This item is only available as the following downloads:
1 PERSONALITY CHARACTERISTICS PREDICTING SUCCESSFUL PARAPROFESSIONAL CRISIS INTERVENTION TRAINEES By DAVID J. HANNAH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE RE QUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 201 1 David J. Hannah
3 To those who actually read this thing I wish you courage
4 ACKNOWLEDGMENTS I am incredibly appreciative and thankful to my advisor and comm ittee chair, Dr. Ken Rice, for countless hours of guidance and encouragement throughout my development into a fully functioning professional. I am also grateful for the help and support of my committee members, Dr. Greg Neimeyer, Dr. Wayne Griffin, and Dr. John Chambers. I am forever indebted to the staff and volunteers of the Alachua Country Crisis Center who have pushed me to do things in this life that I never thought I was capable of doing. There are not enough pages of text available for me to thank the people who have touched my life throughout my seven years as a volunteer, and so instead I would like to leave it with I will be proud to say, I would also like to thank my parents and Jason who have g iven me boundless love, support, sacrifice, opportunities and encouragement to be anything and everything I only hope that one day I can have the courage and ability to return what they have given
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 ABSTRAC T ................................ ................................ ................................ ................................ ..... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 10 2 LITERATURE REVIEW ................................ ................................ ................................ ....... 15 The Nature of Crisis ................................ ................................ ................................ ................ 15 History of Crisis Centers ................................ ................................ ................................ ........ 17 Models of Paraprofessional Crisis Intervention ................................ ................................ ..... 19 Basic Rogerian Principles ................................ ................................ ................................ ....... 19 Research on Crisis Intervention Models ................................ ................................ ................. 21 Research on Trends in the Use of Volunteer Paraprofessionals ................................ ............. 23 Effectiveness of Training Programs ................................ ................................ ....................... 25 Summary of Training Literature ................................ ................................ ...................... 28 Outcome and Effectiveness of Services ................................ ................................ ................. 29 Personality Effects on Performance ................................ ................................ ........................ 31 Volunt eer Personality Characteristics ................................ ................................ ............. 32 Volunteer Demographics ................................ ................................ ................................ 33 Volunteer Screenings ................................ ................................ ................................ ....... 34 Empathy and Personality ................................ ................................ ................................ 35 Emotion Regulation ................................ ................................ ................................ ......... 37 Overview of Personality Constructs ................................ ................................ ....................... 38 Predicting Effective Volunteers ................................ ................................ .............................. 40 Importance of Training and Empathy ................................ ................................ .............. 40 Personality ................................ ................................ ................................ ....................... 41 Formal Personality Assessments ................................ ................................ ..................... 42 Inherent Characteristics ................................ ................................ ................................ ... 43 Current Study ................................ ................................ ................................ .......................... 45 Summary of Hypotheses ................................ ................................ ................................ ......... 49 3 METHOD ................................ ................................ ................................ ............................... 52 Particip ants ................................ ................................ ................................ ............................. 52 Center Recruitment ................................ ................................ ................................ .......... 52 Individual Participants ................................ ................................ ................................ ..... 53
6 Measur es ................................ ................................ ................................ ................................ 55 Personality ................................ ................................ ................................ ....................... 55 Empathy ................................ ................................ ................................ ........................... 55 Perfectionism ................................ ................................ ................................ ................... 56 Emotion Regulation ................................ ................................ ................................ ......... 57 Procedures ................................ ................................ ................................ ............................... 57 4 RESULTS ................................ ................................ ................................ ............................... 60 Descriptive Statistics and Preliminary Analysis ................................ ................................ ..... 60 Primary Analyses ................................ ................................ ................................ .................... 61 Hypothesis 1: Trait Ba sed Empathy ................................ ................................ ............... 62 Hypothesis 2: Personality Characteristics ................................ ................................ ...... 62 Hypothesis 3: Perfectionism ................................ ................................ ........................... 63 Hypothesis 4: Emotion Regulation ................................ ................................ ................. 64 Hypothesis 5: Failure Group ................................ ................................ .......................... 65 Exploratory Analyses ................................ ................................ ................................ .............. 66 5 DISCUSSION ................................ ................................ ................................ ......................... 71 Implications of Results ................................ ................................ ................................ ........... 78 Limitations and Future Dire ctions ................................ ................................ .......................... 81 Conclusions ................................ ................................ ................................ ............................. 84 APPENDIX A DEMOGRAPHIC QUESTIONNAIRE ................................ ................................ .................. 86 B MINI INTERNATIONAL PERSONALITY ITEM POOL ................................ ................... 87 C INTERPERSONAL REACTIVITY INDEX ................................ ................................ ......... 88 D ALMOST PERFECT SCALE REVISED ................................ ................................ ........... 90 E EMOTION REGULATION QUESTIONNAIRE ................................ ................................ .. 91 F ONLINE INFORMED CONSENT ................................ ................................ ........................ 92 REFER ENCE LIST ................................ ................................ ................................ ....................... 93 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 100
7 LIST OF TABLES Table page 4 1 Sample means and standard deviations by group ................................ .............................. 68 4 2 Scale correlations and reliability estimates ................................ ................................ ........ 69 4 3 Sample means and standard deviations for failure group ................................ .................. 70
8 LIST OF FIGURES Figure page 2 1 Moderator model ................................ ................................ ................................ ................ 51
9 Abstract of Dissertation Presented to th e Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PERSONALITY CHARACTERISTICS PREDICTING SUCCESSFUL PARAPROFESSIONAL CRISIS INTERVENTION TRAINEES By David J. H annah August 2011 Chair: Kenneth G. Rice Major: Counseling Psychology Telephone crisis intervention services are heavily used in communities that have them available, and research has shown that those counseling services are effective for clients, but t here is no empirical evidence as to why. Some authors have suggested that good training efforts produce good volunteers, but this process is not well understood. Many crisis centers have difficulties maintaining a volunteer pool large enough to offer aroun d the clock services, and mostly because they struggle to find trainees who are capable of passing the training program. Previous research has suggested that personality traits may affect the outcome of training, and there are some consistent findings rega rding personality traits, however only of those who have already completed training. The proposed study explores how individual differences on personality characteristics will affect the actual outcome of training. Results are expected to show that persona lity traits in the Five Factor Model, empathic tendencies, perfectionism, and emotion regulation can predict the outcome of training.
10 CHAPTER 1 INTRODUCTION Telephone c risis intervention centers have become a very necessary aspect of mental health servic es over the years. This trend is particularly true due to increasing demands for services compared to the number of available trained professionals. In addition, most professionals do not have very extensive training in crisis intervention and suicide prev ention, which makes crisis centers a particularly valuable resource. The trends in using telephone crisis intervention services has been increasing over the years, especially for underserved populations and afterhours client care (Feehan & Wade, 1998; Gott esfeld, Rhee, and Parker, 1970; Lamb & LaFave, 1977) Crisis centers have been a realized concept for over a century; however, they have not been such a widespread, researched, or respected service provide r until a little more than 50 years ago (McGee, 197 4). Today, every state has at least one call center to serve local populations, including some areas that have multiple centers providing services to very specific groups, cultures, and crises. In addition, there are nationally supported hotlines and accre ditation guidelines that govern the standard of care provided by telephone crisis centers. There are several different models of crisis intervention, but the most validated and widely used model is based on the Rogerian principles of empathy, warmth, and g enuineness as the core of helping skills (Hill et al., 2008; Mishara et al., 2007a,b). Since crisis centers are rooted in different states, regions, and cultures, there are some variations in crisis intervention skills. Research has shown that despite thes e differences, most crisis intervention services appear to be effective, and again have a foundation in empathy, warmth, and genuineness. Although these slight individual and cultural differences are known to exist, the literature has yet to come to a cons ensus on what those differences are (Hill & Lent, 2006)
11 The importance of facilitative skills is well documented, which has shed light on the incredible significance of training, yet there are no agreed upon standards of training. Despite a lack of standa rdization, the literature on training has concluded that empathy appears to be the main focus of almost al l forms of crisis intervention skills (Avery, 1978). Effects of training have especially compared to someone without any training experience. In fact, trained volunteers appear to do as well, if not better, than professional counselors. The style of training that produces such effective volunteers is a combination of didactic learn ing and experiential role plays to cultivate facilitative skills 1976 ; Hill & Lent, 2006; Ottens, 1984) Research on the effectiveness of volunteer crisis intervention services show that clients have positive therapeutic interac tions and good outcomes ( Getz, Fujita, & Allen, 1975 ; Mishara & Daigle, 1997 ; Stein & Lambert, 1984) It appears that the personality characteristics of the volunteers have an effect on their performance with callers, but the effects need further investiga tion ( Bobevski & McLennan, 1998; Eriksen & McAuliffe, 2003; Hill et al., 2008; McLennan Culkin, & Courtney 1994 ; Wehr & Wittmer, 1973) The only empirically validated characteristic appears to be empathic tendencies, which are highly associated with effe ctive volunteers versus less effective volunteers ( Paterson, Reniers, & Vllm 2009 ) In addition to experiencing empathic reactions, emotion regulation seems to be an important factor in effectively handling clients in crisis. Emotion regulation mechanism s allow a person to change how they experience emotion by avoiding emotional situations, changing the environment, cognitive restructuring of emotion, and emotional suppression (John & Gross, 2007) The mechanisms that are most applicable to crisis interve ntionists are cognitive
12 restructuring and suppression because the volunteer does not have the opportunity to avoid, select, or change the environment, only their experience of it. The literature on personality traits such as empathy and emotion regulation s tate that the se are stable characteristics that cannot change Instead, people can develop ways to better adapt their personality to the environment (Harkness & Lilienfeld, 1997 ; Seligman, 2007 ). For the current study, this concept means that a trainee ca n have an empathic personality, but through training develop mechanisms to use that tendency for helping skills. In comparison, a trainee with the same type of empathic personality may not have had an opportunity to develop skills to best utilize their cap acity The training process for crisis intervention skills requires a lot of resources such as time, money, and person power. Thus, having a person in a training program that has severe limits on developing empathy skills can take those resources away from another trainee. This drain on resources is particularly noticeable if the person does not complete the training program (Kalafat, 2002). To investigate the differences between volunteers and non volunteers, some researchers have used formal assessments s uch as the Myers Briggs Type Indicator and Minnesota Multiphasic Personality Inventory. These studies concluded that effective volunteers tended to be more emotionally stable compared to non volunteers and ineffective volunteers ( Evans, 1977 ; Heilig, 1967; Lester, 1971; McCarthy & Berman, 1971 ) More recent research has investigated the relationship between the modern Five Factor Model of personality traits and paraprofessional volunteers. This particular model of personality provides an appropriate framewo rk for describing inherent traits of effective volunteers, as well as how these characteristics may be cultivated during training. Results have shown a strong
13 relationship between characteristics of high agreeableness, extraversion, and openness to experie nce to effective interventionists ( John & Gross, 2007 ; Paterson, Reniers, & Vllm 2009 ). Other characteristics about volunteers have shown to be potentially important factors in skills training but reviews of the literature are inconsistent and inconclus ive. The factors that seem most significant are at a core level of personality such as empathic capacity and perfectionism (Hill & Lent, 2006; Hill et al., 2008; Mishara et al., 2007a,b) In addition, research has shown that crisis centers tend to produce effective volunteers, services, and client outcomes; however, volunteer numbers are low, which has been related to low ratios of those who complete training versus those who do not. The current study aim ed to examine th is training process and investigate t raits that differentiate successful from rejected trainees. The literature points to the Five Factor Model (FFM) of personality traits, empathic tendencies, perfectionism, and emotion regulation as the main factors that differentiate effective from ineffec tive volunteers. The current study posits that these traits also have a major affect on the training process. More specifically, that FFM factors and empathic tendencies will be directly related to the outcome of training, and that relationship may be mode rated by perfectionism and emotion regulation. These personality aspects were measu red of trainees before they bega n a training program to examine if they predict ed success or rejection from training. More specifically, the first hypothesis predicted that trait empathy will be a strong predictor of success in training, with those high in empathy being more likely to successfully complete training than those low in empathy The second hypothesis stated that high scores on the FFM subscales of positive perso nality characteristics would predict success, and those lower scores and/or higher scores on negative characteristics would not complete training. It was hypothesized that maladaptive perfectionism would moderate these relationships to predict
14 failure in t raining T he emotion regulation construct of cognitive change was hypothesized to moderate the conditional effects of personality and empathy to predict failure and that response modulation would modify the relationships to predict success. The final hypo thesis predicted that those who were screened out at the interview or earlier in the training process would score in a negative direction for these constructs versus those who successfully completed training. Chapter 2 provides an extensive literature revi ew of crisis intervention history, skills, training, training effectiveness, volunteer effectiveness, and volunteer characteristics. Chapter 2 concludes with hypotheses regarding the expected abilities of these personality aspects to predict the outcome of training. Chapter 3 covers participant recruitment, data collection method, and measures used. Chapter 4 reports on statistical results of the study, and Chapter 5 provides interpretation, discussion, limitations, implications, and conclusions of those re sults
15 CHAPTER 2 LITERATURE REVIEW The Nature of Crisis Traditionally in the field of mental health counseling there have always been a limited number of trained, available professionals compared to the actual demand for therapy services. In fact, it is not uncommon that those who are most in need of services never seek treatment out of fear of stigma or a lack of access to mental health care Research has shown numerous times that social stressors and mental health issues are increasing and yet the avai lability of resources for positive change cannot keep up with the negative social changes (Albee, 2000 ; Albee & Ryan Finn, 1993; Kenny & Hage, 1993 ). In addition to th e discrepancy between services needed and services provided, most therapists do not recei ve extensive training in handling clients in a suicidal crisis, and experiencing the suicide death of a client is one of the biggest fears among clinicians (Brown, 1987; Grad & Michel, 2005) As a result, counselors place more trust in paraprofessional ser vices to triage cases, provide after hours support, and handle emergency crisis and suicide intervention. Many communities have developed telephone crisis centers, often staffed by volunteers, to meet this need. The use of trained volunteers has been a val ued service of crisis centers for decades, and with very positive results. Over the years, volunteer crisis centers have become an integral part of community mental health care ( Feehan & Wade, 1998; Gottesfeld, Rhee, and Parker, 1970 ; Lamb & LaFave, 1977 ). The basis of crisis theory is that people are generally in a state of homeostasis, meaning they are able to maintain a functional ratio between cognitive and emotional stressors compared to coping skills and resources. This equilibrium becomes unbalanced coping abilities are ineffective or insufficient to manage stress, and this imbalance can lead to a
16 state of emotional crisis. Crises are typically referred to as an emotional state because other interventions are often ineffective until the person can begin to regulate his or her emotions. At this point of crisis, the danger is that a person may come to a maladaptive resolution which may include suicide. Due to the nature of crisis, the primary target of intervention is primarily em otional rather than focused on behavior, environment, or situation. The goal of crisis intervention is to resolve the emotional crisis and help restore homeostasis by providing emotional support and additional coping resources (Baldwin, 1979). The major hi storical trend in providing crisis intervention counseling has come in the form of telephone hotlines and crisis centers which offer free services to a community. Research has shown that these centers offer very unique services that appeal to a wide range of people. More specifically, Stein and Lambert (1984) found that telephone crisis centers provide access to counseling services for typically underserved populations. By using the telephone as a medium for connection to services, crisis centers provide sa fety in anonymity which is attractive to many groups that may not otherwise enter counseling due to the social stigma of receiving mental health services. Many crisis centers also offer services after typical business hours, which allow people to access se rvices more conveniently and when they may need help the most. More recently, telephone crisis intervention services have expanded from community specific centers to include national organizations. The two most prominent organizations are the National Hope line Network and the National Suicide Prevention Lifeline which have established the 1 800 SUIC I DE and 1 800 273 TALK toll free lines, respectively. Both of these organizations have a set of standards developed by the American Association of Suicidology us ed to evaluate and accredit established local crisis centers, allowing them to answer the national hotlines. When a person calls one of these numbers, the call is automatically routed to
17 the nearest certified crisis center. The setup of these organizations again demonstrates the integral role of local telephone crisis centers. History of Crisis Centers The first inception of crisis centers began with a focus on understanding suicide. The first program in the U.S. started in 1906 in New York City with the Na tional Save A Life League. This organization was not very influential in the field, but it did remain in operation for many years and influenced the creation of the Los Angeles Suicide Prevention Center (LASPC) in 1958. The LASPC and other early centers we re established for the purpose of researching the psychological process of suicide, suicide intervention, and suicide prevention. In addition, the LASPC began to systematically study the services being offered for consistency and effectiveness (McGee, 1974 ). With the findings that crisis intervention services were actually rather effective, the early crisis centers became focused on providing additional after hours services, sometimes with coverage for 24 hours a day. In order to have such extensive service s available, the phones were often being staffed by trained volunteers from the community due to a lack of professionals that could cover that many hours. As crisis centers became popular among professionals for after hours care, money became available thr ough National Institute of Mental Health (NIMH) grants. being offered (M cGee, 1974). With the success of the LASPC, federal grants became available to establish other centers across the country based on the same model. As the money was accessible and research on training programs continually validated the models, crisis center s began to flourish. By 1972 there were 185 centers across the country that were certified by the American Association of
18 Suicidology (McGee, 1974). Over the next decades, many centers developed other free services such as face to face counseling and emerg ency outreach services, making them epicenters for community mental health, and always using a volunteer model. Today there are over 140 centers certified by the National Suicide Prevention Lifeline for 1 800 273 TALK, and 137 centers certified by the Amer ican Association of Suicidology for 1 800 SUICIDE. Many centers are certified by both organizations, some by one, and some by none, but there are now multiple crisis centers in every state. Centers today continue to develop outreach services, target specif ic populations in need of services, improve on training programs, and train volunteers to use more sophisticated intervention skills (Feehan & Wade, 1998; Lamb & LaFave, 1977). Overall, the available literature on the use of trained paraprofessional volunt eers to provide crisis intervention services is scarce. When conducting a search using PsycINFO in early 2010 that encompassed the past 60 years, there were only 203 peer reviewed articles, books, and dissertations available with combinations of the follow ing search terms: paraprofessional crisis, paraprofessional suicide, volunteer crisis, volunteer suicide, crisis training, suicide training, crisis intervention, helper training, helping training, personality/personalities trainees, and personality/person alities crisis. It is important to note that this search also removed literature related to Crisis Intervention Team training with the police and law enforcement, hostage negotiators, professional clinical training programs, and professional supervision. S ome of these publications also include other types of crisis intervention training that is limited to specific populations such as rape victims and community disaster relief. In addition, there were less than 10 articles that were actually directly related to paraprofessional trainee personalities. The majority of the literature on counselor trainee personalities looks at professional trainees, which were used to inform the current study.
