi been EXpe assistanc Chu: attempted COmmunit) this esCa damand fc perhaps c utilizati training. charches. psych0109 net ‘3390).; (BufEOrd ’h Pu‘ health changes. He outlined these changes as the response to increasing demands for clinical treatment from a growing population of deserving people. Siegel identified the Community Mental Health Centers (CMRC) as the central response to this developing need for mental health workers. Siegel hinted at the other new areas where services have been expected, suggested, and demanded such as, employee assistance programs, and church communities. Churches are only one of many organizations that have attempted to respond to the mental health needs of the community as well as their own congregations. One result of this escalation of available services has been an awakened demand for mental heath service providers. The demand has, perhaps out of necessity, been addressed by increased utilization of individuals without traditional academic training. Furthermore, many organizations, including churches, have an interest in providing their members with psychological, rather than relying on professionals who may not espouse their particular viewpoint on mental health (Bufford & Buckler, 1987). i‘ 0. l' (0, 0.5 0." I. 4" 0 0", 2.113,! The church (Note 1) has has its own motivation for becoming involved in the movement to train paraprofessionals in mental health treatment aside from the current shortage of professionals in the church and community. Counseling faculty and counseling courses offered at United States accredited Protestant seminaries have more than doubled in the peri by only percent percent counselir changes t number of increasin Note 2), Church 1111; Programs ‘- Pastc applicatic “Var: bee Dentally 1 1n the Car of healing to draw dt1 the period from 1965 to 1988, while enrollment has increased by only one third (Linebaugh & Devivo, 1981). Of the 72 percent of the seminaries that responded to the survey, 53 percent had a requirement for a course in pastoral care or counseling, and over half of the respondents were planing changes to increase training and practica in this area. The number of books available on Christian counseling, and the increasing number of lay counselor training approaches (see Note 2), speak to the increasing focus on this aspect of the church mission (see Tan, 1991 for a review of current programs and books). Pastoral care was, and is still, seen as the practical application of theology (Glasscock, 1971). The church has always been called to be active in its compassion for the mentally ill (Clinebell, 1965). The Christian church's role in the care of mental health has roots in Christ's mission of healing and teaching. Often referenced by those who want to draw attention to the church's mission in this regard is Christ's parable of the good Samaritan (Luke 18:29-37) (e.g., Glasscock, 1971). Johnson (1973) notes that this mission was transferred to the disciples, and then sponsored by the church through the middle ages by way of the church's involvement in universities and hospitals. This involvement has taken a professional character since Anton Boisen, a former hospitalized psychiatric patient, became the first chaplain in a mental hospital (Boisen, 1968). Boisen later established the clinical pastoral education program in the f V Ucrchest experien‘ of his b1 1936), pa for this Row. Proponent mental he 1969), he the funda "men an: Stated tn "lated, ‘ Heldenti ChUrCh th “darn 1a Vbil. Meds Of ChUrch as pr°blems . 42 Perch. Clergy f0: Worchester State Hospital in 1925. The story of his own experience with mental illness, outlined in the introduction of his book MW (Boisen: 1936), parallels the growth in the church's sense of mission for this population. Roward Clinebell, Jr., one of the more prominent proponents for the church's involvement in the area of mental health ministries (1965; 1971; Seifert & Clinebell, 1969), has argued that mental health is directly linked to the fundamental purpose of the ChUIChf‘tO increase among women and men the love of God and neighbor. Clinebell stated that mental and spiritual health are inseparably related, each dependent on the other for full expression. Re identified mental health as a concern of the Christian church throughout the centuries, and called mental health ”a modern label for an ancient concern" (1965, p. 21). While the church has felt the call to minister to the needs of their community, communities have also seen the church as a source of consolation and assistance for problems of mental health. In 1968, it was estimated that 42 percent of people with emotional problems turned to clergy for help, compared with 29 percent for physicians and 18 percent for psychologists and psychiatrists (JCMR, 1961). More recent estimates of the percentage of people who would first turn to clergy is in the range of 39 to 78 percent (Cerling, 1983). Within the church it has been clear that the church is called to provide for the mental health needs e r of members and non-members alike. When judged by such referral trends, the recognition of the church as a provider of mental health assistance by the secular mental health community has been, in contrast, somewhat limited (Gorsuch & Meylink, 1988, Meylink & Gorsuch, 1987). Tan (1991) suggested that the church could also serve as a more effective source of care for those who are religiously committed. It is proposed that a Biblical approach to counseling (whether conducted by lay or professional counselors) that explicitly utilizes Christian religious values or perspectives and interventions (e.g., prayer and the use of scriptures), and relies on appropriate spiritual gifts and the power and ministry of the Holy Spirit, makes unique contributions to counseling effectiveness, especially with religious Christian clients. (Tan, 1991, p. 188) Additionally, churches are increasingly being asked by their congregations to provide these services. Tan (1991) noted that while professionals providing mental health services may be reluctant to use religious values and interventions with clients, people are seeking out services for emotional and psychological problems with a distinctive Christian emphasis. Those involved in the emerging lay counseling W.) t) *2 Par pro 1351 the: ministries are particularly eager to provide these integrated interventions. The manner in which these services are provided has not, however, been marked by a consensus as clear as the perceived need. Conflicts regarding the manner in which counseling services are provided range from espousing prevention versus counseling (Prater, 1987) to the controversy of using carefully identified semi-professionals versus training larger numbers to minister from within a congregation (see Emerson, 1986; Estadt, 1986; and Tan, 1986 for a full discussion of these controversies). W As a result of forces from the community and the church, the paraprofessional has emerged. One suggestion from the President's Commission on Mental Realth (1978) was a sharper definition of the role of paraprofessional training, supervision, credentialing, utilization, and compensation. They stated that ”No one can ignore the contribution they have made or the need to increase the effectiveness of that contribution." (p.48) The identification of who qualifies as a paraprofessional or a non-professional has been less clear than the obvious need for additional mental health service pmoviders. Karlsruher's (1974), solution to the problem of labeling was to refer to anyone with a degree in a mental health field beyond a bachelor's degree as a professional therapist. All others, regardless of credentials, would then be non-professional therapists. The President's Commission on Mental Health (1978) identified the non- professional as a catchall term for mental health workers who have less training than the traditionally identified professionals, but more training or experience than a custodial caretaker. Paraprofessional Identity and Training; Paraprofessionals are often trained in context specific training programs devised by professionals or agency administrators. Programs differ in the number of hours in training and continued supervision. Likewise, subjects covered or emphasized differ greatly. Some organized training programs do exist--many offered as skill courses through school district adult education or community colleges. The President's Commission on Mental Realth (1978) cited the existence of 345 paraprofessional training programs, many in community and four-year colleges, graduating 18,888 students a year. The different types of paraprofessionals are generally recognized as falling in two categories: ”ubiquitous" and ”indigenous” (Reiff & Riessman, 1965.) Ubiquitous paraprofessionals are people who do not have professional training, but are from the same social strata as professionals, and more readily identify with the professionals than the clients. Siegel (1973) described this type of relationship as "the professional therapeutic role with its built in features of one-way intimacy, E1 distancing, and a supraordinate-subordinate relationship (which) may prove to be less effective in the long run than a more egalitarian, peer-to-peer relationship” (p.153). Ubiquitous paraprofessionals, encouraged to think and behave in ways that emulate professionals, are typically trained in a formalized, conventional atmosphere. Indigenous paraprofessionals are typically similar in background and perspective to the clients they serve. The roles the indigenous paraprofessional can fill range from advocate and expediter to direct care provider and companion. Indigenous paraprofessionals are typically employed in a setting that values their worth for the type of person they are already. As a result, training with this type of paraprofessional is usually limited to helping them get acquainted with services and resources agencies offer so they can act as referral agents in their community. MW Siegel (1973) stated that community volunteers have been found to be useful as change agents in a variety of settings, and that no research had been published to indicate that volunteers were harmful. Re concluded that "Volunteers, then, may be seen as an important future source of mental health manpower” (p.151). Other reviews suggest that, far from being benign, paraprofessionals may demonstrate effectiveness equal or superior to professionals. Karlsruher's (1974) review of non-professional effectiveness found overwhelming support for the capacity of paraprofessionals in inpatient settings to influence many indices of dysfunction like overt behavior, performance on psychological tests, and discharge rates from treatment. Karlsruher noted that, the lack of controlled studies made it difficult for firm conclusions to be drawn about outpatient groups, but it seemed improvement from previous levels of functioning were possible with paraprofessional treatment providers. According to Karlsruher's conclusions, paraprofessionals seem to have a beneficial effect over no- treatment control groups regardless of the mode of intervention, the length of treatment, or the criteria used to judge the outcome. Perhaps the most controversial review of paraprofessional effectiveness was that of Durlak (1979) with the reply by Nietzel and Fisher (1981) and rebuttal by Durlak (1981). Durlak discussed the reviews of Siegel (1973) and Karlsruher (1974), as well as some other reviews, and noted generally positive findings, and a surprising lack of attention directed towards the effectiveness of paraprofessionals as compared to professionals. Durlak's (1979) results were very supportive of the utility of ubiquitous paraprofessionals as mental health therapists. Overall, outcome results in comparative studies have favored paraprofessionals. ... In terms of measurable outcome, there were no 18 significant differences among helpers in 28 investigations, but paraprofessionals were significantly more effective than professionals in 12 studies. The central finding from these comparative studies is that the clinical outcomes that paraprofessionals achieve are equal to or significantly better than those obtained by professionals. The provocative conclusion from these comparative investigations is that professionals do not demonstrate superior therapeutic skills, compared with paraprofessionals. Moreover, professional mental health education, training, and experience are not necessary prerequisites for, an effective helping person. (Durlak, 1979, pp. 84-85.) Durlak (1979) limited this generalization to ubiquitous therapists because of the limited number of theoretically sound and controlled experiments involving indigenous paraprofessionals. Interest, enthusiasm, and a capitalization on non—specific treatment influences have been suggested as reasons that paraprofessionals have such a comparatively strong influence, but no research has been reported that makes any conclusive statements about the differences. Durlak suggested that it would be a mistake to continue using paraprofessionals without understanding their unique skills, limitations, and deficiencies. Iql In!!!» ”I NH. ..... Dr: 11 Durlak's (1979) conclusions were questioned by Nietzel and Fisher (1981) who charged that Durlak did not effectively address the issue of the difference in treatment effectiveness of professionals and paraprofessionals. Nietzel and Fisher claimed the threats to the internal validity (like confounded variables, choice of outcome variables, non-random assignment, and equal consideration regardless of quality of experimental design) overwhelmed the conclusions of the review. In many cases Nietzel and Fisher revealed inconsistencies in the use of the terms professional and paraprofessional. They went on to point out that Durlak accepted the failure of may of the studies to find difference as implying equal performance, when the statistical power to make such a claim was adequate for only the counseling and guidance studies. In his reply, Durlak (1981) addressed many of Nietzel and Fisher's (1981) concerns. He argued that the professional paraprofessional distinctions he used were the practical and realistic ones used in practice, and that methodological criticisms have sound replies. Durlak appeared to accept much of the critique by Nietzel and Fischer as reasonable, but emphasized that the most salient issue that still remained was the lack of justification for professional training, experience, and outcome when it comes to explaining the lack of significant differences by professionals. of me 12 In an attempt to resolve the somewhat muddled picture resulting from the box-score analysis of the paraprofessional effectiveness research, Rattie, Sharpley, and Rogers (1984) applied the more sophisticated techniques of meta-analysis to the issue. Rattie et a1. (1984) concluded that ”clients who seek help from paraprofessionals are more likely to achieve resolution of their problem than those who consult professionals” (p.534). Just as Durlak (1979) found opponents to his methods (Nietzel & Fisher, 1981), Rattie et al. (1984) were accused of methodological errors that biased their results in favor of paraprofessional effectiveness (Berman & Norton, 1985). Berman and Norton questioned various inclusion criteria and statistical techniques, running their own analysis on a more stringently screened set of data. This analysis revealed no effect size differences explained by therapist training regardless of presented symptoms, orientation of therapist, or group versus individual counseling. Professionals were found to be more effective at follow-up (as opposed to post- treatment measurement) and in shorter treatment than paraprofessionals, and were more effective in studies with older clients. Paraprofessionals demonstrated better effectiveness with younger clients. The effectiveness of lay religious counselors, has not received near the attention given to the study of paraprofessionals in general. Collins (1987) noted that he .knew of no competent study which investigated the De: on: W the Co; COr Prj PE: 13 effectiveness of lay counseling among Christians. Re quoted Propst (1985) stating, "So many people have become involved in paraprofessional training that the idea of its effectiveness seems to have been forgotten" (cited in Collins, 1987, p. 7). Tan (1998) noted one unpublished study which had showed some "generally favorable results" (p.63), but suggested that effectiveness research with random assignment and appropriate control groups was a significant priority for future research efforts. Neither Tan (1998) or Collins (1987) referenced the study by Harris (1985) which, with some methodological limitations, addressed the issue of effectiveness of "lay- persons who have experienced crises similar to the client” (p. 168) among 44 pastoral counseling clients. This is the only effectiveness research on lay counseling that could be identified for this review. In the study, Rarris compared the changes on a semantic differential scale measuring self- concept for clients assigned to one of four treatment conditions. Clients received either private therapy, private therapy and a group, private therapy plus a resource person, or private therapy plus both the group and a resource person. There was not a ”no treatment” control or a group which received the resource person only. While the statistical design was rather crude, utilizing t-tests and simplistic comparison techniques when a 2x4 (time x group) repeated measures analysis of variance was more appropriate, Harris's results suggested that the presence of the resource . a”. q r no a1! Pi< 311$ 90;; Uh; Pro, Pro} Effe 14 person in the treatment made the effect of the therapy more powerful. Clients who received the resource person showed more dramatic change on the self-concept measure than those who received only private therapy, or therapy plus a group. While a number of methodological shortcomings limit the conclusions of this research, the results provided a suggestive starting point for research in paraprofessional pastoral care. The last word has not been written on this subject, yet some conclusions can be drawn. Some of the initial zeal over paraprofessional effectiveness promoted through reviews, such as that of Karlsruher (1974), and predictions, like those of Albee (1959) and Cowen et al. (1967), has worn off. What has been left, however, is the realization that non-professionals are a realistic mental health care alternative for secular or religious contexts. The present picture of budget cuts and increasing demand for services suggests non-professionals present realistic alternatives to traditional service models, particularly for circumscribed populations, problems, or services. Even critics of reviews which suggest paraprofessional superiority over professionals (e.g., Berman & Norton, 1985) concede that professional training has not been demonstrated to increase effectiveness for some types of problems. we: .4! OI! in' tm .1: IQV'U'! W ..1 .183 "or u! ‘="0 a VOAV‘U‘J WW Certainly one of the primary motivating forces in the church is the perception of God's will. Relevant to paraprofessionals, Glasscock (1971) mentions two passages of scripture that speak to the issue of the increasing involvement of laity in traditional pastoral care responsibilities. Glasscock referenced Ephesians 4:11-12, where personal qualities are seen as gifts from Christ with a purpose of strengthening God's people. Glasscock (1971) also references Exodus 18:13-23. This passage is a description of Moses being advised that it is prudent that he appoint judges to share the burden of the Lord's work so that the people would be served without wearing out the ordained clergy. The Exodus passage suggests lay involvement as a practical and responsible solution to keep the church leaders from being overwhelmed by their tasks. Estadt (1986) observed the increasing role of the laity in all aspects of the Roman Catholic church short of administering sacraments. The limited availability of ordained priests has made the practical utility of pastoral care by clergy an impossibility in many communities. In some remote areas lay ministries are not simply supports, but often the only available resource. Clinebell (1965) traced the increasing involvement of the laity in missions to ”the rediscovery of the New Testament truth that every Christian has a ministry simply rr 1“. [1’ 0'1 v“ C} 02 Q) 1a Ch th 16 because he is a Christian." (p.262) Clinebell suggested that laity are not second-class or second-string, but an essential and central part of the ministry of the church. By 1971, Clinebell said it was probable that this rediscovery is taking hold. He states that it is probable that: - M- a :‘I Ministry will increasingly be seen as the function of the entire congregation; laymen will become deeply involved in the church only if they are trained for significant person-centered ministries of pastoral care and social change; a central task of the clergyman will be to recruit, train, and coach lay task forces for these ministries. (Clinebell, 1971, p.35) Lay counseling (or caregiving) is one area at the forefront of recent increases in lay involvement. Lay Christian counseling, as described by Tan (1991), is helping or counseling that is ”done by Christians, usually with an explicitly Christian or biblically based model or framework of people helping" (p. 174). Churches have quickly adopted lay ministries programs. Tan (1998) suggested that Christian lay counseling already poses a significant part of the church outreach ministry and promises to develop and mature even more in the 1998's. '5 C‘? Pa; De] Sc} CCU bro it: hen) 17 e o va b e a un lin M dels and a Eregrams Collins and Tornquist (1981) estimated from the results of an informal North American survey that 28 published evangelically-oriented lay counselor training programs exist. Tan (1991), in a review of the most popular training approaches noted 21 books published since 1978 on the topic. After a review of the current state of lay counselor r" 13*.“ -' training, Clinebell (1971) identified two basic types of programs: "the personal growth and awareness approach," and the ”plunge in human needs approach.” The ”personal growth and awareness approach" addresses a trainee's experience of self and interactive style, particularly as related to attempts to be helpful towards others in relationships. Participants in this approach then apply this self-knowledge to circumstances where they feel they can be beneficial. The ”plunge in human needs" groups focus on specific therapeutic skills. Practica-like cases are used by participants in supervision to hone skills formed in a period of initial training. Additional regular training is scheduled for skill building and personal growth for needs made evident through experience with helping relationships. Tan's 1991 review of the state of lay Christian counseling identifies three basic types of lay counseling programs. The “informal spontaneous" programs and efforts attempt to educate large numbers of leaders and interested members in helping and caring skills in the hope that 18 participants can employ new effectiveness in the context of naturally occurring informal relationships. Sometimes offered as specific training, it could also be taught as a unit of a Sunday school or similar class. This approach is most similar to Clinebell's (1971) ”personal growth and awareness" type of program. The second type of lay Christian counseling program Tan (1991) described was the ”informal organized” model. In this model, applicants for training are screened, education is more extensive, and the caring relationships which are formed are supervised by someone of greater experience or professional training. While the nature of the contacts may have an informal quality and occur in a variety of settings, appointments are arranged, referrals are solicited, and identified pairings of caregivers and recipients are made. Tan noted that this second model works well in some ethnic- dominated congregations where receiving traditional counseling services is a sign of weakness and carries a strong stigma. Often the goal of counseling in these first two models is simply to provide effective and empathic support for recipients. This approach, which emphasizes "standing with” (Lukens, 1987; p.11) recipients in times of distress capitalizes on the particular value of lay caregivers. These first two training models typically attempt to facilitate and focus these skills as resources for caregivers. C1 de 95 re aIn la: in: ”(5 195 19 The third model of lay Christian counseling Tan (1991) identifies is the "formal organized" type. This model uses trained, supervised lay counselors alongside professionals in a formal ministry. Those receiving services from this third type of ministry typically go to the church, counseling center, or para-church organization for scheduled appointments. It is this third model which has come under the most critical scrutiny. Clearly ethical and legal implications of conducting treatment such as counseling by a non-licensed or non-certified paraprofessional raises many legal and ethical questions (Becker, 1987; Collins, 1987; Emerson, 1986; Worthington, 1987). Theological issues have also been raised about the role of the lay, or for that matter the role of the church in providing such formal counseling (Prater, 1987). It is particularly in this third model where churches are confronted with the challenge of determining the role of lay leaders in the church in general, as lay counselors engage in the same style of relationships traditionally reserved only for clergy (Emmerson, 1986; Estadt, 1986). To provide a more complete picture of a comprehensive and structured training program typical of current trends in lay ministries training, the program objectives, content, and structure of the Stephen Ministries' ”Stephen Series” (Raugk & McKay, 1985; Smith, 1986; Stephen Ministries Inc., 1983) are outlined in Appendix D. I: fix he an 28 :‘et-;: few 3-.) s Volu e- or _$ _ - __-- an: '1 Many factors exist which motivate volunteers to become involved with mental health service. A pioneer in developing volunteers as paraprofessional mental health care providers is Margaret Rioch. Rioch (1967) utilized a group of people with high motivation and ability become involved with a volunteer mental health project. Rioch and the others involved with the project (Rioch, Elkes, Flint, Blanche, Newman, & Silber, 1963) worked with women described as housewives with grown children. Describing these women as a vast under-utilized population of helpers, she saw this population as particularly attractive as volunteers because of their ability to realize their own developmental changes by utilizing the natural skills acquired through the experience of parenting children. Many other groups of possible natural helpers exist in today's culture. Adolescents, retired people, what remains of the "empty- nest” mother, and some subgroups of disabled adults are ideal candidates for the increased sense of purpose and meaning afforded by participation in peer counseling programs. Rioch et al. (1963) advocated that organizations recruit individuals with the natural motivation and availability who might be willing to volunteer for mental health service. Consequently, there has been a considerable amount of speculation about the natural motivations that 9! '1 1 EX: ‘5 in; C01 Che di: nee Sc: C01 tn 21 might predispose people to offer their time (c.f. Cerling, 1983; Rolzberg, Knapp, & Turner, 1967), but little evidence of the actual differences in people who have volunteered exists. Knapp and Rolzberg (1964) were among the first to examine these differences when they compared 85 male college students volunteering for the ”companion program" at an inpatient psychiatric hospital (see Rolzberg, Gewirtz, & Ebner, 1964). Volunteers were compared to 85 student controls on personality tests. Negative personality characteristics that might reflect clinically significant differences were not found among volunteers using such measures as the MMPI and Edwards Personal Preference Schedule. Volunteers presented themselves as more compassionate, morally concerned, introverted, and religious than the control subjects. Differences were not found for measures of intellectual ability. Rersch, Kulik, and Scheibe (1969) also measured differences in 151 college volunteers at a mental health facility. The difference with this study from the research of Knapp and Rolzberg (1964) is that the participants in the Rersch et al. study were committing to a 488 hour summer experience. Rersch et al. compared these volunteers to 142 demographically matched control summer school students. Subjects took a battery of psychological tests including the California Psychological Inventory (CPI), Adjective Checklist (ACL), a biographical questionnaire, and other 22 research instruments. Significant differences were revealed on the CPI with the group of volunteers appearing more mature, tolerant, controlled, achievement oriented, and flexibly minded (Rersch et al., 1969). Male volunteers were noted to be more nurturant and less socially poised than non-volunteer males. ACL variables were noted to confirm this pattern, with females exhibiting more independence and less succorance than peer controls. Male volunteer ACL variables showed more qualities of abasement than their peer controls. Biographical data and interest inventories suggested volunteers were more service and volunteer-minded and are more likely to pursue mental health related career paths than the control participants. Rersch et al. concluded that volunteer behavior reflected "a controlled drive for independent achievement and sensitivity to human problems” (p. 34), rather than an over-concern with personal problems. W Snuneelora. Glasscock (1971) and others (Clinebell, 1971; Osborn, 1983; Smith, 1986) have noted that the people ministering to the needs of others have reported changes in way they see themselves which transcend the counseling relationships. This phenomena is not restricted to the therapeutic interaction, and has been reported to occur simply as the result of training. 