Development of the Psychotherapy Supervisor: Review of and Reflections on 30 Years of Theory and Research
Abstract
Introduction
“You are,” answered the male supervisor… . “We all begin as frauds,” he continued … “That is the nature of the learning of psychotherapy [and psychotherapy supervision]. As we read, apply theories and techniques, and learn the vicissitudes of the patient-therapist-supervisor interaction, we grow to be less and less fraudulent, and eventually mature into the real thing”(Chagoya & Chagoya, 1994, pp. 189-190).
Squarely within the supervision context is the person of the supervisor–his or her development as, and process of “becoming,” a supervisor. Of the many issues and concerns within the supervision context, none seems more important than psychotherapy supervisor development and its ramifications” (Watkins, 1995c, p. 157).
Psychotherapy Supervisor Development Models: 1983-1995
Stage Names and Descriptions Across Models | ||||
---|---|---|---|---|
Model | 1 | 2 | 3 | 4 |
Alonso (1983, 1985) | Novice: Anxious, confused, conflicted; feels fraudulent; quest for identity; draws on experiences as supervisee to inform efforts | Midcareer: Internally settled; ideal mentor; generative, secure; stable sense of self; supervisory altruism in evidence | Late Career: Adjusting to getting older personally/professionally; “The choice is between hope, wisdom, and integrity, and clinical despair and boredom” (Alonso, 1985, p. 75). | |
Hess (1986, 1987) | Beginning: Often lack formal training; selfconscious, unaware, anxious; draws on experiences as supervisee for guidance | Exploration: Increasingly aware of importance of supervision and its impact on supervisees; needs of supervisees given priority; fluctuating though much improved performance; informal power base | Identity Confirmation: Solidified supervisor identity; heightened sense of confidence and professionalism; high level of skillfulness | |
Rodenhauser (1994, 1997) | Emulation: Draws on previous experiences as supervisee to provide direction; uses memories of past supervisor to guide work | Conceptualization: Conceptual foundation for supervision takes form; rough practice guidelines established; search for a “system” predominates | Incorporation: Increasingly aware of self as supervisor and impact on supervisees; increasingly sensitive to/respectful of differences and diversity; attention to parallel process phenomena emerges | Consolidation: Solidified identity; theoretically grounded; able to effectively address parallel process and supervisee counter-transference phenomena |
Stoltenberg & McNeill (2009; cf. Stoltenberg & Delworth, 1987) | Level 1: Highly anxious, uncomfortable, confused; concerned about doing the “right thing”; can resort to “flight into structure”; draw on past supervisee experiences for direction | Level 2: Conflicted, confused, frustrated; fluctuating motivation and affectivity; assertion of autonomy mingled with lapses into dependency | Level 3: Consistent motivation; autonomous, comfortable, committed; able to engage in realistic self-appraisals; consults with colleagues on “as needed” basis | Level 3 Integrated: “Master supervisor” highly skilled and integrated; able to effectively work with wide variety of supervisees |
Watkins (1990, 1993) | Role Shock: Anxiety, confusion, turmoil; feelsfraudulent (impostor phenomenon); limited to no supervisory confidence, identity, or self-awareness; great need for support from others | Role Recovery/Transition: Some recognition of supervisory strengths and abilities; nascent identity core begins to take form; tension, anxiety, confusion, and turmoil abate; vacillating cognitive and affective experience | Role Consolidation: Increasingly realistic sense of self and supervisory impact; general sense of confidence, solidified identity core; stable, consistent, secure; theoreticallygrounded; able to recognize and address supervisory transference, countertransference, and parallel process | Role Mastery: “Master supervisor” highest levels of awareness, skill, and identity achieved; consistency, stability, and consolidation predominate; highly committed to supervision as lifelong learning experience; highly adept at working with supervisory transference, countertransference, and parallel process |
There is one commonality cutting across… the developmental models that is particularly relevant …. There is anxiety, self-doubt, and feelings of being overwhelmed when initially assuming the role of supervisor. Whether called role shock, imposter syndrome, or another label, the experience of angst and struggle are common when a supervisee transitions to becoming a supervisor. The new supervisor should find comfort in the awareness that these feelings are not unique and that they will gradually subside over time. In fact, developmental theories suggest that supervisors will transition through phases and eventually feel confident, comfortable, and integrated … (p. 86).