19 Publications related to volunteer paraprofessional crisis interventio n training begins around 1968 and continued rather steadily until about 1980 when research seems to stall out until a resurgence around 2005. Crisis intervention researchers may not be fully aware of the gaps in the literature and limited knowledge that is actually available from this research. Models of Paraprofessional Crisis Intervention Throughout the history of crisis intervention centers, numerous philosophical and theoretical models have been utilized. However, Baldwin (1980) discovered the trend tha t professional and paraprofessional crisis intervention styles, philosophies, and theoretical frameworks have actually converged. More specifically, he found that there is a common underlying theme of Rogerian and humanistic principles, but intervention be haviors depended on the resources available to the systems (e.g., level of sophistication of crisis intervention skills training). This is a trend that has continued today and appears to be a rather stable practice among paraprofessional crisis interventio n centers. The earliest published models of paraprofessional crisis intervention tended to center around basic facilitative helping skills that might be taught to beginning therapists in training rather than actual therapy skills. Just as in the previously mentioned study, Hill and Lent (2006) were able to trace the history of helping skills training centered theory (Rogers, 1942; 1951; 1957). This has become the standard philosophy of helping skills training for peer counsel ors and undergraduates today (Aiken, Brownell, & Iscoe, 1974; Cowen, Gardner, & Zax, 1967; Guerney, 1969; Hill et al., 2008; Leventhal, Berman, McCarthy, & Wasserman, 1976). Basic Rogerian Principles The main hypothesis of person centered therapy explains therapeutic change in a broad, simplified way. The theory posits that when a genuine, nonjudgmental, caring, and empathic
20 therapeutic relationship is created, it will be followed by a self directed growth process. This simplified view of therapeutic change encompasses all of the basic concepts and processes of development important to the theory (Rogers, 194 7 ). Beyond the therapeutic relationship, the person centered model is entirely nondirective in nature. In the therapy process, the client leads as the e xpert, and the therapist follows. Taking on a nondirective attitude involves allowing the client to lead the direction of therapy, and doing directed towards self actualization. It is important to development and currently relates more to being nonjudgmental, less advice oriented, and not imposing an agenda (Rogers, 1951). A word that is commonly used to describe the co ncept of congruence is genuineness. When a therapist is congruent, there will be a noticeable association between thoughts and behaviors. In this condition, there is no personal or professional mask for the therapist to hide behind, and they are forced to remain as transparent as possible to the client. The client would be in a position to observe the inner workings of the therapist. Through this the therapist can model an awareness and acceptance of self (Rogers, 1957). Crisis intervention skills training utilizes genuineness as a method of psychological connection to callers. Unconditional positive regard is a complete acceptance of all aspects of the client including cognitions, emotions, and behaviors. Rogers noted that conditions of worth are placed on our everyday lives by others and ourselves, which can lead to distress. The therapist provides unconditional positive regard to create a safety in the therapeutic relationship that allows the client to work freely without fear of rejection, judgment, etc. Unconditional positive regard coming from the therapist does not waver (Rogers, 1957). The crisis intervention literature
21 commonly refers unconditional positive regard as warmth, which during skills training generally is portrayed in the tone of voice. Emp athy is a method through which the conditions for therapy are communicated and the therapeutic relationship is built. By responding empathically, the therapist reflects his or her appreciation, understanding, attitude, and interest for the client and their world. This also emotional world. Empathy is not considered a tool to be used, but rather true empathy is a way of being, and allowing the client to lead while understand the client, communicating the understanding, and allowing for the client to correct the the rapist as needed (Rogers, 1957). Empathic reflection exactly as Rogers suggested is the most basic and common skill used in crisis intervention training. Research on Crisis Intervention Models Despite having a foundation in person centered theory, Mishara et al. (2007a) recognized that there has actually been very little research done to develop theoretical models specifically for telephone crisis intervention. The authors found that there are two prevailing models which originated at the Samaritan organiz ation and the Los Angeles Suicide Prevention Center. The Samaritans follow a very nondirective model focused on active listening. The Los Angeles Suicide Prevention Center uses a model that is much more involved in suicide assessment, intervention, prevent ion, and postvention from a problem solving philosophy. Postvention is a term used in the suicidology field to describe intervention services provided to those affected by a completed suicide. In their research of crisis intervention models, Mishara et al (2007a) conceptualized the efficacy of models to include technical aspects and the subjective quality of work measured
22 behaviors that should be part of any interaction with a suicidal client such as lethality assess ments and safety planning. The study consisted of surveying crisis center directors about which models they adhere to and subsequently monitoring 1,431 calls involving 782 different interventionis ts. They found a degree of discrepancy between what the directors said their volunteer should do on the phones and what actually happened; however there was still a very high degree of effectiveness for both models, which is later reviewed in the current p aper. The research on crisis intervention models seems to show that there are some slight variations in the actual clinical activities, but the philosophical core is deeply rooted in the basic facilitative conditions of empathy, warmth, and genuineness. S ome recent research has developed around creating and implementing training programs for paraprofessional services. Although there is not much continued research or consensus on the most effective methods for training volunteers, some authors have speculat ed that this is due to the concepts of training being already well established and validated (Hill & Lent, 2006). A meta analysis of training models suggested that a combination of didactic and experiential methods were the most effective at imparting hel ping skills. Although this was clearly the majority trend among crisis intervention training programs, there appeared to be no published consensus on exactly how these programs work. However, the meta analysis did show that there are also many indefinable characteristics about the trainers and trainees that are essential to a successful training program. Unfortunately Hill and Lent (2006) could not come to many solid conclusions about subtle nuances of the training process, and have thus called for more inv estigation into training programs, methods, effectiveness, and outcomes.
23 Research on Trends in the Use of Volunteer Paraprofessionals In the available literature, there are several interchangeable terminologies used to describe those who volunteer to provi in educational programs, but instead volunteered to be trained to interact with callers in crisis. For th e purpose of this document, all of these terms will be used interchangeably to refer those who have not yet graduated from a program. The crisis center system ha s traditionally relied on trained volunteers to provide services to the community. From early on, the majority of centers were in close proximity to colleges and h staffed systems have at least one professional who acts as a trainer and consultant to the workers (McCarthy & Berman, 1971). When examining one particular crisis center in New York, Gottesfeld, Rhee and Parker (1970) found that these volunteers had a wide range of jobs and skill levels, ranging from simple clerical work to conducting basic therapy sessions. Although they found the relationship between the volunteers and professionals was rather equa l, there was certainly a limit to the responsibilities that the volunteers were given. Over time it would appear that different systems began to engage in training volunteers using comparable models of providing services. Later research found similar resul ts when interviewing the professional and paraprofessional staff at university counseling centers (Lamb & LaFave, 1977). A related and more recent study from Feehan and Wade (1998) examined how professionals viewed paraprofessional involvement. The general consensus was that volunteer services allowed for more immediate needs to be met, as well as for much more diversity in the
24 staff and programming ideas. The authors found that the most effective paraprofessional programs had a strong match to the sponsori relationships, and a personal investment in the organization. Although a paraprofessional program can be a strong asset to an organization and professional psychologists, there must be a need from the com munity being served. Ideally, the volunteer program would be designed to meet those needs to receive the optimal support and provide appropriate referrals to clients. It is important to avoid areas where the paraprofessional services may overlap with profe ssional services and instead be complimentary. Researchers have pointed out that thorough training, continued education, evaluation, and program evolution are absolute necessities for strong programs (Maierle, 1973). McCord and Packwood (1973) surveyed 253 crisis centers about the services they offered their respective communities. They found that the majority of centers had phone counselors available 24 hour per day that were volunteers. The centers they surveyed used extensive screening and training progr ams to obtain the most qualified volunteers, with most centers requiring an average of 50 hours of training before being allowed to answer the phones. With many similar programs being established, it appears that the most common characteristics centers loo k for in their volunteers are natural inclinations towards empathy, warmth, and genuineness (Sawatzky & Paterson, 1982). More current researchers have been aware of the lack of continued investigation into the use of paraprofessional programs. Hill and Len t (2006) speculated this gap in the literature was related to the concepts of using trained volunteers being well legitimized in the psychological community; however they also noted that there is little meta analytic consensus about the best
25 methods of tra ining. They found support for the effectiveness of training volunteers, and that most programs focused on training similar methods of crisis intervention skills. Effectiveness of Training Programs The earliest literature on volunteer training programs foc used on the trainer and trainee as the predominant variables that affect training. More specifically, Carkhuff (1969a) speculated that the trainer must have a strong ability to use the helping skills, and that trainees bring their own natural abilities, in stincts, and experiences into the process. Carkhuff saw the trainer as the most critical variable in the equation because the trainer must not only be able to use helping skills, they must also be good at teaching them. This research found that trainees wi ll model after the trainers, so the trainers need to be a high level of sophistication with these skills. The results also showed that a larger skill discrepancy between trainees and trainers led to stronger skills at the conclusion of training. In other w ords, the more highly skilled your trainers are, the more highly skilled volunteers you can train. In addition to having skilled trainers, the training program itself should be rather rigorous and include a system for program evaluation and policy change (Maierle, 1973). Despite knowing the makeup of a good training program, Avery (1978) pointed out that there is no standardized level of competency for this type of training. In response, the author attempted to compare different programs and found this was not feasible because they were too different Instead, the research found that programs were simply effective or ineffective. In order to assess what determines this dichotomy, the author evaluated a training program that appear the training of empathic skills was the main variable that seemed to determine an effective versus ineffective program.
26 Considering that crisis centers offer such a unique and im portant service to communities in need, skills training is an absolute requirement since activities range from basic referrals to actual crisis counseling. Researchers have also looked at the methods of training that appear to best prepare volunteers for t his wide range of services. The outcomes show that hands on experiential training is the best preparation, especially when using role plays in training. It appears that having an experiential training model also has positive effects on the crisis center sy stem as a whole by having stronger volunteers (Margolis, Edwards, Shrier, & Cramer, 1975; Teevan & Gabel, 1978). of trained helpers to discriminate between the differen t facilitative skills (empathy, warmth, and actually responded to their callers. The researcher found that compared to untrained people, the trained volunteers had a better command of facilitative skills in addition to a better ability to discriminate between them. These results suggest that effective training not only imparts the skills, but provides a better understanding of them. Interestingly enough, the auth or also found that some volunteers could verbally discriminate between skills without actually being able to produce them, which was a sign of an ineffective helper. Carkhuff (1969b) also made a within group comparison between volunteers trained in differ ent environments. The results showed that those who were very systematically trained performed much better than those who came from a less structured training program. Thus, having an organized and structured training program is definitely superior. The st udy also compared the unsystematically trained to non trained helpers and found that at least some training, even if not entirely effective, was better than no training at all.
27 One study in particular took a more in depth look at the outcomes of successfu l training. From the previously mentioned research it seems that there are differences between effective and ineffective training, so the goal of this study was to investigate how training affected the actual verbal responses of volunteers The results sho wed that after training there was an expected increase in positive helping responses, but also a major decrease in negative helping responses. Thus it would seem that training has this dual effect that does more than just cultivate the desired behaviors, i t also decreases unwanted behaviors that may hinder the therapeutic process (Danish, volunteers handled helping situations. To do this, the authors measured trainees on helping skills before the training program began. They found that the untrained volunteers were much more directive, leading, and used closed questions that did not encourage further communication as their main modes of response compar ed to trained volunteers Emotional or empathic responses were rarely and unintentionally given by the untrained volunteers. These types of responses were mostly focused on gathering information rather than providing conditions for building rapport, and we re the main unwanted behaviors that decrease with training. up study to replicate these findings. The authors found the same pattern of increasing empathic responding and decreasing negatively impactful responding du e to training. In addition to this training effect, they also investigated the longer term impact of training and found that there is a slight degradation in the empathic accuracy of these skills once training ends. The authors suggested that a very system atic training program with carefully selected volunteers may help the longevity of training effects. It was later confirmed that the effects of training are not only long lasting, but that veteran volunteers
28 actually tend to improve on their skills over ti me compared to those who more recently completed training (Barz, 2001; Taylor, Brady, & Swank, 1991). Summary of Training Literature The consensus in the literature seems to be that an extensive training program is imperative to having volunteers who are effective with helping skills; however there is also a need to examine the process for selecting trainees to go through the program. This is extremely important because there are some centers without the resources for appropriate training as described here and must rely solely on selection. Methods of selection are generally based on personality, interpersonal interactions, and clinical intuition, but there are no published criteria, definitions, or research to validate this method. Although there is still no consensus on methods of selection, research has shown that training is much more effective, but a combination of upfront selection/screening and training is the most effective (Hart & King, 1979). Most research has viewed these constructs through train ing programs that were already well established. A few other studies have used these results to subsequently create training programs and measure effectiveness. The results were replicated that a combination of didactic and experiential methods were the be st for increasing helping skills, clinical judgment, and crisis intervention skills. These results were also upheld when comparing trained and untrained volunteers (Schinke et al., 1979; Selfridge et al., 1975). In addition to understanding that thorough training has the desired effects on helping skills, it is also important to know how impactful these changes are. When training volunteers to be paraprofessional crisis interventionists, the standard of comparison would be professionals. An evaluation of t he mechanics of training programs showed that they do have the capability to train paraprofessionals to function at therapeutic levels similar to professional therapists, despite the programs not being designed to train therapists (Carkhuff & Truax, 1965). Furthermore, one
29 study gave a questionnaire to paraprofessional volunteers and professionals in the field to assess facilitative skills. The results showed very high degree of similarity between the responses of trained volunteers and professionals (Otten s, 1984). As previously mentioned, Hill and Lent (2006) have more recently called for more research and replication on the specific mechanics of training programs, methods, effectiveness, and outcomes. Current research from Mishara et al. (2007a, b) has r esponded by examining well established and standardized national training programs. The results are again consistent and validate the similar studies from nearly 40 years prior. The difficulty in conducting this research is the lack of consistency among tr aining programs. There are some crisis centers that employ very stringent and intensive training methods, while others lack the resources for extensive training and rely mainly on selection methods. Thankfully nearly all methods have been shown to be effec tive; however, there is no available research known to this author that has examined the differences in effectiveness between more and less stringent training programs. Outcome and Effectiveness of Services At this point, there is considerable evidence t o suggest the components of a good training program, and that said program can produce trained paraprofessionals that are highly capable utilizing helping skills of empathy, warmth, and genuineness. That being the case, the use of lay people from the commu nity to provide such an important service has raised questions about the possible effectiveness of these centers and services. There is extensive literature that addresses the effectiveness of therapy, and a subset of research devoted to volunteer based cr isis intervention services that utilize basic helping skills. Early research on the effectiveness of telephone crisis centers was mostly inconclusive, saying that they may provide some valuable points of contact, but the therapeutic impact was questionabl e (Davies, 1982; Genthner, 1974; Neimeyer & Pfeiffer, 1994; Rosenbaum &
30 Calhoun, 1977). Over the course of years of research, the literature has come to the conclusion that these crisis centers are indeed effective services. For example, Stein and Lambert (1984) used counselor self reports, in session behaviors, and client evaluations to investigate the effectiveness of paraprofessional workers. When rated by other professionals in the field, the rs attributed this to effective training in Rogerian facilitative skills, but overall had difficulties operationalizing the criteria for same positive effect In a similar study, Getz, Fujita, and Allen (1975) were interested in program evaluation for crisis intervention centers and models. More specifically, they wanted to investigate the impact that paraprofessional crisis interventionists had on clients. T he results show ed that clients who engaged with the crisis counselors felt they had positive and helpful interactions, thus findings the authors also found that the counselors were highly empathically accurate with their skills to create rapport and assess situations for danger. Although the outcomes of crisis ce nters appear to be positive, the s e data are based solely on volunteers who are trained to use facilita tive helping skills. It is important to note that these training methods are not directed towards actual therapy for long term client interactions and change. However, trained volunteers can create short term therapeutic change in the moments of crisis. In fact, many paraprofessionals have been found to be just as good, if not better, than professionals at crisis intervention. It appears that trained volunteers have such strong clinical outcomes compared to professionals because the training is focused enti rely on those helping skills (Carkhuff, 1968).