23 Collins and Tornquist (1981) note that the emphasis given to programming oriented towards counselor personality changes or personal insight varies with the model of training employed. Most published programs, however, place little emphasis on personal growth in training or applicant screening (Cerling, 1983). Training counselors under a skill-based training model would de-emphasize such experiences, while programs more experiential-didactic in approach might exploit personal growth in hopes that insight could be used in the therapeutic interchange. Among the literature on Christian lay counseling, such observations of personal change have not been empirically evaluated. Observations of change have often been offered in good faith, yet it is hard to imagine promoting a program which had a negative effect on the participants. Additionally, Utterback and Reyman (1984) suggest that volunteers would be unlikely to report a negative picture of the programs in which they have invested large amounts of their extra time. Despite the lack of research, the perceptions of positive change exist. Lukens (1883) noted that 88 percent of the 35 respondents that he surveyed indicated a sense that the program had been personally meaningful for their own spirituality and relationships. Some have suggested that the extraneous benefit of personal change could even be sufficient justification for offering training (Siegel, 1973). Clinebell (1971) reports of trainees being "turned CI Ste 3 1 ex; in 24 on" to relationships (p.37) and parents reporting new empathy and understanding in their relationships with adolescent children. The Stephen Series promotional materials state that, ”By participating in Stephen Ministry, Christian lay persons will-- ... grow in a sense of self-worth and personal accomplishments as they use the gifts God has given them ... experience growth in their personal lives as they use newly acquired relational skills with people whom they meet every day ... grow spiritually, as they experience the Lord's loving presence in fellowship with other Christian caregivers (Stephen Ministries, not dated, p.17). Smith's (1986) review of the implementation of the Stephen Series in the Reformed Church in America quoted a pastor as reporting "the Stephen series is a growing experience for the volunteers in terms of their faith in God and their abilities to understand others" (p.15). Osborn (1983) paraphrased Clinebell (1966), saying ”an awareness of ministry gives the Christian lay person a new self-image” (p.57). In her family therapist training program Osborn observed that ”training opened up the trainees to considerable growth and development in their own lives and personalities” (p.58,). t? 25 Among secular volunteer training programs some empirical validation of claims have provided confirming evidence of therapist changes from the combination of training and experience with helping relationships. One of the earlier programs for training volunteers for working with patients in a psychiatric hospital directly addressed this issue of personal growth. Jules Rolzberg and his associates (Rolzberg et al., 1967; Knapp & Rolzberg, 1964) trained college students in a "companion program" in preparation for interacting with patients at a state hospital. Aside from the benefit they believed the patients would gain from the interpersonal contact, one of the program goals was to provide the students with "a personal maturing experience that would equip them to deal with their own adjustments to life” (Rolzberg et al., 1967, p.92.) They focused on changes that might occur in moral judgment, self acceptance, and other personal qualities of the volunteer. The students' orientation focused on their role as a "friend” to the patients (Rolzberg et al., 1964), and relationships consisted primarily of talking and other one- on-one activities with patients in the lounge, canteen, or other hospital environments. Part of the training included a group with other volunteers and a professional who provided education on mental illness and allowed the volunteers to process their experiences. Rolzberg et al. (1964) examined changes in attitude on the part of students as they compared to control volunteers vorkin< health a more sexual psychie items : Vere c: behav1( initiaj trainix their a Contro] tine De relatiy {ElatiV f°r the Ho using a and ide AcceDta and 1dee Intreasq ”hile Cc acceptan ”portEG BHXiety J 26 working with the Big Brother program and other non-mental health volunteer obligations. Student volunteers changed to a more accepting moral attitude regarding aggressive and sexual behaviors after spending time working in a psychiatric ward. Moral acceptance was measured using 36 items from the Edwards Personal Preference Schedule that were concerned with the expression of sexual and aggressive behavior. While the companion program volunteers were initially less accepting than the control group, the training and intervention experiences significantly changed their attitudes to achieve the level of acceptance of the control group, who did not change significantly over this time period. The Companion Program volunteers' original relative moral severity might be partially explained by relatively greater religious value centrality known to exist for these volunteers (Knapp & Rolzberg, 1964.) Rolzberg et al. (1964) also measured self-acceptance using a 48 item scale (Lesser, 1958) that measured perceived and ideal self-concept on a variety of statements. Acceptance was reflected as the difference of the perceived and ideal. Companion Program volunteers significantly increased in self-acceptance over the course of the program while control volunteers significantly decreased in their acceptance over this same time period. Rolzberg et al. also reported changes on MMPI scores that suggested increased anxiety for the volunteers after the companion program. ex; 85: UK Vol Sel Con Jud fin. m0r. tOle relE vne, earl thre that Si"In: Othez 27 This change was interpreted as a positive indication of increased introspectiveness among volunteers. King, Walder, and Pavey (1978) conducted a similar experiment complimenting those of Rolzberg. King et al. wanted to determine whether it was patient contact alone, or the supervision with the professional that accounted for the differences in moral judgment and self—acceptance. Using the same measures as Rolzberg et al. (1964), King et al. confirmed the finding that volunteers showed greater self- acceptance after involvement with psychiatric patients ' regardless of the presence supervision with a professional. Unlike Rolzberg et al.'s control volunteers, the non- volunteers in King et al.'s sample showed no decrease in self-acceptance. King et al. failed to confirm the conclusions of Rolzberg et al. (1964) with regards to moral judgment of sexual and aggressive behavior. King et al. did find differences in moral judgment, with underclassmen found more tolerant than upperclassmen and volunteers more tolerant than non-volunteers, but volunteers did not seem to relax their moral judgment with exposure to the patients when compared to the non-volunteers. There were some differences in the methods of these two early studies done with inpatients. A big difference which threatens the validity of the comparative conclusions is that the control group in the King et al. (1978) study were simply psychology students and not necessarily involved in other volunteer activities. This would make it difficult to h De Cc 0! re 28 determine if changes (or their absence) were due to correlates of volunteerism. Since volunteers have been demonstrated to have different qualities form non-volunteers (Rersch et al., 1969; Knapp & Rolzberg, 1964), this difference is critical to the interpretation of the King et al. results. The other (intended) difference was that the volunteers in the King et al. study did not have concurrent supervision. King et al. used this difference to suggest supervision had a strong impact on volunteers, and may pose as an active agent of change. It was suggested that the supervision with the hospital staff may have modeled tolerant behaviors and verbal instructions that may have encouraged the participants to understand or accept the behaviors of patients. Unfortunately, no research which includes equal control, supervised, and unsupervised groups has been conducted which focuses on the change experienced in the volunteer. The only study to empirically address personal change among a religious lay counselor training group was the uncontrolled survey by Tan (1987) which suggested participants increase in confidence and a sense of perceived competence due to training. A fairly extensive collection of literature is, however, available on changes due to religious caregiver training with professionals. Derrickson (1998) has provided the most recent and complete review currently available of changes in students taking Clinical Pastoral Education (CPE) programs. tn the of ex; {HM Stu: Obs) 29 Derrickson listed 39 studies which have used standardized instruments to assess the change that results from the training experience. In the absence of other research on change by participants of lay caregiver training, this area of investigation was the closest analogue for developing expectations and hypotheses. Among the 136 studies which reported on the impact of pastoral care training on participants, Derrickson (1998) found 14 which used their own instruments, and 39 which used a total of 51 different standardized tests a total of 84 times. The Minnesota Multiphasic Personality Inventory (MMPI) was the most common instrument, and was used in 11 studies. Only equivocal and inconsistent results were observed in studies which employed the MMPI, with many studies not revealing any change at all. Only one study had results which were suggestive of change of sufficient magnitude as to be notable in behavior, but no control group was used (Lucero & Currens, 1964). In fact, Derrickson detailed only one report of the 136 which utilized a control group (Geary, 1977). Dayringer and Pavia, (1986) noted 3 unpublished doctoral dissertations which used controls (Keller, 1961; Kim, 1968; Reid, 1972), but methodologically sound research was clearly the exception. The Personality Orientation Inventory (POI) and Edwards Personal Preference Scale (EPPS) were the next most popular instruments employed in CPE research used in six and five aft, (197 Adie 38 studies, respectively. Significant but inconsistent results were regularly identified with these instruments. Three of the studies noted by Derrickson (1990) report changes on the Adjective Check List (ACL) that were the presumed result of training. The sequential nature of these studies, the variety of concerns addressed in the series, and the relevance of this research to the question of the change which might occur with laypersons exposed to similar training experiences make these three (Dayringer & Pavia, 1986; Grant, 1975; Thomas, Stein, & Klein, 1982) ideal for further elaboration here. Grant (1975), the first to use the ACL to measure change among CPE participants, evaluated the personality changes of seven theology students and seven ordained ministers over the course of an 11 week educational program in pastoral counseling. Measurements were taken before and after the program. No control group was employed. Grant (1975) found the majority of significant changes on the Adjective Check List (ACL), where mean scores on eight of the scales were significantly changed after the program. The scales where significant increases were observed were: Number of Adjectives Checked, Defensiveness, Achievement, Endurance, Reterosexuality, Personal Adjustment, Dominance, and Order. When viewed as a whole Grant thought the changes indicated that: The students, as a group, were more confident in self and others, more enthusiastic and ambitious, 31 more interested in life, experience most things around them in a wholesome and outgoing manner but not always on a deep level of involvement." (Grant, 1975 p. 256) Differences were also observed on the Sixteen Personality Questionnaire (16PF), where scores revealed a decrease in factor B and an increase in factor F. Grant (1975) explained that factor B is a measure of general intellectual ability, and he attributed the decrease to fatigue from the demands of the course. Factor F is a measure of assertiveness or humility. Grant attributed the increase in assertive qualities at the end of training to the extensive group interaction. Grant found no differences with the a modified version of the Osgood Semantic Differential scale which measures perceptions of various classifications of people. Grant (1975) administered the Minnesota Multiphasic Personality Inventory (MMPI) only at the pre-test. Drawing on MMPI findings, Grant suggested that the group mean profile, characterized by sophisticated defensiveness, some difficulties in forming relationships, and wishes to be seen in a good light, suggested the changes on the ACL might be superficial attempts to change in the appropriate direction rather than lasting personality differences. Grant suggested comparison with a control group and follow-up assessments to measure this change. 32 Thomas et al. (1982) attempted to replicate Grant's (1975) findings with a larger sample, extending the hypothesis to suggest that different placements might cause different effects. Thomas et al. also conducted a three- month follow-up to address the concern that changes were transient. The 81 subjects were not randomly assigned to groups, and, like Grant, Thomas et al. did not employ a control group. Thomas et al. (1982) found significant post-training increases in the ACL scores for the following sub-scales: Number of Adjectives Checked, Exhibition, Autonomy, and Aggression. Additionally, Deference was found to have significantly decreased. Thomas et al. referred to these as ”positive” changes that indicate a complex view of self with a more independent ”up-front” aggressive style with others. Grant's (1975) speculation that effects are transient turned out to be quite well founded. While statistical procedures were not very clear, only the ACL Number of Adjectives Checked scale appeared to demonstrate a persisting effect at three months. Only a few difficult to interpret results regarding the setting of the CPE placement were identified, but selection effects, and the differences in course content and extraneous community variables appear to confound any clear conclusions. The most recent of CPE research to utilize the ACL was .an examination of 79 health care professionals by Dayringer (and Pavia (1986). Dayringer and Pavia elected to examine if CPE trai p: fess; ccunsel: or £011 chaplai 33 CPE training interventions would produce the same changes in professionals of various disciplines such as nurses, nuns, counselors, and others. In this experiment no control group or follow-up assessment were employed. Subjects took a chaplain taught accredited CPE continuing education course titled ”Clinical education in psychosocial care”. The students were found to experience some changes which were attributed to the contents of the training. Notably, the ACL scales of Defensiveness, Favorable Adjectives Checked, Self-Control, Lability, Achievement, Order, and Intraception all showed gains following the training. Dayringer and Pavia (1986) assert that CPE methods are as effective with pastors as other professionals in eliciting personal change. Thomas et al. (1982) observed that they found different scales to reflect change in personality than those found by Grant (1975). Thomas et al. attributed this difference to possible differences in training emphasis in the groups. When compared to the pre-post effects in Dayringer and Pavia (1986), even more differences emerge. A review of the descriptions in the ACL manual (Gough & Reilbrun, 1983) suggests that despite the differences, conclusions that could be drawn about the type of participant personality changes are relatively similar. Aspects of assertiveness, confidence, independence, and increased self-valuing were found in all three studies. The combined picture of the three studies suggested that change due to CPE training resulted in increased adaptive behaviors for counseling . I' win...“ C5 IE (15 EX trl "P gre pec ex' toe use Ihv beg Inn of J 9101 (now 34 capacities with few non-adaptive personality changes as a result of CPE training programs. Only one published CPE study has been identified that used a control group (Geary, 1977). While still not a true experimental design in the absence of random assignment to groups, this still represents a great improvement over most studies. Geary used 31 students taking their first CPE training and compared their pre-test, post-test, and follow- up test results with 21 advanced students in CPE, 18 graduates of CPE instruction, and 31 control career church people that had not received CPE instruction. Geary explored the hypothesis that students would move more towards Maslow's description of self-actualization. Geary used the Personal Orientation Inventory and the Self-Concept Inventory to measure self-actualization. Geary found beginning CPE group participants had increased scores on Inner Directed, Existentiality, Self-Acceptance, and Nature of Man due to training on while those of the Non-CPE control group did not. These differences were suggestive of movement towards self-actualization. In response to Grant's (1975) undocumented observation that results might be transient, Geary's design employed three-month follow-up testing and revealed personal changes decreased with time away from the program. Geary also reported that personal growth appears the strongest among the female members of that beginning CPE class. While Geary expected to find less actualization in advanced CPE students is c! the ; Pastc Maybe excep Which Iesul Spec) £01101 extern vhich traini 35 because of their roles as supervisors and administrators, advanced CPE students showed more evidence of consistent and increasing growth than other groups. Geary concluded that the people who experience great personal change from CPE are those who are already willing, and in many cases desirous, for personal change. Whether the CPE class encouraged this change or just capitalized on it was unclear. This sample of the available research in CPE education is characteristic of approach, methodology, and findings in the published literature (Derrickson, 1998), and research in pastoral care in general (Gartner, Larson, & vachar- Mayberry, 1998). The use of control groups, with the single exception of Geary (1977), is notably absent. The extent to which personal changes are lasting, or are specifically the result of training and not pre-existing factors or non- specific group effects, is questionable. With the longest follow-up studies consisting of only three months, the extent to which conclusions can be made regarding changes which result, or last, from less intensive lay caregiver training is particularly questionable. y,, ..s ._,-~ . - . ,. :—~, ..~» -. . .-..- Beyond sample limitations of lay religious caregivers or even clergy caregivers a much larger range of studies have explored the types of changes that result from training. While some are directly related to therapeutic skills, such as empathy, others have a less direct 36 relationship. A sample of some of these qualities are considered below. WW Frauenfelder and Frauenfelder (1984) found changes in empathic listening for crisis line volunteers after a training program. Using pre and post-tests with student crisis intervention hotline volunteers and college student controls, the Hogan empathy scale (Hogan, 1969) and another empathy test specifically suited to hotline work were administered. The hotline volunteers were given a 26 hour training seminar where 2 sessions were devoted to empathy instruction. Rotline students increased in empathic responses as measured by the hotline-oriented empathy test, but did not change on the empathy variable. Frauenfelder and Frauenfelder speculate that the Hogan empathy scale measures empathy as a personality trait, and as a result, it would be more resistant to alteration from a simple training experience. While Frauenfelder and Frauenfelder (1984), have demonstrated that specific behaviors such as empathy can be altered, their research is not supportive of the premise that training programs can alter the personalities of the participants. Frauenfelder and Frauenfelder suggested that their results were more supportive of strict screening policies, rather than identifying empathy as a characteristic to be developed in training. 37 o ontr c n es w h r Hospice Training is another common form of para- professional training, and observations of changes as a result of hospice training would likely be applicable to other forms of caregiver training. When preparing volunteers for working with dying patients, locus of control becomes an important issue since a hospice worker will be confronted with the loss of control in the eventual death of patients. Hayslip and Walling (1985) hypothesized that training could increase the external locus of control for the volunteers as a way of giving them strength to handle the stresses of the volunteer commitment. Twenty-nine participants and 38 controls from the hospice mailing list were given pre and post-tests. The eight-week training program failed to make any impact on the locus of control scores, although some measures of death anxiety appeared to decrease for the volunteers. WWM mining... Martin and Carkhuff (1968) examined the change in the interpersonal dimensions of empathic understanding, positive regard, genuineness, concreteness, and depth of elicited expression that resulted as a result of a daily graduate psychology summer practicum. A control group of child development students was used for comparison. The five interpersonal dimensions were measured on the basis of a 58 item scale (18 questions per dimension) filled out by 2 e-m+ 38 raters based on a 45 minute analogue sessions. Training provided in the practicum appeared responsible for changes observed at follow-up. It can probably be argued that these 5 dimensions do not represent interpersonal behavior as the authors claim, but the dimensions do span a number of relationship variables that are important in helping contacts, and it was clear that something changed in the way the students interacted in relationships. ,,. » ... .. .-. - ,.,.-s ,- . -:A .p 7. , ,. Slight (1985) addressed the problem of separating out the critical components of a personality change resulting from immersion in training and the experience of practica. While Slight's subjects were speech pathology clinician trainees, her results addressed the issue of personality (or at least trait) changes as a result of training. Slight (1985) identified four fears reflective of typical beginning mental health trainee concerns. These fears included concern for living up to a supervisor's standards, having responsibilities for the well-being of future clients, the ability to apply theoretical knowledge in a helpful way, and a fear that they might be harmful to the clients somehow. She developed a 48 item Likert-type scale to measure anxiety resulting from the four outlined fears. This test, referred to as the ”Slight Clinician Anxiety Test (SCAT)” (Slight, 1985), was administered to 22 experimental subjects, 18 matched student controls who had taken the same coursework as the experimental subjects, and 39 8 students enrolled in a class where they observed other students doing therapy. All groups received the test before and at the end of the term. Slight's (1985) results indicated that participation in practicum decreased the anxiety about the experience as measured with the SCAT. These anxiety reductions were attributed to decreases in fears about supervision and general fears about practicum. Since 9 students repeated the practicum, and then student-taught, there was the opportunity to test further reductions with additional experience. No further reductions were found. Anxiety reduction occurred for the observation and non-practicum groups as well, suggesting that anxiety decreases as the experience of practicum grows closer. While all groups showed a post-test score reflective of reduced anxiety, results should be regarded with some skepticism as test- retest reliability information and validity information on the SCAT are not yet convincing. Among lay Christian counselors, one study has been conducted which addressed confidence and self-perceived. confidence with counseling skills. Tan (1987) administered a self-report questionnaire to 27 participants before and after a 32 hour six month training program which taught a biblically based cognitive behavioral Christian counseling model. Participants use 188 point rating scales to indicate their perceptions on six dimensions. No control group was 48 employed. Mean post-test scores for all six questions reflected improved confidence. The questions were: 1. How much do you know about counseling? 2. How much do you know about Christian counseling? 3. How competent (or effective) do you think you are in counseling? 4. How confident or certain are you of your competence in counseling? 5. How competent (or effective) do you think you are in Christian counseling? 6. How confident or certain are you of your competence in Christian counseling?. (Tan, 1987 p. 58-59) While some have questioned if training provides any skills related to effectiveness (Hattie et al., 1984), confidence and anxiety about counseling seem sharply influenced by training for those who seek it. ,“u, . .- -.,, ,,,.-- ;- . --- . . . .. Evidence for the purported personality changes that accompany training experiences and initial practicum experiences in mental health related fields is inconclusive. While some studies document changes in moral attitudes and self-acceptance (Rolzberg et al. ,1967; King et al., 1978), others (Frauenfelder & Frauenfelder, 1984; Hayslip & Walling, 1985; Slight, 1985) provide an unclear picture. If taken at face-value, the personality changes documented in 41 the CPE group training experiences are inconsistent and appeared short-lived at best (Dayringer & Pavia, 1986); Geary, 1977; Grant, 1975; Thomas et al., 1982). Methodological limitations are acknowledged to limit the generalization, and perhaps even reliability of the results (Grant, 1975), as few CPE studies have used appropriate control groups. Hhat_Qualities_are_Essentia1_te_the_Theranentic W In order to evaluate the changes that might occur for volunteers involved in a lay caregiving training program, it would help to know which variables of personality and behavior have been found to be related to the effectiveness of counselors. Effective training programs that seek to produce change among participants would hope to influence these variables in efforts to increase counselor effectiveness. What follows are some brief summaries of opinions and research with some of the more salient variables thought to characterize effective counselors. BMW. If one variable was considered the most important quality a therapist could possess, empathy would be it. Many regard empathy as essential to, or even sufficient for, effective therapeutic intervention (Brenner, 1982; Hogan, 1975; Patterson, 1984). Brenner thought empathy was the necessary precondition that allows a person to explore and reveal themself in therapy. 