MAT deficiency

This research addresses the extent of student impairment in Commission on Accreditation for Marriage and Family Therapy Education (COAMFTE) accredited marriage and family therapy programs, indicators of impairment used by program directors, faculty time devoted to impaired students, and the frequency of student dismissal. The data come from a survey of 44 COAMFTE program directors that was conducted during the spring of 2001, as well as responses to an open-ended item asking for a description of "the most troublesome student" to come to the attention of the respondent. The authors discuss the findings with respect to similar studies in counseling psychology and the larger literature on clinical training. Finally, the authors offer suggestions for future research.

Marriage and family therapy (MFT) training programs are in the business of preparing therapists who are self-aware, conceptually sound, ethically sensitive, and effective in the work that they do with clients. Faculty in Commission on Accreditation for Marriage and Family Therapy Education (COAMFTE) accredited programs are gatekeepers to the profession, assuring that those who graduate from their programs have the knowledge, skills, and personal characteristics to practice competently. Therapist competence, in general terms, is a therapist's ability to bring about desired change with clients (Herman, 1993; Shaw & Dobson, 1988). In his assessment of the predictors of therapist competence, Herman (1993) summarized the relevant research in counseling psychology and identified three general areas of therapist competence. These include knowledge, training, and experience; theoretical orientation; and therapist personal characteristics (Herman, 1993). A deficiency or ineptitude in any of these areas could be considered to be "impairment."

A significant component of knowledge, training, and experience would include the ability to recognize when circumstances pose an ethical dilemma and to generate responses that arc informed by professional ethics. Training sensitizes students to the importance of the principles of autonomy, beneficence, and non-maleficence, as well as justice and fidelity (Kitchener, 1984) and teaches them to apply these principles with the consultation and support of supervisors. Students who are unable to apply these principles appropriately would be considered to be deficient.

Herman's (1993) second category of therapist competence, theoretical orientation, includes the ability to translate theory into practice. Students who do well in didactic classes but are unable to apply the material in clinical situations would show a deficiency in theoretical orientation.

Herman's (1993) final area of competency is receiving increasing attention in the MFT literature and is sometimes referred to as "the person of the therapist" (Aponte, 1994, p. 5). This component of training is of special concern because of its impact on therapy outcome. Asay and Lambert (1999) suggest that of the four common factors that contribute to variance in therapy outcome, the therapeutic relationship accounts for 30%, a contribution more potent than anything else that happens within the session, including technique. Students who graduate from accredited programs must be prepared to engage clients from backgrounds similar to and different from their own in a meaningful way. The emotional well-being of therapists-in-training is also a legitimate concern of program faculty. Beutler, Crago, and Arizmendi, in their 1986 review of studies related to therapist emotional well-being, concluded that therapist emotional well-being facilitates both the process of treatment and the outcome. Eight years later, Beutler, Machado, and Neufeldt (1994) reviewed an additional set of studies related to therapist emotional well-being and client outcome and concluded that "high therapist distress or disturbance levels may not only prevent client growth, but actually induce negative changes" (p. 238). The authors go on to state "the possibility that emotional problems on the part of the therapist may negatively affect even relatively well functioning clients should be given considerably more attention" (p. 238).

Although rates of student impairment in clinical training programs are generally reported to be below 5% (Bernard, 1975; Biaggio, Gasparikova-Krasnec, & Bauer, 1983; Boxley, Drew, & Rangel, 1986; Burgess, 1994; Forrest, Elman, Gizara, & Vacha-Haase, 1999; Keppers, 1960; Olkin & Gaughen, 1991; Sweeney, 1969; Tedesco, 1982), the potential risk to clients, the drain on faculty and program resources, and the stress experienced by trainees are all significant and legitimate concerns. Only one of the studies conducted in clinical training programs cited above included MFT programs in their sample (Olkin & Gaughen, 1991), and even then only included a small number of MFT programs and did not analyze data for MFT programs separately. The research presented here rectifies that problem by surveying all programs listed as accredited on the COAMFTE web site as of spring of 2001. The data was collected in the spring of 2001.