31 The available research seems to support the idea that Rogerian facilitative skills as an approach to crisis intervention results in better outcomes. Few studies have looked at client changes during calls; how ever Knickerbocker (1973) and Echterling and Hartsough (1983) both found significant positive changes in multiple emotional, cognitive, and behavioral dimensions as calls progressed. In addition, Mishara and Daigle (1997) showed that crisis intervention le d to more decreases in depression and higher rates of safety contracts with highly lethal clients. More recently, Mishara et al. (2007b) spoke with crisis center directors about the services they offered, particularly what they considered to be good and b ad qualities of volunteers. The authors then examined those variables in real life calls to see how they affected the outcome. Using outside observers to rate levels of empathy, they found a significant relationship between higher levels of empathy and pos itive interactions with callers, such as the caller remaining on the line and obtaining safety contracts. The main finding was that positive changes in callers was related to the expression of empathy and respect, which may be more innate qualities of volu nteers that are cultivated in during the training process. Personality Effects on Performance Knowing that crisis intervention services and personnel are effective, we turn our attention to what qualities these people possess that make them effective. Thi s topic has captured a lot of attention, especially when using lay people from the community instead of trained and/or licensed professionals. According to the literature, the main aspects of effective crisis counselors are an awareness of personal strengt hs, weaknesses, attitudes, judgments, biases, etc. T hese characteristics are what the training process is designed to develop and foster. Ideally during the training process, skilled trainers will be able to see trainees develop these areas. The difficulty lies in trainees possessing these qualities while being able to collaboratively engage an individual in crisis without taking responsibility for their problems (Kalafat, 2002).
32 Kalafat (2002) made the point that these characteristics also match with the h elping skills philosophy previously outlined which is inherent in most crisis center models. In addition to looking for these positive qualities in trainees, the author also noted the importance of screening for potential volunteers that have a significant ongoing crisis that may impair their skills or ability to handle feedback. Feedback is an incredibly important part of the training process, and if the trainee becomes defensive they are less likely to hear and incorporate the feedback. Thus personality c haracteristics of trainees are important factors, particularly perfectionism which is Volunteer Personality Characteristics Several studies have addressed questions regarding the personalities of people who do paraprofessional counseling, especially how personality may affect their work. Wehr and Wittmer (1973) noted that there are clear connections between the personalities of professionally trained counselors and th eir effectiveness, but no research had been done related to paraprofessionals. The focus of their research was to compare the personality traits of paraprofessionals in training to students training in professional counseling programs using the Sixteen Per sonality Factor Questionnaire (Cattell & Eber, 196 8 ). Results showed that the paraprofessionals scored lower in areas related to structured education about helping skills, maturity related to age and experience, and self confidence in delivering skills. Ho wever, they found paraprofessionals to be higher than professionals on sensitivity to others. Despite these personality differences, the authors did not find a difference in actual effectiveness between the two groups and concluded that paraprofessionals a re competent with therapeutic interactions (Wehr & Wittmer, 1973). A similar study also found that there are no personality characteristics that seem to impact the effectiveness of volunteer counselors. Although there are no apparent differences in
33 perfor mance, this research only examined those who had completed training, and there is no mention of personality affecting the training itself. Currently, there is not enough published research to draw conclusions about personality and training (Bobevski & McLe nnan, 1998; Eriksen & McAuliffe, 2003; Hill et al., 2008; McLennan, Culkin, & Courtney, 1994 ). One possible explanation is that effectiveness and performance is being measured after unsuitable volunteers with maladaptive personality characteristics have be en screened out in the training process. An earlier study seems to support this idea by researching how volunteers are initially screened prior to training. The researchers looked at paraprofessional volunteers involved in a wide range of jobs and skill l evels ranging from simple clerical work to conducting therapy. They found that volunteers were often hired for personal and interpersonal qualities (Gottesfeld, Rhee, & Parker, 1970). McClure et al. (1973) looked specifically at unscreened volunteer counse lors for personality issues. Compared to the general population, people who volunteer for these services tended to report a higher than average incidence of experiencing emotional problems. Thus, the author suggested that screening potential volunteers for emotional problems may be appropriate. Volunteer Demographics Tapp and Spanier (1973) noted that most volunteers tend to come from college populations, so they made a similar comparison between volunteer counselors and average college students. More speci fically, the authors looked at constructs of self concept, self actualization, and openness to self disclosure. They found little differences between the groups with self concepts; however, volunteers scored higher on self actualization and disclosure, bot h of which map on to the concept of genuineness that is important in intervention work
34 Research has been done to investigate the type of people who are drawn to volunteer crisis intervention programs. Sawatzky and Paterson (1982) found that in some progr ams, many people have natural abilities to relate to people and natural tendencies towards positive helping skills that cannot be formally taught in school. The authors noted that these were often personal characteristics such as empathy, warmth, and genui neness. Similarly, Hirsch and Stone (1982) believed that the training process would be greatly affected by these aspects and attitudes that trainees bring to the program. More specifically, the authors speculated that performance during training would be r The results showed a significant relationship, but the authors were skeptical of the accuracy of these results because there were no reliable measures for these constructs at t he time. Volunteer Screenings Due to a lack of reliable measures, many centers rely on other screening methods. In a survey of 253 crisis centers, McCord and Packwood (1973) saw where the majority of centers screened volunteers using a combination of one o n one interviews and the training program itself. Out of the surveyed centers, most required an average of 50 hours of training before answering the phones, providing numerous opportunities for screening unsuitable trainees. The authors also found that som e centers did use formal assessment, which most often included the Minnesota Multiphasic Personality Inventory (MMPI; Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989). The notion that continually comes from the research on characteristics of effecti ve volunteers is that screenings are necessary throughout the training process. Additionally, it appears that some people have a natural capacity for crisis intervention work, which is less researched in the literature. Mishara et al. (2007b) suggested tha t crisis center programs should take more control over the quality of the interventions they provide, which can begin with basic
35 screenings for potential trainees. In particular they recommended that centers seek out helpers who exhibit qualities of empath y and respect towards others. are people who have innate relational skills that happen to match well with the facilitative skills of empathy, warmth, and genuineness. Th e authors speculated that natural helpers are more likely to enter helping professions than others, but could not find any consensus in the literature about how to measure such a concept. After developing a measure, they did find that natural helpers have a higher propensity to enter the helping profession and become involved in paraprofessional counseling services compared to the general population. These results provide some evidence that there is a self selection process in signing up for training progra ms. Empathy and Personality It appears the consensus in the literature is that empathy is one of the major characteri stics of effective volunteers. Davis (1983) conceptualized empathy as a complex construct that is multidimensional, including both cognitiv e and emotional components in its experience. The emotional experience of empathy is an affective impact on the person, and the cognitive component involves the validation of the emotional experience with the other person. The author investigated natural e mpathic tendencies and found significant individual differences in experiences of empathy. In addition, he found that higher levels of empathy were related to a higher probability of helping others, which suggests that natural empathic tendencies can affec t helping behaviors. Natural empathic tendencies appear to be very stable across time and may be more of a trait like quality than an induced state (Pickett, 2006). Just as Tapp and Spanier (1973) had noted more than 30 years ago, Paterson, Reniers, and V llm (2009) found that college students are still the typical volunteer population for crisis centers. The literature about college students shows that they are often overwhelmed and under a
36 lot of emotional distress due to things like major life transitio ns, academic pressures, academic stress, and personal losses (Levine & Cureton, 1998). These findings have made researchers curious about the appropriateness of college students as crisis interventionists. Currently there is not much available research on the personality characteristics of volunteers. To address this gap in the literature, Paterson, Reniers, and Vllm (2009) compared active paraprofessional volunteers to non volunteers on empathy, personality traits, and mental health problems. The authors used the Interpersonal Reactivity Index (IRI; Davis, 1980) and a measure of the Five Factor Model personality traits known as Openness C onscientiousness, E xtraversion, A greeableness, and N euroticism (Goldberg, 1993). The results showed that volunteers wer e significantly higher on agreeableness and empathy, but there were no differences in mental health issues between the two populations. It is important to note that the group had not completed the entire training p rogram, so the authors speculated from their results that empathy is actually n ot a learned skill but rather a personality trait (Paterson, Reniers, & Vllm 2009). Although Paterson, Reniers, and V llm (2009) did not obtain findings regarding any of the other Five Fact or Model components consistent with previous research they did expect at least some differences in fa ctors such as Openness and Extra version One possibility for this restriction of significant findings is that the Five Factor Model components are not as sensitive as more specific aspects of the larger dimensions. The authors alluded to this possibility by suggesting that empathy is tied into the larger Five Factor Model dimensions and may be an important personality trait to in successful volunteer. Their study highlights the inconsistencies and general lack of research support of these factors in the literature.
37 Emotion Regulation Reviews of the literature on personality traits as they relate to emotion have sh own possible individual differences in experience and regulation. In fact, humans actually have some ability to control their emotions using different mechanisms. The theory is that people may avoid emotional stimuli, change the environment, or actively se lect which stimuli to focus on and block out others. The mechanisms by which a person may then regulate their emotions is through situation selection (avoidance), situation modification (self assertion), attention deployment (distraction), cognitive change (reappraisal), and response modulation (suppression). Researchers have suggested that these components of emotion regulation are part of a developmental adaptation in a social world rather than personality traits themselves (John & Gross, 2007). Accordin g to research on emotion regulation from Gross and John (2003), there is a particular process of emotional cues and engagement that trigger emotion regulation mechanisms. First there is an emotional stimulus or situation, which is then followed by attentio n, appraisal, and response to that context. The authors outlined antecedent focused mechanisms which are more of pre emptive environmental manipulation rather than emotional response (e.g., situation selection, situation modification, and attention deploym ent). Thus, response focused mechanisms take place after an emotion has been engaged or experienced in some way (e.g., cognitive change and response modulation). The distinction between antecedent focused and response focused emotion regulation mechanisms is particularly important because in crisis intervention work, the most effectively trained volunteer is skilled in using emotional response, previously referred to as empathy and genuineness. Gross and John (2003) more specifically noted that those who en gage response modulation tend to be more in tune with the negative emotions of crisis, whereas those who utilize cognitive change were able to alter their cognitive experience of negative emotion and
38 thus not be as empathic with those emotions. Additionall y, crisis intervention theory and skills are nondirective in nature, meaning the volunteer does not manipulate the environment or emotional process of the call. In their research, John and Gross (2007) examined potential correlations between emotion regul ation mechanisms and the Five Factor Model personality traits. Multiple studies have commented on these personality traits in crisis center volunteers and found high agreeableness, extraversion, and openness to experience because of the characteristics inh erent in conducting counseling interventions (Paterson, Reniers, & Vllm 2009; Tishler, Bartholomae, & Rhodes, 2005). That being said, crisis center volunteers would likely employ high levels of situation modification and lower levels of situation selecti on, meaning they are more likely to jump into a situation with the goal of being emotionally open. These strategies are more focused on the downregulation of negative emotions rather than the upregulation of positive emotions, which is consistent with cris is intervention work that shows injecting hope to be ineffective ( Granello & Granello, 2007; John & Gross, 2007). Overview of Personality Constructs The previously mentioned research has focused more on environmental adaptations than personality traits; h owever, Harkness and Lilienfeld (1997) provide an overview and review of the literature on individual differences in personality. The authors noted that aspects of personality are stable traits rather than passing and unstable states. These major personali ty traits are often given different labels depending on theoretical orientation, but the most common names that come out of the literature are related to the Five Factor Model dimensions of personality. Personality traits are not very tangible, but are of ten inferred by their resulting behaviors. Researchers generally talk about these traits as being predispositions to certain thoughts, behaviors, and emotions which can sometimes be predicted. In the context of crisis intervention
39 training, this would mean that certain personality traits may be able to predict successful trainees. However, there is a large degree of individual differences within those traits because genetics and the environment both play a large role in the stability and malleability of per sonality traits (Harkness & Lilienfeld, 1997). Related to adaptation and malleability, the literature on individual differences discusses the phenomenon of adapting personality to an environment, which often has costs that are emotional, social, etc. Hark ness and Lilienfeld (1997) expand this theory to say that a person not only learns to adapt their personality to their environment, but also learns to adapt to their own inner works and the associated costs. Basically, this means that the same personality traits may manifest differently due to these individual differences. The example the authors give is a criminal versus a firefighter, both of whom have high levels of sensation seeking personality traits, but have adapted to their own basic tendencies diff erently. Harkness and Lilienfeld (1997) concluded that the goal of change should not be inherent traits such as neuroticism, but rather the characteristic adaptations which are the thoughts, behaviors, and emotions that adapt the trait to the environment. More importantly, the authors suggest that changes cannot be made with inherent traits except within a restricted range, meaning one can vary the extremity of a trait or its characteristic adaptations. Th us in this case of the current study, crisis interv ention training should not be focused on changing a personality trait such as empathy, but instead should cultivate the expression of empathy. In similar research and review, Seligman (2007) likened personality to things like intelligence, talents, and br ain chemistry which are linked to biology and are thus unchangeable. However, he cites examples of people overcoming their biology within a degree of variance. The research seems to show that people are able to change state like aspects of themselves, but not
40 deeper aspects of self and identity. Again, in the context of the current study, this would mean that some trainees will either possess the personality characteristics of effective interventionists to be cultivated in training, or not at all, suggestin g an element of predictability. Predicting Effective Volunteers Importance of Training and Empathy As noted in the previous research studies, measuring effectiveness of systems and individuals can be more tangible and behaviorally measured. The current cha llenge is to look at more trait like constructs that have to be operationalized in self reports for the purpose of actually obtaining effective individuals. Kalafat (2002) noted that resources such as time, money, and personnel are low these days, so upfro nt screenings to conserve these resources is incredibly necessary. The author suggested that screenings should happen at an initial interview to look at interpersonal functioning, any serious trauma history that might adversely affect their work, and their ability to handle feedback. Kalafat pointed out that screening on these levels should be a continuous process throughout training in order to keep only the most competent trainees. ecificity of skills required to be a crisis interventionist, and to acquire those skills in a brief period of training, having a very systematic training program that carefully selects and screens trainees is necessary. This is especially important because of the knowledge that skills degrade somewhat after training, so trainees need to possess a high capacity for these skills. The authors reiterate the knowledge that most programs are well equipped to effectively train a high level of skill, but there is l ittle research about individual characteristics of the trainees that might predict their success. Similar research from Ickes (1993) focused on a model of empathy related to empathic accuracy. This model distinguishes between empathic understanding, expres sion, and
41 communication, and suggests that accuracy must happen at all three points to be effective. In conjunction with the personality lite rature, empathy would thus be a personality trait but with many individual differences in accuracy. The subsequent purpose and outcome of successful training would then be to cultivate mechanisms that use empathic capacity and improve on the Personality In the earlier history of crisis centers, completion of a training program was not always a requirement. At these centers, volunteers would be selected from basic interviews based on their personality. This method was worrisome to researchers because there was no published definition, method, criteria, or re search to support this process. Research has shown that selection can be an effective method compared to no standard at all, but not more effective than a training program. In fact, results have also shown that those who go through training and have a bett er pre training baseline of facilitative skills become more effective interventionists. In contrast, those with lower pre training baselines were never very effective. Thus it seems that there are some pre existing factors that are incredibly important to the training process (Hart & King, 1979). As previously mentioned, the trainer is one of the more important variables in the training process; however, Carkhuff (1969a) suggested that the trainee variable is important because it includes a degree of pre tr aining self selection. In other words, people who tend to have a higher natural ability for helping skills are more likely to become involved in crisis center training. The author believed that attrition rates in training programs are highly related to tra inees having personality styles that are not congruent with nondirective facilitative skills. Carkhuff suggested that the best predictor of future skills may be pre training abilities. Barz (2001) attempted to follow up with this idea by looking at the exi sting significant relationships in the lives of
42 volunteers and found that they did not engage in higher levels of empathy, warmth, or genuineness more so than non volunteers. Formal Personality Assessments Some researchers have investigated the use of form al instruments as screening and selection methods. Lester (1971) figured the Myers Briggs Type Indicator (MBTI; Myers, 1962) might be a more appropriate instrument. The author measured pre training volunteers using the MBTI and compared their scores to a w ithin group measure of counselor effectiveness. Results showed that the most effective volunteers produced profiles consistent with stronger empathy and clinical judgment skills. Heilig (1967) found that the MMPI could be used as a screening measure, but w as limited to detecting significant mental health problems that were not noticeable during a one on one interview. Since then, the MMPI had become the most used screening measure, and later research attempted to discriminate between effective and ineffecti ve volunteers. After completing the MMPI and the training program, volunteers answering the phones were unknowingly and randomly given a scripted role prediction of effective versus ineffective volunteers was rough ly 90% accurate, however these results do not clearly show if this was related to psychopathology or other personality aspects such as defensiveness (Evans, 1977). It appears that the MMPI is mostly useful to look for emotional stability to handle the subs equent training process (McCarthy & Berman, 1971). This conclusion suggests that in addition to deeper personality issues that affect the training process, there are variables which moderate the impact of core personality traits on outcome. The research by Harkness and Lilienfeld (1997) and Seligman (2007) support this idea by showing that personality cannot change, but rather the degrees of extremity of expression and environmental adaptations will
43 ng program is likely related to their personality traits, but moderated by the way in which they have adapted their personality to their environment. In a study of trainee differences, Schoenfeld, Preston, and Adams (1976) used the MMPI and a measure of in telligence to examine differences between current volunteers and those who were screened out of the systems during the training process. The researchers were able to compare volunteers who completed training but were inactive, those who were rejected from training, the most effective currently active volunteers, and the least effective currently active volunteers. Significant differences were found between each of the groups on intelligence and various MMPI scales. Results favored the most effective active volunteers and those who completed training but were currently inactive, showing higher intelligence scores and less potential psychopathology on the MMPI. Although their results show that trainees who are rejected from training are likely different in int elligence and personality, there is not enough specific information on how these traits are affecting their performance in training. In herent Characteristics As Mishara et al. (2007a) suggested, there are likely personality characteristics of volunteers th at are more important than trained skills. This means that the most successful trainees already have some capacity for helping skills, rather than trying to learn a completely new skill that does not match with their philosophy for helping others. Mishara et al. (2007b) showed that positive changes in calls were related to the expression of empathy and respect, which highlights the importance of trainees possessing these skills. The authors suggested that the actual ability to experience empathy and express it to the client are trait like qualities and cannot be learned through the training process from scratch, which was even echoed by Rogers
44 (1957). They also recommended that crisis centers look for volunteers that possess these natural qualities, which ha s been less researched in the literature. In addition to natural empathic tendencies Hill and Lent (2006) were interested in trainee variables that may mediate the outcome of training. From their literature review, there was not much consensus on the most appropriate variables, but did at least find student status of undergraduate or graduate did not affect training. Overall, the authors were not able to draw any significant conclusions from the meta analysis, further demonstrating a lack of consensus in t he literature about which variables are most important to the training process. Additionally, this may also be evidence that the mediational model does not best suit these variables, and a moderator model may better describe these relationships. These auth ors also suggest that researchers begin to investigate natural helping abilities. A follow up study followed the same notion that there appears to be a lot of difficulty in identifying predictors of effective trainees. Hill et al. (2008) speculated that hi gher levels of initial empathy, higher grade point averages, and personality characteristics would affect the outcomes of training. In a previous study, Hill, Sullivan, Knox, and Schlosser (2007) identified high levels of perfectionism as negatively impact ing training in a qualitative investigation of authors conceptualized perfectionism as high or low, rather than maladaptive versus adaptive as the measure is intended to be used. None of these variables was found to be significant predictors of the effectiveness of volunteers, which may be due to the conceptua l use of the variables within a mediational model.