42 Hogan (1969; 1975; Greif & Hogan, 1973) has written extensively on empathy as a personality variable that characterizes the ability of a person to take the perspective of another. Typically this perspective taking is thought to include a degree of compassion (Hogan, 1975), personalized through tolerance, even-temperance, social self-assurity, and humanistic social values (Greif & Hogan, 1973). Patterson (1984) emphasized the centrality of empathy, warmth, and genuineness (EW&G) to the process of change in therapeutic interaction. Patterson was sharply critical of other reviewers who relegate EWSG to "non- specific or non-causal class of factors” (p.346). He claimed that the efficacy of EW&G as relationship variables may be at the root of the reported effectiveness of all methods of counseling and psychotherapy. With application to religious counseling approaches, Schlauch (1998) paraphrased Kohut (1977) and suggested that empathy is the "essence" (p. 14) of pastoral psychotherapy. In order for pastoral counselors to accomplish the goal of helpfulness in the pastoral relationship, Schlauch suggested pastor's relationships with clients reflect the theological and ethical responsibility to act with empathy as persons created in the image of God. Research has been done which supported Patterson's (1984) and Schlauch's (1998) claims that effective empathic interventions are critical to therapy success. Kepics (1979) rated 37 ten-minute therapy segments for tracking 43 errors. Tracking errors were defined as incorrect statements by therapists that lead a client off track of efforts to deepen emotions and self-understanding. Examples of client responses indicative of a tracking problem included helplessness, hostility, anxiety, guilt, intellectualization, and denial of dependency. Kepics found that tracking errors revealed by client's subsequent response were disproportionately associated with these subsequent negative and oppositional responses. In contrast, non-tracking error responses were followed by positive responses like insight, mastery, and assertion. Kepics reported that the oppositional qualities of clients decreased when therapists were taught to decrease the frequency of tracking errors, although empirical evidence was not offered for this claim. WWW Many have argued that it is not exclusively personality variables or skills that determine therapist effectiveness, but the way the personal qualities of the therapist and client interact which crystallizes the qualities of EWSG. Rowe, Murphy, and De Csipkes (1975) suggested the irregular findings of attempts to identify essential individual therapist characteristics reflected differential emphasis and revelation of personality characteristics as they are evoked in the dynamic context of interpersonal relationships. 44 Loesch, Crane, and Rucker (1978) ran experiments to test the conclusions of Rowe et al. (1975) for a sample of counselor trainees. Loesch et al. used a multiple regression analysis of eight counselor characteristics and three criterion variables. Their results yielded no combination of counselor characteristics alone were predictive of effectiveness. While this was only indirect support for the importance of relationship variables, Rudy, McLemore, and Gorsuch (1985) have directly demonstrated that interpersonal style is related to therapeutic effectiveness. Using Benjamin's (1974) Structural Analysis of Social Behavior (SASB) to measure the dimensions of power and affiliation in relationships, Rudy et al. (1985) asked 28 clients and their 11 therapists to rate each other's interpersonal style. With the self-report measures of the SASB Rudy et al. reported the ability to account for up to 65% of reported variance in therapy progress. Therapist and client ratings of perceived progress indicated the most change when when therapists and clients rated the therapist as friendly and freeing. Clients also rated their progress as improved in instances where therapists were seen as helping and protecting; a helping style characterized by advice giving. While these friendly and helpful therapist styles were perceived as helpful, they were less clearly related to symptom abatement. Hostile or blaming interactions were, however, clearly associated with negative perceptions and no symptom reduction. r' s ”WW a: r‘f ha ta 63 he 81' 11h. the cilia: 45 Rudy et al. (1985) also found that therapist's ”introject" (way of relating to self) ratings were very powerful in predicting client improvement. Among other relationships, therapist negative self-concepts were equated with a lack of client symptom reduction. When therapists rated clients as affiliative and clients saw therapists in the same light, therapists rated therapy as more successful and clients rated it as more effective. They conclude: Relationship factors, which theoretically can be translated into specific interpersonal behaviors, are at least catalysts that energize "therapeutic technique” and may even lie at the heart of perhaps all but the most arcane treatments. (Rudy et al.,1985 p.277) Although empirical support is lacking, VanKatwyk (1988) has proposed a supervision tool for religious counseling he called the ”Helping Style Inventory". The tool is essentially an interpersonal orientation model that draws .heavily from models of Leary (1957), and Benjamin (1974), although specific credit was given only to Rogers (1977). While untested, VanKatwyk suggested the inventory as a theory on which to base psychological and theological interventions. W111 Brenner (1982) suggested tolerance and acceptance are «qualities essential to developing a relationship with a 46 client which permits disclosure and exploration. In order for tolerance and acceptance to exist in the therapeutic relationship, they must first be within the realm of accessible qualities of the therapist. Jackson and Thompson (1971) measured cognitive flexibility and tolerance for ambiguity in fully trained school counselors. Counselor effectiveness was determined by a judgment of success in videotaped sessions for helping clients move towards therapeutic goals and adaptive skills. No significant differences were found which would have suggested effectiveness depended on the level of flexibility or ambiguity tolerance, however differences were found which suggested it may be easier for counselors to be flexible in client contacts where ambiguity is low. . In another examination of cognitive style Tinsley and Tinsley (1977b) compared the Omnibus Personality Inventory profiles of 74 trainees in counselor education for differences between groups later judged by supervisor evaluations to be effective, ineffective, or undetermined. Trainees rated effective tend to be more introspective, flexible, tolerant, and independent than the ineffective group. Whether these differences are transferred into meaningful differences in therapy was not clearly evaluated, and it is likely that the supervisors rating the students were simply responding to these qualities without realistic basis for estimating effectiveness. 47 Fry and Charron (1988) examined variables of cognitive style in a sample of 32 graduate therapist trainees. These investigators measured serialism-holism and field dependence-independence dimensions of cognition in the trainees and in 32 students who would serve as two-session clients in this semi-analogue study. Fry and Charron a I l matched and miss-matched the cognitive styles of the client and therapist in treatment pairs intended to focus on personal growth and development. Analysis of client ratings of changes in self-awareness and self-satisfaction after 2 sessions suggested the serialism-holism dimension was irrelevant to effectiveness as a counselor. Therapists who displayed a strong field dependent cognitive style were, however, shown to have vastly different influence on client ratings of growth in therapy depending on the cognitive style of the client. Field dependent therapists were not as useful with field dependent clients as they were for field independent clients. Field independent therapists seemed to work well with either kind of client, but they had vastly discrepant ratings from clients regarding improvement. Field independent therapists ratings of client improvement were in much greater agreement when the client also held an independent orientation. These are just three studies of cognitive styles and qualities, and they all present different conclusions about ‘the role of cognitive style in therapist effectiveness. The 48 Fry and Charron (1988) research even seems to suggest the involvement of interpersonal factors with this variable. W In another publication from their research, Tinsley and Tinsley (1977a) examined the differences in interests, needs, and abilities of counselor trainees. While the same flaws of their 1977b research are evident here (the effectiveness rating is only a supervisor impression), differences were found for the "effective” and "ineffective” groups on vocational interests. While no differences in verbal aptitude were revealed, ”effective" trainees were more interested in writing and academic achievement, and less interested in some vocational areas of the Strong Campbell Interest Inventory that could be thought of as routine. Ineffective counselors were found to have higher values on a number of Minnesota Importance Questionnaire scales measuring interest in ability utilization, achievement, authority, supervision-technical, and working conditions. WW Jay Haley's (1969) article, which detailed how to be a failure as a therapist, concentrated a good deal of sarcastic humor on a therapist attitude that could be described as conceited, contemptuous, and detached. While Haley's (1969) article was only theoretical, some evidence was been collected that supports the concept that attitude can influence effectiveness. 49 Jackson and Thompson (1971), in a study described in the section on cognitive style, also examined therapist attitudes. They measured attitudes towards self, most people, most clients, and counseling using the Semantic Differential Scale. Their results revealed that effective rated therapists had significantly more positive attitudes towards self, most clients, and counseling. "The most effective counselors were more positive in viewing mas; nnnnlg and mgst glienns as friendly, able, and worthy, while viewing ggnnsgling as freeing, altruistic, and important.” (p. 252, original emphasis reflects scale names) Mc Lennan (1985) concluded that although many personal qualities have been successfully identified as important for effective counseling, personality scales had not proven to be a reliable, valid, or quick way to determine which therapists would prove effective. Mc Lennan proposed a resolution to this problem with a 14 item inventory of beliefs about counseling (items can be seen in Appendix 0). The Helping Beliefs Inventory (RBI) discriminated counselors from non-counselors with an interest in helping. Mc Lennan administered the RBI to participants in a 38 hour training program prior to training. After training, trainers (blind to the scores) rated the new therapist's abilities as high, medium, or low skilled. Therapist skill was positively related to the pre-training RBI scores, with all differences between means found significant. These results indicated. that the beliefs possessed prior to a training experience “ilfi'iii—fiw ‘ S (A ‘- 58 influenced how counselors were later judged in terms of therapeutic skill. Values certainly hold an important place in therapy as well. Since clients who improve have been noted to take on the values of their therapist (Rosenthal, 1955), the therapist's values would be critically important when client adaptation is measured. Reiff and Riessman (1968) have argued that it is the inherent attitudes and values held by indigenous non-professional mental health workers which make them so valuable. Reiff and Riessman state that: The poor often seem more willing to help each other in a pezsnnal way. ... The closer relationship with a nonprofessional peer may prove to be the most effective way of reaching and involving the poor. It provides a human link to the professional service and makes possible new kinds of programs which are better designed for those who need them and are therefore better utilized by them. (Reiff & Riessman, 1968, p. 9- 18, emphasis original) W Some recent attention has been focused in religiously oriented journals to qualities referred to as "gifts of the (Holy) Spirit" (Richard, 1987; Tan, 1991, 1998). The gifts of the Spirit, described in Romans 12:8 and 1 Corinthians 12:8-18 and 29, include exhortation, healing, wisdom, “H I: know} also inter and 1 t0 C1 ofte; in t; and ‘ C181 98a} cher COur tlai and Chri: train (13w 51 knowledge, discerning of spirits, and mercy. Tan (1998) also lists prophecy, teaching, faith, miracles, tounges, intercession, prayer, leadership, and administration as other gifts of the Spirit. These gifts offer many direct counseling skills for Christian therapists. While spiritual gifts have not been empirically pursued, parallels (and contradictions) to secularly described.essential therapist i qualities are easily observed. u o s ter t c . This section has reviewed a small sample of research and theory examining characteristics thought to be related to counselor effectiveness. Counselor training programs often claim to have a commitment to helping the trainee grow in themselves and in their ability to engage in meaningful and helpful relationships with clients. Since there are qualities believed to be helpful in therapeutic interactions, it is expected that training programs which claim to teach helping skills would foster these same qualities. Tan (1987), in the only known research to date on changes experienced by participants of lay Christian counseling training, identified that upon the completion of training, participants report a stronger sense of competence and greater knowledge of both counseling in general and Christian counseling in particular than at the beginning of training. While self-report pre and post measures were used, no control group was employed and the instrument was Re: 52 not standardized or anchored in any behaviors that would correspond to any guarantee of skill or effectiveness. Recognizing these shortcomings, Tan (1987) suggested that: There is a need for more sophisticated research to be done, employing more than one evaluation measure. Other evaluation measures (e.g., personality inventories, ratings of counseling skills, based on role plays or audiotapes or videotapes of real life situations, made by qualified observers, or ratings of counselor's competence made by clients) should be used as well in future studies, and appropriate statistical analyses performed. (Tan, 1987, p. 68) Such a study, with an appropriate control group, methodology, instruments, and appropriate statistical procedures is proposed here. Hmtheees When individuals involved in caregiving training activities are compared with others who have chosen to be involved in other church-related group activities, certain differences would be expected: 1. Some initial differences in personality qualities would be expected which would distinguish people volunteering to become trainees in a Lay counseling programs from participants in other church-related activities . 53 Based on similar research with volunteers by Knapp and Rolzberg (1964), Kulik and Scheibe (1969), and Hersch et al. (1969), differences in self-report personality measures would be expected which reveal volunteers for caregiving training as more compassionate, mature, independent, sensitive to human problems, and somewhat socially withdrawn. There are a number of qualities presumed to be related to effective counseling. It would be expected that the content of an effective lay training program would influence these qualities. Changes would be expected from pre—test to post-test in the caregiver training group that would not be evident in adult Christian education control groups. These personality variable changes would be expected to be most prominent for new volunteers, but continued growth would be expected for volunteers who completed training previously and are now engaged in continuing supervision, training, and caregiving relationships. Participants would be expected to show changes as a result of training experiences on the following constructs: 2a. Interpersonal orientation shift: Research by Rudy et al. (1985) suggested therapists perceived as effective adopted interpersonal perspectives where their perception 1:: 2b. 2c. 54 of themselves in relation to others is characterized by disclosing and expressing, and enjoying friendly autonomy. Changes on such an interpersonal dimension are also consistent with interpersonal theories expressed in reference to pastoral counseling by VanKatwyk (1988). Adoption of process helping beliefs would be expected: Changes in beliefs relevant to helping relationships would be expected to occur for participants completing training as caregivers. It would be expected that graduates would learn a new set of beliefs which allow them to avoid problem solving techniques characterized by distracting, diverting, directing, or controlling those for whom they provide caregiving (Mc Lennan, 1985). It is expected that additional experience as caregivers would strengthen and emphasize these beliefs and values about helping. Self-concept shift: Therapeutic effectiveness is related to the self-concept or self-image 0f the therapist (RUGY et al., 1985; Tinsley & Tinsley, 1977a). CPE research also identified personality change as the result of counseling training consistent with a self-concept shift (Derrickson, 1998; Geary, 1977; Grant, 1975; Thomas et al., 1977). As such, it 26. 55 would be expected that experimental groups would experience changes on measures of self-concept. Changes would also be expected to be seen in personality measures which indicate personal growth in the direction of greater openness to feelings, and self-actualization (Thomas et al., 1977). Changes in self report found significant in CPE training (Dayringer & Pavia, 1986; Grant, 1975; Thomas et al. 1982) would be expected to generalize to the lay training context. Self-esteem shift: Since positive self-perceptions were found related to a sense of personal effectiveness (Jackson 5 Thompson, 1971), self-esteem would be expected to increase as a result of the training experience. This hypothesis also tests numerous speculations, literary descriptions, and anecdotal reports of personal growth from training experiences. While changes would be expected in the caregiver training group, relatively few changes would be expected in the Christian education control group. Changes due to simple group participation would not be anticipated. Variables such as age, gender, and socioeconomic status would be expected to have relationships with a 56 number of the variables. So as not to attribute changes due to these extraneous variables to the training experience, the impact of these variables should be determined and statistically controlled. ch pa no tr. or dr- 13 tel W "11 METHODS Barnum Participants were 287 members of 33 groups from 28 church congregations. The 28 congregations were solicited from a list of 36 mainline Christian churches in the West Michigan area with active Stephen Ministries programs (see Appendix A). Eight Stephen Ministries participants and 33 Bible Study participants approached about participation chose not to complete materials. Churches contacted but not participating either no longer had an active program, were not conducting training, or had significantly altered the training format suggested by the Stephen Ministries organization. Group denominational affiliations, and dropout rates can be found in Appendix B and Table 4. The mean age of participants was 45.8 years (SD = 13.8). Sixty one participants were male and 146 were female. Participants had a mean Index of Social Position (Hollingshead, 1957) score of 36.29 (SD = 14.86), consistent with a social class rank of III on a scale of I to v. Participant scores on self-report measures of personality style suggest that this group was fairly conventional and not an extreme group. Some isolated possible differences from normative samples were identified, namely that this sample was more inclined to express self-critical perceptions and assume a deferential posture. These differences these were thought to be reflective of characteristic Protestant religious qualities and would not 57 58 compromise the applicability of results to other church populations. For a comparison of this sample with normative samples on the self-report measures of personality style see Appendix C. Participants were chosen from three types of church- offered groups. The participants chose to participate in these groups as part of Fall and Winter educational opportunities, independent of knowledge of the research project. The groups included 134 participants in new Stephen Ministries training classes (NSM). This program, described in Appendix D, is a structured training program for skill development in "distinctively Christian supportive caregiving" (Voss, 1988). The second group was comprised of 29 participants from a continuing Stephen Ministries group (CSM). These participants had completed the SM training at least one year prior and were actively engaged in caregiving relationships and monthly supervision. A third group, 44 participants of Bible Study classes (BS), was included as a way to account for test-retest effects, and the non-specific effects that could result from participation in any group in church on a regular basis. BS classes included a variety of structured educational series including ”Kerygma", the 'Bethel Bible Series", and a religious book study. Of the 287 total participants, 186 NSM participants, 26 CSM participants, and 22 BS participants completed both pre and post tests (see Appendix B, Table 4). Notable differences in both participation and retention were observed when the 59 BS group was compared to SM groups. The differences might warrant some question of the extent to which this sample is representative of Bible Study groups in general. BS groups were selected only from those churches which had active SM programs. No BS group members had SM training. The CSM group was assembled by using a random procedure to select one CSM participant from a CSM group in a given church for every three NSM participants in that same church's new training class. Groups training their first class of SM's resulted in an inflation of the 3:1 (NSM:CSM) ratio. Procedure Contact of the group members was made through contact with the SM training leaders. A letter of reference from Gary Voss, M.Div., training Director of Stephen Ministries Inc., was included as part of this contact to assure groups of the Stephen Ministries sanction of the study (see Appendix E). Consenting trainers approached members of their respective groups to assess group willingness to participate in the study. Trainers informed their group of the nature of the research and expectations according to standard guidelines (see Appendix F). Group leaders were sent a follow—up letter explaining expectations and requirements, including provisions for anonymity (see Appendix G) and voluntary involvement (see Appendix H). Materials prepared by the experimenter were mailed to trainers for administration. An accompanying in: 68 instruction sheet provided details of how to present the materials and respond to concerns which might arise (see Appendix 1). Participants were informed by both the consent sheets (see Appendix H) and trainers that participation in the pre-test did not deny the right to refuse to participate in the post-test without reprisal, and that status in the group would not be affected by participation in the study. The pre-test was administered on the day of the first NSM class, while the post-test, administered only to those who had completed pre-tests, was given after the 58 hour Stephen Series training. The same test sessions were used to assess CSM participants. CSM participants did not repeat the training, but continued to be engaged in at least one meeting per month for continued training, consultation, and supervision. BS participants completed testing materials on the day of their first class, with the post-test at the end of a term. While some BS classes continued for a second term, the time period roughly corresponds to the same period of time required to complete the NSM training, although total hours in the BS group were less than in the NSM group (see Appendix B, Table 4). Participants were told that the research intended to measure changes in the way people see themselves that might result from being involved with a group experience. In return for involvement, group leaders were given a written summary of the research findings to present to the participants (see Appendix J). 