RESEARCH QUESTIONS

The following questions guided the research reported here:

1. What is the extent of student impairment in COAMFTE accredited MFT programs?

2. What indicators do program directors think should be used in assessing student impairment?

3. How much time do faculty and program directors invest in addressing "problem students"?

4. What strategies do faculty use to respond to student impairment?

5. How do program directors perceive the institutional support they receive in addressing student impairment?

6. How often are students dismissed and what are the outcomes of dismissal decisions?

7. How do program directors describe their "most troublesome problem student"?

METHOD

Our 5-page questionnaire, a modification of those used by Olkin and Gaughen (1991) and Burgess (1994), was sent to directors of all 80 COAMFTE accredited programs in the spring of 2001. The questionnaire included both closed- and open-ended questions. No code numbers were used to link returned questionnaires with a master list, and respondents were assured that there was no way to link their responses to their program. Respondents were asked to send copies of their grievance and dismissal policies (to be analyzed at a later date). These were removed from the return envelope and separated from the questionnaire so as to protect the anonymity of questionnaire responses. Requests for a summary of the research were separated from questionnaires in a similar manner. Responses to open-ended questions were collapsed into categories and paraphrased to further protect confidentiality. In addition, respondents were told that they could remove their institution's name from any printed material they shared with us. Two follow-up mailings to the initial mailing were conducted using Dillman's (2000) "tailored design method," including a reminder postcard and a second packet of materials. The use of this strategy yielded a return rate of 55% (N = 44).

RESULTS

Description of Programs

Half of the questionnaires returned (n = 22) were from accredited master's programs, with another 20% from programs that offered both the master's degree and a PhD or a master's degree and a postdegree certificate (n = 9). Eleven percent of the questionnaires were from doctoral programs (n = 5) and 14% were from postdegree programs (n = 6). Two respondents failed to indicate level of program. More than two-thirds of the programs interview prospective students before entry into their program and require the Graduate Record Exam (GRE) or the Miller Analogy Test (MAT). Our questionnaire did not distinguish between face-to-face and telephone interviews. Eighty percent (n = 35) reported requiring a personal biography of applicants. Only two programs reported requiring objective personality measures. Eighty-six percent of the programs (n = 38) do a global evaluation of each student at least annually. Three of the program directors reported requiring personal psychotherapy for all their students, although others noted that they "strongly encourage" it during supervision.

Frequency of Impairment in Last 5 Years

Using Burgess's (1994) definition of impairment as "an objective and negative change in a student's functioning that is pervasive and affects more than one area of the student's functioning, goes beyond problems typically expected in professional training, is emotional, physical or academic and is impervious to feedback" (p. 10), program directors were asked how many impaired students their program had dealt with in the last 5 years. A total of 108 impaired students were identified by our respondents. The range was 0-8 per program. The mean was 2.5, and the modal response was 2 impaired students in the last five years.

Indicators of Impairment

Using the survey of clinical programs by Olkin and Gaughin (1991) as a guide, we listed 16 indicators of student impairment and asked respondents to rank their top five concerns. The five concerns checked most often were, in order of perceived importance: ethical violation, unprofessional conduct, suicide attempts, substance use/abuse, and possible signs of a personality disorder. Table 1 lists the 16 concerns, together with the number of respondents who ranked each item among their top five concerns. One respondent failed to answer the question. Thus, the total number of respondents for this item was 43. Two of these 43 respondents "checked" rather than ranked their top five concerns. In these two cases, the checked responses were each given a rank of three for coding purposes.

Remediation Methods

The most popular remediation methods (those being used by one-half or more of the programs) included referral to therapy, increased supervision, leave of absence, increased contact with faculty advisor, and repeating academic coursework. Other choices included on the questionnaire but checked less often included tutoring, special seminars or extra coursework, peer support groups, special assignments, and referral to an ombudsperson. Our respondents generated several more strategies, including: (a) faculty meeting with student to discuss problems and concerns; (b) slowing or postponing practicum work; (c) letter of remediation cosigned by faculty and student; (d) "shadowing" a peer mentor; (e) cotherapy; and (f) a "counseling out" process that includes a written recommendation plan that could include extra coursework, independent testing, therapy, and more live or video supervision.

Outcome for Students Identified as Impaired in Last 5 Years

Directors were asked to report how many of the students identified as impaired took a leave of absence, dropped out, were dismissed from the program, or transferred to another program. Four participants (who reported eight cases of impairment) failed to provide information on outcome. Thus, outcome is reported for 100 of the original 108 cases. Two outcomes were reported for one of the 100 cases, making the total number of outcomes reported 101. The majority of the 100 students either dropped out (n = 39) or took a leave of absence (n = 35). Seventeen were reported to have been dismissed from the program. Ten students reportedly transferred to another program.