45 Other variables that have been explored as potentially related to training outcomes are self efficacy and motivation. Hill et al. (2008) attempted to measure self efficacy in trainees, but found that they could not give an accurate report of self efficacy before training began. This finding makes sense considering that trainees are likely to experience a low level of self efficacy with a skill they have not yet been trained in using. Similarly, altruistic m otivation is a moot point for a volunteer training program because there is an inherent self selection process that implies motivation (Barz, 2001) While there may be other variables that impact the training process such as frustration tolerance, anxiety, and stress to name a few, the literature appears to be calling for a better initial understanding of more basic personality constructs. Current Study Most of the literature thus far has shown that crisis centers are effective, have good client outcomes, a nd tend to employ effective volunteers. Additionally, it seems these volunteers have common characteristics and skills in empathy, warmth, and genuineness, plus have personalities that are suited for providing psychological services that match well with pr ofessionals in the field. T his research has been done with the idea of hopefully improving services; however, there is very little research that looks at the front end personality characteristics of volunteers that may foster the acquisition of effective c risis intervention skills. It appears that there are many methods for screening and training potential crisis center volunteers, but clearly not everyone who enters a training program is a good fit for the system. Those who are not able to attain helping s kills within the short training period are generally removed from training. For many centers, this can be difficult because of the loss of potential volunteers, thus the goal of the current study wa s to hopefully prevent high levels of attrition during tra ining and potentially increase the accuracy of application/interview screening (McCord & Packwood, 1973).
46 The available research very clearly shows that an intrinsic degree of empathy is one of the more important personality characteristics that predict su ccessful and effective volunteers. In addition, there appears to be a strong relationship between the Five Factor Model of personality traits and effective interventionists (Paterson, Reniers, & Vllm 2009; Tishler, Bartholomae, & Rhodes, 2005). According to personality theory and research, these are inherent traits and mostly world. However, people are able to moderate their relationship to the world by utilizing e nvironmental adaptations (Harkness & Lilienfeld, 1997). The current study extend ed these theories to paraprofessional crisis intervention training. Congruent with the aforementioned the ories and research, this study wa s aimed at the personality characteris tics of trainees that lead to the successful completion of a t raining program. However, this wa s not a direct relationship and wa s actually moderated by environmental adaptations that allow a trainee to utilize his or her natural tendencies The general di rection of these hypotheses were that higher levels of positive adaptations would increase the likelihood of success in training, whereas fewer adaptations or more negative ones would decrease the chance of success. which was defined as being screened out by a trainer of if the trainee drops out on their own volition. More specifically, the personality characteristic that seem s to be the most strongly related to the successful completion of crisis intervention t r aining i s intrinsic empathy. A personality trait of empathy has consistently been shown to be correlated with skill development, sophisticated use of facilitative skills, therapeutic change, and client out come (Hill et al., 2008). Intrinsic empathy as a personality trait is also inclusive of environmental variables, such as life
47 Hill, 2008). However, trait empath y has only been investigated as affecting the effectiveness of volunteers, not the ou tcomes of training. It wa s expected that there would be a positive relationship between natural empathy and the successful completion of training. Conversely, lower levels of empathy would be strongly associated with being rejected from training. To measure the core of personality, it would seem that the Five Factor Model best suites the literature on crisis intervention training. In replication of Paterson, Reniers, and V llm (2009), it is expected that successful trainees will have higher levels of Agreeableness than rejected trainees. Although Tishler, Bartholomae, and Rhodes (2005) did not find significant results for other personality traits, they would likely argue tha t successful trai nees would score higher on Extra version, Openness, and Conscientiousness because those dimensions match well theoretically with facilitative skills. Prior, albeit limited, research has not revealed training to be significantly affected by Neuroticism. However, because rejected trainees have been shown to have more maladaptive characteristics consistent with Neuroticism ( Heilig, 1967; McClure et al., 1973; McCord & Packwood, 1973 ), in the present study, higher levels of Neuroticism were expe cted to be associated with a lesser likelihood of training success. Related to maladaptive characteristics, the current study will also extend the research on how perfectionism relates to skills training. Qualitative resea rch from Hill, Sullivan, Knox, and Schlosser (2007) found that perfectionism is negatively related to helping skills, and Hill et al. (2008) discovered that high levels of perfection ism were more likely among less effective volunteers. However, the operationalization of perfectionism on a used by Hill and colleagues did not match with the measure of perfection ism they employed ; rather, the measure is designed to distinguish maladaptive from adaptive
48 perfectionists. In addition, the current study followed a similar procedure consistent with Hill et al. (2008) to investigate the relationship between perfectionism and the completion of training. It wa s expected that maladaptive perfectionism would moderate the outcome of training specifically resulting in fail ure wherea s adaptive perfectionism and the absence of perfectionism would not have any major effect on outcome. In addition to the personality traits that appear to be related to the outcome of training, this relationship may be moderated by other variabl es. Despite the importance of other variables being supported by the literature, there are no clear variables that have been identified One environmental adaptation that may help a trainee to utilize a natural empathic reaction is emotion regulation (John & Gross, 2007). The hypothesis of the current study wa s that successful trainees would utilize less maladaptive and more adaptive skills compared to those who d id not complete the training program. Emotion regulation research indicates that the response m odulation mechanism is related to a deeper understanding of negative emotions similar to crisis, whereas cognitive change is negatively related to empathic reactions to negative emotions (Gross & John, 2003) Thus, it wa s expected that those who scored low er on empathy and the positive personality traits would be rejected if they scored higher on cognitive change (Reappraisal) whereas those scored lower on Reappraisal would be successful. In contrast, trainees with higher empathy and positive personality t raits would be rejected if they score lower on response modulation (Suppression), compared to those who score higher on Suppression that would be more likely to succeed The available literature and crisis center certification criteria suggest that trainee s be screened with application, interview, and experiential processes designed to monitor for the helping skills outlined in this study. Most trainee screening s happen at the application or
49 interview level because this is where the most egregious issues re lated to personality, empathy, or intervention philosoph y are noticed and m ore subtle issues tend to be discovered during the training process ( American Association of Suicidology, 2010; Hart & King, 1979; Kalafat, 2002; McCord & Packwood, 1973; Mishara e t al., 2007b) Therefore it wa s also anticipated that there would be significant differences in personality traits and natural empathy among those screened out at different points More specifically, that those who were screened out during the interview pr ocess would be lower on Extr a version, Openness, Agreeableness, Conscientiousness, and empathic tendencies compared to those who were screened during the training process or complete d the program The overall goal of this research wa s to extend the literatu re on understanding the training process for paraprofessional crisis interventionists. There is an apparent consensus in the literature that personality affects the outcome of training but with the exception of some limited research on higher order dimens ions derived from the Five Factor Model, there has been very little attention to the study of specific personality traits and their association with training outcome. The literature also suggests that the relationship between empathy and training outcome i s related to moderator variables, but again the exact variables are unclear If these variables are found to be strong predictors of training outcomes, then they may be able to serve as a screening measure for centers that do not have the resources to inve st in a stringent training program. There is only a limited amount of available literature on which to base this research model, therefore this is more of a beginning conceptual model that may drive future research. Summary of Hypotheses To summarize the expected direction of the results, the following list and Figure 2 1 reviews each research question and hypothesis:
50 1) Is a personality trait of empathy related to training success ? I t was expect ed that stronger empathic tendencies would be related to a great er likelihood of completing training, whereas lower or non existent empathy would predict failure 2) Will broad personality traits affect the outcome of training? I t was expect ed that successful trainees would be higher on Extra version, Openness, Agreeablen ess, and Conscientiousness than non successful trainees. Because research has found that rejected volunteers often have more maladaptive traits, it was also expect ed that they would be higher on Neuroticism than those who graduate d from training. 3) How does perfectionism relate to volunteer training? I t was expect ed that perfectionism would be a significant moderator between training outcome and both personality characteristics and empathy Although more maladaptive personality characteristics and lower leve ls of empathy were expected to decrease the likel ihood of training success, th ose effect s w ere expected to be even worse for maladaptive perfectionists. In contrast, adaptive perfectionism may compensate for maladaptive personality characteristics and lowe r levels of empathy; other things being equal, adaptive perfectionists with lower empathy or more maladaptive personality characteristics were not expected to have low levels of completed training. There were no specific expectations for non perfectionists in terms of predicting training outcome. Likewise, at higher le vels of empathy or positive personality characteristics perfectionism wa s not expected to play a role in training outcome. 4) What role will emotion regulation play in training outcome? I t was expect ed that emotion regulation would be a significant moderator between training outcome and both personality characteristics and empathy In thi s case, lower levels of the positive personality characteristics and empathy would each decrease the likelih ood of success. In addition, higher levels of the cognitive change mechanism (Reappraisal) would further decrease the likelihood of success. In contrast, higher scores on response modulation (Suppression) would compensate for those initial levels of person ality and empathy to increase the likelihood of success in training. More s pecifically lesser empathic tendencies and maladaptive personality traits would predict failure when higher level s of the cognitive change mechanism (Reappraisal) were present 5) Wi ll there be any differences in personality traits and empathy depending on when a trainee is screened out? I t was expect ed to find that those who were screened out during the interview process would score lower on Extr a version, Openness, Agreeableness, Con scientiousness, and empathic tendencies when compared to those who were screened out during the actual training process or successfully complete d the program.
51 Figure 2 1. Moderator model
52 CHAPTER 3 METHOD Participants Center Recruitment As Hill and Lent (2006) noted previous research on training outcomes was weakened by data collected from only one site representing a single training method, program, philosophy, and trainee population. In response to this critique, the first level of recruitment for this study wa s on a programmatic level, meaning the involvement of crisis centers. To improve the external validity of the study, 7 c risis centers each from different states and geographic regions were recruited to obtain a diverse cross section of traine es. By collecting data from several centers, the results were also less likely to contain regional, cultural, and organization biases that were inherent in the training program philosophies. Crisis centers were recruited through the American Association of Suicidology (AAS) email listserv, which is directly linked to AAS certified crisis center directors, as well as educators and researchers specializ ing in crisis intervention and suicide prevention. Each center completed two training classes during the dat a collection phase of this research that were included in the analyses. Crisis centers were recruited based on their match to a stringent training process. As previously mentioned, the most effective training programs utilize multiple levels of screening, which were the inclusion criteria for programmatic participation. Additionally, this uniformity among participating centers ensure d the highest level of constraint and increase d experimental validity. In the current study, appropriate programs screened pot ential trainees through application, interview, and role play training process es as well as a probationary period after graduating from training where trainees were closely monitored by experienced staff Included centers ha d some individual differences in their exact skill use and emphasis depending on local
53 mental health and cultural needs, number of active volunteers, total hours of required training, and timing of training classes; however, all recruited centers follow ed AAS guidelines for accreditation The AAS accreditation guidelines dictate organizational structure, the training process, high standards of service, and program evaluation (American Association of Suicidology, 20 1 0) Despite these differences, the literature suggests that trainers are l ooking for the same basic skills of empathy, warmth, and genuineness ( Hart & King, 1979; Kalafat, 2002; Mishara et al., 2007a,b) It was also expected that most centers, if following consistently similar training guidelines, would experience comparable att rition rates during the training process. Presently there are no available studies to document exact or average attrition rates among crisis center training programs; however anecdotal observations across several crisis centers would suggest that between 1 0% and 30% of trai ning applicants are screened out before the training program begins, and another 10% to 40% are screened out during the training process. The current study was not aimed at confirming these rates because not all trainees were required to participate Individual Participants The second level of recruitment involve d the trainees of the recruited crisis centers. Trainees were given the opportunity to participate in this research before the training process began Participants were recruited t hrough a standardized recruitment letter given out by a volunteer of the respective center rather than the training director to avoid any perceived coercion or obligation to participate. All potential participants went through the application and were invi ted for an interview which should replicate the self selection and initial screening processes mentioned in the literature A total of 97 trainees consented to participate, and 91 trainees completed the battery of measures. Participants were given embedde d validity items explicitly directing them to answer
54 the item in a specific way to be sure they were following instructions and reading the items No participants were removed from the final sample for these validity checks ; however, six particip ants were removed for having completed less than half of the measures Because these six participants did not provide answers to most items, there was not enough data to check for any significant differences between them and those included in the final sam ple. Demographic information was collected through a self report survey (see Appendix A). The final sample of 91 trainees consisted of 32 men ( 35 .2%), 58 women ( 63 7 %), and 1 with missing gender data (1. 1 %). Participants ranged in age fr om 18 to 56 ( M = 29 39 SD = 1 0 40 ). Approximately 72 5 % of the sample self identified as White/Euro American, 9 .9% Black/African American, 3 3 % Asian Indian/Asian Indian American, 1.1% Vietnamese, 2.2% Filipino, 2.2% Multiracial, 3.3% Other, and 4.4% with missing data. Abou t 9.9% of participants self identified as being of Mexican, Puerto Rican, or other Hispanic/Latino/Latina descent Of the total sample, the highest completed education included 8.8% high school diploma, 6.6% associates degree, 39.6% some college, 28.6% bac helors degree, 13.2% masters degree, 2.2% doctoral degree, and 1.1% with missing data. Approximately 51.6% of participants were currently enrolled in college, whereas 46.2% were not, and 2.2% who did not report. Among those enrolled in college, 47 particip ants had a grade point average to report, which ranged from 2.0 8 to 4.0 ( M = 3.43, SD = 0. 4 5) based on a scale from 1.0 to 4.0 (participants who did not yet have an official GPA or were not enrolled were asked not to provide an answer). About 37.4% of part icipants had engaged in some form of human relations or helping skills training prior to volunteering, 61.5% had not, and 1.1% did not report. At the point of entering training, 64.8% of trainees were employed, 33.0% were not, and 2.2% did not provide an a nswer.