61 W Training courses for both NSM and BS groups were not directly audited for content. The BS groups, which did not directly cover materials specific to helping relationships, were structured programs of directed study lead by lay leaders and clergy with a goal of enhancing knowledge of theological ideas and Biblical understanding. Since inclusion criteria restricted the NSM groups to those groups whose leaders reported that they closely followed the fairly inclusive and comprehensive SM training materials, it was assumed that the NSM classes were roughly equivalent. Variation between NSM groups exists as each NSM group had different trainers. For more information on the SM training materials, see Appendix D. 835211311 All participants were administered identical materials at pre and post-test. The individual coded test packets included a cover sheet and consent form containing instructions for participation (see Appendix H). A participant information sheet which asked the participant's age, gender, and a question as to the group in which they chose to participate was also included (see Appendix K). Instruments: W The Adjective Check List (ACL) (Gough & Reilbrun, 1983) a popular test in which participants indicate which adjectives, from a list of 388, describe them best. The ACL 62 is a quick (28 minute) instrument with great theoretical flexibility for the prediction and observation of numerous personality features. Normed on over 9,888 subjects the test has over 37 scales, many derived with factor analytic techniques. The range of test-retest coefficients at one year for males was from .34 to .77 with a median of .65. The female sample had higher test-retest correlations with a median of .71. (Gough & Reilbrun, 1984). A consensus of reviewers suggests the ACL is a valid and useful instrument for the measurement of self-concept (Zarske, 1984). Derrickson (1998) noted 5 studies of CPE related change which employed pre-post assessment of students on the ACL. Derrickson describes the ACL as a good measure for assessing change among CPE groups as it is sensitive to the qualities that such courses seem to influence. If for no other reason, the ACL was an interesting measure to include to compare the results of lay caregiving training with the literature on CPE training induced change. For additional information on the characteristics of the ACL see Appendix L. W The Gordon Personal Profile (GPP), a standardized self- report instrument for the measurement of self-concept and self-esteem related variables, is a brief instrument which requires about 15 minutes to complete. The GPP attempts to measure useful and meaningful aspects of self-concept using forced choice item on scales developed with factor analytic 63 techniques. The scales purport to measure Ascendancy, Responsibility, Emotional Stability, and Sociability. The Ascendancy scale reflects qualities of a self- assured, independent, and active style. The Responsibility scale measures qualities of determination, persistence, and reliability. Those who score high on the Emotional Stability scale report relative freedom from tension and a balanced lifestyle. Sociability is a scale which measures gregarious and social interests as a personality trait. A Self-Esteem score can be derived as the total of the four component scales. Self-esteem is regarded as the evaluative feeling about oneself in comparison with some internally held ideal. The GPP is well suited to measuring this quality as it measures peoples self-perceptions in a variety of proposed social contexts and internal emotional states. The typical high intercorrelations of the four component scales also suggests the presence of an underlying construct such as self-esteem. GPP scale internal consistency is in the range of the high .88's (Gordon, 1963). Reliability of the scores is attested to by the stability of scores over various training experiences. Test retest coefficients for 127 men enlisted in a 29 week Navy training program ranged from .58 to .78 while coefficients from three graduating classes of 4 year optometry training was consistently in the range of .58 to .68. Other reliability information is included in the manual and has been referred to as "satisfactorily high" 64 (Dickens, 1965). Concurrent validity with counselor trait ratings revealed three correlations above .58, and concurrent peer ratings of college students ranges from .47 to .73 (Dickens, 1965). Dickens noted, however, that external validities in the non-student and non-counseling situation rarely exceed .38 to .35. Given these qualities of reliability and validity, the GPP appeared to be a good measure to address claims of changes in self-esteem, self- concept, emotional health, and social functioning among SM candidates. Additional information on the GPP can be found in Appendix M. W The SASB (Benjamin, 1974) is a well developed instrument used for interpersonal assessment. Based on the interpersonal circle (Leary, 1957) the instrument measures the dimensions of affiliation and interdependence in three perspectives (called ”surfaces") of the interpersonal relationship. The surfaces are: (Surface 1) Other, where the respondent describes the behaviors of another person in relation to the self, (surface 2) 5211, Where the respondent describes the behaviors of the self in relation to others, and (Surface 3) W. or W, where the participant responds to questions dealing with the result of the self focusing on the self (Benjamin, 1979). Scores are coefficients of fit to patterns drawn from mathematical models based on the theory that respondents can not be both dominating and submissive nor loving and 3‘! r: 65 hateful. Examples of these pattern coefficients and their clinical manifestations can be found in Appendix N. The short form versions of the SASB (Benjamin, 1988) offer decreased test time requirements by limiting the focus of the assessment and removing redundant assessment of clusters of behaviors. The short form relies on the internal consistency of the larger model but requires only about 28 minutes to complete. Predictive validity for assessing counselor qualities is adequate in that the SASB was found to correlate with a number of variables related to the perception of therapists, clients, therapy variables and outcome measures of psychotherapy (Rudy et al., 1985). With regard to the long form, from which the short form was derived, Benjamin (1974) stated that the high internal consistency should suggest that test-retest should be at least adequate, and a number of subsequent investigations (e.g., Benjamin, 1979, and Rudy et al. 1985) have used the test with good results suggesting the SASB is as reliable as people. For more information on psychometric properties of the SASB see Appendix N. Participants completed the INTREX Short form version ”C/She/Present/l” to indicate their responses, which were subsequently scored to yield the 3 pattern coefficients for each of 4 aspects (self and presumed other perspectives of the first two surfaces) of the interpersonal relationship. These aspects are elaborated in Appendix N. Coefficients 66 for these data were calculated using the ELQMAS Program available through Intrex interpersonal Institute. The_flelnlns_fieliefs_1nrenterx_iflflll In an attempt to measure changes that might occur in beliefs relevant to counseling, the Helping Beliefs Inventory (RBI) (Mc Lennan, 1985) was employed. The RBI was developed to distinguish effective from ineffective trainees based on beliefs about helping relationships. Items solicit ratings of helpfulness of strategies which distract the person from their immediate experience, control and direct people in solving their problems, and divert people away from exploring their situation (see Appendix 0). While the ideal use of the instrument as a valid and reliable screening measure has not been fully achieved, it appears to be increasingly validated as a reliable way of screening effective from ineffective counselors. Adequate test-retest reliability (1 = .74) was established with a group of 43 college psychology students over a 18 week interval. validating research has suggested the measure can discriminate counselors from non-counselors as well as discriminate between counselor of high, medium, and low rated skill. The criterion in this validation trial was blind counselor ratings of skill, leaving some question if the measure assesses effectiveness in securing client change or adopting a professional posture. More information on the RBI can be found in Appendix 0. Wen 67 A measure of socio-economic status was employed in this study to control for whatever influence this variable may have on the variables of study and to aid in the determination that the control groups are of similar composition. The traditional choice for a quick, but valid measure of socio-economic status has been the Hollingshead (1957) Two Factor Index of Social Position (hereafter referred to as SES out of convention). The measure is composed of an employment classification and ranking of level of educational achievement. The two factors are weighted and the resulting product is recorded. Derived scores can be transformed into one of five social class scores that compose the actual Index. A score of I represents the lowest social class, while V represents the highest. This final step assures that there are meaningful differences between groups, but is not always used as the non- transformed variable (used in this study) affords more statistical power. Reliability is dependent on the accuracy of the rating process for occupation and education, and the stability of these factors in the population being measured. Hollingshead and Redlich (1958) found that the SES corresponded with meaningful differences in social behaviors. RESULTS WW Pre and Post-test Means and standard deviations for all the dependent variables displayed by group type are available in Appendix B. A series of 2 x 3 (persistence x group type) Analyses of Variance tests (ANOVA's) were used to determine if the 53 subjects who completed only pre- testing (non-persistent) were different from the 154 subjects who completed both pre and post-testing (persistent). The ANOVA also served to establish if the characteristics of participants of the different groups, were similar at pre-test. Age, SES, and gender were observed to have significant relationships to a number of the variables in the study (see Table 16 in Appendix 0), including persistence and group membership. Participants who completed both pre-test and post-test materials were of higher SES than the 53 participants who did not not complete materials (E(l,20l) = 9.88; p < .002). Tukey Honestly Significant Difference (BSD) comparisons of the ANOVA main effect for group type suggested participants in the BS groups were more likely to have lower SES standing than NSM participants (E(2,20l) = 6.53, p < .002). No differences were observed for age and gender (9 > .05) for either group membership or persistence. Because of these relationships, the three demographic variables were included as covariates for the remaining analyses. 6 8 69 When SES, age, and gender were used as covariates, no differences emerged between persisting participants and those who did not persist (p > .05). For additional specifics on the persistence main effect see Appendix R. At .pre-test groups were not different on any of the variables measured by the GPP, and differences were observed for only one ACL variable (Low Origence, Low Intellectence). Five SASB variables suggested differences existed between those who chose SM training and those who chose BS groups. Details of these pretest group differences can be found in Appendix R. Examination of the interaction effects of the ANOVA's for the Group type x Persistence revealed persisting participants had different qualities depending on the type of group. Relatively few non-persisting CSM resulted in a reconsideration of the data in a 2 X 2 (group type x persistence) ANOVA with only BS and NSM participants. BS subjects who did not persist to complete the post-test, and NSM who remained to complete post-test materials, score higher on the GPP responsibility variable and the ACL variables for Deference, Self-control, and Nurturing Parent. NSM participants who do not persist to complete post-test materials and BS participants who did persist to complete post test materials score higher on the ACL measures Autonomy, Aggression, Change, Creative Personality Scale, and Free Child. Means, standard deviations, and E-test values can be found in Table 17 in Appendix R. 70 The Hypotheses of Change The change that results exclusively from training and extended participation in the SM program over time was explored in this mixed design experiment with both longitudinal and cross-sectional data. CSM participants provide a cross-sectional view of the effects of . participation in the SH program over time, while pre and post-testing provided a longitudinal view of the effect of the training course. The longitudinal view of the BS participants made a useful contrast for the effect of NSM training to compare the generic effects of group participation in religious education. Since none of the variables, with the possible exception of the GPP, were based on principles which lead, naturally, to multi-varlate analyses, a series of 2 x 3 (time x group) repeated measures ANOVA's, with age, SES, and gender as covariates, were used to explore data. To control for experiment—wise error a conservative criterion of p < .01 was adopted for hypothesis testing. When the GPP and ACL variables (listed in Appendix P) were subjected to the repeated measures ANOVA procedure with covariates, no significant main effects or interaction effects were observed at either the p < .01 or even the p < .05 level. Personality change, at least as observed by these two self-report measures, does not occur as a result of 8" training or experience. 71 The scores of the SASB proved somewhat more challenging to decipher. While the absence of any hypothesized interaction effect at either the p < .01 or the p < .05 level suggests little benefit results from training, some post-hoc group type main effects were observed. A pre-post post-hoc main effect was also identified with one SASB variable. Examination of these post-hoc effects is explored in more detail in Appendix 8. The 2 x 3 (time x group) repeated measures ANOVA (with covariates) of HBI scores showed evidence of change suggestive of the adoption of helpful beliefs by NSM participants (E (2, 138) = 8.80, p < .000). Heans and 95 percent confidence intervals for the group scores at pretest and post-test are available in Table 1. Table 1 !° 0° - , n 90 °_ 0 V 09 1- W Pre-test Post-test Group M SD 95% Confidence M SD 95% Confidence Interval Interval BS 24.9 6.3 21.9 to 27.8 24.2 6.0 21.3 to 27.0 NSM 24.6 6.7 23.3 to 26.0 20.4 6.2 19.1 to 21.6 CSH 19.2 4.8 17.1 to 21.2 19.7 6.6 16.9 to 22.5 72 Figure 1 graphically displays the data from Table l and shows how NSM participants are equal to BS participants on the RBI at the Pre-test. The CSM participants, however, have significantly lower scores than the BS and NSM group indicating a more helpful set of beliefs, presumably by virtue of their previous training. Once the course is complete the scores of the BS participants do not change, while the NSM participant scores decrease, and are similar to the previously trained CSM group. The similarity of the scores on the RBI for CSM participants at pre-test and post- test, and the post-test scores of the NSM participants suggests that the helping beliefs are fostered in the SR coursework and not out of experience as a caregiver (which may well serve to reinforce and maintain these beliefs). Variables found related to the RBI, and a comparison of RBI values with values obtained with other samples are discussed in Appendix T. 73 MWMIWmnpnmpofl-m I ”MW 28 O Newman-m k WWIW 24 F 4 22 20 k d 74 U n h adi ional Method for Ana sis of Chan e. Covariates and control groups are useful ways of removing extraneous influences on test scores. However, no previous study with religious caregiving training programs (including most CPE training programs) were identified which employed these controlled methods of data analysis. In an effort to relate this study to previous findings, the data was also analyzed in the traditional, albeit less controlled, fashion. For the purposes of this analysis the CSM group and the BS group were excluded. Multiple paired t tests were used to compare NSM scores at pre-test with scores at post-test for those subjects who completed both pre and post-testing. No covariates were used. Of the 55 analyses, 9 variables had significant differences (p (2-tailed) < .05) between pre-test and post- test scores. These variables, mean scores at pre and post test, and the t-test values are listed in Table 2. 75 Table 2 W W Variable Pre-test Post-test t (df) p‘ 5 (£2) u (52) Gordon Personal Profile Variables Emotional Stability 22.58 (5.94) 23.88 (5.90) -3.11 (94) .003 Adjective Checklist variables Dominance 3.99 (6.04) 4.85 (5.35) -2.20 (102) .030 Intraception 14.90 (4.25) 15.89 (3.32) -2.56 (102) .012 Succorance 1.78 (3.83) 1.15 (3.57) 2.18 (102) .031 ‘Self-confidence 8.47 (5.71) 9.84 (4.78) -3.20 (102) .002 Adult scale 8.65 (5.52) 9.72 (5.06) -2.51 (102) .013 Adapted child -8.17 (5.95) -9.73 (5.47) 3.34 (102) .001 A4“ 13.45 (4.16) 14.36 (4.12) -2.28 (102) .025 Relping Beliefs Inventory 24.65 (6.71) 20.36 (6.17) 8.09 (96) .000 SASB Variables No Significant Differences (p >.05) ' Probabilities are 2-tailed " Low Origence Righ Intellectence 76 It must be remembered that variables with significant effects noted in Table 2, except the RBI, did not show any significant interaction effects suggestive of benefit from training when procedures which delivering greater effect discrimination were implemented. With the exception of the RBI, none of the results with variables in Table 2 were significant when control groups and the covariates age, gender, and SES were utilized. DI SCUSSION Summarx_2f_EindinsS The first hypothesis suggested that some initial differences suggestive of a compassionate, independent, somewhat socially withdrawn person with sensitivity for human problems should distinguish participants volunteering for training as Christian caregivers from participants choosing to attend BS groups. One personality variable and three interpersonal variables distinguished the BS group from the others. The findings suggest the BS group had a stronger drive for contentment, conventional thinking and action, as well as a reasonably optimistic outlook in comparison to the SM group, providing limited support for the hypothesis that participants drawn to SM experiences are more independent in their thoughts and actions than BS participants. SASB scores suggest NSM and CSM participants more strongly report that they see themselves, and believe they are seen by others, as adopting a friendly, affiliative style in their interpersonal interactions than the participants choosing the BS group. While this might be construed as limited support for the idea that NSM are more sensitive or drawn to address the concerns of others, it would not be evidence for a socially withdrawn style among NSM candidates. Post-hoc analysis of the characteristics of participants who did not complete post-test materials, and presumably did not persist in training experiences, revealed 77 78 some noteworthy relationships. While age and gender did not seem to be a factor in retention or choice, participants of high SES were more likely to choose caregiver training, and were also more likely to drop-out. BS groups have better retention for participants who report original and different thinking, less concern for social convention, and a somewhat impulsive style. Individuals reporting these same qualities dropped out of SM training. The results also suggest that participants of SM courses who stayed through the training and Bible study participants who did not complete post—tests were less likely to take an expressive or impulsive style of interacting, and instead report assuming a deferential style of interacting. These participants were more more likely to be constrained by expectations of others and more reliant on routine, reporting less desire to think originally or different. One would question if SM courses, with their structured curriculum and tightly organized supervision, prove challenging and ultimately frustrating for ”free- spirited original thinkers” who might find more welcome in the didactic atmosphere of most Bible study classes. The hypothesis that new participants would not be found different on a measure of useful helping beliefs was not disconfirmed. BS and NSM participants had essentially similar mean scores on the RBI prior to participation in the groups. The second hypothesis suggested SM experiences of training and caregiving create changes in personality, 79 relationship style, and beliefs of the participants which improve their capacity as caregivers. It was thought that an interpersonal orientation shift to an affiliative, disclosing style would be expected. Additionally, shifts in self-concept suggestive of increased self-awareness and comfort were expected along with a general improvement in self-esteem. Finally, changes which indicated increased personal awareness of experienced and perceived emotions was expected as a result of SM participation. None of the self-report variables on the ACL, GPP, or SASB showed any evidence of change between groups that could be attributed to the training or caregiving experiences. As such, the hypothesis that the SM training course somehow changes the personality or basic interactional style of participants was not supported in this study. The suggestion that the experience of involvement in helping relationships over time could change the helper was not supported either. One hypothesis regarding changes that could occur as a result of training was supported. The RBI scores of NSM participants, which were not different from BS participants at the pre-test, had changed to the level of CSR participants by the post-test. At post-testing NSM and CSM participants had scores significantly different from BS participants, whose scores remained unchanged from pre-test. It can be said that NSM participants learn helping beliefs not held prior to the course. Additionally, NSM's maintain 80 these beliefs after experience with caregiving relationships. The observation that training appears responsible for the adoption of helpful beliefs about caregiving relationships provides objective support for Tan's (1987) finding of increased perceived competence among training graduates. Limitations in the validation of the RBI suggest that conclusions should be limited to suggesting the caregivers learn accepted beliefs about helping that may or may not be related to client change. The third hypothesis suggested that changes observed in participants would be the direct result of training or caregiving activities, and not the result of participation in groups, the effect of covariates, or changes that would naturally occur with the passage of time. This hypothesis dictated the use of an appropriate control group, the implementation of covariates, and the observation of a group of trained SM's already involved in caregiving. Only the change in helping values was observed using these controlled procedures; when less stringent methods were employed, the rationale for the use of such controls became clearly apparent. When less stringent statistical and methodological methods were used which were typical of those used in the majority of personal change research (Derrickson, 1990), a number of "changes" in personality functioning could be construed. In the absence of control groups and covariates, NSM participants showed post-training changes on a number of 81 self-report variables (see Table 2). If taken at face value, these results suggest NSM training develops personality characteristics of increased personal confidence, ambition, relationship skill and confidence, and a sense of directedness and purpose. Many of the qualities assessed by the scales where change was observed are the very qualities proposed to change in the second hypotheses. When the controlled procedures were used the changes were no longer apparent. When these "uncontrolled" results were compared to those of Grant (1975), Thomas et al. (1982), and Dayringer and Pavia (1986) some overlapping scales and a number of overlapping concepts emerge (see Table 3). This experiment used more of the available scales on the ACL, including the Transactional Analysis scales and Origence Intellectence .scales, so the number of scales found significant with this technique which are unconfirmed in other similar studies may be somewhat misleading. Overlap among the four studies yielded no scale changes common to all samples. Previous researchers have suggested that the ACL would reveal different results in different studies due to sample inconsistencies resulting from different leaders, topics, or agendas (Thomas et al., 1982). The absence of results with statistical rigor which were present in unsophisticated analysis suggests that differing results might simply represent chance variations which would 82 likely disappear with the introduction of control groups and covariate adjustments. 83 Table 3 -uu: 0- ‘7'! . -nt 1 _ S ale -n°f: 8f_, - ,_O C E 1 ca e iv te Scale Grant (1975) Thomas et al. Dayringer & on ACL (1982) Pavia (1986) (g = 14) (fl = 81) (fl = 63) Current Study (N = 103) Number of Adjectives Checked *** * Defensiveness *tt * Favorable Adjectives Checked an: Self-control * Achievement xxx *tt Endurance *** Reterosexuality *** Personal Adjustment * Dominance * Order * * Intraception * Exhibition * Autonomy ** Succorance ** 84 Deference #0 Lability * Self-confidence *** Adult tr Adapted Child #88 Low Origence Righ Intellectence ** * Sig. increase 2 < .05 # Sig. decrease p < .05 ** Sig. increase 9 < .025 #l Sig. decrease p < .025 *** Sig. increase 2 < .01 #88 Sig. decrease p < .01 (Adapted from Derrickson, 1990) 85 l c i h le St he Mi 1 t ies tr 4 . Stephen Series courses effectively teach participants useful helping information which supplements available resources for pastoral ministries of a congregation. The use of SM training to teach this information to laity‘ extends caregiving skills to people who might otherwise struggle with how to be helpful. If the GPP, ACL, and SASB are considered measures of mental health or personality, the course does not serve as a counseling "growth” experience. The possibility exists, and cannot be easily verified, that a few specific individuals might benefit greatly from the experience. However, the program on the whole does not appear to promote any noteworthy personality or relationship changes for participants outside of changes that any group involvement might bring. Churches which offer the course suggesting that participants might personally benefit from the experience, ”feel better about themselves", or "experience increasing joy and satisfaction in all of life" (Stephen Ministries, no date), particularly if there is a choice among SM versus other group activities, operate on assumptions that have no current empirical support. The changes in personality and interactional style that people have reported (Smith, 1986) could still be present, but are likely incorrectly attributed to the training. If present, these personality changes could well be the result of group experiences in general, the natural process of 86 maturation, or self-reflection precipitated by the testing experience. Stephen Series training does cause modest changes in the participant's personality, but then so do other group experiences. The conclusion one must draw suggests that participation in anything may cause changes that are often positive, and being active is better than not participating. The possibility exists that SM training produces personality changes that are simply not measured with the instruments used in this study. This, however, would seem unlikely given the variety of scales and the emphasis on self-report. Since many of the participants are reportedly quite enthusiastic about their experience and are likely eager to present their significant time investment in training in a good light, it would reason that self-report measures would be the most sensitive to change. Observational or indirect assessment of personality characteristics would likely reveal less change as the motivation to appear somehow different would be presumed to be stronger in the self-report. The opportunity to examine the qualities of participants who did not complete the post-test portion of the research allowed some speculation on applicant screening which would maximize completion of the training. This would be particularly important for churches faced with restricting participation due to limited resources. Conclusions based on examination of those who did not 87 persist through training presumes that participants who did not complete post-test materials did not complete the course. The way the experiment was structured there were no assurances that participants who did not complete post- testing dropped out of the course. It is clearly possible that participants may have completed the training, but chose to exercise their freedom to not complete the testing materials at post-test. The presence of only 2 CSM participants failing to complete follow-up materials suggests that compliance of participants completing post- test materials after completing the course was likely quite high. Participants most likely to drop from the SM training report more original and divergent thinking, show less concern for social convention, and report a somewhat carefree and impulsive style. The SM course allows for latitude in self-expression, emphasizes an active listening style, and a helpful accepting posture towards others. The SM format, as well as values of effective helping relationships, however, do not fit well with opinion giving, impulsivity, and forceful therapist self-expression. These same qualities which make SM experiences frustrating for participants would, however, make a person a vital and central participant in a BS course, therefore encouraging participation and persistence in that context. This study has focused on a interesting anecdotal observation of a non-focal phenomena associated with SM 88 training. The absence of observed personal change in participants should not in any way detract from the central purpose of the course-~to train Christian caregivers. The effectiveness of the training in achieving this primary objective is in part supported by the one significant finding which showed new SM graduates learned specific helping beliefs. The accurate assessment of the effectiveness of the Stephen Series graduate as a caregiver would require a separate study altogether. To date only one moderately rigorous study on the effectiveness of lay Christian caregivers has been conducted (Harris, 1985). Clearly, this is one area which begs for more research attention. It could be argued that the mere presence of laity as caregivers to supplement the pastoral services is reason enough to continue offering the course, but this assumes effectiveness, and requires generalization of other research with paraprofessionals (e.g., Durlak, 1979). Specifics of the Christian lay counseling context may yield unique results from other paraprofessional research, and much could be learned about the most effective and useful way to structure lay counseling programs. Tan (1991) has identified numerous models and theories of counseling. The relative effectiveness of these program models and theories, as well as identification of vital elements of training could provide fertile research with practiCal applications for the thousands of churches seeking to implement programs. Short of such rigor, a number of alternatives exist which 89 can continue where this study ends by documenting if changes in helping beliefs transfer to meaningful behaviors. Tan's (1987) suggestion of ratings of counselor skill from vignettes and role plays is one step towards measuring the effectiveness potential of the intervention. Eventually research which directly challenges the effectiveness of Christian caregiving training and relationships in comparison to traditional pastoral counseling, traditional support of worship and Christian fellowship, and no interventions at all will be necessary. Ropefully such research will employ adequate controls, methodology and statistical sophistication as to account for all relevant factors; such rigor has not been traditionally characteristic of research with these populations (Gartner et al., 1990). Such research will also need to consider the special constructs of health or valued changes among these populations, as the constructs targeted for intervention may differ from traditional therapy goals. The results of this study find some limitations by virtue of sample qualities. The use of a rather conservative collection of mostly Protestant participants reflects the current implementation of the Stephen Series in. West Michigan, but may not be indicative of the effects of such a program in other populations or with other religious groups. Selection bias in the control group where a disproportionate number of non-participants and higher drop- out rate was observed also lend some question to the 98 reliability of these results. For now, generalizations might best be limited to Protestant congregations in conservative communities until further validation of the phenomena is documented. 