Student-Initiated Lawsuits

Eighty-six percent (n = 38) of the program directors reported that there had been no student-initiated lawsuits in the last 5 years. Three programs reported one suit in the last 5 years, two programs reported two suits in the last 5 years and one program had six suits in the last 5 years. Of the 13 reported lawsuits, three involved alleged ethical or professional conduct violation by the student, three contested terminations, two contested failure of the internship or practicum portion of training, two alleged discriminations against a student or student applicant, one contested grades, and one contested denial of admission. One suit, the grounds for which were not described, was dropped. Although we do not know the size of the program that reported six lawsuits, we do know that it was a program offering more than one level of credentialing. We did not ask program directors to report the outcome of lawsuits.

Description of Most Troublesome Student

In an open-end question, we asked participants to describe the "most troublesome" student to come to their attention, including how the problem came to their attention and the outcome. Descriptions of problems included reference to students' inability to accept supervisory feedback or to consider their own part in issues, unprofessional and inappropriate behavior (including moralistic comments to clients, racist and sexist comments to colleagues, poor boundaries with clients, missing appointments), dishonesty, ethical concerns (including not following confidentiality guidelines and failure to complete case notes), declining academic performance, poor clinical skills, lack of assertiveness and inability to retain clients, self-identified personal problems and depression, eating disorders, alcohol abuse, and apparent thought disorders (including suicidal ideation, impaired judgment, and impaired impulse control). One student reportedly committed suicide.

The program directors described four general ways in which the problems listed above came to their attention. These included observation by on-campus faculty, feedback from off-campus supervisors, classroom performance, and concern expressed by fellow students.

Some of the outcomes described were positive in that students were able to accept feedback, take responsibility for their issues and address areas of concern so that they could stay in the program and successfully graduate. Some students took leaves of absence to address personal issues. Another common outcome was a process described as "counseling out" of the program. This consisted of faculty presenting concerns to the student as they emerged over time and eventually discussing the possibility that a clinical profession was not a good fit for the student. In the "counseling out" process described, students voluntarily chose to leave the program, sometimes transferring to a nonclinical program. Others dropped out when it was clear they were about to be dismissed. However, some situations eventuated in formal dismissal. For a few of these, the action of dismissal resulted in student-initiated lawsuits or the threat of a lawsuit.

Impact on Program Faculty

We were interested in knowing how much faculty time was devoted to "problem students" each month. More than one-half (56%) reported this to be less than one-half hour per month. But comments such as the following, written on the questionnaire, imply great variation. One respondent who checked "less than 1/2 hour" noted, "As there's usually no problem-when there is a problem, then it's 3 to 5 hours per month." Another who did not give a quantitative response said, "Depends on what is going on. Triangles little as 1/2 hour, but at times as much as five." We asked program directors how often problems get taken beyond the program or department, how much support they get from upper administration, and how much personal stress was involved in dealing with "problem students" for themselves and for their faculty. Thirty-nine percent reported that issues are never taken beyond the program or department, whereas 50% reported that issues are taken outside the program or department one-fourth of the time. Only four program directors reported issues being taken beyond the program or department one-half of the time or more.

We asked program directors how much support they feel from the university or institute in "identifying and responding to student impairment." Forty percent reported perceiving support from the larger unit all of the time, and another 28% reported support three-quarters of the time, and 32% perceived support one-half of the time or less. With regard to the personal stress involved in dealing with "problem students," on a scale from 1 to 5, the mean response for program directors was 3 and the modal response was 2. Program directors rated the personal stress involved in dealing with "problem students" for their faculty very similarly (M = 3, mode = 2, range = 1-5).

DISCUSSION

Clearly, faculty in COAMFTE accredited programs take their gatekeeping function seriously. Students are routinely evaluated in most programs and given feedback about their performance. When impairment is identified, students are informed of faculty concerns, and these are addressed through a variety of interventions, including remediation and dismissal. A significant number of students are, in fact, dismissed from their training programs, When student impairment becomes an issue, faculty time devoted to the problem can be considerable. Although not a frequent outcome, programs sometimes face legal action on the part of a student who feels unfairly evaluated. Nearly one-third of our participants reported perceiving a lack of support from the larger institution during such stressful events. Yet our data show that programs persist in their efforts to guard the public interest.

The research on student impairment in clinical programs is not informed by a single definition of impairment (Forrest et al., 1999). We used the definition offered by Burgess (1994) so that we could compare student impairment in MFT programs with the impairment that Burgess reports for doctoral programs in counseling psychology. However, we also wanted to obtain rich descriptions of a range of training problems. Therefore, we used a less restrictive concept in the open-ended portion of our questionnaire. We simply asked program directors to describe the most "difficult problem student" they had encountered.