55 Measures Empathy The Interpersonal Reactivity Index (IRI; Davis, 1980) was used to measure intrinsic empathy as a personality trait. The IRI is a 20 item self report measure. Items are rated on a 6 to take anothe (Empathic Concern), tendency to respond empathically towards fictitious characters (Fantasy), and reactions of distress to the experiences of others (Personal Distress). Higher summed scores on each subscale are consistent with higher levels of empathy. Davis (1980) found the internal consistency of the IRI ranged from .71 to .77 and evidence for strong convergent, discriminant, and criterion validity which has been replicated by later research using college age populations (Hill et al., 2008; Stahl & Hill, 2008). Alter man et al. (2003) supported a parsimonious latent structure of the IRI that was used for this study. In a factor analysis, they found that items from Empathic C oncern and Perspective an d Fantasy subscales were not a central component of this latent construct Therefore, based on their recommendations, items from the Perspective Taking and Empathic Concern subscales were combined to create a total empathy score that better captured the conceptualization of empathy adopted in the present study. Alterman et al. (2003) reported that empathy scored in this manner from the IRI was sig nificantly related to positive personality characteristics and other measures of empathic tendencies.
56 Personality The Mini International Personality Item Pool (Mini IPIP; Donn ellan, Oswald Baird, & Lucas, 2006) was used to assess core personality traits c onsistent with the Five Factor Model of personality. This measure is a 20 item short form version of the International Personality Item Pool Five Factor Model that originally consisted of 50 items (Goldberg, 1999). The Mini IPIP has an item pool to measu re Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness/Intellect/Imagination with four items each. Participants rate each item on a 5 point T he Mini IPIP subscales of Extraversion, Agreeableness, and Neuroticism demonstrated adequate reliability among college student populations with subscale alphas ranging from .72 to .85; the Conscientio usness and Imagination subscales have shown reliability alphas of .63 and .67 which may need to be interpreted with caution. In previous research t he Mini IPIP also demonstrated good convergent validity with the original longer measure having correlations ranging from .55 to .85, and strong criterion validity. Perfectionism The Almost Perfect Scale Revised (APS R; Slaney, Ashby, & Trippi, 1995; Slaney, Rice, Mobley, Trippi, & Ashby, 2001) was used to measure dimensions of perfectionism. The APS R contains 2 3 self report items using a 7 and standards for performance (Standards), the subjective degree of discrepancy between expectations and actual performance (Discrepancy), and a need for order and structure (Order). The APS R subscales are adequately reliable among college student populations
57 (Standards), .92 (Discrepancy), and .86 (Order). Construct validity for the measure has ranged from .49 to .83 (Rice & Slaney, 2002; Slaney, et al. 2001). The Standards subscale i s used to differentiate perfectionists from non perfectionists, and the Discrepancy subscale then distinguishes maladaptive from adaptive perfectionists. Although this measure in its current form includes the Order subsc ale, a review of perfectionism research has suggested that across several measures of perfectionism, this particular construct does not seem to help in distinguishing groupings of perfectionists (Stoeber & Otto, 2006). Emotion Regulation The Emotion Regul ation Questionnaire (ERQ; Gross & John, 2003) was used to measure emotion regulation mechanisms. The ERQ includes two subscales that measure cognitive change (Reappraisal) and response modulation (Suppression) strategies using 10 self report items rated on a 7 that higher scores are indicative of a stronger tendency to use the relative strategy. The ERQ has demonstrated high internal consistency in undergraduate samples with alphas ranging from .75 to .82 for the cognitive change subscale and .68 to .76 for the response modulation subscale. Gross and John (2003) also found the ERQ to have strong convergent validity with other measures of affect and affect regulation, as well as adequate divergent validity from measur es of other similar constructs Procedures Par ticipating crisis centers were provided with instructions for the training director and instructions for participants. Training applicants were given a link to a website containing the informed consent contract and measures. All individual participants wer e given an online
58 informed consent page describing the nature of the study as research ing characteristics on crisis intervention trainees and were prompted to provide an electronic signature before being permitted to proceed (see Appendix F ) Participants complete d the Mini IPIP, IRI, APS R, ERQ, and answer ed questions about general demographic information (see Appendices A E) According to prior research most training programs are rather efficient at screening out grossly inappropriate trainees before th e training process begins (Evans, 1977; Hart & King, 1979; Heilig, 1967; Kalafat, 2002; McCord & Packwood, 1973; Mishara et al., 2007b) thus m easures were completed after the initial app lication but before any screening interview, exposure to the training process or intervention philosophy. The goal of timing da ta collection at the interview wa s to systematically account for those who were removed from the application pool due to extraneous reasons, such as time commitment issues, and allow the applicatio n process t o be part of the error term in the analysis. The accuracy and integrity of responses were checked with Pilot data found that there were no significant sequencing effects of the measures Once all measures were administered, training and screening processes continue d as they were intended according to their respective systems and philosophies. Training directors were asked to keep data on outcomes for individual trainees, such as if they gradua te d from the program, drop ped out on his or her own, or were screened out by a trainer and for what reason Outcome data were collected from the directors after the training program commence d Although the directors knew to keep records of this information the training team members were blind to the experiment al purpose and design Additionally, the training directors were not given feedback regarding the responses of the trainees.
59 As previously mentioned, the outcome variable for this study wa s based on t he individual wa s fairly simple in that a person applied for training, interviewed, and successfully completed the entire training process to become an official volunteer. O pposite of success is failure; however t he s more complex in that it include d a large degree of variability of reasons for not completing training. a s an individual that does not comple te training d id not fit the conceptual model of wa s defined as an individual that began the training program, but wa s screened out due to lack of skill acquisition philosophical mismatch, and other reasons that are rooted i n personal dynamics. own volition for various reasons. The re were also those that were excluded for extraneous variables such as being too young (many centers hav e age requirements) or not being able to make a required time commitment; thus these individuals were screened out during the application process and not included in the actual data collection.
60 CHAPTER 4 RESULTS P articipants were grouped based on the dic hotomous outcome of training according to the There were 57 participants in the 33 Within the failure group there were 17 participants who dropped out of tra ining, and 16 who were screened out. Because data were collected aft er the application process, consist ed of those who were screened out of training due to legitimate skill and empathy issues and those who dropped out for personal reas ons Because this study was aimed at only those who attempted training and failed, those who were screened or dropped at the application step did not participate in the study For the proceeding analyses, the dichotomous outcome of training was coded as 1 for success and 0 for f s the reference point for the logistic regression analyses Descriptive Statistics and Preliminary Analysis Table 4 1 displays the range of scores, means, and standard deviations for the individual sc ale scores. When compared to other research, the means and standard deviations obtained fro m these participants were mostly identical ; t he exception s were Mini IPIP and ERQ scores which were all higher on average. Table 4 2 includes correlations among all measures and subscales, as well as internal consistency coefficients for each measure The internal consistencies reported in this study were comparable to what previous research has shown, with the exception of .77 o bserved for the Standards subscale compared to .85 ( Rice & Slaney, 2002; Slaney, et al., 2001 ). The correlations among subscales were generally low to moderate The s ignificant correlations among subscales did not have a high magnitude and tended to relat e in directions that are conceptually congruent. The data were investigated for potential outliers, and according
61 to Tab achnick and Fidell (2006) this wa s done by examining the z score distribution of each subscale per outcome grouping. Following their rec ommendation, there was one outlier that surpass ed the critical z score of 3.29 in the Agreeableness subscale data. Upon further exploration, this appeared to be a univariate outlier that was about 4 standard deviations below the next smallest Agreeablenes s score within the successful group, and almost 6 standard all of his or her other subscale scores were about average for the group, and he or she correctly at tended to all of the validity check items. It seem ed this score was a true outlier and errant oddity, and was thus removed fr om the subsequent analyses. Mahalanobis distance was calculated for each group and confirmed that the data did not include any mult ivariate outliers. The data fell within the accepted guidelines for normality assumptions (skewness and kurtosis values less than 2) Initially, the Agreeableness subscale scores of the Mini IPIP were negatively skewed When the aforementioned outlier was removed, the skewness and kurtosis values for this subscale were acceptable, thus no data transformation was necessary. The agreeableness construct is operationalized as personality qualities that involve being good natured and cooperative, which is partic ularly fitting for this population (Paterson, Reniers, & Vllm 2009) Primary Analyse s To tes t the hypothesized relationships between personality traits and training outcome, a series of binary logistic regression analyses were performed Binary logistic regression analyses are robust enough not be affected by variances in the independent variables or measurement error, so the unstandardized coefficients were used in the analysis and no statistical corre ctions were necessary. In addition to this binomial a ssumption, a logistic regression analysis also assumes a linear relationship between continuous predictors and the dichotomous outcome.
62 Hypotheses that included moderating variables were entered in sequential steps to test conditional effects and interacti ons. Hypothesis 1: Trait Based Empathy The first hypothesis was that stronger natural empathic tendencies would predict successful trainees, versus those with lower empathy who would be more likely to fail. To assess the relationship between trait based e mpathy and success in training, the logistic regression model used the total score for the two included subscales as the independent variable to predict the dichotomous outcome. A test of the full model against a constant model was not statistically signif icant 2 ( 1 N = 9 1 ) = .32 p = 5 7 This result indicated that the operationalization of trait based empathy with this measure did not contribute to the outcome of training. Hypothesis 2: Personality Characteristics The second hypothesis predicted that h igher scores on Extraversion, Openness, Agreeableness, and Conscientiousness would be characteristics of successful trainees, compared to those with lower scores on these variables and/or higher scores on Neuroticism who would be more likely to fail. To as sess the relationship between the F FM of personality traits and training outcome, the logistic regression model included Mini IPIP subscales as variables to predict training success or failure. The full model suggested that the Five Factor Model of persona lity traits did in fact help predict the outcome of training, 2 ( 5, N = 90 ) = 11.19, p = .0 5 and could accurately predict 71.1 % of cases More specifically, the Wald criterion showed that the Agreeableness construct was the only factor to significantly pr edict the outcome ( b = .45, p = .0 1 ) The odds ratio wa s 1.55 for this predictor (95% confidence interval of 1.12 to 2.13) suggesting that a one unit increase on the Agreeableness subscale score increases the likelihood of passing training by about 55% T he conditional effect for Openness represented a trend effect
63 ( p = .07) and could not be confidently interpreted; however, the odds ratio for Openness is .82 which suggest ed higher score s may be associated with failure Thus it would seem that the FFM pers onality factors d id Agreeableness scores were more likely to succeed and complete the training Hypothesis 3: Perfectionism The third hypothesis was that perfectionism would sig nificantly moderate the relationship of natural empathy and FFM to training outcome, in that maladaptive perfectionism would negatively affect the relationship, and both adaptive and non perfectionism would not modify the relationship. Testing the relative adaptiveness of perfectionism requires the use of two subscales (Discrepancy and Standards), therefore s eparate 3 step logistic regression model s w ere created to test the moderating role perfectionism has on the relationship between FFM factors and traini ng success, and on the relationship between empathy and training success With the model of empathy factors, t he first step included centered APS R and IRI total score as the conditional effects, follow ed by a second step of including 2 way interaction s be tween scores The final step added 3 way interaction terms between Discrepancy, Standards, and the IRI total score The first model of conditional effects of empathy and perfectionism was non significant 2 ( 3 N = 91) = 1.2 7 p = .7 4 The second step of 2 way interactions was also non significant 2 (3 N = 91) = 2 84 p = 42 ], as was the third step with 3 way interactions 2 (1 N = 91) = 44 p = 51 ] These results indicate d that adding perfectionism as a moderating variable did not affect the relation ship between empathy and training outcome The first model of conditional effects of FFM factors and perfectionism was non significant 2 (7, N = 90) = 12.18, p = .10 The second step of 2 way interactions 2 (11, N = 90) = 14.53, p = .21 ] and the third st ep with 3 way interactions 2 (5, N = 90) = 7.52, p = .19] also
64 resulted in non significant effects. I t is interesting to note that the earlier analysis of FFM factors as conditional effects was a significant model, which was lost simply by adding the APS R subscale scores As with the previous analysis, Agreeableness remained a significant predictor throughout this model ( b = .23, p = .03); however, this was not sufficient to make the entire model significant, which may have been suppressed by the addition of perfectionism variables. The results of both models suggest ed that in this conceptual model, perfectionism was not a n influential variable associated with ed or enhance d his or her ability to comp lete a training program. Hypothesis 4: Emotion Regulation The fourth hypothesis was that emotion regulation would significantly moderate the relationship between the measured personality constructs and the outcome of training M ore specifically the expec tation was that higher levels of cognitive change (Reappraisal) would negatively affect the relationship and higher scores of response modulation (Suppression) would modify the conditional effect to predict success. A 3 step logistic regression model was used to examine emotion regulation as a moderating variable in the relationship of empathy to training outcome, and the FFM factors to training outcome. The initial model of conditional effects for empathy and emotion regulation was not able to significant ly predict the outcome of training [ 2 (3 N = 91) = 6. 10 p = 11 ]. Emotion regulation also did not significantly moderate the relationship between empathy and ou tcome in the second block of predictors [ 2 ( 3 N = 91 ) = 3.85 p = 28 ] The final block showed that adding the 3 way interactions did not lead to a significant step in the 2 (1, N = 91) = 04 p = 84 ]. With the relationship between FFM constructs and outcome, the initial model was significant [ 2 ( 7, N = 90 ) = 14.53, p = .04 ] and consistent with the earlier finding the effe ct for
65 Agreeableness was positive and significant ( b = .41, p = .0 2 ). The 2 way interaction among the variables was not significant [ 2 ( 11, N = 90 ) = 15. 70 p = .15 ] nor was the 3 way interaction [ 2 ( 5, N = 90 ) = 4.9 2 p = .4 3] Although none of the 3 way interactions was significant, the same 2 way interactions between Reappraisal and Conscientiousness ( b = .06, p = .05) and Suppression and Extraversion ( b = .06, p = .01 ) were significant throughout the steps and may again indicate the importance of emoti on regulation as a moderating variable Hypothesis 5: Failure Group The final hypothesis was that trainees with the lowest empathy and negative personality characteristics would be screened out during the interview process, and those with relatively lowe r scores would be screened earliest in the training process. collecting data from multiple sites, the different sites were not all able to provide such specific This limitati on resulted in being able to collect data only about the training program on his or her own volition or w as asked to leave by the training staff Although the data are informative, the number of participants in these groups create d such small cells that smaller effects could not be detected. Statistical inferential tests could be used to explore these data, but they would only reveal significant results for those effects that were substantially large. A lthough the data are informative, they d id not allow for substantial confidence in testing the hypothesis regarding the failure group using inferen tial statistics T he levels of the personality scores of interest within the failure group s could have be en compared but the strength of those relations was probably best gauged with effect sizes rather than inferential statistics
66 A series of Bonferroni corrected t tests were conducted and d ( measure of effect size ) was calculated for each comparison to better understand these relations As might be expected given the small sample sizes, n o ne of the t tests showed statistically significant differe nces between the groups Table 4 3 displays the means, standard deviations, and the resulting effect sizes for differences between the tw o failure groups The effect sizes were typically of trivial size ( d < .2), with a few small to moderate effects rangin g from .29 to .56 The pattern of effects suggest ed that some of the FFM factors, empathy, and perfectionism may have play ed a role in differentiating those trainees who are likely to drop out of training versus those who are screened out by the trainers. It seem ed that those who dropped out scored lower on empathy related factors and higher on personality factors that were predicted to motivate helping tendencies. T hose who were screened scored higher on average with empathy measure, but also higher on Neu roticism and Discrepancy, and lower with other positive personality factors. These trends may indicate that those who dropped out of training had more of the characteristics that the previous analyses suggested are associated with success, versus those who were screened that did not score as high on those particular characteristics. This finding may mean that those who dropped out were aware that they did not empathize as much compared to those who needed to be asked to leave training rather than recognizin g the lack of fit on their own. Exploratory Analyses T here are a large number of extraneous variables that can affect training beyond those examined in this study (e.g., age, education, experience, and others, see Hill & Lent, 2006 ). Although there is no f irm grounding in the literature to support directional hypotheses, it was important to explore some of these variables. A series of chi square analyses were conducted to compare the success and failure groups for differences in terms of training site, gend er race/ethnicity, level of education completed current enrollment in college employment status,
67 and previous helper skills training experiences. The results of these analyses and show ed there were no significant differences among the variables ( p > .05) ; however, there seemed to be a significant difference between the two groups with level of 2 (5, N = 90) = 11.25, p = .05]. The standardized residuals of the cells suggest ed that trainees with a high school education or associates degree tend e d to fail at a significantly higher rates with a medium effect size ( = .35) There were no significant differences between the outcome groups with age [ t (88) = 1.35, p = .18] or grade point average [ t (45) = 1.24, p = .22].