1 or her res a h Since only one study could be identified which attempted to measure the change of participants in SM or similar layministries courses (Tan, 1987), hypotheses were drawn from research with CPE training courses. The presence of significant findings in this study using methods similar to the existing CPE research, which were absent with adequately controlled research methods, calls into serious question the validity of the results of the investigators of this phenomena in CPE education. Investigators measuring change which results from CPE training without appropriate control groups or methods could well be measuring the effect of taking a great deal of coursework with the same people, being displaced from families for 6 or more weeks, or making a career shift, rather than the specific effect of the training content. Better controlled, methodologically sound research will be necessary to insure the accuracy of conclusions of existing CPE courses claiming personal change. Conclusions of the summary by Derrickson, 1990 are particularly challenged by this, as all but one study (Geary, 1977) did not use an adequate control group. An additional problem that has been apparent in a number of conclusions of CPE research is the assumption that 91 change is occurring and all one has to do is search around until an instrument sensitive to the change is identified. The ACL has been the target of such faulty logic. Just a sampling can be found below: The ACL clearly has proven to be a very sensitive device for measuring change and personal growth. The SIV [Survey of Interpersonal Values] is also somewhat productive. On the other hand, the Rokeach Dogmatism Scale was not useful in this study. (Dayringer and Pavia, 1986, p. 39) The ACL obviously proved to be the most sensitive instrument used for the measuring of change in the three studies. (Thomas et al., 1982, [mislquoted in Derrickson, p. 352) The ACL was the most useful instrument in detecting positive changes and hence its use is highly recommended. If a control group is not available it would be worthwhile to have the students take the test on three or four occasions in order to evaluate more meaningfully the changes observed between testings. (Grant, 1975, p. 261) Good research assumes that no change exists until convincing evidence to the contrary is presented (Kazdin, 1980). Without the convincing evidence, the assumption of the absence of change stands. All these quoted comments run counter to logic required to reject the "null hypothesis.” 92 Additionally, Grant's (1975), suggestion that repeated administrations might substitute for control groups is misleading. This practice would only emphasize re-test effects, and clearly would not substitute for even a mediocre control group. There was little suggestion in this study that re-test effects were very prominent, but the control group did appear critical for the correct interpretation of the results. All of the authors cited above seemed to subscribe to the assumption that change is present and one simply needs to find the scale that measures it. If a measure with poor enough test-retest reliability or noteworthy practice effects is used, the uncontrolled methodology typical of these studies will yield consistent, but invalid conclusions. The Qeurnal ef Pesterel gere continues to accept CPE research on program induced change with inadequate methodology (Derrickson & Ebersole, 1986), and despite printing scathing reviews (e.g. Schmidt, 1986), the quality of empirical research remains meager with impressionistic reports maintaining the majority of the printed page (Gartner et al., 1990). The conclusions of this study, particularly the vastly different conclusions when using poor quality uncontrolled methods, should provide a convincing argument that methodological rigor is more than just being picky, but can save a researcher from misleading conclusions. 93 fiBmfl§£11 Research on effectiveness of church based programs such as caregiving ministries has been quite scarce in comparison to the apparent popularity of the movement (Tan, 1990). It might be that churches or creators of programs perceive they have little to gain and much to loose by documenting the impact of their interventions. This study, which failed to confirm perceptions of personal change in self-esteem and relationships for training participants, might even be taken as evidence of the potential risk of such exploration. Yet the benefit of this type of research, such as this study‘s convincing documentation of the effectiveness of teaching helpful beliefs about caregiving relationships, is information that can help convince congregations, ministers, and boards of the utility of expensive training and resource allocation. Clearly the next step in research of lay Christian caregiving is identification of the qualities of change among recipients which result from the caregiving relationship. A study which uses appropriate control groups, measures impact over time, assesses appropriate and meaningful constructs with reliable and valid instruments, and utilizes sound statistical and methodological criteria will add valuable and needed information to this emerging phenomena of lay caregiving. Note 1 The use of the words ”church" and "churches" in this paper is intended to be inclusive of most mainline Christian religions. While many of the inferences could be generalized to a variety of religious groups, the majority of the work presented in this paper was concerned with traditional Protestant denominations, and on some occasions the Roman Catholic church (see Appendix A, Table 4). Note 2 : ”Lay" or ”Laity" are words used to describe church members who are not members of the ordained clergy. There are some who are very particular about the usage of the word when it is used in relation to pastoral care (Cerling, 1983). The strict usage of the term ”lay” would include those who have advanced degrees in counseling or psychology, perhaps even licensure, yet are not ordained church officers. The more common usage of the term ”lay," and the usage that will be employed here, is more analogous to the term "paraprofessional." This would include those members of a church without ordination or advanced degrees relevant to the focus of the counseling mission. 94 APPENDIX A Churches from which the sample was drawn. LAEELEQ_AREA Redeemer United Methodist Church 13980 Shavey Rd. De Witt, MI 48640 First United Methodist Church 411 Harrison St. Grand Ledge, MI 48837 Rolt United Methodist Church 2321 Aurelius P.O. Box 168 Rolt, MI 48842 St. John's Lutheran Church 511 E. Sturgis St. John's, MI 48879 Pennway Church of God 1101 E. Cavanaugh Rd. Lansing, MI 48910 Trinity United Methodist Church 7533 W. St. Joseph Lansing, MI 48917 Central United Methodist Church 215 N. Capitol Ave. Lansing, MI 48933 Grace Lutheran Church 239 E. North St. Hastings, MI 49058 First Baptist Church 210 Church St. Howell, MI 48843 QBAND.RAELQE_ABEA Holy Family Catholic Church 9669 Kraft Ave. Caladonia, MI 49316 Peace Reformed Church 6950 Cherry Valley Rd. Hiddleville, MI 49333 95 96 Fair Haven Reformed Church 2025 Baldwin Jenison, MI 49428 Twelfth Ave. Christian Reformed Church 7551 12th Ave. Jenison, MI 49428 Peace Lutheran Church 1225 Twelve Mile Road Sparta, MI 49345 Roly Redeemer Catholic Church 2700 Baldwin Jenison, MI 49428 Rillcrest Christian Reformed Church 3617 Rillcrest Rudsonville, MI 49426 Roly Cross Lutheran Church 1481 Baldwin Jenison, MI 49428 First Reformed Church 3060 Wilson Ave. Grandville, MI 49418 Zion Lutheran Church 582 Lamerauex NW Comstock Park, MI 49321 La Grave Ave. Christian Reformed Church 107 La Grave Ave. Grand Rapids, MI 49503 Righland Rills Christian Reformed Church 1015 Westend Dr. NW Grand Rapids, MI 49504 Rememberance Reformed Church 1915 Maplerow NW Grand Rapids, RI 49504 Calvin Christian Reformed Church 700 Ethel Ave. SE Grand Rapids, MI 49506 Fifth Reformed Church 2012 Griggs St. SE Grand Rapids, MI 49506 97 Oakdale Park Christian Reformed Church 1313 E Butler SE Grand Rapids, MI 49507 Calvary Christian Reformed Church 3510 Byron Center SW Wyoming, MI 49509 Third Reformed Church 2060 Michigan NE Grand Rapids, MI 49503 Fairview Christian Reformed Church 1465 Three Mile Rd. NW Grand Rapids, MI 49504 Central Reformed Church College & Fulton Grand Rapids, MI 49503 Aldersgate United Methodist Church 4301 Ambrose NE Grand Rapids, MI 49505 St. Michael's Episcopal Church 2965 Wycliff Dr. SE Grand Rapids, MI 49506 St. Thomas the Apostle Parish 1449 Wilcox Park Dr. SE Grand Rapids, MI 49506 Alger Park Christian Reformed Church 2655 Eastern SE Grand Rapids, MI 49507 Sacred Reart Parish (Roman Catholic) 156 valley Ave. SW Grand Rapids, MI 49504 Eastminster Presbyterian Church 1700 Woodward SE Grand Rapids, MI 49506 Aberdeen Reformed Church 1000 Aberdeen NE Grand Rapids, MI 49505 APPENDIX B Group descriptive statistics Table 4 WWW. 4 Denomination type of number of participants Did Total affiliation group not hours ---------------------- part- pre-test post-test icipate 1. CRC NSM 4 4 0 50 CSM 2 2 0 2. RCA NSM 17 15 0 48 3. RCA NSM 14 13 0 36 4. Lutheran NSM 3 3 0 52.5 CSM 2 2 3 5. United Methodist NSM 6 6 0 50 CSM 2 1 0 6. CRC NSM 11 10 0 50 7. CRC BS(BNT) 4 2 9 20 BS(BOT) 6 6 3 21 8. RCA BS(K) 14 6 4 32 9. Lutheran NSM 5 5 0 48 10.Roman Catholic NSM 5 5 na 50 CSM 3 3 na 11.Presbyterian BS 6 3 5 14 12.Lutheran NSM 8 5 0 50 CSM 5 2 0 13.Roman Catholic NSM 11 10 1 52 CSM 4 4 0 l4.RCA NSM 8 0 0 50 15.CRC NSM 5 5 0 50 CSM 1 1 0 16.CRC NSM 9 9 0 50 l7.Episcopal NSM 4 4 0 50 CSM 2 2 0 18.United Methodist NSM 10 6 0 50 CSM 3 3 0 BS(BK) 14 5 16 13 19.RCA NSM 10 10 0 59 CSM 4 4 0 20.Roman Catholic NSM 5 0 na 51.5 Total NSM 134 106 participants CSM 29 26 * BS 44 22 207 154 98 99 Table 4 (cont.) Note: Total hours are not applicable for CSM groups as they are no longer actively in training. CSM groups receive ongoing supervision at least one meeting per month. NSM = New Stephen Ministries group. CSM = Continuing Stephen Ministries group. BS = Bible Study group. BNT = Bethel Bible Series New Testament course. BOT = Bethel Bible Series Old Testament course. K = Keregma Bible Study course. BK = Study of a Christian Book. RCA = Reformed Church in America. CRC = Christian Reformed Church. na = data not available APPENDIX C Comparison of participant scores on the ACL and GPP with available norms. Participants in this study were fairly conventional, and did not appear to represent an extreme sample, although some differences were apparent. Participant scores from the initial testing on the ACL were compared with a normative sample identified in the ACL manual (Gough & Reilbrun, 1983). Participant scores represent raw score values. ACL raw scores are often transformed to standard scores with a mean of 50 and standard deviation of 10. However, standard scores were not used in this experiment as the hypotheses were focused on observing change as a result of training rather than the identification of individual profile patterns or comparison of individual scores with normative groups. Norms available in the ACL manual are divided into male and female groups. For purposes of comparison, the participants in this study were divided and displayed by gender. Means and standard deviations of the participants on the 37 most commonly used scales ACL are presented in Table 5. 100 181 Table 5 l'-.!' 11' ’ 540.2 0. Q‘V,-. 0!‘ 0 0 -I 0 W Scale Present Sample Normative Sample' Mean Male Female Male Female (Standard Deviation) (N = 61) (N = 144) (N = 5,238) (N = 4,164) Number of Adjectives Checked 113.93 115.13 93.40 97.37 (30.77) (27.29) (36.36) (34.64) Favorable Items Checked 53.02 53.06 44.06 46.23 (12.91) (11.11) (15.74) (15.03) Unfavorable Items Checked 7.61 7.97 8.37 8.29 (9.51) (7.19) (8.84) (8.37) Commonality 12.36 12.43 12.90 13.48 (5.27) (3.10) (4.18) (3.82) Achievement 13.03 12.51 10.85 10.65 (5.56) (5.41) (6.36) (6.21) Dominance 5.52 4.10 5.52 4.64 (6.57) (5.45) (5.64) (5.93) Endurance 11.28 11.31 7.86 7.69 (5.66) (5.69) (6.63) (6.48) Order 10.82 10.38 6.98 6.28 (5.50) (5.88) (5.83) (5.94) Intraception 15.13 14.81 12.56 12.96 (4.65) (4.07) (5.27) (5.11) Nurturance 15.16 16.71 10.55 12.92 (6.61) (5.03) (6.81) (6.34) Affiliation 22.77 24.10 17.96 19.95 (6.52) (5.88) (7.39) (7.21) Reterosexuality 8.62 9.18 6.62 8.19 (4.38) (3.93) (4.25) (4.20) Exhibition -0.95 -0.83 1.78 1.34 (6.08) (5.22) (5.45) (5.72) Autonomy 0.46 -0.85 3.35 2.24 (4.85) (4.10) (5.11) (5.12) Table 5 (cont.) Aggression Change Succorance Abasement Deference -6.21 2.61 (3.88) 0.34 (3.58) 6.77 (3.69) 4.29 (5.79) Counseling Readiness Scale Self-control Self-confidence Personal Adjustment Ideal Self Scale 0.90 (2.17) 2.31 (3.78) 9.11 (5.86) 9.44 (4.30) 7.64 (5.91) Creative Personality Scale 2.92 (3.59) Military Leadership Scale 11.61 (5.74) Masculine Attributes Scale 9.31 (4.14) Feminine Attributes Scale Critical Parent Nurturant Parent Adult 12.51 (3.91) 6.39 (5.62) 11.67 (6.00) 9.72 (5.65) 102 -6.51 (5.51) 2.14 1.92 7.79 (3.89) 5.65 (4.28) 3.83 (3.38) 2.01 (3.63) 8.82 (5.15) 9.84 (3.60) 5.66 (6.10) 2.35 (3.67) 12.21 7.47 (3.49) 14.56 (3.05) 6.17 (6.13) 11.86 (5.49) 8.73 (5.82) -2.79 (5.49) 4.60 (3.80) 0.26 (3.51) -0.22 (4.81) 0.91 (5.26) -6.91 (6.02) -0.19 (4.23) 8.41 (5.21) 7.48 (4.47) 4.97 4.11 (3.98) 9.97 (5.42) 8.62 (4.41) 9.78 (3.99) 6.81 (5.81) 7.44 (6.80) 7.23 (6.20) -3.39 (5.51) 5.01 (4.13) 1.47 (3.87) 1.22 (5.07) 1.99 (5.39) 3.41 (3.64) -0.33 (4.40) 8.06 (5.33) 8.41 (4.30) 8.77 (6.49) 3.55 (4.08) 10.20 (5.05) 6.71 (3.90) 12.51 (4.04) 6.46 (6.01) 8.27 (6.60) 6.10 (6.28) Table 5 (cont.) Free Child Adapted Child Righ Origence Low Intellectence Righ Origence Righ Intellectence 1.62 (6.04) -9.90 (5.40) 6.67 (2.90) 6.84 (3.32) Low Origence Low Intellectence Low Origence Righ Intellectence 10.59 (3.43) 13.39 (4.38) 103 2.13 (5.99) -8.08 (6.13) 6.76 (3.31) 6.49 (3.72) 11.10 (2.79) 13.46 (4.28) 2.21 (5.60) -7.05 (6.33) 5.27 (3.24) 6.97 (4.44) 8.23 (3.61) 10.65 (4.84) 2.83 (5.88) -6.06 (6.43) 5.37 (3.33) 6.85 (4.39) 9.06 (3.45) 10.37 (4.80) ' From Gough & Reilbrun (1983) 104 Scales which seemed to be more heavily endorsed by this sample when compared to norms were: Number of adjectives checked, Favorable adjectives checked, Achievement, Order, Nurturance, Abasement, Deference, Self-control, and Nurturance. Less strongly endorsed among the sample were the scales Aggression and Change. Men appeared to more strongly endorse items on the counseling readiness scale than their male normative counterparts. If these differences were indeed significant and meaningful, it would be uncertain if the discrepancy between the sample and the norms was the result of the qualities of volunteers, church members, religiousness, or a willingness, to be affiliative enough to attend an education experience. The qualities of Abasement and Deference stand out as particularly noteworthy differences, and likely represent the religious qualities of the sample. Willingness to make objective and sometimes critical self-evaluation and submit to a forgiving God for mercy is presumably a quality of Christian religious values. A similar comparison between initial GPP scores and available norms was also conducted. While good norms are available for high school and college populations, much of the data on the GPP for adults is available only broken down by occupation. One sample of general female adult scores was available and is compared to the present sample in Table 6. Table 6 'H'? ‘94 0 ’Hr...‘ °-.., ’2! _ 9 " . !-. "4' WW Sample Ascendancy Responsibility Emotional Sociability Stability M (32) R (an) R (an) M (52) Female Participants N = 138 19.9 (6.7) 25.9 (5.0) 22.3 (6.0) 20.5 (6.3) General Female Adults‘ n = 179 20.2 (6.3) 24.7 (4.2) 21.6 (6.5) 22.7 (6.0) ' from Gordon (1978) Unfortunately, a corresponding sample of general adult males was not available. A fairly large sample of males in lower management at public utility was thought to be the best available normative information. The lower level manager sample is roughly consistent with the average SES ranking of III for the present sample. The GPP scale means 105 and standard deviations for male participants are compared with this sample of managers in Table 7. Table 7 01103. ‘01.. a 0- __°al (3?" 0"W !- 1310'. W Sample Ascendancy Responsi- Emotional Sociability bility Stability M (52) n (52) a (£2) u (52) Male Participants N = 53 20.7 (6.7) 26.5 (4.3) 24.4 (6.2) 18.5 (5.9) Male lower level managers“ N = 1,384 21.6 (5.8) 27.9 (4.1) 26.2 (5.3) 21.4 (5.8) ' from Gordon (1978) No outstanding differences between the participant samples and the normative samples were identified with the GPP. There appears to be a suggestion that Sociability among males was less than that for the low level managers used for comparison, but given the requirements of interaction with people required in most management jobs, this can be somewhat dismissed. Means for the comparison of participant self-report interpersonal interaction style as measured by the SASB could not be compared to normative information as a suitable control group was not available. While the SASB has established norms for ”normal" populations, the specific relationship-targets of ratings used in this survey were not among the normed information. Presentation of norms based on ratings of other types of interactions for comparison would have been misleading. APPENDIX D Outline of the Stephen Ministries training program and course content. The Stephen Series is a commercially available training product of Stephen Ministries, 8016 Dale, St. Louis, MO, 63117-1449. The Stephen Series is described as a "complete system for training and organizing lay persons for caring ministry in and around their congregations" (Raugk, 1991). This program is typical of what Tan (1991) described as an informal organized approach. The goals of the program are to provide churches with the opportunity and means to implement ”quality Christian supportive caregiving" (Voss, 1988) within their congregations. The program is quite popular. Since 1975 over 3,200 churches representing 65 denominations having implemented the program in their congregation. Over 13,175 leaders have been trained in the Leader's Training Courses which are conducted in a 12 day seminar format by the Stephen Series staff (Stephen Ministries, 1991). These trained leaders return to their home congregations to teach congregation members to implement caregiving skills in and around their congregations. Most congregations implement the program by enrolling with the Stephen Ministries organization and sending at least one pastor or another church staff person and a few selected lay leaders to the leaders training seminars. Upon completion of the seminar, recruitment of potential caregivers and the educational preparation of congregation to understand the role of the program is implemented. Volunteers are screened to identify those who understand and are ready for the challenge of the training and subsequent helping relationships. The course itself is composed of 20 modules. The contents of the modules are listed in the Table 8. Table 8. '°!'!rl‘ 450’ H00 “ 911.1404va 1. It's Not Easy 2. God as the Curegiver 3. God, You, and Me 4. Why Care? 5. Family Ties 6. Move over, Freud! 7. Touching Spiritual Depths 8. Ministering to the Whole Person 9. Servanthood vs. Servitude 10. A Surprise Gift: Forgiveness 11. Confession and Absolution over the Back Fence 12. Tools of Your Trade: Their Use and Abuse 13. Prayer 106 107 Table 8 (cont.) 14. The Bible 15. Sharing a Blessing 16. A Cup of Cold Water 17. The Evangelism-Caring Connection 18. Celebrating Results 19. Rope-Full Caregiving 20. The Thrill of it All Various options are suggested as a way to complete the 20 modules of the course. Twenty one-hour training experiences, 10 meetings for 2 1/2 hours, or 7 meetings for 2 1/2-hours with a weekend retreat are all suggestions offered in the trainers manual (Raugk & McKay, 1985). Retreats are encouraged both as a way to shorten the number of weeks required to complete the formal training and as a way to solidify group process. Attendance requirements are suggested to be very firm as the creators of the program considered group participation and interaction a high priority. Assignments for the participants to complete at home are included as part of the program. Readings include the recommendation that participants work through the next module on their own to develop questions and comments. Many groups choose to add additional books to the training materials (Raugk & McKay, 1985). Each of the twenty modules outlined in Table 8 is designed to be run with a similar format. This format can be found in Table 9. Table 9. °ll.- 0 0 , 'OI‘I ‘ “ -,!I°"“oo: 1. Quenins_£raxer.(or opening devotion)--1 to 5 minutes. 2. Lead;in--1 to 2 minutes. 3. Diseueeien_nneetiene--l0 to 20 minutes. 4. WWW-~20 to 40 minutes. 5. Gethe;ing_zegetheL--1 to 10 minutes. 6. Qleaing--1 or 2 minutes. (Raugk & McKay, 1985, p.15) 108 In addition to the individual units, participants are presented with a number of topical units in the trainee manual that are intended as educational resources for relevant counseling situations. The Table of contents for these materials are presented in Table 10. Table 10. WW - Introductory Resources - First Relping Contact Resources - Feelings Resources - Listening Resources - Telecare Resources - Christian Caring Resources - Assertiveness Resources - Professionalism Resources - Confidentiality Resources - Small Step Approach Resources - Community Resources - Termination Resources - Crisis Theory and Intervention Resources - Hospitalization Resources - Grief Resources - Death and Dying Resources - Depression Resources - Suicide Resources - Older Persons Resources - Shut-In Resources - Divorce Resources - Inactive Member Resources - Childbirth as a Family Crisis Resource - Jo-Rari Window Resources - Everything Stephen Ministers Want to Know About Caring... - Supervision Resources - Stephen Series Verbatim Resources - Referral Resources - vacancy Resources - The Future of Stephen Ministries (Stephen Ministries, 1983) Once training is complete Stephen Series Graduates are "commissioned" and referrals are sought and assigned by a coordinator. Coordination and supervision is typically the responsibility of a pastor of a given church. Stephen Series graduates typically take only one or two referrals at any given time. The recommended frequency of contact ranges, but the standard expectation is about 1 hour once a week. 109 One of the central features of the Stephen Ministries groups, relatively unique among lay counseling programs, is the emphasis on continuing supervision. Even after they are fully trained, participants in the Stephen Ministries programs are expected to participate in at least monthly supervision meetings with the trainers. Some of the changes that may occur for the participants as a result of their involvement can be expected to be continually supported and maintained through these supervision and continuing education meetings. For further information, please contact: Stephen Ministries 8016 Dale St. Louis, MO 63117-1449 (314) 645-5511 APPENDIX E Letter of reference from Stephen Ministries STEPHENMINISTRIES 1325 Boland . St. Louis, Missouri 63117 - 314/645-5511 Dear Stephen Leaders and Stephen Ministers: In my travels for Stephen Ministries and over the telephone, I often hear Stephen Ministers make statements like: "Stephen Ministry has changed my life." "I get more out of this ministry than I give.” "I've grown so much as a result of the Stephen Series." "The Stephen Series has been a turning point in my life." Maybe you, too, have heard such a statement from your Stephen Ministry colleagues, or maybe you yourself have expressed such an observation. Martin J. Wealkes, the grandson of a Stephen Leader in Grand 'Rapids, a member of a Grand Rapids church using the Stephen Series (Fifth Reformed Church), and a doctoral student in clinical psychology at Michigan State University, is attempting to assess and evaluate the change Stephen Ministers experience in the way they think and feel about themselves as a result of their participation in the Stephen Series training process. In other words, he is trying to document scientifically and academically the personal growth so many Stephen Ministers experience through Stephen Ministry training. Stephen Ministries of St. Louis, the producer of the Stephen Series, has approved the purpose and design of Mr. Waalkes' research and has provided him with a selected list of congregations using the Stephen Series. We believe the outcome of Mr. Waalkes' study will be of significant value to you and to all congregations using the Stephen Series. When you receive the research results summary, you will see how the Stephen Series training process has affected a larger group of individuals. .When we at Stephen Ministries receive the summary, we will share its results with our constituent congregations. They, in turn, will be able to share this information with their own members thus enhancing congregational preparation and Stephen Minister recruitment. 