The student problems our respondents found most concerning were clinical, rather than academic in nature (see Table 1). These problems require subjective evaluation and likely draw on data across a variety of settings. Programs accredited by the COAMFTE are at least as good as their counseling psychology counterparts at identifying impaired students. For instance, Burgess (1994) reported 2.4 impaired students per program in a 5-year period in her research compared with the 2.5 impaired students per program in a 5year period reported in this study. The fact that our program directors reported slightly more impairment per program than the Burgess (1994) study may be related to program size. Marriage and family therapy programs typically are smaller than clinical and counseling psychology training programs. Being responsible for smaller programs, MFT faculty may be in a better position to identify student impairment. This hypothesis is supported by Burgess's comparison of "large" (75-378 students) and "small" (12-25) programs in her data set. The smaller programs reported a 4% rate of student impairment over the past five years compared with a 2% rate for the large programs. Using data from the COAMFTE (Kaveny, January 29, 2002), it is clear that the average MFT program is more similar in size to the smaller clinical and counseling psychology programs in the Burgess study. The average COAMFTE master's program has 29 students (range 4-86), whereas the average doctoral program has 24 students (range 10-58). The average postdegree program has 17.7 students (range 4-36).

It is possible that some of the most concerning problems identified by program directors, such as thought disorders, suicidal ideation, and substance abuse, evolved from issues that students brought into the training program. Ideally, these are problems that would be identified and addressed during the admission process, through personal biographies, interviews, or perhaps personality testing. Only two programs in our sample reported using personality testing as a part of the admissions process. Widespread use of personality testing for MFT programs is an issue that would challenge deep philosophical foundations of our field and might not be a comfortable option for many program faculty. However, clearly defining impairment in program materials may help both faculty and students identify impairment early and address it within a clear structure, whether or not the impairment evolved from an issue that was present at the time of admission.

The finding that some students drop out or transfer to another program when it becomes clear that they are not likely to succeed in their present COAMFTE program suggests that programs should directly inquire about previous training experiences and ask for a release to talk with any prior supervisors. This is consistent with the best practices recommendations of Storm, Todd, Sprenkle, and Morgan (2001).

Closely associated with therapist impairment is the issue of therapist self-care. For some students, the stresses inherent in clinical training co-occur with unusual personal Stressors, such as a marital or family crisis. It is important that programs teach about the importance of therapist self-care, that the connection between self-care and professional ethics be addressed, and that personal efforts to enhance self-care be supported in the training program. This includes supporting students (and faculty) in acknowledging self-impairment when it occurs and taking steps to address the problem, such as personal therapy, a reduced caseload, or time away from the program.

The program directors in our sample described students who were suspected to have thought disorders, personality disorders, eating disorders, substance abuse disorders, and one who actually committed suicide. Under what circumstances should programs deny entry to or dismiss students who have a history of mental disorder? What guidelines can we provide to MFT programs and supervisors who work with students who develop an emotional disorder during the course of supervision?

Although we did not ask this question of our respondents, we were struck by how many of the accounts in response to our open-ended question about "problem students" revolved around psychiatric symptoms. Four of the 37 cases described suspected thought disorders or depression severe enough to lead to suicidal ideation. None of the program directors reported use of a leave of absence for students coping with psychiatric issues, although leaves of absence were reported in cases where students were coping with other life Stressors, such as loss of a relationship or care-giving for family members. It is not clear if this is because the impairment was not thought to be responsive to treatment, rendering the student permanently inappropriate for clinical training, or if mention of leave of absence as a possible component of remediation was considered, but omitted from the response to our open-ended question.

Little is written about how to go about the gatekeeping function of supervision, especially when it involves mental health issues of supervisees. Brady and Post (1991) insist that educators must address the "emotional stability" (p. 108) of students when making retention decisions but give little guidance about how to do that. Wilcoxon (1992) describes such decisions as more "art than science" (p. 23) and cautions against reliance on "standardized procedures" (p. 23). However, Wilcoxon also reminds us of the potential for bias and cautions supervisors to seek consultation as they make decisions regarding retention or dismissal of students.