68 Table 4 1. Sample means and standard deviations by group Fail (N = 3 3 ) Success (N = 57) Measure Min. Max. M SD Min. Max. M SD Discrepancy 13 75 37.24 15.28 14 79 33.95 13.55 Standards 30 49 41.65 4.84 30 49 41.90 4.89 Extraversion 4 19 12.79 3.65 6 19 13.72 3.18 Agreeableness 10 20 17.71 2.42 15 20 18.66 1.28 Conscientiousness 9 20 15.71 3.43 8 20 15.65 3.10 Neuroticism 4 16 10.38 2.88 5 16 9.95 2.72 Openness 9 20 16.65 2.90 11 20 16.54 2.53 Reappraisal 20 42 32.18 5.80 14 42 31.81 6.06 Suppression 4 22 13.59 5.62 5 21 11. 21 3.59 Empathy 42 63 54.50 5.26 45 65 55.14 5.26
69 Table 4 2. Scale correlations and reliability estimates Measure 1 2 3 4 5 6 7 8 9 1 0 11 1. Discrepancy .93 2. Standards .10 .78 3. Extraversion .32** .14 .81 4. Agreeableness .05 .16 .07 .65 5. Conscientiousness .25** .23* .15 .12 .77 6. Neuroticism .39** .00 .16 .11 .27** .61 7. Openness .23* .29** .35** .35** .02 .03 .73 8. Reappraisal .26* .09 .09 .14 .20 .41** .25* 84 9. Suppression .40** .06 .38** .28** .12 .01 .25* .05 .76 10 Empathy .21 29 ** 08 43** .2 0 .2 0 18 .3 9 ** 18 .79 11 Outcome .11 .03 .13 .25* .01 .08 .02 .03 .25* .0 6 -Note: ficient, except for Outcome which is a point biserial correlation. coefficient s alpha appear in italics on the diagonal. p < .05. ** p < .01.
70 Table 4 3 Sample means and standard deviations for failure group Dropped (N = 17) Screened (N = 1 6) Measure M SD M SD d Discrepancy 36.94 12.32 38.50 18.28 0.10 Standards 42.35 4.02 40.44 5.34 0.42 Extraversion 13.29 3.41 12.25 4.04 0.29 Agreeableness 17.82 2.70 17.69 2.21 0.06 Conscientiousness 15.77 3.44 15.44 3.54 0.10 Neuroticism 9.59 2.67 11.13 3.03 0.56 Openness 17.29 2.57 15.88 3.20 0.51 Reappraisal 32.82 5.49 31.75 6.32 0.19 Suppression 13.71 5.43 13.94 5.84 0.04 Empathy 53 24 5 71 56 13 4 4 9 0. 57
71 CHAPTER 5 D ISCUSSION The purpose of this study was to investigate the personality characteristics of paraprofessional crisis intervention trainees that lead to successful completion of a training program. Previous research has suggested that the most successful interventionists tend to have in common certain qualities withi n the Five Factor Model of personality and a high degree of intrinsic empathy ( Paterson, Reniers, & Vllm 2009; Tishler, Bartholomae, & Rhodes, 2005 ). This same literature has demonstrated that these personality traits tend to be unchangeable characterist ics for the way in which an individual engages the world; however, this relationship can be moderated by employing environmental adaptation skills ( Harkness & Lilienfeld, 1997 ). It seems the most relevant potential moderating adaptations that have come out of prior research are perfectionism and emotion regulation (Hill et al., 2008 ; Hill, Sullivan, Knox, & Schlosser, 2007 ; John & Gross, 2007). The main goal of the present study was to investigate how these personality characteristics related to the outcome of training, and how the suggested moderating variables may alter that association In the current study, crisis center training programs experienced about a 37% attrition rate of trainees which is similar to anecdotal information provided by crisis cent er directors It is also interesting to note that the range s of scores on some of the Mini IPIP and ERQ subscales were higher than most previous studies. Conceptually this seems consistent with a helper population that research has suggested are likely to more skilled with managing emotions ( Sawatzky & Paterson, 1982 ; Stahl & Hill, 2008 ). Based on available information the data seemed representative of many crisis intervention centers around the count ry and thus may provide helpful information that can generalize to most training programs.
72 Regarding the conditional effects of the variables it was surprising to find that empathy did not significantly predict the outcome of training. Previous research h as noted that empathy is a key factor in distinguishing successful and effective crisis interventionists, and is a staple of crisis intervention skills ( Hill et al., 2008; Mishara et al., 2007a,b ). Throughout these studies a trait level of empathy has been shown to be the most essential component of an effective interventionist, and was projected to be a predictor of trainability. However, it appears that trait empathy did not have a strong conditional effect as a predictor for training outcome. Based on th e literature this finding was unexpected, but may be in part due to the conceptualization and measurement of empathy as a trait like construct within this population. Examining empathy as a trait may legitimately distinguish people who identify as helpers from those who do not but within a population of identified helpers, state based empathy might yield more accurate results ( Bobevski & McLennan, 1998; Eriksen & McAuliffe, 2003; Hill et al., 2008; McLennan, Culkin, & Courtney, 1994 ) In other words, peopl e with high trait like empathy might be naturally drawn to crisis intervention work, but the distinguishing factor is with those who can have a higher state of empathy when engaging in training role plays This interpretation would be consistent with the n otion that empathy involves both the internal emotional reaction and a behavioral output to demonstrate it ( Ickes, 1993 ) T he conditional effects of the Five Factor Model of personality characteristics predicting training outcome showed that the Agreeablen ess subsc ale was a significant predictor which is consistent with expectations based on the literature (Paterson, Reniers, & Vllm, 2009 ) H igher scores on this subscale indicate that a person has a tendency to be compassionate and prosocial. T he items of this subscale are worded in a way that is conceptually very close to empathic tendencies and there was a si gnificant correlation with empathy ( r = 43 p < .01) to indicate a
73 degree of conceptual overlap, but not enough to be concerned about multicolline arity This finding appears almost contradictory to the aforementioned results that relied on a direct measure of empathy. One possible explanation is that the Agreeableness subscale as part of the Mini IPIP, is a more robust and parsimonious measure of e mpathic tendencies as a personality characteristic, whereas the IRI is focused on specified components of empathy and may have degraded the effect If this discrepancy between measures is correct, it could mean that a personality trait of empathy does sign ificantly pred ict trai ning outcome. T he importance of the Agreea bleness construct also suggests that being open to feedback is integral to the training process which can give trainers something s pecific to look for when deciding ogram (Kalafat, 2002). There may be a range of Agreeableness scores, particularly the highest end of scores, within which an applicant is likely to be a good fit for training, and anything outside of that range may be a cause for concern in terms of the pr obability of success. The odds ratio associated with Agreeableness is 1.55, meaning that the odds of a trainee who is relatively higher on Agreeableness is about one and a half times higher than a trainee who is lower on Agreeableness. It is important to n ote that the predicted probabilities of success in this model do vary depending on scores of the other FFM dimensions; for example, a score of 1 on Agreeableness was associated with predicted probabil ities of success ranging from 37 % to 71 %. However, since the other FFM dimensions were non significant in this model, the predicted probabilities associated with Agreeableness alone provides a clearer picture of increasing trend of success with higher Agreeableness scores. For example, when taken alone, a score of 18 on Agreeableness is associated with a 61% predicted probability of success. Therefore, trainees who score in the higher range of this measure may not require any additional help or training than they currently receive; conversely, those who score in
74 the lower range may need extra attention because the crisis interventionist skills may not come as naturally for them. The Openness effect suggested a possible trend such that an important factor in successful trainees might be described as having an open appreciation of and curiosity for new experiences. As previously mentioned, an open and non judgmental attitude towards callers is important in helping others deal with difficult issues such as suicidal ideation, which means this construct may have more of an effect than was shown in these results. Neuroticism never emerged as a statistically significant predictor of outcome ; however the effect sizes of mean differences in scores between the groups is consistent with prior research which suggested that t hose who score on the high end of Neuroticism are generally unsuited for crisis intervention work and typically screened out ( Heilig, 1967; McClure et al., 1973; McCord & Packwood, 1973 ). Contrary to the hypothesis, Conscientiousness and Extraversion were not significantly related to outcome. However, as explained with the inspection of the failure group, some of these factors may be helpful in differentiating reasons for lack of success in the training than overall training outcome, per se For example, th ose who were screened out during the training processes tended to score lower on positive personality characteristics such as Agreeableness, and higher on negative characteristics such as Neuroticism. Perfectionism did not significantly moderate the relati onship between either empathy or personality and training outcome as initially p redicted. In this case, higher scores on Standards and Discrepancy would denote perfectionists and Extraversion describes a desire to be connected with other people. T hese int eractions cannot be interpreted with any degree of empirical confidence; however, a n examination of subscale means and the exploratory chi square test suggest that those with more maladaptive perfectionistic tendencies and less outgoingness
75 may negatively affect a chance at successfully completing a training program. Although this is not a statistically significant finding, it is consistent with prior research that pointed to perfectionism as a possible factor affecting trainees (Hill et al., 2008 ). Because perfectionism continues to be on the margin of importance across multiple studies, it may be that perfectionism is an individual difference characteristic than a common characteristic tha t most trainees share Emotion regulation also did not significantly affect the relationship between empathy and training outcome. As previously mentioned, this result was surprising given that a majority of the published literature suggested that empathy was a strong factor that distinguished effective crisis such as emotion regulation could help to enhance differences between successful and unsuccessful trainees One possible explanation for the lack of significance in this model is that emotion regulation as a construct does not adequately capture how a trainee effectively utilizes his or her natural empathic tendencies to be successful in training. Therefore, o ther factors may be more important in understanding h ow trainees use empathy and would need to be further investigated by future research. T here were strong predictors that stood out amongst these models, but it is important to also look at their actual pr edictive power. Each full model, regardless of significance, was skewed in its ability to predict outcome, typically towards those who successfully completed training. Consistent throughout the analyses was the finding that the models could predict those w ho would pass training with about an 80% or higher success rate; however, the models could typically only accurately predict around 55 % of t hose who failed. Thus, the full models would be much more reliable for predicting those who could successfully compl ete training, versus
76 weeding out those who are likely to fail. Based on the odds ratios, the effect sizes of nearly all predictors in all models were about 1.0, indicating the effect of these variables to predict outcome is small Examining roup, it was unfortunate not to be able to follow through with the level of specificity of the original hypothesis; however, preserving the external validity of the study by surveying multiple sites across the country may provide stronger groundwork for fu ture research At this point, the results have shown that there are significant differences in these variables among those who passed versus those who failed training primarily involving Agreeableness and some select interactions involving emotion regulat ion and empathy In better understanding those who did not complete training, the available data were able to show that some of the constructs may influence whether a person dropped from training or was screened out. A majority of the effect sizes tended t o be small according to generally accepted guidelines (Cohen, 1988). The handful of medium effects might be more important to consider because they highlight the differences between groups with a larger percentage of the standard deviation between them. Th ose medium effects that stood out involved Standards, Neuroticism, Openness, and Concern. Based on the mean scores, these effect sizes suggest that those who dropped out were more open to feedback than those who were screened Those who dropped out also ha d higher personal standards for their performance in training and may have dropped because they interpreted feedback as not meeting their standards. Those who were screened scored higher on less prosocial personality characteristics, which may have affecte d their abilities to be genuinely empathic. There were no large effect sizes found. Data were also collected about the reason why a person did not complete training. As previously mentioned, this information was collected from the training directors at the out set of
77 their training cycles and included how the person left training (dropped versus screened) and why. T here we re some clear trends in what caused people to leave the training programs if they dropped or were screened out. Most of those who dropped out of training on their own volition had scheduling issues that would not allow them to commit to the time required for training; a few of these participants dropped out because they recognized they were struggling with attaining the skills. Only one pers on was screened out of training because they could not keep up with the time commitment and did not take it upon his or her self to drop. Of the others who were screened out, about half were asked to leave because they struggled with skill acquirement and could not keep up with their fellow trainees. These trainees may have been able to begin training by learning a very basic emotional paraphrase, but could not develop that skill further to handle more sophisticated emotions and content. The other half of the screened trainees were asked to leave because of philosophy issues. The philosophy of crisis centers are often informed by Rogerian principles and AAS guidelines, both of which take an approach that directly addresses emotion and suicidal ideation in a non judgmental manner. Those trainees that were screened because of philosophy issues tended to have difficulties agreeing with this philosophy, likely for reasons varying from discomfort to personal experiences that affected their abilities to engage in that way. At this point it is unclear whether or not the personality constructs measured in this study had an effect on their being screened. Although the variables in these statistical models have consistently appeared in the literature as being important characterological factors of trainees that affect their success in training, this is not an exhaustive list (Hill & Lent, 2006). The current study collected data on some of these other variables such as training site, gender, age, race/ethnicity, educatio n, enrollment status, GPA, employment status, and previous helper skills training experiences. Of
78 these factors, education was the only one t o be significant. A n examination of the educational groupings replicated typical rates of success to failure for ea ch level of education It is interesting to note that those with a high school diploma or associate s degree had higher failure than success rates ; these are small numbers of participants, but this is also referring to a phenomenon that involves small numb ers of people in a training class. The most glaring difference occurred within the group of participants who had completed their degree. This particular group of trainees had a much higher pass rate ratio in comparison to other groups. One poten tial explanation is that those with at least some college experience or degree have enough education maturity, development, or life experience to be successful with training compared to those without college experience In short, those who have completed their undergraduate studies are the most likely to complete the program. Implications of Results Results of this study have helped to provide a better understanding of the role personality plays in the training of paraprofessional crisis interventionists. Previous research has demonstrated some connections between personality characteristics and effective interventionists, but has not been able to produce consistent or conclusive results to explain how those same constructs may affect the training process. Most of the literature predictor of success, and suggests that there may be other variables that cannot be directly measured which might aff ect that relationship (Hill & Lent, 2006) As previously mentioned, some of these variable s have been hypothesized to be issues that impact empathy and performance, such as emotion regulation and perfectionism respectively. Although the current study has been grounded in this literature b ase which has suggested that there are certain potentially immeasurable variables, this project should be considered to
79 support more of a conceptual model that needs further testing and exploration than a definitive statem ent of primary predictors of training success More specifically, it seems that the Agreeableness construct of personality overlaps with empathic tendencies and is the strongest contributing factor in trainee success; however, other variables such as perfe ctionism and emotion regulation seem to moderate that relationship and other conditional effects in very specific directions. This exploratory finding is informative, but it does not provide an exhaustive list of variables or preclude other factors that ma y have an impact such as childhood experiences, past and current relationship development, exposure to crisis, and cognitive rigidity to name only a few examples. Hopefully the results of this study provide a different way to examine the variables in quest ion because a trait of empathy alone or broad personality traits are almost too parsimonious to explain such a complex phenomenon. The results of this study may be better suited to provide training programs with a profile for trainees that are most likely to be success ful rather than those who are more likely to fail. Consistent with prior findings, the typical successful trainee would be college age d with either some college education, and would more preferably have completed their degree ( Sawatzky & Pate rson 1982 ; Tapp & Spanier 1973 ). Upon talking with this person there would be obvious interpersonal skills, a natural and observable level of compassion towards others, an inclination towards being cooperative, and an excitement to try new experiences. D uring the training process, this person would show that they can tolerate and regulate intense emotional experiences by either holding back and cognitively modifying his or her external responses to be more appropriate for the situation, or not modifying a t all but being willing to openly share his or her internal emotional experience of empathy.