110 111 We invite and encourage your participation in this important and worthwhile study by Mr. Waalkes. Should you have any questions or concerns about the purpose or process of this research as it relates to Stephen Ministries of the Stephen Series, please feel free to call or write me at our office in St. Louis. With you, in the Service of Christ, The Rev. Gary C. Voss Director of Training Appendix F Letter to trainers describing terms of participation Dear Stephen Ministry Training Director; Thank-you for your expressed interest in the caregiver training research program. This letter is just a reminder about some of the details involved with helping us implement the research in your church. I've also enclosed a copy of a personal reference so that you have some understanding of who I am. I hope that when the research is complete that I can make some useful statements about the personal changes that occur as a result of involvement with this type of a ministry. This letter will explain the commitment should you choose to become involved. You are our main contact. I will expect that you will communicate the intent and expectations of the project to your Stephen Ministry class. I want you to be able to explain to them the nature of the research in a manner which they can understand the importance, but not so much that they unintentionally bias their responses to ”help me out”. I also hope that you will feel comfortable enough with my intentions and procedures that you may be able to answer any questions that will arise during the presentation of the research to your class. Basically, you can present the research to your class (or anyone else who may need to approve it) in this way: "The researchers are a doctoral student in psychology at Michigan State University and his professors. They want to understand what changes, if any, that people make in their lives, and the way they think and feel about themselves as a result of a training program like this one. In order to do this they need to measure those qualities before the training begins, and again after it ends. These testing sessions will occur in a group before the first meeting, and after the last meeting. Each testing will take about an hour. Your participation will be completely voluntary, and if you choose, you are not obligated to finish. If you choose not to participate it will not effect your welcome in the training program. Your names will not be used on the test data and will only be used on a separate sheet to keep things straight until the second testing. I (i.e. you) will be doing the actual testing and the testing will occur ... (you set the exact time). What they hope to do with their project is make some useful statements about what happens to people when they go through a program like this one. They will provide us with a summary of these results when the data is reviewed, probably sometime in the Fall of next year." ' 112 113 So that you can best plan the research into your programming for your pending class, here are the requirements of the research program: 1. A pre-test session: This will be a group testing session for all the volunteering participants that need to occur PRIOR to the first meeting. I understand that your enthusiasm for beginning a new class may be high, but it is essential that an accurate measurement be obtained BEFORE any of the components of the course have begun. The exact timing of the testing is up to you, but it is likely that the most representative results will be collected if it happens just prior to your first session, perhaps separated by a short break. In this pretest session, I would like you, as the leader or trainer, to present the research project to your class as a strictly voluntary opportunity, making certain that they understand that their decision to be involved is unrelated to their welcome in the training. It is necessary that they understand that their participation can be discontinued at any time, including at the post- test. You should also tell the class that they will be completely anonymous in their responses, although certain code procedures will be employed to assure the match-up of post-testing tests. The participants will all receive a consent form explaining these things to them. I want you to be my administrator for the testing, and I will send you all the materials that you will need to distribute, based on the number of participants you expect. Since you will be administering the tests, I would like you to be familiar with how to answer questions your class might have, so you will receive more specific instructions and the ”answers” to some commonly asked questions. 2. A post-testing session: This session is identical to the first session in content, and it needs to occur just following the last session you meet with the group as part of their formal training. I understand that you may be less interested in the project and finishing a course with a test may not be your preferred style, but the necessity of both the post-test and the relative timing of this last administration are important. While the exact time of the final administration is up to you, it must follow the last session and would probably be best 114 administered immediately after the conclusion (on that same night or day). The problem with meeting at another time is that people, even though they may want to contribute, cannot get motivated to return for a hour of testing. The same conditions apply for this second test as for the first one. You will be the administrator of the instruments and you will inform the class that this is a voluntary, anonymous opportunity. You will answer any questions they might have according to the provided instructions. When the testing is complete, you keep the name list and return the instruments to me. 3. A feedback session: I am certain that this will be the best part for you and the class. Sometime next Fall I hope to have made sense of the research. I will send you a brief summary of the results, and you will than share this information with your class. This group feedback reflects the only way I will use the results, so any individual scores and performances on the tests cannot be relayed because anonymity would prevent their identification. This summary, based on the results of all the participating churches will also be sent to the Stephen Ministries Headquarters since they have been very helpful in the early stages of this project. I also want to get some idea about what happen for people who have already been trained and are continuing under supervision as Stephen Ministers. What I would like is for you to invite to these testing sessions one graduate for every three current trainees to take the testing with them. These people can be told essentially the same thing as the trainees, but you may want to emphasize to them that I am trying to get the whole picture of the influence of the program, including the people who are already trained. These people would need to participate in both testings as well, but could drop out if they did not want to continue. It is VERY important that you choose these additional people randomly. It would not be representative if you choose the people you thought would be most interested. I suggest that you take your list of trained Stephen's Ministers and begin by asking the first person on the list. Next ask the third, fifth, and so on until you have one graduate participant for each three trainees. I know this makes some extra work for you, but the extra information they provide will be very valuable in understanding the lasting effects of the program. I hope this has been helpful in elaborating on the nature of our project. I will be contacting you soon to make firm your commitment and make arrangements for the 115 actual testing and other details. If you have any questions, or want to talk about the research, please call me, Martin Waalkes (616-243-5655). If I'm not home, you can leave a message and I'll call you back. Sincerely, . Faculty Supervisors: John Powell, Ph.D. Martin J. Waalkes, M.A. Norman Abeles, Ph.D. Doctoral Candidate APPENDIX G Code correspondence form CODE to NAME correspondence key This information is very confidential. After participants hand in their test, place their name after the code number on their test. When everyone had returned their test and filled in their name, this page should be sealed in the provided envelope and the group leader should seal that envelope. The group leader Should sign his/her name across the envelope flap to assure confidentiality. The envelope should be returned with the rest of the materials. When the follow-up testing is conducted, this envelope will be returned to the group and opened so participants will take the materials which correspond to their original code numbers. Code Number Participant Name 116 APPENDIX R Consent cover letter for research participants Dear Group Member, You have been asked to participate in a research project about personality changes resulting from group participation in church-related activities. This research is being conducted by Martin J. Waalkes, M.A, through the department of psychology at Michigan State University, under the supervision of Norman Abeles, Ph.D. and John Powell, Ph.D. This research involves two testing sessions that will take about an hour each. One of these testings will occur before you begin your group, and the other after you finish. Your participation in this research is strictly voluntary, If at any time, for any reason you with to discontinue your involvement, you are free to do so and you will continue to be welcome in the group. Your participation in the first testing does not obligate you to complete the second testing. Your responses will be completely confidential. The tests will have identifying codes on them, but this is only to match your first test with the second test. It works like this: 1. You will get a test packet with a code number on it. 2. After you fill the test out, hand it in to the leader. 3. The leader will have a sheet with the code numbers on it. 4. Place your name next to the code number on your test. 5. The leader will seal this in an envelope and return it with the materials. 6. When the time of the second testing comes, the leader will re-open the sealed envelope and you will get a testing packet that has the same code number as the first test you took. 7. When you finish the second testing the leader will destroy the list and return your materials to me. Since the records are completely anonymous, please feel free to be completely honest in your responses. Sometime in the Fall of 1989, I will report the results of the research to your leader who will then share them with you. Unfortunately, because of the confidentiality restrictions, feedback on individual results will not be available. There is no guarantee that you will receive any personal benefit from your involvement in this research. The results may not even agree with what your own experience of the groups was. 117 118 Your signature on this form is only an indication that you are aware of the terms of the research and does not obligate you in any way to participate in the research. The form is for records purposes only, and cannot be used to identify your test responses. An additional copy of this form is provided to allow you to keep a copy. The research project has been explained to me and I understand this explanation, including what is expected of me and my rights in relation to those expectations: Signed: Date: APPENDIX I Cover letter to trainers providing instructions for administration of instruments. These instructions should explain everything you should need to know to administer the instruments to your group. SECTION 1 Details about how and when to administer. 1. On this initial assessment you will need to set aside about 1 hour and 15 minutes to 1 and a half hours to administer the tests. .Be certain this time comes before any of the content of your course so that a true assessment of participants state upon entering the group can be obtained. You may choose to administer this at any time, but it works best if you do it immediately prior to your first group meeting, perhaps separated by a brief break before introducing the content of your course. 2. Find enough 42 pencils for your group. I apologize for not providing these, but the cost to mail them was prohibitive. 3. Present the materials to the participants saying something like this: I'm going to hand out the packets of test materials that I told you about when you expressed interest in the class. This is research that is being done by Martin Waalkes, a doctoral student in psychology at Michigan State University. He is studying the changes people experience when they participate in a group such as this one. Inside these envelopes you will find a consent form which details the expectations of the research. There are two copies of this form. Keep one copy of the consent form for your own records and sign the other copy, returning it with your test responses. -Since your responses are anonymous, please be frank and do not put your name on anything but the consent form. -Work quickly, don't labor over a single item. Just give the first answer that comes to mind. -Read the instructions that accompany each test and fill out the answers accordingly. -Make your marks on the computer forms dark and make them fill the circle with no stray marks. -Start with the top test and work through all the forms without skipping any until you get to the 119 ' 120 end. -Don't forget about the back sides of the tests, particularly the first six computer sheet pages. After giving these instructions and handing out the tests, wait with the participants until they have completed the test so you can answer any questions that might arise. SECTION 2 Row to answer questions from participants about the tests and items. The purpose of the tests is to obtain an accurate measure of how the participant feels at the moment. It is important to stress that there are no correct answers and everybody will respond differently. If you get asked any questions at all they will likely be one of the following. On page 1 of the blue computer form: 0: ”I don't know what to put for occupation.” -— encourage the participant to give a concise generic description of their job. Homemakers and retired people put the spouses or past occupations here. Q: ”What does this 'new student' mean?" -- This applies only to Stephen Ministries classes. If the person is just beginning Stephen Ministries training, then check new student. On the Adjective Check List (the 300 item checklist that starts on page 1): 0: ”How many can I check?" -- participants can check as may as they feel describe them. Q: ”Where do I put my answer?" -- Participants should blacken the 41 circle for those adjectives they consider to be self-descriptive. Q: "What does mean?” -- ”Even if I did know, I couldn't tell you. Just go by what you think it means.” The Helping Beliefs Inventory (on the back of the last page of the computer forms): 0: ”How do I do this?" 121 -- ”Just read each statement and blacken one of the circles from 1 to 5 next to the item that shows how helpful that would be to the person. Use the scale in the instructions to decide which number is most appropriate." 0: ”Is helpful?" —- ”There are no right or wrong answers. Just mark what you think is most helpful". 0: ”Do I put down what I actually do or what I think I should do?" -- ”Put down what you believe will be most helpful". The SASB (this is the folded 32 item inventory): 0: ”Do I need to fill out this stuff at the top?" -- ”Please do not. Just fill in your ratings for the questions.” Q: "What if I am rating a man?" -- "That is fine, just read he instead of she in the sentence" 0: "What does he mean by 'remember the person you rate'?" -- "When you take this test again later on he wants you to rate the same person over again.” For the Gordon Personal Profile (the wide test with the bold blue banner across the top and the squares on the back): 0: ”Do I fill out anything on the front side?” -- ”No, just read the instruction and fill out the test on the back" 0: ”Do I answer each statement?” -- ”No, just pick one in each box of four that is most like you and mark 'M' and pick one that is least like you and mark 'L'.” Q: ”what if none of the statements is most (or least) like me?" -- ”Just choose the one(s) that fit better than the others. remember, it only means most or least, not exactly like me" SECTION 3 Row to conclude the testing and what to do next. 122 1. As the participants finish their tests, have them come up and write their name on the code-name key form provided with the materials. 2. Have the participants replace all their tests in the coded envelopes. 3. Collect the signed consent forms separately. 4. Put the code-name correspondence key in the provided envelope and sign the back. 5. Put all the materials in the postage paid return envelope. Including: All signed consent forms All used and unused tests in their respective individual envelopes. The sealed and signed name-code correspondence key 6. Mail the envelope 7. Call me if there are any problems: MARTIN J. VAALKES (616) 243-5655 (home) (517) 355-9564 (MSU Psychological Clinic)’ APPENDIX J Narrative description of results provided to participants. Personal Change Resulting from Lay-Ministry Christian Caregiving Training. Martin J. Waalkes, M.A. October 4, 1991 Churches have been recognizing the value of the often under-utilized pool of talent and support that exists among the members of the congregation. As pastors are often burdened with the volume of the demand for individual caregiving, some churches have turned to a more purposeful development of members of the congregation as skilled Christian caregivers by implementing educational programs to teach caregiving skills. Most churches stop short of calling these counseling training courses, but an examination of the content often reveals a strong emphasis on listening skills and a non-directive, supportive approach that addresses the individuals emotional, spiritual, and in some cases, physical needs. One popular version of these skill training courses is the Stephen Ministries course. The benefits of a helping relationship have long been recognized as bi-directional. That is to say both the person providing support and the recipient typically report mutual experiences from the interpersonal exchange. Likewise, those involved in Stephen Series courses have reported their perceptions that they feel that they had gained something from the participation in the course alone that changed the way they feel about themselves and their relationships with important people in their lives. Reports range from improvements in self-esteem, patched-up relationships, improved self-confidence, and emotional reflectivity. While not intended for this purpose, these observations have lead some to believe that the program itself, while intended to teach students how to be helpful to others, could benefit those in need of help. These conclusions are not without basis. The lenrnel ej_£ee§ere1_§e1e has printed several articles identifying the change on self-report measures for pastors undergoing Clinical Pastoral Education (CPE) courses. These studies have measured students before and after the CPE training and found statistically significant changes on measures of sensitivity, autonomy, achievement, and self-control. Presumably, these same types of changes could occur in laity who receive such specialized training. 123 124 METHOD To assess the possibility that those receiving training as Christian caregivers directly benefit from the experience, 207 volunteers were recruited from Christian congregations in the West Michigan area. Of the 207 volunteers, 134 were new students in the Stephen Series. To get a rough estimate if the experiences built over time, or were enhanced by practicing as a caregivers, 29 experienced Stephen Ministers were assessed at the same time. Finally, the participating churches were polled as to other educational opportunities, and 44 volunteers were included who had enrolled in Bible study series such as ”Keregma” or ”Bethel Bible Series". This last group served as an important "control" group to serve as a comparison for those natural changes that could be expected to occur over time, and as a result of being in a group on a regular basis in church. Of the 207 volunteers, 106 new Stephen Ministry students, 26 experienced Stephen Ministers, and 22 Bible Study participants completed the pre and post tests. The pretest was administered on the day of the first group, while the post-test, administered only to those who had completed pre-tests, was given after the 50 hour Stephen Series training. The testing materials consisted of the Gordon Personal Profile (GPP), the Adjective Checklist (ACL), the Structural Analysis of Social Behavior (SASB), and the Helping Beliefs Inventory (RBI) in addition to a few demographic questions. The GPP is a self-report measure which includes measures of 4 personality traits and a measure of self esteem. The ACL is another self-report measure that can be scored to yield 37 scales measuring style if responding, need and drive states, relative valuing of emotional and rational qualities, and styles of interacting with others and perceiving oneself. The SASB is a theoretically-grounded, complex, self-report instrument for the measurement of the various aspects of an interpersonal relationship. The RBI is a short, but effective, set of questions which can reliably discriminate therapists from non-therapists based on the presence of non-directive counseling beliefs. RESULTS First analyzed were the concerns which allow confidence that the data analyzed for the effect can be trusted. Comparisons of the subjects who completed pre-tests only (presumably dropping out before the end of the course) were different in a number of ways than those who completed both pre and post testing. Those who completed pre and post testing had higher socio-economic status (SES), and the two Stephen Series groups had higher SES as well. Age or gender did not factor as a concern with regard to the presence or absence of post testing or group affiliation, but did play a role in some later analyses. As a result, all the remaining 125 analyses were conducted with the effect of age, gender, and SES on the data statistically controlled. The Bible study group participants who did not complete post testing were different in a number of ways from other participants. They were more likely to be constrained by expectations of others and more reliant on routine. They scored lower on a measure of desire to think originally or different. These Bible study participants who did not complete post-tests were also less likely to take an expressive or impulsive style of interacting, and instead report assuming a deferential style of interacting. In contrast, new Stephen Series students who did not complete post-testing had higher scores for measures of original and different thinking, showed less concern for social convention, and report acting with a more impulsive style. Perhaps these students had a more difficult time adjusting to the non-directing style of the caregiving, or simply grow restless quickly and move on. The next data analyzed was the focus of the research. When the data from the participants who completed both pre- test and post-test was examined the hypothesis would require an "interaction" which showed scores changed over time depending on which group was examined. For the GPP, SASB, and the ACL (the measures which were to assess a change in personality or interactional style), no support was found to suggest that Stephen Series participants, new or experienced, changed as a result of participation in the training. One noteworthy hypothesis was supported. There is clear evidence that new Stephen Series participants are as unaware as Bible Study students when it comes to helpful counseling values prior to the start of the course. The experienced Stephen Series participants, however, possess these beliefs already (presumably by virtue of their previous training). Once the course is complete and the post-test administered, the Bible Study students express the same, less effective, beliefs about helping, while the newly educated Stephen Series group now express values and beliefs about helping akin to their experienced counterparts. The similarity of the scores on the RBI for continuing Stephen Series participants at pre-test and post- test and the post-test scores of the new Stephen Series participants suggests that the helping beliefs are fostered in the Stephen Series coursework and not out of experience as a caregiver (which may well serve to reinforce and maintain these beliefs). Out of curiosity, the data was analyzed one final way. Pre-test scores were simply compared to post-test scores for only the new Stephen Series group. This is not a recommended procedure for this particular research problem, but is nonetheless the way all but one of the 39 known CPE research projects were conducted and analyzed. In the absence of control groups and covariates several "changes” in personality and interpersonal style measures were 126 observed. When analyzed in this loose fashion, new Stephen Series participants appear to change to express more emotional stability confidence, and a decreased need to solicit support for themselves. They showed increases in interests for taking a leading role in relationships, understanding their own or others' behavior, and assuming an analytical and logical style. These qualities were identified as having grown in addition to the increase in identifying the useful helping values mentioned above. DISCUSSION AND CONCLUSIONS The hypothesis that the Stephen Ministry training course somehow changes the personality or basic interactional style of participants was not supported in this research. Additionally, the suggestion that the experience of involvement in helping relationships over time could change the helper was not supported either. New Stephen Series graduates do, however, learn useful and effective helping values that they did not know prior to the course, and they maintain these values after experience with caregiving relationships. These findings have several implications: 1. Stephen Series courses teach valuable skills that supplement the pastoral ministries of a congregation and extend caregiving skills to people who might otherwise struggle with how to be helpful. 2. The course does not serve as a counseling experience. Churches which offer the course to members suggesting that they might personally benefit from the experience operate on assumptions that have no current support. 3. The changes in personality and interactional style that people have reported could still be present, but are likely miss-attributed or are not measurable in conventional means. If present, these personality changes could well be the result of increased church attendance and belonging to a group. It certainly would not be the first time positive behaviors were associated with increased church attendance. 4. The presence of findings when using methods similar to the existing CPE research which disappear with adequately controlled research methods calls into serious question the validity of the results of the investigators of this phenomena in CPE education. CPE change investigators could well be measuring the effect of taking a great deal of coursework with the same people rather than the effect of the content. This entire project has focused on a interesting observation of a non-focal phenomena associated with Stephen 127 Series training. The absence of personal change should not in any way detract from the central purpose of the course-- to train Christian caregivers. The effectiveness of the course in achieving this primary objective is in part supported by the one significant finding which showed new series graduates learned important information about helping relationships. The assessment of the effectiveness of Steven Series graduates as caregivers would require a separate study altogether. It could be argued that their mere presence as caregivers is reason enough to continue offering the course. APPENDIX K Demographic information collection form Cover Letter for Testing Materials and the Information Sheet CODE Dear Group Member: What you have here is a packet of questionnaires. These tests measure aspects of the way you feel about yourself, others, and some ideas you may have about different topics. Since you are anonymous in your response, feel free to be completely honest in your responses. DO NOT PUT YOUR NAME ON ANY OF THE QUESTIONNAIRES OR THIS PAGE. Follow the instructions on each test, and answer every question, even if you have to give an answer that is just close. If you have questions, your group leader should be able to answer them for you. When you are done, bring your tests to the leader and the leader will record your name on the code sheet so you can get the correct second test when the time comes. CIRCLE ONE: Male Female AGE:‘____ HOW MUCH FORMAL EDUCATION HAVE YOU HAD? WHAT TYPE OF A JOB DO YOU HAVE? (What do you do for a living)? (put husband‘s occupation if you are a homemaker and indicate previous occupation if you are retired) WHAT KIND OF A GROUP ARE YOU IN HERE: (Check One) ___ Bible Study Prayer Group Fellowship Group Stephen Ministry Training' '(Indicate if: New trainee , or 1 year grad , or 2 or more year graduate ) Other (List: ) 128 APPENDIX L Description of the Adjective Checklist. The Adjective Check List (ACL) (Gough and Reilbrun, 1983) is an established personality assessment instrument with research and clinical applications. The history of the use of adjective lists in personality assessments is long, and the ACL itself has undergone numerous revisions and has had many new scales contributed from diverse theoretical positions. The test itself is a list of 300 adjectives that participants read to indicate those which describe them best. Scales are scored based on the pattern of endorsement. The scales are in many cases the product of research with factor analytic techniques. Some of the scales have been refined through theoretical means and subsequently verified and improved with research. Gough (1960) calls the former "Empirical" and the later "Rational” scales. The ACL has been normed on over 9,000 subjects from various age, functioning level, and socio-economic backgrounds. These scores have been used to develop normative tables so that individuals can be compared to norms using T-scores. In some cases ACL scores have been used to describe others, ideal-self, past-self, or other constructs. Internal consistency and test-retest correlations vary depending on the scale used, and values for each of the scales are published in the manual with separate values for males and females. The median internal consistency coefficients for the scales was .76 for females, and for males was .75. The range of test-retest coefficients at one year for males was from .34 for high origence-low intellectence to .77 for Aggression with a median of .65. The female sample had higher median test-retest correlations with .71. (Gough and Reilbrun, 1983). Such figures seem reasonable for a self-report personality instrument. Gough and Reilbrun suggest that reliability is variable within people as well, and that certain personality types will report more irregularly, both within the test and over time, than others. The validity and utility of the instrument can, in part, be attested to by the large number of applications it has been successfully employed. The manual itself, however, makes no specific mention of validity information, and validity must be inferred from the various research that has been done on the individual scales. The general consensus of reviewers is that the ACL is a valid and useful instrument for the measurement of self-concept (Zarske, 1984). 