It is our belief that most clinicians will, at some time in their career, experience a period when stress is high, support is low, and demands are high enough that they begin to show signs of impairment. For some, this may occur during training. Following Kerr and Bowen's (1988) typology of symptoms, the signs of a developing impairment may be somatic, social, or emotional. Our belief is that openness on the part of a clinician regarding his or her impairment and communication regarding how it is being addressed is a sign that the impairment may be manageable within the context of a well-structured and supportive program environment. Such a training environment would have clearly written policies regarding student impairment, grievance and dismissal policies, and students would be aware of those policies' existence. A valuing of self-care would become a part of the culture of the program, and students and faculty would model taking responsibility for self-care in decisions they make about personal boundaries, including the size of their caseload, the proportion of crisis cases in their caseload, honoring time with family and friends, and willingness to negotiate vacation times so that clinical responsibilities are equitably and responsibly covered during therapist absence. Todd (1997) notes that the gatekeeping function of training and supervision require(s)

a painstaking, gradual process where there are periodic benchmarks and detailed discussions of progress and possible means for remediation. When handled with such care, it may be possible to have students reach the desired conclusion that they do not belong in the field or need to take extensive time off for personal therapy, (p. 251)

Our limited data would appear to support Todd's conclusion. The process that program directors described was often gradual and sometimes resulted in the student making the final decision about how to address impairment in a way that was respectful of the needs of all stakeholders: the student, clients, the program, and the profession.

LIMITATIONS AND SUGGESTIONS FOR FUTURE RESEARCH

The prevalence of student impairment can be expressed as a frequency (number of impaired students per program) or as a rate (number of impaired students in program divided by all students in the program). Both expressions can be found in the training literature. Our quantitative data are limited in that they include only a report of frequency of student impairment. We selected frequency, rather than rate, to make it easier for program directors to respond to the questionnaire. We were concerned that busy program directors would be put off by a questionnaire asking them to dig into program files. An additional reason for our decision to ask only for the number of impaired students in the last 5 years was that Burgess (1994) was able to obtain more complete data for her sample on frequency than rate. Thus, we were able to compare our frequency data with similar data from counseling psychology. Nevertheless, we wish we had asked respondents to report on the size of the typical entering class as a way of estimating program size. We also could have asked program directors to estimate the extent of impairment in a typical training year.

It is important to note that the objectives of training at the master's and doctoral levels are different. Master's programs prepare students to be clinicians, whereas doctoral programs prepare researchers and clinical supervisors. Accordingly, one might expect impairment to develop and be identified differently at these two levels of training. Our data are not adequate to explore that possibility, because they are heavily weighted with master's programs, and because level of program confounds data from dual degree-granting institutions. Program directors were not asked to report data separately for each degree offered. This is a limitation of our research that may be addressed in future research.

We need additional research to determine what remediation methods are most appropriate for which kinds of student impairment. For instance, when is personal therapy useful, and how might it be structured so that concerns of the program are addressed while still respecting the privacy of the remediated student? How effective is increased supervision in enhancing the clinical skills of students? Can some impairments, such as ethical lapses, be addressed successfully via coursework? We need guidance on how to balance the needs of impaired students with our obligations to the public and to the profession. Additionally, we need to know if the extent of impairment varies by type of program (master's, doctoral or post-degree) and if the type of impairment varies by program type.

Finally, we need research on the experience of remediation from the students' perspective. How do they perceive the feedback they get from faculty about their professional performance, and the way issues of concern are identified and addressed within the program? For those in remediation, how open are they with their peers about the process, and do they feel fairly treated? What impact does the identification of impairment and the remediation process have on the culture of the program? What role do students see for themselves in addressing the impairment of student colleagues, and do they have models for how to confront and support a colleague who is impaired? How often would students, and faculty, admit to being impaired at some point in their career? How do students and faculty handle personal distress that falls short of impairment but nevertheless affects their professional functioning? All of these are questions that deserve further study.

CONCLUSION

Burgess's (1994) definition of impairment is a rigorous one, in that it requires that negative change in student functioning be "pervasive" and that the student be "impervious to feedback." Using that definition, COAMFTE program directors reported an average of 2.5 impaired students over a 5-year period. This . suggests that every other year the typical program invests considerable energy in addressing serious issues of student impairment. The good news is that programs do identify and respond to instances of student impairment. Impaired students are not allowed to "pass through" the program. However, there is much that we do not know about how to address impairment in a fashion that effectively guides students identified as impaired while protecting the well-being of the program, the profession, and the public.

REFERENCES

Aponte, H. (1994). How personal can training get? Journal of Marital and Family Therapy, 20, 1-5.