80 According to previous research, the actual services being offered by crisis centers produce considerably favorable outcomes for clients because the training and sc reening process are done effectively. The larger issue at hand is that many crisis centers struggle to obtain enough effective volunteers to provide those services because they often screen out a large number of trainees (McCord & Packwood, 1973). Results of this research may provide a screening battery for crisis centers that do not have the resources for intensive training programs. The goal of the screening battery is to reduce attrition rates during the training process, resulting in more effective volu nteers, which hopefully leads to crisis centers being able to provide more around the clock services for their respective communities. In addition, these results may also offer trainers the awareness that certain personality characteristics such as high ag reeableness significantly increase the probability that the trainee will be able to complete the program with proper training, versus someone who may have a lower degree of agreeableness that may require comparably more time and energy to train. The way in which these results came from such a diverse sample of training programs may allow them to be more generalizable across programs. In the hopes of also improving the number of potentially effective trainees per training class, centers may also want to cons ider more focused volunteer recruitment efforts. The main significant demographic variable was amount of education, and indicated that the most successful trainees had at least some college experience Therefore, it seems that college campuses may be the b est places to advertise for obtain ing trainees. The results of this research also showed that successful trainees have a natural tendency to be helpful, honest, compassionate, cooperative, and open to trying new experience s; all of these characteristics ar e commonly seen in those studying h elping professions, which may be the best college departments to recruit from
81 Limitations and Future Directions A main limitation of this study is the generalizability of results to other helping populations. Crisis inte rvention training is rather specific, and although the skills tend to overlap with other therapeutic and helper skills, the results found here may not be applicable beyond crisis centers, for example to 911 dispatch centers. However, another possible appli cation of this research is to transpose the model to other training related programs, such as for 911 operators or graduate school programs to find the combination of these factors that predict the outcome. In addition, it is important to note that this p articular niche of providing mental health services tends to follow one specific Rogerian philosophy of client interaction, which limits the generalizability of results to the few centers that may follow other philosophies such as solution focused models o f intervention T he training process in these other programs may be interesting to understand in relation to the model set forth in this study ; however future research may still want to continue to focus on the majority group because there are still so ma ny factors that are difficult to measure in a controlled way. Using the Rogerian model that informs crisis intervention skills as a framework for this study provides a strong foundation for the research, but also limits the interpretation of results. For e xample, under this model there were assumptions made that emotion regulation is mostly negative with a few positive qualities, rather than putting more emphasis on the positive. This framework is also partial to emotional experiences over cognitive or beha vioral ones, which may limit the way in which the personality constructs were viewed and interpreted. Because most crisis intervention philosophy are centered around Rogerian principles it may be difficult for future studies to follow a different theoretic al framework, but they could be more intentional in stretching the bounds of the model when operationalizing the constructs of interest for measurement.
82 Most training programs have a strong degree of uniformity because they have to follow the American Asso ciation of Suicidology guidelines for certification as a call center (American Association of Suicidology, 2010). The process of the current study was based on this assumption and did find evidence to support consistency among a random sample of crisis cen ters. T his information is beneficial in confirming the integrity of programs ; however the external validity of results came at the sacrifice of more intense specificity of the data. Future research may want to take an approach of capturing a more intimate picture of a single training program. This could be accomplished by application process, continually measur e the progress of each trainee and record the point at which they leave training in a time to f ailure statistical model. The data should also include enough participants to further investigate personality factors that may affect why a person is screened out of training. It could also be important to follow up with participants to monitor how they fa re after completing training. Research of this nature can involve a level o f specificity that is very data intensive, but may provide a deeper understanding of the internal process of a person attempting to go through crisis interventionist training. Futur e studies may also want to consider the number of variables building the predictive models, because in the current study the true influence of the significant variables may have been lost in the static of a multitude of non significant variables and intera ctions. For example, most of the full models were accurate at predicting those who would be successful with training, but could not be as confident about predicting those who would fail. Results of this study seem to show that there are a small number of s pecific variables that very significantly relate to training success but are often contained in a large group of other minor factors. Thus, it may be important for future work to test more parsimonious models that would allow for a better
83 understanding of those variables that have a more impactful role. By taking this approach the hope would be to find a model that had high predictive success for both pass and fail groups. Another limitation of the current study is the conceptualization and measurement of empathy. Although this study remained consistent with previous research in the literature, it may be important to explore other ways of understanding empathy. The way in which empathy was s natura l capacity to experience the internal emotional impact of an empathic reacti on. This study also looked at emotion regulat ion as a mechanism by which people interact with their empathy and might affect their ability to complete training. Because we understa nd that successful trainees have higher levels of intrinsic empathy, future research may want to take a similar direction with exploring a behavioral differentiating factor. For example, it may be important to better understand differences between those wh have the internal experience. After all, Rogers ( 1951 ) often noted that having an internal empathic reaction is not as important as being able to express it. A related limitation is with the based model. While the goal of this study was to better understand trait level characteristics that influence the outcome of crisis interventionist training, this provides a limited view of the training process that precludes state l evel experiences. Future research may want to focus on better understanding how trait level characteristics may influence state level experiences in training that affect performance and outcome. For example, a trainee may have a low trait level baseline fo r empathy that previous intense role play in training may trigger a higher state level of empathy. As previously suggested, the outcome of training may then be mo level
84 expression of empathy rather than an innate trait. Future studies may also want to focus on how state empathy impacts overall trait empathy. There was also a limitation on the multicultural applications of this research because the general demographics of volunteer crisis interventionists tend to be a more heterogeneous which was reflected in this sample Recruiting participants from multiple states and communities was hypothesized to improve on the rep resentation of different cultures, ages, etc. in the sample, but could be more intentionally addressed in future research There is also the common limitation associated with using self report measures which are open to the possible misinterpretations by p articipants and regional biases which can only be accounted for in the results and not controlled Related to the analysis, the moderator model also poses a potential limitation because both the predictors and moderators are measured simultaneously, which does not create the strongest case for significant moderation. However, this method does fit within the theoretical scope of this study due to the hierarchical nature of the constructs (e.g., perfectionism fits under the broad umbrella of personality), an d the outcome variable is measured at a later time to infer a more linear relationship. Conclusions The results of this study have shown continued support for the effect of empathy on the success of crisis interventionists in training. T his research has al so demonstrated that different understandings of empathy, along with the inclusion of additional empathy related variables, will have a significant effect on the outcome of training. It seems that previous researchers may have been struggling to find signi ficant conditional effects that help predict the outcome of training because other variables are moderating those relationships. While this research has been able to explore some of these variables, there are many more that need to be investigated and addr essed.
85 The literature on emotion regulation and perfectionism show that these constructs map on well to the process of empathic interactions that are evaluated in training programs. When empathic tendencies are measured, we see that there is a minor degree of predictive power in the outcome of training. However, when emotion regulation in particular moderates that relationship, we get a better picture of how intrinsic empathy predicts outcome depending on what the trainee actually does with his or her empat hy both internally and externally. As researchers, clinicians, trainers, and supervisors we are building a fuller understanding of the training process, including personality factors of the trainee, but there is still a lot to learn and explore about addit ional variables that influence the number and efficacy of trained interventionists to help those in crisis.
86 APPENDIX A DEMOGRAPHIC QUESTION NAIRE 1. Please circle the number next to your gender: (1) MALE (2) FEMALE 2. How old are you? Years 3. Please circle the n umber that best describes your Hispanic origin: (1) Not of Hispanic, Latino/a, or Spanish origin (2) Mexican, Mexican American, Chicano (3) Puerto Rican (4) Cuban (5) Other Hispanic, Latino/a, or Spanish origin 4. Please circle the number that best describes your race, or please describe the specific group that you identify with the most: (01) White (02) Black, African American (03) American Indian or Alaska Native (04) Native Hawaiian (05) Asian Indian (06) Japanese (07) Chinese (08) Korean (09) Vietnamese (10) Filipino (11 ) Guamanian or Chamorro (12) Samoan (13) Multicultural Mixed Race (14) Not listed, please specify 5. What is the highest degree you have completed? (1) High school diploma or GED (2) Associates degree (3) Some college (4) Bachelors degree (5) Masters de gree (6) Doctorate degree 6. Are you currently enrolled as a student? (1) YES (2) NO 7. Are you currently employed? (1) YES (2) NO 8. If you are a student, please indicate your Grade Point Average: GPA (4.0 scale)
87 APPENDI X B MINI INTERNATIONAL P ERSONALITY ITEM POOL scale below to describe how accurately each statement describes you. Describe yourself as you generally are now, no t as you wish to be in the future. Describe yourself as you honestly see yourself, in relation to other people you know of the same sex as you are, and roughly your same age. So that you can describe yourself in an honest manner, your responses will be kep t in absolute confidence. Please read each statement carefully, and then circle the number that indicates how accurately each statement describes you. Response Options VERY INACCURATE 1 MODERATELY INACCURATE 2 NEITHER INACCURATE NOR ACCURATE 3 MODERATEL Y ACCURATE 4 VERY ACCURATE 5 1. 1 2 3 4 5 2. 1 2 3 4 5 3. 1 2 3 4 5 4. Have fr 1 2 3 4 5 5. 1 2 3 4 5 6. 1 2 3 4 5 7. 1 2 3 4 5 8. 1 2 3 4 5 9. 1 2 3 4 5 10. 1 2 3 4 5 11. Ta 1 2 3 4 5 12. 1 2 3 4 5 13. 1 2 3 4 5 14. 1 2 3 4 5 15. 1 2 3 4 5 16. 1 2 3 4 5 17. 1 2 3 4 5 18. Make a mess 1 2 3 4 5 19. 1 2 3 4 5 20. 1 2 3 4 5
88 APPENDIX C INTERPERSONAL REACTI VITY INDEX The following statements inquire about your thoughts and feelings in a variety of situations. For each item, indicate how well it describes you by choosing the appropriate number on a scale have decided on your answer, fill in the letter on the answer sheet next to the item number. READ EACH ITEM CAREFULLY BEFORE RESPONDING. Answer as honestly as you can. DOES NOT DESCRIBE ME WELL AT ALL 1 DOES NOT DESCRIBE ME WELL FOR THE MOST PART 2 NEUTRAL OR MIDDLE GROUND DESCRIPTION OF ME 3 DESCRIBES ME WELL FOR THE MOST PART 4 DESCRIBES ME VERY WELL 5 _____ 1. I daydream and fantasize, with some regularity, about things that might happen to me. _____ 2. I often have tender, concer ned feelings for people less fortunate than me. _____ 3. I sometimes find it difficult to see things from the "other guy's" point of view. _____ 4. Sometimes I don't feel very sorry for other people when they are having problems. _____ 5. I really get involved with the feelings of the characters in a novel. _____ 6. In emergency situations, I feel apprehensive and ill at ease. _____ 7. I am usually objective when I watch a movie or play, and I don't often get completely caught up in it. _____ 8. I try to look at everybody's side of a disagreement before I make a decision. _____ 9. When I see someone being taken advantage of, I feel kind of protective towards them. _____ 10. I sometimes feel helpless when I am in the middle of a very emo tional situation. _____ 11. I sometimes try to understand my friends better by imagining how things look from their perspective. _____ 12. Becoming extremely involved in a good book or movie is somewhat rare for me. _____ 13. When I see someone get hurt I tend to remain calm. _____ 14. Other people's misfortunes do not usually disturb me a great deal. _____ 15. If I'm sure I'm right about something, I don't waste much time listening to other people's arguments. _____ 16. After seeing a play or movie, I have felt as though I were one of the characters. _____ 17. Being in a tense emotional situation scares me. _____ 18. When I see someone being treated unfairly, I sometimes don't feel very much pity for them _____ 19. I am usually pretty effective i n dealing with emergencies. _____ 20. I am often quite touched by things that I see happen. _____ 21. I believe that there are two sides to every question and try to look at them both. _____ 22. I would describe myself as a pretty soft hearted person. ____ 23. When I watch a good movie, I can very easily put myself in the place of a leading character. _____ 24. I tend to lose control during emergencies. _____ 25. When I'm upset at someone, I usually try to "put myself in his shoes" for a while.
89 _____ 26. When I am reading an interesting story or novel, I imagine how I would feel if the events in the story were happening to me. _____ 27. When I see someone who badly needs help in an emergency, I go to pieces. _____ 28. Before criticizing somebody, I try to imagine how I would feel if I were in their place.
90 APPENDIX D ALMOST PERFECT SCALE REVISED The following items are designed to measure certain attitudes people have toward themselves, their performance, and toward others. It is important that yo ur answers be true and accurate for you. In the space next to the statement, please select a number from "1" (strongly disagree) to "7" (strongly agree) to describe your degree of agreement with each item. STRONGLY DISAGREE 1 DISAGREE 2 SLIGHTLY DISAGREE 3 NEUTRAL 4 SLIGHTLY AGREE 5 AGREE 6 STRONGLY AGREE 7 _____ 1. I have high standards for my performance at work or at school. _____ 2. I am an orderly person. _____ 3. _____ 4. Neatness is impo rtant to me. _____ 5. _____ 6. My best just never seems to be good enough for me. _____ 7. I think things should be put away in their place. _____ 8. I have high expectations for myself. ____ 9. I rarely live up to my high standards. _____ 10. I like to always be organized and disciplined. _____ 11. Doing my best never seems to be enough. _____ 12. I set very high standards for myself. _____ 13. I am never satisfied with my accomplishments. ____ 14. I expect the best from myself. _____ 15. I often worry about not measuring up to my own expectations. _____ 16. My performance rarely measures up to my standards. _____ 17. I am not satisfied even when I know I have done my best. _____ 18. I am se ldom able to meet my own high standards for performance. _____ 19. I try to do my best at everything I do. _____ 20. I am hardly ever satisfied with my performance. _____ 21. _____ 22. I have a strong need to strive for excellence. _____ 23. I often feel disappointment after completing a task because I know I could have done better. _____ 24. Using the scale above, please rate the degree to which you agree that you are perfectionistic.
91 APPENDIX E EMOTI ON REGULATION QUESTI ONNAIRE We would like to ask you some questions about your emotional life, in particular, how you control (that is, regulate and manage) your emotions. The questions below involve two distinct aspects of your emotional life. One is you r emotional experience or what you feel like inside. The other is your emotional expression or how you show your emotions in the way you talk, gesture, or behave. Although some of the following questions may seem similar to one another, they differ in im portant ways. For each item, please answer using the following scale: STRONGLY DISAGREE 1 DISAGREE 2 SLIGHTLY DISAGREE 3 NEUTRAL 4 SLIGHTLY AGREE 5 AGREE 6 STRONGLY AGREE 7 _____ 1. When I want to feel more positive emotion (such as joy or amusement), I change _____ 2. I keep my emotions to myself. _____ 3. When I want to feel less negative emotion (such as sadness or anger), I change what _____ 4. When I am feeling positive emotions, I am care ful not to express them. _____ 5. think about it in a way that helps me stay calm. _____ 6. I control my emotions by not expressing them _____ 7. When I want to feel more positive emotion, I chang about the situation. _____ 8. I control my emotions by changing the way I think _____ 9. When I am feeling negative emotions, I make sure not to express them. _____ 10. When I want to feel less negative emotion, I about the situation.