129 APPENDIX M Description of the Gordon Personal Profile The Gordon Personal Profile (GPP) is a popular multi- scale self-report instrument for personality trait assessment. A companion instrument to the Gordon Personal Inventory, the GPP was developed from factor analytic techniques on forced choice items in an attempt to represent useful and meaningful measures of personality. The test consists of 18 items. Each item consists of a tetrad of statements from which the subject must choose one statement that is ”most like me" and one statement that is "least like me”. Each tetrad includes items individuals typically regard as equally complimentary, and two that are equally uncomplimentary. The positive and negative pairings underwent five revisions before they were judged to be similar in social desirability in large administrations. As a result, some of the influences of social desirability are disarmed with the expectation to choose between equally good and equally poor choices. The instrument still has limitations in detecting a faked presentation. Braun and Farrell (1974) have demonstrated that college students, when given instructions to appear as an ideal management candidate, would inflate all but their Sociability Scale scores in comparison to an administration that asked that they be honest and forthright. These inflated scores were obtained even though subjects were told that there was a (fictional) scale to detect faking. While this supports the position of that intentional efforts to assume a fabricated identity cannot be detected, it does suggest that some subtle score adjustment from social desirability exists. While various methods of scoring the scales have been devised (c.f., Van Tassel, 1972), including weighting item contributions depending on their content, scoring popular responses, and relating scores with other tests, the more traditional method (used in this study) involves assigning the value +1 for positive statements, and -1 for negative statements on a given scale. The four basic scales, the profile measures, are Ascendancy, Responsibility, Emotional Stability, and Sociability. One additional Scale, Self:£stssm (SE), a recent addition to the test, is derived from the other four Personal Profile Scale scores. While it can be argued that social desirability may be the underlying construct, the perception of what is socially desirable and the process of the incorporation of those qualities in the personality is possibly central to the development and manifestation of self-esteem. The reliability of the GPP appears sufficient to regard the test as useful. A split-half procedures show adequate reliabilities for the four component scales (see Table 11). 130 131 Table 11 9 . '!-__ Te 19! 1 t7? 0 ne 0 9°! " 79!: ’_0 _l' $£§l£§ Trait Reliability Ascendancy .88 Responsibility .85 Emotional Stability .88 Sociability .84 (Gordon, 1963, p.21) Stability of the scores for 127 men enlisted in a twenty-nine week Navy training program ranged from r = .50 for Emotional Stability to r = .70 for Ascendancy. The stability of scores over four years of optometry training for three classes of 50 or more students was consistently in the range of .50 to .60. Considering the amount of change presumed characteristic of medical school and military training, these seem to be respectable test-retest reliability values. Scores from the GPP are typically transformed into T- scores based on published norms. Normative information is available for different populations, particularly college students, but Reilbrun (1965) notes the absence of a general population norm that would not be biased with youth and education. Examination of the items indicated that the GPP has face validity. For example: "takes the lead in group discussions" (Ascendancy), "sees a job through despite difficulties" (Responsibility), "calm and easygoing in manner” (Emotional Stability), and "enjoys having lots of people around” (Sociability). After reviewing studies published at that time, Reilbrun (1965) concluded that the GPP, like all tests of this sort, is somewhat limited in its power to predict behavior, yet some significant relationships with job performance could be identified. APPENDIX N Description of the Structural Analysis of Social Behavior The SASB (Benjamin, 1974) is one of the best developed instruments available for interpersonal assessment. Benjamin's three surface operationalization of the interpersonal circle has quickly become regarded as the heir to Leary's (1957) interpersonal checklist for measuring the dimensions of power and affiliation in the interpersonal relationship. Essentially, the full battery assesses three aspects (called ”surfaces”) of interpersonal self-image. They are the (Surface 1) other, Where the respondent describes the behaviors of another person in relation to the self (See Figure 2), (Surface 2) 521:, where the respondent describes the behaviors of the self in relation to others (see Figure 3), and (Surface 3) lnllflnfixgniQ, or QLDQL: Intgeieet, where the participant responds to questions dealing with the the result of the self focusing on the self (Benjamin, 1979). The interpersonal circle theory on which the instrument is based has two axes on which ratings of perceptions for each surface are made. The vertical axis is the power or control dimension, and the horizontal axis is the affiliation or affect dimension. It is theorized that interpersonal behavior can be thought of in terms of the combinations of these two variables (Kiesler, 1983). It is also thought that a pattern of consistency will characterize responses within each surface such that item values result in a pattern of responses which can be characterized by a single style. This is based on the assumption that one cannot be both nurturing and attacking at the same time, just as one cannot be freeing and managing. There are 3 basic patterns of behavior which are identified, drawn from a possible 12 (the discrete nature of the assessment limits what could theoretically be a continuous model). These patterns correspond to mathematical models when responses are charted by cluster. Examples of these pattern coefficients and their clinical manifestations can be found in Table 12. 132 133 Figure 2 Freeing and Forgetting I I I I I Ignoring and I Affirming and Neglecting I Understanding Invoke I Encourage Hostile I Friendly Autonomy I Autonomy Attacking and --------------- + -------------- Nurturing and Rejecting l Comforting Hostile I Friendly Power I Influence I Belittling and I Helping and Blaming I Protecting III I IV I Watching and Managing (Rudy et al., 1985 p.267) 134 Figure 3 WW Asserting and Separating I II I I Walling off and I Disclosing and Avoiding I Expressing Take I Enjoy Hostile I Friendly Autonomy I Autonomy Protesting and -------------- + -------------- Approaching Withdrawing I and Enjoying Hostile I Friendly Comply I Accept l Sulking and I Trusting and Appeasing I Relying III I IV I Deferring and Submitting (Rudy et al., 1985 p.267) 135 Table 12 Focus and Pattern sign of Coefficient Attack Control Conflict Focus on OTHER (Surface 1) POSITIVE Attack Control Power Double-bind NEGATIVE Friendly Give autonomy Attachment Double-bind Focus on SELF (Surface 2) POSITIVE Protest & Submit Ambivalent withdraw differentiation NEGATIVE Friendly Take autonomy Ambivalent attachment adapted from Benjamin (1988, p.77) Pattern coefficients for individuals can then be calculated by determining the goodness of fit between the individual subject score and a given ideal mathmatical pattern. Using the pure values for coefficients, rather than the pooled values of 3 approximated pattern values, an individual's pattern coefficient could be said to be indicative of the style of interacting if the absolute value of the coefficient was greater than .71. Coefficients for this data were calculated using the FIGMAS program available through the Intrex Interpersonal Institute. The short form versions of the SASB (Benjamin, 1988) offer decreased test time requirements by limiting the focus of the assessment and removing redundant assessment of clusters of behaviors. Instead of several questions, the short form relies on only one item per cluster and relies on the internal consistency of the larger model and the treatment of data as indicative as a pattern of responses to sustain reliability. The short form requires only about 20 minutes to complete. The 2 surfaces of the SASB that will be employed in this study are the interpersonal focus on other surface (Surface 1) (see Figure 2) and the interpersonal with a 136 focus on self surface (Surface 2) (see Figure 3). Since subjects rate both themselves and the other person on both these surfaces, the result is 12 coefficient scores (3 coefficients for self on the two surfaces and 3 coefficients for ratings of how others are perceived to relate to self on the two surfaces). The aspects are elaborated in Table 13. Table 13 o e , g e 0,. - o! g o . o 0 Q , - ,_- - ir . . Aspect Surface Description Aspect 1 Rating ”other" Self (Surface 1) Rater's perception of other's focus on rater. Aspect 2 Rating ”other" Other (Surface 2) Rater's perception of . other's focus on self in relation to rater. Aspect 3 Rating ”self" Self (Surface 1) Rater's focus on other. Aspect 4 Rating "self" Other (Surface 2) Rater's focus on self in relation to other. APPENDIX 0 Description of the Helping Beliefs Inventory The HBI (Mc Lennan, 1985) was developed to distinguish trainees that held beliefs about helping known to be conducive to effective counseling relationships from ones that held beliefs not conducive to a therapeutic relationship. HBI items ask participants to rate the effectiveness of helping strategies that distract the person from their immediate experience, control and direct people in solving their problems, and divert people away from exploring their situation (see Table 14). Table 14 2. Tell the person what you think they should do 3. Try to cheer the person up 5. Tell the person to look how lucky they are in other things 6. Tell the person about similar experiences you have had 7. Tell the person not to worry about the problem 8 Give the person sensible advice about how to handle the problem 10. Tell the person what is the right thing to do 11. Try to get the person's mind off the problem and onto something else 13. Tell the person what you think you would do 14. Tell the person what someone else did in that situation Distractor Items: (not scored) 1. Listen to what the person wants to say 4. Offer the person support and sympathy 9. Suggest alternatives for the person to try 12. Get the person to clarify their feelings about what is happening Scoring scale: almost never likely to be helpful seldom likely to be helpful sometimes likely to be helpful often likely to be helpful almost always likely to be helpful U'IuwaI-f IIIIIIIII (Mc Lennan, 1985) 137 138 The original version of the RBI was pilot tested on 120 diverse subjects as 10 helpful and 10 non-helpful items. The 10 non-helpful items were found to cluster and account for 36 percent of the variance on the test. These non- helpful items, in conjunction with 4 non-scored distractor items to counter a response-set effect, compose the present version of the RBI (Mc Lennan, 1985). Test-retest reliability was established with a group of 43 college psychology students over a 10 week interval. Reliability was adequate (r = .74), but in light of information the students may have learned over the course of the training, the indicated decrease in mean scores was not unreasonable to expect. As a result, actual test-retest reliability could be higher than this report. For validation, the Inventory was administered to 30 counselors and 230 members of recreational clubs who indicated an interest in helping others, but had no experience. Mean scores for counselors (M = 20.2, an s 6.26) were lower than those of the non-counselor group (M = 29.3, 52 = 6.48) suggesting stronger endorsement of an active and directive interactional style by the non- counselor group (Mc Lennan, 1985). Non-counselors were more likely to strongly endorse each individual item than the counselor group. Among the non-counselors, the higher (less helpful) RBI scores were found to correlate with measures of authoritarianism, inflexibility, and low psychological mindedness. In another validation study Mc Lennan (1985) administered the RBI to 45 counselor trainees prior to training. At the end of training, the trainers, who were unaware of the trainees' scores, rated the trainees into groups of high, medium, and low skill. Highly rated therapists had the lowest pre-training scores (22.8; SD = 2.80), followed by the medium skill rated group (27.4; SD = 2.64), with the low rated skill level group having had the highest scores prior to training (30.7; SD = 3.28). All the differences in the means were significant (p < .05). The RBI appears to represent a good means of reliably assessing beliefs about helping relationships that are relevant to conventional perceptions of skillful caregiving interactions. The extent to which these beliefs are actually related to client change or perceptions has yet to be established. The validation process of comparing scores to supervisor ratings of competence leaves the possibility that the test measures capacity to adopt conventional, although not necessarily helpful beliefs about helping. APPENDIX P Means and Standard deviations for subjects completing both pre-test and post-test E (2, 140) = 0.041 139 variable N Bible Study New Stephen Continuing Pre / Post Minister Stephen Minister (Levene Statistic) Mean (SD) Mean (SD) Mean (ED) (N = 22) (N = 106) (N = 26) Demographic Measures N = 207 Age Pre 44.84 (15.25) 44.67 (13.52) 46.48 (13.16) Socioeconomic Status Pre 30.98 (13.56) 37.49 (14.97) 38.79 (14.84) E (2, 204) = .378 Helping Beliefs Inventory N = 141 Pre-test 24.85 (6.28) 24.65 (6.71) 19.17 (4.79) E (2, 200) = 0.631 Post-test 24.15 (6.06) 24.36 (6.17) 19.71 (6.60) E (2, 141) = 0.326 I Gordon Personal Profile variables N = 135 Ascendancy Pre-test 22.53 (7.92) 19.65 (6.90) 21.44 (6.83) E (2, 188) = 0.341 Post-test 20.71 (8.14) 20.34 (6.13) 22.48 (6.50) E (2, 140) = 1.05 Responsibility Pre-test 24.35 (5.98) 26.00 (4.56) 26.52 (4.76) E (2, 188) = 0.650 Post-test 24.47 (6.19) 26.07 (5.35) 26.91 (4.86) E (2, 140) 8 0.495 Emotional Stability Pre-test 22.24 (7.30) 22.58 (5.94) 22.91 (6.30) E (2, 188) = 0.262 Post-test 22.06 (7.16) 23.88 (5.90) 24.09 (6.07) Sociability Pre-test E (2, 188) = 0.488 Post-test E (2, Self-Esteem Pre-test E (2, 188) = 0.268 Post-test E (2, 140) = 0.1656 ADJECTIVE CNECKLIST VARIABLES Number Checked Pre-test E (2, 202) = 0.585 Post-test E (2, 145) = 0.453 22.29 (7.86) 21.18 (8.42) 140) = 4.809" 91.41 (17.84) 88.50 (21.47) N = 148 114.00 (32.37) 111.48 (31.33) Favorable Items Checked Pre-test E (2, 202) = 0.149 Post-test E (2, 145) = 0.551 49.62 (14.06) 50.19 (11.45) Unfavorable Items Checked Pre-test E (2, 202) 8 1.122 Post-test E (2, 145) - 2.161 Commonality Pre-test E (2, 202) = 0.023 Post-test E (2, 145) = 0.449 Achievement Pre-test E (2. 202) Post-test E (2, 145) 0.032 0.841 Dominance Pre-test E (2, 202) Post-test E (2, 145) 0.114 2.619 9.67 (9.47) 7.86 (9.06) 11.95 (3.33) 11.86 (2.80) 11.76 (5.97) 12.86 (5.97) 3.48 (6.76) 4.29 (7.50) 140 19.45 (5.92) 19.78 (5.92) 87.82 (16.33) .35 (15.14) 117.75 (26.64) 117.73 (30.15) 53.60 (11.15) 54.64 (10.01) 8.72 (9.12) 7.43 (6.35) 11.42 (4.68) 12.67 (2.87) 12.61 (5.53) 13.39 (4.78) 3.99 (6.04) 4.85 (5.35) 20.61 (4.74) 20.48 (4.91) 91.17 (14.30) 94.54 (15.98) 110.04 (30.54) 109.13 (28.84) 53.75 (12.45) 52.54 (12.32) 5.29 (5.94) 4.83 (4.74) 12.38 (2.30) 12.25 (2.29) 12.88 (5.83) 12.38 (5.40) 4.83 (5.47) 5.21 (5.01) Endurance Pre-test 10.33 (6.14) E (2, 202) = 0.111 Post-test 11.33 (6.00) E (2, 145) = 1.308 Order Pre-test 9.81 (6.24) E (2, 202) = 0.174 Post-test 11.00 (5.22) E (2, 145) = 0.264 Intraception Pre-test 13.95 (5.46) E (2, 202) = 0.604 Post-test 14.00 (4.53) E (2, 145) = 3.625‘ Nurturance Pre-test 14.33 (5.95) E (2, 202) = 0.348 Post-test 15.05 (4.33) E (2, 145) = 0.036 Affiliation Pre-test 22.43 (7.05) E (2, 202) = 0.668 Post-test 22.29 (6.09) E (2, 145) 8 2.658 Reterosexuality Pre-test 8.38 (4.31) E (2, 202) = 0.974 Post-test 7.91 (4.00) E (2, 145) = 0.173 Exhibition Pre-test -1.00 (6.72) E (2, 202) = Not available Post-test -1.00 (5.44) E (2, 145) - Not available Autonomy Pre-test 1.48 (4.93) E (2, 202) = Not available Post-test 0.24 (4.78) E (2, 145) = Not available Aggression Pre-test -4.43 (7.15) E (2, 202) I Not available Post-test -5.57 (6.45) E (2, 145) = Not available 141 11.16 11.67 10.66 10.92 14.90 15.89 16.24 17.09 23.95 24.71 8.86 9.10 -1.07 -0.82 -0061 -0.78 -6.74 -6.70 (5.56) (4.81) (5.21) (5.14) (4.25) (3.32) (6.33) (4.51) (6.03) (5.07) (4.17) (3.84) (5.80) (5.72) (4.83) (4.35) (5.45) (4.83) 12.08 11.17 10.75 10.25 16.21 15.63 17.92 17.71 24.38 24.00 8.79 8.50 -0004 0.96 -2013 ’1.50 -7021 -5.88 (5.88) (5.93) (7.03) (5.97) (4.10) (4.28) (4.17) (4.36) (6.47) (6.72) (3.56) (4.25) (4.10) (4.66) (3.49) (4.01) (4.55) (4.10) 142 Change Pre-test 3.43 (4.24) 2.22 (3.71) 1.33 (3.33) E (2, 202) = 0.795 Post-test 3.14 (5.12) 1.75 (3.90) 2.13 (3.52) E (2, 145) = 1.495 Succorance Pre-test 1.95 (4.85) 1.78 (3.83) 0.92 (3.89) E (2, 202) = 0.010 Post-test 2.14 (4.81) 1.15 (3.57) 0.25 (3.96) E (2, 145) = Not available Abasement Pre-test 7.95 (4.94) 7.75 (3.65) 6.67 (3.77) E (2, 202) = 2.276 Post-test 7.62 (5.06) 7.21 (3.79) 6.88 (3.76) E (2. 145) = 1.135 Deference Pre-test 3.14 (4.88) 5.54 (5.18) 6.21 (3.54) E (2, 202) = 1.813 Post-test 3.95 (4.48) 5.82 (4.71) 6.42 (3.43) E (2, 145) = 0.790 Counseling Readiness Scale Pre-test 4.05 (2.96) 2.82 (3.24) 3.21 (3.67) E (2, 202) = 0.489 Post-test 3.76 (3.95) 2.64 (3.40) 2.88 (3.28) E (2, 145) = 0.472 Self-Control Pre-test 1.24 (4.06) 2.32 (3.91) 2.13 (2.51) E (2, 202) = 3.436‘ Post-test 2.05 (4.40) 2.22 (3.73) 2.38 (2.68) E (2, 145) = 2.928 Self-Confidence Pre-test 7.76 (5.99) 8.47 (5.71) 10.29 (5.11) E (2, 202) = 0.171 Post-test 8.00 (6.54) 9.85 (4.78) 9.63 (5.22) E (2, 145) = 2.484 Personal Adjustment Pre-test 9.14 (3.98) 9.49 (4.15) 11.21 (3.62) E (2, 202) = 0.262 Post-test 8.91 (4.39) 10.11 (2.80) 10.38 (3.52) E (2, 145) = 2.148 Ideal Self Scale Pre-test 4.67 (6.04) 6.04 (6.11) 7.67 (6.82) E (2, 202) = 0.084 Post-test 5.48 (6.26) 6.62 (6.36) 7.63 (6.91) E (2, 145) 8 0.094 Creative Personality Scale Pre-teSt 3000 (3023) E (2, 202) = 0.773 Post-test 2.76 (3.56) E (2, 145) = 0.705 Military Leadership Scale Pre-test 11.38 (4.61) E (2, 202) = 0.024 Post-test E (2, 145) = 1.296 12.19 (5.00) Masculine Attributes Scale Pre-test 7.62 (3.65) E (2, 202) = 0.225 Post-test 7.19 (3.62) E (2, 145) = 1.593 Feminine Attributes Scale Pre-test 13.38 (4.56) E (2, 202) = 1.477 Post-test E (2, 145) = 0.696 12.71 (3.48) Critical Parent Pre-test 8.05 (6.96) E (2, 202) = 0.384 Post-test 6.91 (6.39) E (2, 145) = 1.139 Nurturing parent Parent Pre-test 9.05 (6.54) E (2, 202) = 0.247 Post-test 10.19 (5.90) E (2, 145) = 0.994 Adult Pre-test 8.38 (6.75) E (2, 202) = 0.588 Post-test 9.62 (5.64) E (2, 145) = 0.352 Free Child Pre-test 2.52 (6.12) E (2, 202) = 0.416 Post-test 1.95 (6.73) E (2, 145) = 0.249 Adapted Child Pre-test -7.10 (7.31) E (2, 202) = Not Available Post-test -7.95 (7.32) E (2, 145) = Not Available 143 2.26 (3.77) 2052 (3051) 11.97 (4.98) 12.88 (3.54) 8.08 (3.95) 8.65 (4.02) 14.18 (3.39) 14.37 (3.48) 6.35 (6.23) 5.72 (5.74) 11.89 (5.88) 12.78 (5.02) 8.65 (5.52) 9.73 (5.06) 1.55 (6.52) 1.55 (6.32) -8018 (5095) -9.73 (5.47) 3.17 (4.01) 3.46 (4.19) 12.67 (4.02) 12.42 (3.51) 7.42 (3.83) 7.42 (3.22) 14.50 (2.81) 13.92 (2.72) 3.96 (5.63) 4.25 (5.27) 13.04 (5.76) 11.96 (6.54) 10.29 (6.60) 10.08 (5.72) 2.75 (5.53) 2.67 (6.10) -9.50 (6.74) -9.63 (6.99) 144 High Origence Low Intellectence Pre-test 6.43 (3.41) 6.96 (3.03) E (2, 202) = 0.447 Post-test 5.95 (2.84) 6.71 (3.63) E (2, 145) = 1.354 High Origence High Intellectence Pre-test 7.52 (3.52) 6.80 (3.76) E (2, 202) = 0.343 Post-test 6.81 (3.75) 6.70 (4.16) E (2, 145) = 0.165 Low Origence Low Intellectence Pre-test 9.86 (3.28) 11.19 (2.85) E (2, 202) = 1.866 Post-test 9.95 (3.06) 11.56 (2.48) E (2, 145) = 1.478 Low Origence High Intellectence Pre-test 13.14 (4.53) 13.45 (4.16) E (2, 202) = 0.090 Post-test 13.57 (3.93) 14.36 (4.12) E (2, 145) = 0.137 Structural Analysis of Social Behavioral Scores N s 148 Non-Transformed Aspect 1 Attack Pattern Coefficient Pre-test -.44 (.50) -.68 (.34) E (2, 201) = Not Available Post-test -.45 (.56) -.68 (.34) E (2, 146) = Not Available Aspect 1 Control Pattern Coefficient Pre-test -.23 (.44) -.22 (.28) E (2, 201) = Not Available Post-test -.15 (.40) -.21 (.26) E (2, 146) = Not Available Aspect 1 Conflict Pattern Coefficient Pre-test .11 (.18) .10 (.25) E ('2, 201) = 0.333 Post-test .17 (.23) .10 (.23) E (2, 146) = 0.893 Aspect 2 Attack Pattern Coefficient Pre-test -.54 (.51) -.71 (.30 ) E (2, 201) = Not Available Post-test -.57 (.49) -.76 (.21) E (2, 146) = Not Available 5.79 (3.30) 5.46 (2.99) 5.92 (3.94) 6.50 (4.34) 11.00 (2.43) 10.92 (2.45) 13.96 (4.81) 13.38 (4.43) -.71 (.24) -.63 (.38) -.30 (.21) -.26 (.24) .04 (.26) .16 (.27) -.69 (.27) -.67 (.31) 145 Aspect 2 Control Pattern Coefficient Pre-test -.33 (.23) -.22 (.26) E (2, 201) = Not Available Post-test -.39 (.24) -.26 (.27) E (2, 146) = Not Available Aspect 2 Conflict Pattern Coefficient Pre-test .05 (.13) -.03 (.19) E (2, 201) = Not Available Post-test .08 (.16) -.07 (.18) E (2, 146) = Not Available Aspect 3 Attack Pattern Coefficient Pre-test -.58 (.42) -.73 (.31) E (2, 201) = Not Available Post-test -.64 (.28) -.74 (.26) E (2, 146) = Not Available Aspect 3 Control Pattern Coefficient Pre-test -.38 (.27) -.22 (.27) E (2, 201) = Not Available POSt-tQSt -037 (027) -027 (020) E (2, 146) = Not Available Aspect 3 Conflict Pattern Coefficient Pre-tCSt 012 (013) 011 (021) E (2, 201) 8 0.311 Post-test .20 (.20) .11 (.20) E (2, 146) = 1.408 Aspect 4 Attack Pattern coefficient Pre-test -.49 (.52) -.74 (.25) E (2, 201) = Not Available Post-test -.53 (.44) -.76 (.23 E (2, 146) = Not Available Aspect 4 Control Pattern Coefficient Pre-test -.28 (.26) -.14 (.27) E (2, 201) - Not Available Post-test -.24 (.28) -.16 (.27) E (2, 146) - Not Available Aspect 4 Conflict Pattern Coefficient Pre-test .07 (.19) -.03 (.19) E (2, 201) = Not Available Post-test .13 (.25) -.03 (.18) E (2, 146) = Not Available -.22 (.23) -034 (022) .02 (.15) .07 (.21) -.81 (.14) -077 (021) -024 (022) -.28 (.26) .08 (.15) .07 (.14) -080 (015) -.72 (.26) -024 (019) -.17 (.32) .03 (.16) .01 (.21) 146 SASB Coefficients Transformed to Absolute Value Aspect 1 Attack Pattern Coefficient Pre-test .62 (.22) .72 (.23) .71 E (2, 201) = 0.567 - Post-test .65 (.28) .74 (.20) .68 E (2, 146) = 5.102“ Aspect 1 Control Pattern Coefficient Pre-test .45 (.21) .30 (.20) .33 E (2, 201) = 1.667 Post-test .36 (.22) .29 (.17) .31 E (2, 146) = 1.332 Aspect 1 Conflict Pattern Coefficient Pre'tOSt 017 (012) 019 (019) 015 E (2, 201) = 0.112 Post-test .20 (.20) .18 (.18) .20 E (2, 146) = 1.353 Aspect 2 Attack Pattern Coefficient Pre'tGSt 069 (025) 074 (021) 071 E (2, 201) 8 0.006 Post-test .70 (.25) .76 (.21) .68 E (2, 146) I 2.151 Aspect 2 Control Pattern Coefficient Pre-tCSt 036 (016) 028 (019) 027 E (2, 201) = 0.120 Post-test .39 (.24) .31 (.21) .34 E (2, 146) = 0.313 Aspect 2 Conflict Pattern Coefficient Pre-test .11 (.09) .15 (.13) .11 E (2, 201) = 3.484‘ Post-test .15 (.10) .15 (.12) .15 E (2, 146) 8 2.053 Aspect 3 Attack Pattern Coefficient Pre-test .68 (.20) .76 (.21) .81 E (2, 201) = 0.018 Post-test .66 (.24) .76 (.21) .77 E (2, 146) = 1.189 Aspect 3 Control Pattern Coefficient Pre-test .42 (.19) .29 (.18) .29 E (2, 201) 8 1.697 Post-test .38 (.26) .30 (.15) .35 E (2, 146) 8 8.605" (.23) (.28) (.18) (.17) (.22) (.24) (.23) (.27) (.17) (.21) (.09) (.16) (.14) (.21) (.15) (.17) 147 Aspect 3 Conflict Pattern Coefficient Pre-test .13 (.13) .17 (.17) .12 (.12) E (2, 201) = 0.214 Post-test .24 (.16) .15 (.17) .11 (.11) E (2, 146) = 2.193 Aspect 4 Attack Pattern Coefficient Pre-test .67 (.21) .75 (.21) .80 (.15) E (2, 201) = 0.804 Post-test .64 (.25) .76 (.23) .72 (.26) E (2, 146) = 0.222 Aspect 4 Control Pattern Coefficient Pre-test .30 (.24) .25 (.18) .25 (.17) E (2, 201) = 0.1802 Post-test .31 (.18) .25 (.19) .28 (.23) E (2, 146) = 0.849 Aspect 4 Conflict Pattern Coefficient Pre-test .15 (.14) .14 (.13) .12 (.11) E (2, 201) = 3.332' Post-test .21 (.19) .14 (.12) .16 (.12) E (2, 146) = 3.258‘ Note: The Levene statistic computed for pretest includes data from those participants who completed only the pre-test. Levene statistics are not available for variables with medians which have a negative value. Probability > .05 unless otherwise noted. ' s p < .05 " = p < .01 Note: For SASB Scores, Aspect values represent the following: Aspect 1: Other Surface 1; Other's focus on rater. Aspect 2: Other Surface 2; Other's focus on Self in relation to rater. Aspect 3: Self Surface 1; Rater's focus on other. Aspect 4: Self Surface 2; Rater's focus on self in relation to other. APPENDIX 0 Table 16 0'0'g0.' - -0 “ ‘ .