Asay, T. P., & Lambert, M. J. (1999). The empirical case for the common factors in therapy: Quantitative findings. In M. A. Hubble, B. L. Duncan, & S. D. Miller (Eds.), The heart and soul of change: What works in therapy (p. 23-55). Washington, DC: American Psychological Association.

Bernard, J. L. (1975). Due process in dropping the unsuitable clinical student. Professional Psychology, 6, 275-278.

Beutler, L. E., Crago, M., & Arizmendi, T. G. (1986). Therapist variables in psychotherapy process and outcome. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change (3rd ed; pp. 257-310). New York: Wiley.

Beutler, L. E., Machado, P. P., & Neufeldt, S. A. (1994). Therapist variables. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change (4th ed.; pp. 229-269). New York: Wiley.

Biaggio, M. K., Gasparikova-Krasnec, M., & Bauer, L. (1983). Evaluation of clinical psychology graduate students: The problem of the unsuitable student. Professional Practice of Psychology, 4, 9-20.

Boxley, R., Drew, C. R., & Rangel, D. M. (1986). Clinical trainee impairment in APA approved internship programs. Clinical Psychologist, 39, 49-52.

Brady, J., & Post, P. (1991). Impaired students: Do we eliminate them from counselor education programs? Counselor Education and Supervision, 31, 100-108.

Burgess, S. L. (1994). The impaired clinical and counseling psychology doctoral student. Unpublished doctoral dissertation, The California School of Professional Psychology, Alameda, CA.

Dillman, D. A. (2000). Mail and Internet surveys: The tailored design method. New York: John Wiley. Forrest, L., Elman, N., Gizara, S., & Vacha-Haase, T. (1999). Trainee impairment: A review of identification, remediation, dismissal, and legal issues. Counseling Psychologist, 27, 627-686.

Herman, K. C. (1993). Reassessing predictors of therapist competence. Journal of Counseling and Development, 72, 29-32. Kaveny, D. (January 29, 2002). Personal communication.

Keppers, G. L. (1960). Selection (if any) of graduate students in guidance and counseling. Vocational Guidance Quarterly, 9, 90-94.

Kerr, M. E., & Bowen, M. (1988). Family evaluation. New York: Norton.

Kitchener, K. S. (1984). Intuition, critical evaluation and ethical principles: The foundation for ethical decisions in counseling psychology. Counseling Psychologist, 12, 43-55.

Olkin, R., & Gaughen, S. (1991). Evaluation and dismissal of students in master's level clinical programs: Legal parameters and survey results. Counselor Education and Supervision, 30, 227-288.

Shaw, B. F., & Dobson, K. S. (1988). Competency judgments in the training and evaluations of psychotherapists. Journal of Consulting and Clinical Psychology, 56, 666-672.

Storm, C. L. Todd, T. C., Sprenkle, D. H., & Morgan, M. M. (2001). Gaps between MFT supervision assumptions and common practice: Suggested best practices. Journal of Marital and Family Therapy, 27, 227-239.

Sweeney, T. J. (1969). Selective retention in secondary school counselors education. Human Education, 7, 53-58.

Tedesco, J. F. (1982). Premature termination of psychology interns. Professional Psychology, 13, 695-698.

Todd, T. C. (1997). Problems in supervision: Lessons from suprvisees. In TC. Todd & C.L. Storm (Eds.), The complete systemic supervisor: Context, philosophy, and pragmatics (pp. 241-252). Boston, MA: Allyn & Bacon.

Wilcoxon, S. A. (1992). Criteria for the selection and retention of supervisees: A survey of approved supervisors. Family Therapy, 19, 17-24.

Candyce S. Russell Kansas State University

Colleen M. Peterson University of Nevada, Las Vegas

Candyce S. Russell, PhD, School of Family Studies and Human Services, Kansas State University; Colleen M. Peterson, PhD, Center for Individual, Couple, and Family Counseling, University of Nevada, Las Vegas.

This research was supported by the Vera Mowery McAninch Professorship. Appreciation is expressed to Allison Joy for her assistance with data collection and to Mary A. Beggs and David Snodgrass for their assistance with the transcription and analysis of data.

Correspondence concerning this article should be addressed to Candyce S. Russell, Kansas State University, School of Family Studies and Human Services, 303 Justin Hall, Manhattan, Kansas, 66506-1403. E-mail: Russell@humec.ksu.edu

Copyright American Association for Marriage and Family Therapy Jul 2003
Provided by ProQuest Information and Learning Company. All rights Reserved