92 APPENDIX F ONLINE INFORMED CONSENT Dear Trainee: The purpose of this study is to measure some personal characteristics and how they relate to crisis center training outcomes. Participation in this study involves two sets of questionnaires that take approximately 15 minutes and 5 minutes to complete. The second set of questionnaires will be made available once your training program has finished. You do not have to answer any question you do not wish to answer. No compensation is offered for involvement in this study. There are no known risks involved in completing the study and many people may find that they learn something about themselves from participating in this research. Nonetheless, if being part of the study makes you feel uncomfortable, you may consider speaking to a counselor who may be able to help you with your reactions. You can speak with the training director of your crisis center for counseling options. You may benefit by parti cipating in this study through increased awareness and self understanding. You will also be contributing to knowledge regarding confidential to the extent provided by law, and your decision to participate will not be shared with the crisis center training director. Your responses on the questionnaires will be assigned a code number and your name will not be used in any report. You can only participate if you are 18 years of age, or older. Your participation in this study is completely voluntary. There is no penalty for not participating and you have the right to withdraw from the study at anytime without consequence. If you have any questions about this research, y ou may contact Joey Hannah at firstname.lastname@example.org or (352) 215 1892 or Dr. Kenneth Rice at email@example.com or (352) 273 2119 Any questions or concerns about your rights as a research participant may be directed to the UFIRB office, University of Florida, Box 1122 50, Gainesville, FL, 32611; or by phone at (352) 392 0433. Thank you for your time. I sincerely appreciate your involvement in this research. By typing your name below, you agree that you have read the procedure described above and voluntarily agree to p articipate in the study. You may print this page for your own records if you wish
93 REFERENCE LIST Aiken, J., Brownell, A., & Iscoe, I. (1974). The training and utilization of paraprofessionals in a college psychological service center. Journal of College Student Personnel 15 (6), 480 486. Alterman, A. I., McDermott, P. A., Cacciola, J. S., & Rutherford, M. J. (2003). Latent structure of the Davis Interpersonal Reactivity Index in methadone maintenance patients. Journal of Psychopathology and Behavioral As sessment, 25 (4), 257 265. Amer ican Association of Suicidology (2010). Organization accreditation standards manual (9th ed.). Washington, DC. Avery A. (1987). Communication skills training for paraprofessional helpers. American Journal of Community Psycho logy, 6 (6), 583 591. Baldwin, B. (1979). Crisis intervention: An overview of theory and practice. The Counseling Psychologist 8 (2), 43 52. Baldwin, B. A. (1980). Style of crisis intervention: Toward a convergent model. Professional Psychology, 11 113 120. Barz, M. (2001). Assessing suicide hotline volunteers' empathy and motivations. Available from ProQuest Dissertations and Theses database. Bobevski, I., & McLennan, J. (1998). The telephone counseling interview as a complex, dynamic, decision proces s: A self regulation model of counselor effectiveness. Journal of Psychology: Interdisciplinary and Applied 132 (1), 47 60. Butcher, J. N., Dahlstrom, W. G., Graham, J. R., Tellegen, A., & Kaemmer, B. (1989). MMPI 2: Manual for administration and scoring Minneapolis MN : University of Minnesota Press. Carkhuff, R. (1968). Differential functioning of lay and professional helpers. Journal of Counseling Psychology 15 (2), 117 126. Carkhuff, R. (1969a). Critical variables in effective counselor training. J ournal of Counseling Psychology 16 (3), 238 245. Carkhuff, R. (1969b). Helper communication as a function of helpee affect and content. Journal of Counseling Psychology 16 (2), 126 131. Carkhuff, R., & Truax, C. (1965). Training in counseling and psychot herapy: An evaluation of an integrated didactic and experiential approach. Journal of Consulting Psychology 29 (4), 333 336.
94 Cattell, R., Eber, H., & Delhees, K. (1968). A large sample cross validation of the personality trait structure of the 16 P.F. wi th some clinical implications. Multivariate Behavioral Research [Special Issue], 107 131. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates. Cohen, J. (1992). A power primer. Psychological Bulletin, 112 155 159. Cowen, E., Gardner E., & Zax, M. (1967). Emergent approaches to mental health problems. New York NY : Appelton Century Crofts. Danish, S., D'Augelli, A., & Brock, G. (1976). An evaluation of helping skills training: Effects on helpers' verbal responses. Journal of Counseling Psychology 23 (3), 259 266. helping behavior: Description and implications for training. American Jo urnal of Community Psychology, 4 275 282. D'Augelli, A., & Levy, M. (1978). The verbal helping skills of trained and untrained human service paraprofessionals. American Journal of Community Psychology 6 (1), 23 31. Davies, P. J. K. (1982). The functioni ng of British counselling hotlines: A pilot study. British Journal of Guidance and Counselling, 10 195 199. Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85. Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44, 113 126. Demidenko, E. (2008). Sample size and optimal design for logistic regression with binar y interaction. Statistics in Medicine, 27 36 46. Donnellan, M., Oswald, F., Baird, B., & Lucas, R. (2006). The Mini IPIP Scales: Tiny yet effective measures of the Big Five Factors of Personality. Psychological Assessment 18 (2), 192 203. Echterling, L. & Hartsough, D. (1989). Phases of helping in successful crisis telephone calls. Journal of Community Psychology 17 (3), 249 257. Eriksen, K., & McAuliffe, G. (2003). A measure of counselor competency. Counselor Education and Supervision 43 (2), 120 133. Evans, D. (1977). Use of the MMPI to predict effective hotline workers. Journal of Clinical Psychology 33 (4), 1113 1114.
95 Feehan, P. F., & Wade, S. L. (1998). The paraprofessional alternative. Journal of Career Development, 25 149 157. Frazier, P., Ti x, A., & Barron, K. (2004). Testing Moderator and Mediator Effects in Counseling Psychology Research. Journal of Counseling Psychology 51 (1), 115 134. Genthner, R. (1974). Evaluating the functioning of community based hotlines. Professional Psychology 5 (4), 409 414. Getz, W. L., Fujita, B. N., & Allen, D. (1975). The use of paraprofessionals in crisis intervention: Evaluation of an innovative program. American Journal of Community Psychology, 3 135 144. Goldberg, L. R. (1993). The structure of phenot ypic personality traits. American Psychologist, 48 26 34. Gottesfeld, H., Rhee, C., & Parker, G. (1970). A study of the role of paraprofessionals in community mental health. Community Mental Health Journal 6 (4), 285 291. Granello, D. H., & Granello, P. F. (2007). Suicide: An essential guide for helping professionals and educators (1 st ed.). Boston MA : Allyn & Bacon. Gross, J., & John, O. (2003). Individual differences in two emotion regulation processes: Implications for affect, relationships, and well being. Journal of Personality and Social Psychology 85 (2), 348 362. Guerney, B. (1969). Psychotherapeutic agents: New roles for nonprofessionals, parents, and teachers. New York NY : Holt, Rinehart, & Winston. Harkness, A., & Lilienfeld, S. (1997) Individual differences science for treatment planning: Personality traits. Psychological Assessment 9 (4), 349 360. Hart, L., & King, G. (1979). Selection versus training in the development of paraprofessionals. Journal of Counseling Psychology 26 (3), 235 241. Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behavior Research Methods, 41 (3), 924 936. Heilig, S. M. (1967). Manpower: Utilization of non p rofessional crisis workers. In McGee, R. K. (Ed.) Planning Emergency Services for Comprehensive Community Mental Health Centers Gainesville FL : University of Florida.
96 Hill, C., & Lent, R. (2006). A narrative and meta analytic review of helping skills t raining: Time to revive a dormant area of inquiry. Psychotherapy: Theory, Research, Practice, Training 43 (2), 154 172. Hill, C. E., Roffman, M., Stahl, J., Friedman, S., Hummel, A., & Wallace, C. (2008). Helping skills training for undergraduates: Outc omes and prediction of outcomes. Journal of Counseling Psychology, 55 359 370. Hill, C. E., Sullivan, C., Knox, S., & Schlosser, L. (2007). Becoming psychotherapists: Experiences of novice trainees in a beginning graduate class. Psychotherapy: Theory, Research, Practice, and Training, 44, 434 449 Hirsch, P., & Stone, G. (1982). Attitudes and behavior in counseling skill development. Journal of Counseling Psychology 29 (5), 516 522. Hoyle, R., & Robinson, J. (2004). Mediated and Moderated Effects in Social Psychological Research: Measurement, Design, and Analysis Issues. The Sage handbook of methods in social psychology (pp. 213 233). Thousand Oaks, CA US: Sage Publications, Inc. Ickes, W. (1993). Empathic accuracy. Journal of Personality 61 (4), 587 610. John, O., & Gross, J. (2007). Individual differences in emotion regulation. Handbook of Emotion Regulation (pp. 351 372). New York, NY : Guilford Press. Kalafat, J. (2002). Training telephone counselors. Crisis intervention and counseling by telepho ne (2nd ed.) ( pp. 260 286). Springfield, IL : Charles C. Thomas Publisher. Knickerbocker, D. (1973). Lay volunteer and professional trainee therapeutic functioning and outcomes in a suicide and crisis intervention service. Available from ProQuest Disserta tions and Theses database. Lamb, D., & LaFave, F. (1977). The use of paraprofessionals in university mental health settings: A national survey. American Journal of Community Psychology 5 (3), 321 326. Lester, D. (1971). The Myers Briggs Type Indicator a s a predictor of counselor effectiveness. Crisis Intervention 3 (2), 44 46. Leventhal, A., Berman, A., McCarthy, B., & Wasserman, C. (1976). Peer counseling on the university campus. Journal of College Student Personnel 17 (6), 504 509. Levine, A., & Cur eton, S. (1998). student. San Francisco CA : Jossey Bass. Maierle, J. (1973). The politics of supporting paraprofessionals. Professional Psychology 4 (3), 313 320.
97 Margolis, C. G., Edwards, D. W ., Shrier, L. P., & Cramer, M. (1975). Brief hotline training: An effort to examine impact on volunteers. American Journal of Community Psychology, 3 59 67. McCarthy, B. W., & Berman, A. L. (1971). A student operated crisis center. The Personnel and Gui dance Journal, 49 523 528. McClure, J. N., Wetzel, R. D., Flanagan, T. A., McCabe, M., & Murphy, G. E. (1973). Volunteers in a suicide prevention service. Journal of Community Psychology, 1 397 398. McCord, J., & Packwood, W. (1973). Crisis centers and hotlines: A survey. Personnel & Guidance Journal 51 (10), 723 728. McGee, R. (1974). Crisis intervention in the community Oxford England: University Park. McLennan, J., Culkin, K., & Courtney, P. (1994). Telephone counsellors' conceptualising abilitie s and counselling skills. British Journal of Guidance & Counselling 22 (2), 183 195. Mishara, B., Chagnon, F., Daigle, M., Balan, B., Raymond, S., Marcoux, I., Bardon, C., Campbell, J. K., & Berman, A. (2007a). Comparing models of helper behavior to actua l practice in telephone crisis intervention: A silent monitoring study of calls to the U.S. 1 800 SUICIDE network. Suicide and Life Threatening Behavior 37 (3), 291 307. Mishara, B. L., Chagnon, F., Daigle, M., Balan, B., Raymond, S., Marcoux, I., Bardon C., Campbell, J. K., & Berman, A. (2007b). Which helper behaviors and intervention styles are related to better short term outcomes in telephone crisis intervention? Results from a silent monitoring study of calls to the U.S. 1 800 SUICIDE network. Suici de and Life Threatening Behavior 37 308 321. Mishara, B. L., & Daigle, M. S. (1997). Effects of different telephone intervention styles with suicidal callers at two suicide prevention centers: An empirical investigation. American Journal of Community P sychology, 25 861 885. Myers, I. B. (1962). The Myers Briggs Type Indicator m anual Princeton, NJ: Educational Testing Service. Neimeyer, R., & Pfeiffer, A. (1994). Evaluation of suicide intervention effectiveness. Death Studies 18 (2), 131 166. Ottens A. (1984). Evaluation of a crisis training program in suicide prevention for the campus community. Crisis Intervention 13 (1), 25 40.
98 Paterson, H., Reniers, R., & Vllm, B. (2009). Personality types and mental health experiences of those who volunteer f or helplines. British Journal of Guidance & Counselling 37 (4), 459 471. Peduzzi, P., Concato, J., Kemper, E., Holford, T. R., & Feinstein, A. R. (1996). A simulation study of the number of events per variable in logistic regression analysis. Journal of C linical Epidemiology, 49 1373 1379. Pickett, L. A. (2006). Trait versus state empathy: Evaluating the stability of empathy and the relationship between negative affect and empathy in an incarcerated sex offender population Available from ProQuest Disse rtations and Theses database. Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers 36 717 731. Rice, K. G., & Ashby, J. S. (2007) An efficient method for classifying perfectionists. Journal of Counseling Psychology, 54 72 85. Rogers, C. R. (1942). Counseling and psychotherapy. Boston MA : Houghton Mifflin. Rogers, C. R. (1947). Some observations on the organization of personali ty. The American Psychologist, 2 358 368. Rogers, C. R. (1951). Client centered therapy: Its current practice, implications, and theory. Boston: Houghton Mifflin. Rogers, C. R. (1957). The necessary and sufficient conditions of therapeutic personality c hange. Journal of Consulting Psychology 21, 95 103. Rosenbaum, A., & Calhoun, J. (1977). The use of the telephone hotline in crisis intervention: A review. Journal of Community Psychology 5 (4), 325 339. Sawatzky, D., & Paterson, J. (1982). The use of paraprofessionals in the helping profession: A look at trends in North America. International Journal for the Advancement of Counselling 5 (1), 27 34. Schinke, S., Smith, T., Myers, R., & Altman, D. (1979). Crisis intervention training with paraprofessio nals. Journal of Community Psychology 7 (4), 343 347. Schoenfeld, L. S., Preston, J., & Adams, R. L. (1976). Selection of volunteers for telephone crisis intervention centers. Psychological Reports, 39 725 726. Selfridge, F. F., Weitz, L. J., Abramowitz S. I., Calabria, F. M., Abramowitz, C. V., & Steger, J. A. (1975). Sensitivity oriented versus didactically oriented in service counselor training. Journal of Counseling Psychology 22 (2), 156 159.
99 Seligman, M. E. (2007). What y ou c an and w hat y ou c an't: The c omplete guide to s uccessful s elf i mprovement New York NY : Vintage. Slaney, R. B., Ashby, J. S., & Trippi, J. T. (1995). Perfectionism: Its measurement and career relevance. Journal of Career Assessment, 3 279 297. Slaney, R. B., Rice K. G., Mobley, M., Trippi, J., & Ashby, J. S. (2001). The revised almost perfect scale. Measurement and Evaluation in Counseling and Development, 34 130 145. Stahl, J., & Hill, C. (2008). A comparison of four methods for assessing natural helping abili ty. Journal of Community Psychology 36 (3), 289 298. Stein, D. M., & Lambert M. J. (1984). Telephone counseling and crisis intervention: A review. American Journal of Community Psychology, 12 101 126. Stoeber, J., & Otto, K. (2006). Positive conceptio ns of perfectionism: Approaches, evidence, challenges. Personality and Social Psychology Review, 10 (4), 295 319. Tabachnick, B. G. & Fidell, L. S. (2006). Using multivariate statistics (5 th ed.). Boston, MA: Allyn & Bacon. Tapp, J., & Spanier, D. (1973 ). Personal characteristics of volunteer phone counselors. Journal of Consulting and Clinical Psychology 41 (2), 245 250. Taylor, R., Brady, M., & Swank, P. (1991). Crisis intervention: Longer term training effects. Psychological Reports 68 (2), 513 514. Teevan, K., & Gabel, H. (1978). Evaluation of modeling role playing and lecture discussion training techniques for college student mental health professionals. Journal of Counseling Psychology 25 (2), 169 171. Tishler, C. L., Bartholomae, S., & Rhodes, A R. (2005). Personality profiles of normal healthy volunteers: A potential concern for clinical drug trial investigators? Medical Hypotheses, 65 1 7. Wehr, M., & Wittimer, J. (1973). Paraprofessional trainees and counselor education students: A compar ison of personality and predicted counseling effectiveness. Counselor Education and Supervision 12 (4), 255 262.
100 BIOGRAPHICAL SKETCH David Joseph Hannah was born in Washington D.C. in 1983. He graduated cum laude from the University of Florida in 2005 wit h a Bachelor of Science in psychology and a minor in general education. David was admitted to the University of Florida Counseling Psyc hology doctoral program in 2006 where he received his Master of Science in 2008 and his Ph.D. in 2011. His current academ ic pursuits include the study o f crisis intervention training suicide prevention perfectionism, defensiveness and therapeutic interventions. His clinical work is focused on providing counseling services for college students, crisis intervention/suicide prevention in the local community and outreach programming After receiving his degree he began a staff psychologist position at the New Mexico State University Counseling Center in Las Cruces, New Mexico.