- ‘ - 0". gender. variable N Pearson correlation coefficient Age SES Gender’ Gordon Personal Profile variables N = 191 Responsibility -.196" Emotional Stability -.146' -.155' Adjective Checklist variables N = 205 Endurance -.209“ Order -.211" Exhibition -.150‘ Aggression -.193" Change -.248" Succorance .185" Deference .185" Counseling Readiness Scale .401" Self-control .137‘ Ideal Self Scale -.149' Masculine Attributes Scale -.223" Feminine Attributes Scale -.272“ Critical Parent -.139‘ Nurturing Parent .164“ Adult .165‘ -.136‘ Adapted Child .140“ High Origence Low Intellectence -.154' High Origence High Intellectence -.208“ Low Origence High Intellectence -.182" Helping Beliefs Inventory N a 203 .148' -.192" Structural Analysis of Social Behavior Scales N - 204 Aspect 2 Control pattern coefficient .148' Aspect 4 Attack pattern coefficient -.141‘ Aspect 4 Conflict pattern coefficient -.148‘ 148 149 ‘ 2-tailed probability p < .05 " 2-tailed probability p < .01 * Note: Values for relationships with the gender variable are Pearson point bi-serial correlation coefficients. Negative values indicate a relationship of higher scores for Female participants. APPENDIX R Participant predispositions and characteristics of non-persisting participants W The pre-test only vs. pre-test and post-test main effect (hereafter referred to as the persistence main effect) of the 2 x 3 (persistence x group) Analyses of variance (ANOVA), with covariates age, SES, and gender, was examined to identify characteristics which might be common to individuals who did not participate in the post-test. Participants who completed the materials (N = 35.03, 52 a 14.06) had a lower SES rating than those who did not complete follow-up materials (M a 35.03, an = 16.58; E(1,20l) = 9.88, p < .002). Age did not differ between persisting and non-persisting participants (2 > .05). Eighteen male and 35 female participants did not complete post-testing, a ratio which was representative of the composition of the groups at the outset (Chi-Square (2) = 1.234, p > .05). When age, SES, and gender were employed as covariates, no main effects for the persistence factor were observed for any of the remaining variables listed in appendix RA (2 > .05). WW When participant scores were compared for group main effects to reveal differences which existed at pre-test irrespective of persistence for post-test, some differences were observed. A main effect for group was observed for SES (E(2,201) = 6.53, p < .002). Comparisons of group means using the Tukey Honest Significant Difference (HSD) procedure (not considering the covariate influences, and using p < .05) suggested mean SES ratings for participants in the NSM group (M = 37.49, an = 14.97) were higher than participants of the BS group (N a 30.98, an 8 13.57). This implied that BS participants had higher socioeconomic status than the NSM's. No main effect for participant age was observed for group choice. Seventy one percent of BS participants were female, while 69 percent of NSM and 79 percent of CMS participants were female. When compared to the overall level of 71 percent female participants, no differences in gender between groups were observed (Chi- Square (2) a 1.302, p > .05). With the 2 x 3 (persistence x group type) Analyses of variance (ANOVA), with covariates age, gender, and SES, no main effects for group choice were observed for GPP variables on pre-test scores. A main effect for group type was observed for the RBI (E(2, 194) a 7.31, p < .001). The Tukey HSD method (with p < .05 and no covariates) identified the CSM group as having a lower HBI mean score (fl = 19.81, 32,- 5.55) than the NSM (n - 25.05, 82 - 6.72) or BS (N - 25.77, an - 6.38) group (which did not differ significantly from each other). This suggests that CSM participants reported stronger identification with non-directive and non- 150 151 distracting helping values at the time of the first session than the NSM and BS participants. The only ACL variable found to have a group main effect (p < .05) was A3, low origence low intellectence (E(2,194) = 3.44, p < .034). Tukey's HSD procedure (not considering the covariate influences, and with p < .05) identified lower scores in the BS group (M = 9.93, £2 = 3.21) than the NSM group (M = 11.21, SD = 2.99). The CSM group (M = 11.24, SD = 2.39) was not found to be different from either group. This suggested greater conventionality, contentment, and optimism among new participants choosing to be involved in SM opportunities. Group Main effects were also observed for five of the pre-test SASB variables. ANOVA's with covariates age and SES for the Aspect 1 Attack pattern coefficient (E(2,195) = 7.71, p < .001), and Aspect 3 Attack pattern coefficient (E(2,194) = 3.39, p < .036), both identified pre-existing differences between groups. Tukey HSD examination of Aspect 1 Attack pattern means suggests the BS group (M = -.494; ED 8 .470) was had less strong correlations to the negative manifestation of the Attack pattern than both the NSM (M = - .694; £2 = .329) and the CSM group (M = -.726; an a .228). The Aspect 3 Attack pattern coefficient main effect for group type did not withstand the removal of covariates and the stringency of the Tukey HSD method to yield differences in group means, despite a significant main effect in the initial analysis. All 3 pattern coefficients for interpersonal Aspect 4, the rater's focus on self in relation to the other, revealed significant pre-existing differences between groups (p < .05). For aspect 4, these results were as follows: Attack pattern coefficient (E(2,194) = 5.86, p < .003), control pattern coefficient (E(2,194) = 3.26, p < .041), and conflict pattern coefficient (E(2,194) a 3.34, p < .038). Tukey's HSD comparisons (not adjusted for covariates, and with p < .05) suggests the mean correlations with the negative version of the attack pattern for aspect 4 is lower for the BS group (M = -.568; an = .420) than the NSM (M = - .734; £2 = .269) and CSM group (M = -.767; SD = .216). is different from the SM groups on many of these identified interpersonal measures. Using the Tukey HSD method The Aspect 4 Control pattern group type main effect did not yield any differences in group means despite a significant main effect in the initial analysis. On the Aspect 4 Conflict pattern coefficient measure, BS subjects (M = .068, an a .234) showed stronger correlations, albeit minimal and likely meaningless, to the Conflict pattern than NSM (M = - .032, an a .184). CSM participants (M = .013, ED 8 .163) did not differ from NSM or BS participants on the Aspect 4 conflict measure. Interactions between the persistence factor and group type suggested a number of features characteristic of persistent BS participants that were not characteristic of 152 non-persisting SM participants. There were 22 BS participants, 28 NSM participants, and 2 CSM participants who did not complete post test materials. The limited number of CSM subjects who did not complete post-test materials restricts comments to BS and NSM groups. To aid clarity of interpretation, those variables with significant interactions were analyzed with a second ANOVA, with age, SES, and gender as covariates, exelnding_§he_§fiM_da§e. Interactions effects of these variables and the interpretations of those interactions with significant ANOVA interaction effects are presented in Table 17. Table 17 Group BS NSM Mean (SD) Mean (an) variable Non-persisting Persisting E values GPP variables N a 164 Responsibility Non-persisting scores 28.91 (5.10) 25.22 (5.70) Persisting scores 24.05 (5.71) 26.02 (4.45) Persistence effect: E(1, 157) = 3.05 Group type effect: E(1, 157) = 0.01 Interaction: E(1, 157) = 9.57“ Sociability Non-persisting scores 19.18 (4.78) 21.39 (4.76) Persisting scores 21.58 (7.86) 19.61 (6.66) Persistence effect: E(1, 157) = 0.00 Group type effect: E(1, 157) = 0.10 Interaction: E(1, 157) = 3.14 (group means are equal) ACL variables N s 176 Autonomy Non-persisting scores -2.10 (4.57) 1.11 (3.29) Persisting scores 1.86 (5.14) -.57 (4.86) 153 Persistence effect: E(1, 169) .01 Group type effect: E(1, 169) - .01 Interaction: E(1, 169) = 9.35“ Aggression Non-persisting scores -7.76 (5.56) Persisting scores -3.95 (7.33) Persistence effect: E(1, 169) .05 Group type effect: E(1, 169) = .64 Interaction: E(1, 169) = 5.75‘ Change Non-persisting scores 0.00 (3.96) Persisting scores 3.41 (4.14) Persistence effect: E(1, 169) = 0.66 Group type effect: E(1, 169) 8 1.67 Interaction: E(1, 169) = 16.00“ Deference Non-persisting scores 7.10 (4.32) Persisting scores 2.91 (4.89) Persistence effect: E(1, 169) a 0.04 Group type effect: E(1, 169) = 0.02 Interaction: E(1, 169) = 10.99" Self-control Non-persisting scores 3.76 (3.74) Persisting scores 1.05 (4.06) 0.03 0.19 Persistence effect: E(1, 169) Group type effect: E(1, 169) Interaction: E(1, 169) = 8.71' Creative Personality Scale Non-persisting scores 1.19 (3.31) Persisting scores 3.18 (3.26) Persistence effect: E(1, 169) = 0.94 Group type effect: E(1, 169) = 1.15 Interaction: E(1, 169) = 6.22“ Nurturing Parent Non-persisting scores 13.57 (3.79) Persisting scores 9.14 (6.39) Persistence effect: E(1, 169) = 0.48 Group type effect: E(1, 169) = 0.73 Interaction: E(1, 169) = 6.66" -5.93 -6063 4.52 2.14 10.78 12.03 (4.54) (5.44) (2.93) (3.78) (3.32) (5.20) (2.69) (3.90) (3.17) (3.80) (4.65) (5.86) Free Child Non-persisting scores Persisting scores Persistence effect: E(1, Group type effect: E(1, Interaction: E(1, 169) = SASB variables N 8 176 Aspect 3 Conflict Non-persisting scores Persisting scores Persistence effect: E(1, Group type effect: E(1, Interaction: E(1, 169) = 154 0.48 (5.35) 2.68 (6.02) 169) = 169) = 4.61‘ 1.07 0.47 .20 (.21) .11 (.14) 169) = 0.09 169) = 1.93 .07 (.18) .10 (.21) 3.18 (Group Means are Equal) Note: CSM values were excluded from these analyses due to the small number of participants (2) who did not complete post-test materials. ' p < .05 “ p < .01 Two variables, Sociability on the ACL and the SASB Aspect 3 Conflict pattern coefficient, did not demonstrate a robust enough effect that the interaction found in the 2 x 3 (time x group) ANOVA with covariates could be observed in the 2 x 2 (time x group) ANOVA with covariates. of effect in the sample, CSM participants in the second analysis, could have been due to the influence of the scores of CSM participants, once it was restricted to exclude or the reduced power with the smaller sample size. The absence APPENDIX 8 Repeated measures 3 x 2 (group x time) ANOVA SASB post-hoc main-effects A number of unexpected relationships were noted between SASB variable scores and the group and time factors. The results reported here were drawn from 2 x 3 (Group x Time) repeated measures ANOVA's with age, SES, and gender as covariates which were performed to explore the primary hypotheses assessing the impact of training and experience on the scores of SR participants. While the primary hypotheses observed the interaction effect to determine if change occurred, some main effects were also observed among SASB variables. Only one variable was observed to have a significant time effect. The Aspect 2 Control pattern coefficient revealed evidence of change over time (1,145) = 5.48, p = .02). Post-test values (M = -.288 £2 = .264) showed stronger correlations to the negative sign of the control pattern than pre-test values (M = -.234 an e .251). These changes on the participant's ratings of the other's relationship with self in relation to the rater showed relatively meaningless changes in a small movement towards a style of taking autonomy (see Table 12). The magnitude of change, as well as the overall value of the larger of the two coefficients still represent correlations to the conflict pattern in the relatively uncorrelated range. Benjamin (1988) suggests individual scores must correlate at least .72 to be considered indicative of the pattern. There were seven variables which revealed main effects for the group-type factor. These variables, and the corresponding E-ratio values are identified in Table 18. 155 156 Table 18 4: :_ -.9 7‘ -!° 0 9°19!- - 0 0 I. 0! ‘0: -.n n.-. l 0 ° 0 0 0 1 0 o u' -.-._ ‘0 II (0 "‘ 0° 0 :0 - o- W SASB Variable F ratio (DF) probability Aspect 1 Attack 5.74 (2, 142) .004 Aspect 2 Attack 4.41 (2, 142) .014 Aspect 2 Conflict 8.38 (2, 142) .000 Aspect 3 Attack 3.61 (2, 141) .030 Aspect 3 Control 3.48 (2, 141) .034 Aspect 4 Attack 8.63 (2, 141) .000 Aspect 4 Conflict 5.31 (2, 141) .006 While it might seem useful to explore the differences between means of these SASB variables, understanding these main effects is complicated by the knowledge that values sum across pre and post-test for each subject. Analyses that reveal a more accurate understanding of the existing differences between groups are the analysis of the 2 x 3 (time x group) repeated measures ANOVA interaction effects (which revealed no differences for any SASB values) and the analysis of pre-test differences (which did identify some differences). The preexisting SASB variable differences at pre-test are elaborated in first portion of the results section. APPENDIX T Examination of variables with a relationship to the Helping Beliefs Inventory and comparisons to previously measured populations Participant scores from the initial administration of the RBI were compared to initial values for other variables in the study. Two of the covariate demographic values had significant relationships with the RBI. Participants of lower socioeconomic status were more likely to espouse helping beliefs thought to be less useful in caregiving relationships (1(201) = .15, 2-tailed p < .035). This effect could well be related to education, which has a lage role in the determination of SES. Females were more likely to have lower initial RBI scores (M = 23.7, SD = 6.75), indicative of stronger endorsement of helpful beliefs than Male participants (M = 26.5, SD = 6.33; E(1, 201) = 7.65, p < .006). On the ACL masculine attributes scale, higher masculinity scores were associated with higher (or less helpful) scores on the RBI (1(200) = .17, 2-tailed n < .017). The corresponding relationship with feminine attributes was not observed (p > .05), suggesting that some of the masculine qualities of directiveness, control, and assertiveness associated with everyday ideas of masculinity might be more responsible for higher RBI scores among males than the idea of the feminine attributes being responsible for lower scores. This is to say that while poor ideas about helping may relate to masculine qualities, being a woman or having feminine qualities would not necessarily predispose one to be helpful. This could be construed as partial corroboration of Mc Lennan's (1985) finding that high scores on the RBI were related with high scores on a measure of authoritarianism, and low scores on measures of flexibility and psychological mindedness. Two other ACL scales were observed to have significant relationships with the RBI: Number of adjectives checked(r(200) = .14, 2-tailed p < .045) and Al, the High Origence, Low Intellectence scale (1(200) a .25, 2-tailed p < .000). The correlation with the ACL A1 scale suggests less helpful RBI scores were also associated with the relative higher valuing of emotions and feelings over logic and reason. Given the variety of traits measured with the SASB, ACL, and GPP, surprisingly few significant results emerged. This relative absence of results suggest the RBI may be measuring something rather unique, and clearly modifiable. That the RBI has been equated with counselor skill (Mc Lennan, 1985), speaks well for the ability to more easily impart helping strategies without needing to change personality characteristics of the helpers. 157 158 WWW annulations As the RBI is a relatively untried scale, few ”definitive” norms are available. In constructing the scale, Mc Lennan (1985) did, however, collect scores from a number of populations. For comparison, these values are presented along with values from this study in Table 19. Table 19 WW aamnleh Population N M ED Mc Lennan (1985) Introductory psychology students 43 29.4 4.0 (10 weeks later) 43 27.9 4.6 High skill volunteer trainees 12 22.8 2.8 Moderate skill volunteer trainees 20 27.4 2.64 Low skill volunteer trainees 13 30.7 3.28 Employed professional counselors 30 20.2 6.26 Non-counselor adults 230 29.3 6.48 Current Study pre-training CSM 27 19.8 5.55 NSM 132 25.1 6.72 BS 44 25.8 6.38 post-training CSM 24 19.7 6.60 NSM 97 20.4 6.17 BS 20 24.2 6.06 The values in Table 19 appear to suggest that the beliefs adopted by SM participants are commensurate with those adopted by the skilled counselor trainees as well as employed professional counselors in the Mc Lennan (1985) study. The scores for SM participants appear substantially lower than the non-counselor adults and the introductory psychology students. The Bible Study participants, while significantly higher (p < .05)than the SM participants in this study still seem to espouse more helpful beliefs than Mc Lennan's non-counselor group. If the Bible Study participants do indeed have more helpful beliefs about helping than the typical non-counselor adult, and BS participants and potential SM participants have equal initial beliefs, then the SM training appears to help those already inclined to hold helpful beliefs to refine them even 159 more effectively. This conclusion assumes that differences from 29 to 24 and from 24 to 20 are meaningful differences from a clinical perspective while validating research of this sort has not yet been collected. Clearly, more normative information is necessary to interpret these values more reliably. REFERENCES Albee. G. w. (1959). Mental_flealth_uanneuer_11ends. New York: Basic Books. Becker, W. W. (1987). The paraprofessional counselor in the church: Legal and ethical considerations. leggne1_ef Baxsh2123x_and_§hr1stianitx, §(2). 78-82. Benjamin, L. S. (1974). Structural analysis of social behavior. Baxshglesical_8exiex. 81, 392-425- Benjamin, L. S. (1979). Structural analysis of differentiation failure. Eeyehiafiry, 42, 1-23. Benjamin, L. S- (1988)- SAaB_Short_Enrm_usetis_Manual. available from: INTREX Interpersonal Institute, Inc., P.O. Box 55218, Madison, WI 53705. Berman, J. S., 5 Norton, N. C. (1985). Does professional training make a therapist more effective? Earshelesisa1_flullet1n. 28. 401-407- Bolsen, A. T. (1936). The_Ean12rat12n_2f_the_1nner_flgrld- New York: Harper & Brothers Publishing. Boisen, A. T. (1960). an_ef_the_nepthe. New York: Harper & Brothers. Braun, J. R., & Farrell, R. M. (1974). Re-examination of the fakeability of the Gordon Personal Inventory and Profile: A reply to Schwab. Esxshglgsisal_aennrts. 34. 247-250. Brenner. D. (1982). The.£ffectixe_fsxshotheranist. New York: Pergamon Press . Bufford, R. K., & Buckler, R. E. (1987). Counseling in the church: A proposed strategy for ministering to mental health needs in the church. 1enrne1_e£_fieyehelegy_end Christianitx, 6(2), 21-29. Cerling, L. C. (1983). Selection of lay counselors for a church counseling center. Jsurnal_2f_£sxshglnsx_and WI 2.: 67-72 - Clinebell. H. J., Jr. (1965). Menta1_flea1th_Thrgush gagieeien_gemmuni§y. New York: Abbington Press. Clinebell. H. J., Jr. (1966). Basis_Txnes_gf_Eastoral genneeling. Nashville: Abbington. Clinebell, H. J., Jr. (1971). Experiments in training laity for ministrY- Eastgral_£sxshglogx. 22. 35-43- 160 161 Collins, G. R. (1987). Lay counseling: Some lingering questions for professionals. Iourna1_2f_2sxcholosx_and Christianitx. 6 (2). 7-9. Collins, G. R., & Tornquist, L. M. (1981). Training Christian people helpers: Observations on counselor education. J9urna1_2f_Esxsholosx_and_Theolosx. i. 69- so. Cowen, E. L., Gardner, E. A., & Zax, M. (Eds.). (1967). . New York: Appelton Century Crofts. Dayringer, R. & Pavia, E. (1986). CPE methods for the education of interdisciplinary professionals. The lourna1_9f_£astora1_§are. in. 30-39. Derrickson, P. E., (1990). Instruments used to measure change in students preparing for ministry: a summary of research on clinical pastoral education students. The Journa1_9f_£astoral_§are, 11. 343-356. Derrickson, P. E. & Ebersole, M. (1986). Lasting effects of CPE: A five year review. The_lournal.of_zastoral_care. ‘1, 5-150 Dickens, C. F. (1965). Review of the Gordon Personality Profile (p. 229-231). in Buros (Ed.) Eix£h_Mental Measurements_xearbook. Highland Park "-3.: Gryphon Press . Durlak, J. A. (1979). Comparative effectiveness of paraprofessional and professional helpers. Earsholosisa1_nu11etin, 36. 80-92. Durlak, J. A. (1981). Evaluating comparative studies of paraprofessional and professional helpers: A reply to Nietzel & Fisher. Earshologisa1_fiulletin. 12. 566-569. Emerson, J. G. (1986). Lay pastoral counseling: Thoughts and response. The_Journa1_2f_2astoral_§are. 4!. 291-292. Estadt, B. K. (1986). The lay ministry megatrend: An inclusive or exclusive stance. The_len;nel_ef_fie§;gzel_ care. 10, 292-296. Frauenfelder, K. & Frauenfelder, J. (1984). Brief Report: The effect of brief empathy training for student hotline volunteers. 911§Lg_1nteryentlen, 11, 96-103. Fry, P. S., & Charron, P. A. (1980). Effects of cognitive style and counselor-client compatibility on client growth. 12urna1_2f_§ounse11ns_asxshologx. 21, 529-538- 162 Gartner, J., Larson, D. B., & vachar-Mayberry, C. D. (1990). A systematic review of the quantity and quality of empirical research published in four pastoral counseling journals: 1975-1984. lenrnel_e£_fiee§e;el Care. 11. 115-129. Geary, T. F. (1977). Personal growth in CPE. Tne_1en;ne1_efi Eastera1_care. 26. 12-17. Glasscock, J. A. (1971). Pastoral action in the church, vintage 1971. Eastera1_2sxshologx. 22. 44-50. Gordon. L. v. (1963). Manua1I_Qordon_Eers9na1_£rof11e (Rev. ed.). New York: Harcourt, Brace S World. Gordon. L. v. (1978). G9rd2n_Eersonal.£rofile:1nxentorx Manna11_1913_3exision. Cleveland. OH: Psychological Corporation. Gorsuch, R., & Meylink, W. D. (1988). Toward a co- professional model of clergy-psychologist referral. Journal of Psychology and Christianity. 1(3), 22-31. Gough, H. G. (1960). The ACL as a personality assessment research technique. Eeyehelegieel_flepezte, 5, 107-122. Gough, R. G., & Reilbrun, A. B. (1983) The_Ad1ee§Lxe_§heek LLgt_Manna1. Palo Alto: Consulting Psychologists Press. Grant, G. (1975). An objective evaluation of an eleven-week supervised pastoral education program. Tne_lenrne1_ef Eastera1_§are. 22. 254-261. Greif, E. B., & Hogan, R. (1973). The theory and measurement of empathy. 1ourna1_of_§ounseling_2sxshologx. 20. 280- 284. Haley, J. (1969). The art of being a failure as a therapist. American_lourna1_of_Qrthonsxshiatrx. 32. 691-695- Rarris, J. (1985). Non-professionals as effective helpers for pastoral counselors. The_Journal_of_£astora1_care. 32, 165-172. Rattie, J. A., Sharply, C. F., 5 Rogers, R. J. (1984). Comparative effectiveness of professional and paraprofessional helpers. Barthological_nnlletin. 25. 534-541. Haugk, K. C. (1991). Introductory letter included in ”Friends Packet" available from Stephen Ministries, 8016 Dale, St. Louis, MO, 63117-1449. 163 Haugk, K. C., & McKay, W. J. (1985). Leefie1;g_§elge_;e W. available from Stephen Ministries, 8016 Dale, St. Louis, MO, 63117- 1449. Hayslip, B., Jr., & Walling, M. L. (1985). Impact of Hospice volunteer training on death anxiety and locus of control. Qmega, 1&, 243-254. Reilbrun, A. B. (1965). Review of the Gordon Personality Profile (p. 231-232). in Buros (Ed.) Sixth_Mental Meegg;emen§e_xeerpeet. Highland Park N.J.: Gryphon Press . Hersch, P. D., Kulik, J. A., & Scheibe, K. E. (1969). Personal characteristics of college volunteers in mental hospitals. MW WM: 3.3.: 30-34- Hobbs, N. (1964). Mental health's third revolution. W. 3.4. 822-833. Hogan, R. (1969). Development of an empathy scale. Jenrnel W. 3.3. 307-316. Hogan, R. (1975). Empathy: A conceptual and psychometric analysis. ThemnnselinLEmhologist. 5. 14-18. Hollingshead, A.B. (1957). W Eeeielen. (available from August B. Hollingshead, 1965 Yale Station, New Haven, Connecticut). Hollingshead, A. B. 5 Redlich, F. C. (1958). seeiel_glaee WW- Nev York: Wiley. Rolzberg, J. D., Gewirtz, R., & Ebner, E. (1964). Changes in moral judgment and self-acceptance in college students as a function of companionship with hospitalized mental gagients. W. 2.8.. 299- Holzberg, J. D., Knapp, R. R., & Turner, J. L. (1967). College students as companions to the mentally ill. In E.L. Cowen, E.A. Gardner, 5 M. Zax (Eds.) Emergent WW (pp. 91499). New York: Appleton-Century-Crofts. Jackson, M., & Thompson, C. L. (1971). Effective counselor: Characteristics and attitudes. Jenzn31_ef_§ennee11ng_ Embologx. 1.8.. 249-254. Johnson, P. E. (1973). THe church's mission to mental health. WM. 12. 30-40. 164 Joint Commission on Mental Illness and Health (1961). benign fen_MenLe1_fleel§n. New York: Basic Books, Inc. Karlsruher, A. (1974). The nonprofessional as a psychotherapeutic agent: A review of the empirical evidence pertaining to his effectiveness. Amezieen W. 2. 61-77. Kazdin. A. E. (1980). SW peyehelegy. New York: Harper S Row. Keller. J. V. (1961). W W ‘.I - one II'.‘ '00 ‘ 0';00 '0 0| Teee. Unpublished doctoral University of Chicago. 1’: 0. dissertation, Loyola Kepics, J. G. (1979). Tracking errors in psychotherapy. MW. 33. 365-377- Kiesler, D. J. (1983). The 1982 interpersonal circle: A taxonomy for complementarity in human transactions. W. 9.0. 185-214- Kim, L. E. (1960). A critical study of changes in protestant theological students with clinical pastoral education (Doctoral Dissertation, University of Southern California. 1960). Waste. 21. 1646. King, M., walder, L. O., S Pavey, S. (1970). Personality change as a function of volunteer experience in a psychiatric hospital. Wanna WI 3.5.: 423’4250 Knapp, R. R., Rolzberg, J. D. (1964). Characteristics of college students volunteering for service to mental patients. WM. 2.8.. 82-85- Kohut. 8- (1977). WWI. New York: International Universities Press. Kulik, J. A. S Scheibe, K. E. (1969). Personal characteristics of college volunteers in mental hospitals. 25121121932. 33. 30-34. Leary. 'r. (1957). Willa. New York: Ronald Press. Lesser, W. (1958). W Unpuhl ished doctoral dissertation, Michigan State University. 165 Linebaugh, D. E., S Devivo, P. (1981). The growing emphasis on training pastor-counselors in protestant seminaries. , 2, 266-268. Loesch, L. C., Crane, B. B., S Rucker, B. B. (1978). Counselor trainee effectiveness: More puzzle pieces? MW. 11. 195-204 . Lucero, R. J., S Currens, W. C. (1964). The effects of clinical training on the personality functioning of the minister. W. 20.. 147- Lukens, H. C., Jr. (1983). Training of paraprofessional christian counselors: A model proposed. lennnel_efi WW. 2. 61-66. Lukens, H. C. Jr. (1987). Lay counselor training revisited: Reflections of a trainer. Isurnal_of_£sxshologx_and W. 6(2). 10-13. Martin, J. C., S Carkhuff, R. R. (1968). Changes in personality and interpersonal functioning of counselors-in-training- MW. 24, 109-110. Mc Lennan, J. P. (1985). Helping beliefs inventory: Brief screening measure for training volunteer applicants in counseling. W. 5.6. 843-846- Meylink, W. D., S Gorsuch, R. (1987). Relationship between clergy and psychologists: The empirical data. lenrnal Winslow. 1(3). Nietzel, M. T., S Fisher, 8. G. (1981). Effectiveness of professional and paraprofessional helpers: A comment on Durlak. WWII. 8.5.. 555-565- Osborn, E. B. (1983). Training paraprofessional family therapists in a Christian setting. 19nrngl_ef WW. 2. 56-61. Patterson, C. H. (1984). Empathy, warmth, and genuineness in psychotherapy, A review of reviews. Egyehethezepy, 21, 431-438. Prater, J. S. (1987). Training Christian lay counselors in techniques of prevention and outreach. Jenrnal_ef Wain. 6(2). “-33. President's Commission on Mental Health (1978). Eepe;§_§e tne_Ere§1den§. Vols. I S II. Washington D.C.: U.S. Government Printing Office. 166 Probst, L. R. (1985). Paraprofessional therapy: in D. G. Brenner (ed.). Baker_encyclopedia_of_psxshologx. Grand Rapids: Baker. Reid. P- w. Jr. (1972). An_exaluat12n_of_the_impact_Qf I .. 0“ O .1 '0‘ - 0’ _00' Q 0' -.2'. O. " mTMTeeeLT Unpublished doctoral dissertation, University of North Carolina at Chapel Hill. Reiff, R., S Riessman, F. (1965). The indigenous non- professional: A strategy of change in community mental health programs. C2mmun1ty_Menta1_Rea1th_lourna1. Monograph No. 1. .‘W Richard, R. C. (1987). The professional counselor and local paraprofessional mental health organizations. {gunnel of_Esxchologxaand_§hristlanltx. 6(2). 35-38. Rioch, M. J. (1967). Pilot projects in training mental health counselors. in Cowen, E. L., Gardner, E. A., S ZaX. M. (Eds.). Emergent_Apnr9ashes.to_flenta1_flea1th Eggpleme. New York: Appelton-Century-Crofts. Rioch, M. J., Elkes, C., Flint, A. A., Blanche, S. U., Newman, R. G., S Silber, E. (1963). National Institute of Mental Health pilot study in training mental health counselors. American_3ourna1_of_Qrthopsxshiatrx. 33. 678-689. Rogers, C. (1977). Qn_pereen§1_pegel. New York: Dell Publishing. Rosenthal, D. (1955). Changes in some moral values following psychotherapy. J9urna1_of_consulting_zsxchologx. 13. 431-436. Rowe, W., Murphy, H. B., S De Csipkes, R. A. (1975). The relationship of counselor characteristics and counseling effectiveness. Eeyieg_efi_Edneatlenal Research. 45. 231-246. Rudy, J. P., McLemore, C. W., S Gorsuch, R. L. (1985). Interpersonal behavior and therapeutic progress: Therapists and clients rate each other. Eeyehiefiry, 1E, 264-281. Schlauch, C. R. (1990). Empathy as the Essence of Pastoral Psychotherapy. The_Journa1_2f_zast2ral_care. 44. 3-l7. Schmidt, W. S. (1986). Critical comments. The_1gn;n11_ef Bastora1_9are. 40. 15-20. 167 Seifert, R., Clinebell, R. J., Jr. (1969). Eereenel growth QMenge_Agen;e. Philadelphia: Westminster Press. Seigel, J. M. (1973). Mental health volunteers as change agentS. Amsrisan_1surna1_sf_£ommunitx_2syshslssr. 1 . 138-158. Sleight, C. C. (1985). Confidence and anxiety in student clinicians. Ths_91inisa1_sunsryissr. 3. 25-48. Smith, T. (1986). Equals in ministry: The Stephen Series trains and organizes lay Christians to perform caring ministry. The_gnnreh_flezelg, October 17, pp. 14-15. Stephen Ministries (not dated). The_Stephen_Seriesi_Christ caring_fsr_people_throush_pssnls. (Available from Stephen Ministries 8016 Dale, St. Louis, MO, 63117- 1449.). Stephen Ministries (1983). stsnhsn.3eries_Traines_Trainse Manuel, available from: Stephen Ministries, 8016 Dale, St. Louis, MO, 63117-1449. Stephen Ministries (1991). Congregations enrolled in the Stephen Series through 1991. carellne, 11(1), 15 (Available from Stephen Ministries 8016 Dale, St. Louis, MO, 63117-1449.). Tan, S. Y. (1986). Training Paraprofessional Christian Counselors. The_Journal_sf_Basrsral_£srs. 40. 296-304. Tan, 8. Y. (1987). Training lay Christian counselors: A basic program and some preliminary data. lenznn1_ef Esxsholsgy_and_christianitx. 3(2). 57-61. Tan, S. Y. (1990). Lay Christian counseling: The next decade. Iourna1_sf_Bsxsholosx_and_£hristianity. 3(3). 59-65. Tan, S. Y. (1991). Religious values and intervention in lay Christian counseling. Jsnrnsl_of_£syshslosx_and Christianity. 11. 173-182. Thomas, J. T., Stein, L. I., S Klein, M. H. (1982). A comparative evaluation of changes in basic clinical pastoral education students in different types of clinical settings as measured by the Adjective Check List and the Experience Scale. The_1eg;nel_efi_eeg§eze1 m, H, 181-1930 ‘ aI "IIIIIIIIIIIIIIIIIII“