Considering Findings From the Most Recent Reviews, 1980-1995
Worthington (1987)
Russell and Petrie (1994)
Watkins (1995b)
Summary
Considering Research on Psychotherapy Supervisor Development, 1996 to 2011
Method
Authors | Setting/Sample | Measures | Analyses | Procedure | Findings/Conclusions | Limitations/Strengths |
---|---|---|---|---|---|---|
Baker, Exum, & Tyler (2002) | 12 doctoral students (6 f, 6 m; 5 African-American, 7 European-American; mean age=34 years) enrolled in clinical supervision practicum (experimental group); 7 doctoral students (6 f, 1 m; 2 African-American, 5 European-American; mean age=40 years) not yet enrolled in supervision practicum (comparison group) | Psychotherapy Supervisor Development Scale (PSDS); Retrospective interview questions | Independent and dependent ± tests; Qualitative analysis of interview questions | All students tracked across one semester, administered PSDS at beginning, midpoint, and end of semester, and administered retrospective interview at midpoint and end of semester | Supervisor development of supervision practicum students increased significantly across semester and was significantly greater compared to nonpracticum students; qualitative results were mixed | Small sample size; development studied only over single semester in single program |
Barnes (2002) | 287 clinical supervisors (58% f; 83 % European-American; mean age=43 years; average years of supervision experience=8.06) | Demographic questionnaire; Psychotherapy Supervisor Development Scale; Counselor Supervisor Self-Efficacy Scale | Correlations Analyses of variance Factor analysis | Three-phase study: Expert evaluations, pilot study, and main study; 1,462 questionnaire sets mailed out to liaison for distribution; 287 clinical supervisors served as final sample; they completed survey packet materials and mailed it back to researchers | “… results provide some evidence of … validity of the Supervisor Complexity Model” (p. 81) and “… indirect validation for other related supervisor development models” (p. 81) | Self-report survey data; instrument development/validation study |
Bencivenne (1999) | 136 American Association of Marriage and Family Therapy (AAMFT) Approved supervisors (70 f, 66 m; 129 European-American; mean age=52.17 years); 116 of supervisors’ supervisees (64 f, 52 m; 104 European-American; mean age=35.8 years) | Supervisors: AAMFT-Approved Supervisor Demographic Questionnaire; Pychotherapy Supervisor Development Scale; Supervisor Emphasis Rating Form-Revised; Personal Preference Scale Supervisees: Supervisee Demographic Questionnaire; Supervisee Levels Questionnaire Revised; Personal Preference Scale | Pearson productmoment correlations; factorial analysis of variance; multivariate analysis of variance | Initial mailing to 500 AAMFTA-Approved supervisors, additional mailing to 200 more supervisors; each participating supervisor asked to randomly select a supervisee to fill out supervisee questionnaire packet; each supervisor asked to complete supervisor questionnaire packet; follow-up PSDS sent to supervisors to assess its test-retest reliability | Supervisor development related to hours of training in supervision with AAMFT-Approved supervisors (increased training associated with higher development); experience not related to supervisor development; four-week temporal stability of PSDS was .863 | Correlational, crosssectional, quasiexperimental survey study |
Culbreth &Cooper (2008) | 232 substance abuse counseling clinical supervisors (57.5% f; 86.5% white; mean age=51.1 years; mean years experience as supervisor=8.6) | Demographic information questions; Psychotherapy Supervisor Development Scale; Role Questionnaire (RQ); Counselor Supervisor Self-Efficacy Scale | Pearson productmoment correlations; multiple regression analyses run for entire group (N=232), recovering supervisors (48%), and non-recovering supervisors (52%) | Survey packet including questions/measures mailed to 491 registered or certified substance abuse counseling clinical supervisors in Midwestern state; reminder postcard and second survey packet mailed out as follow-up | Perceived self-efficacy in supervision theory/techniques most significant predictor of supervisor development across groups (entire sample, recovering supervisors); some variations in supervisor development found between recovering and non-recovering groups; results seen as being supportive of need for training in clinical supervision for substance abuse counseling supervisors | 48.43% return rate; selfreport measures used only; one-state survey |
Kurdt (2001) | 16 clinical supervisors (6 f, 10 m; 15 European-American; mean age= 45 years), 16 supervisees (12 f, 4 m; 15 European-American; mean age=30 years), and 12 clients (9 f, 3 m; all European-American, mean age=30 years; 6 supervisors (2 f, 4 m) from group of 16 served as interview participants for qualitative portion of study | Supervisory Styles Inventory; Working Alliance Inventory; Supervisory Working Alliance Inventory; Supervisory Development Interview | Pearson productmoment correlations Content analysis procedure | Survey packets mailed to 49 potential participants, all actively practicing clinical supervisors affiliated with master’s degree counseling program (33% response rate); supervisors then provided supervisee and client packets to supervisee for distribution/completion | Four areas emerged as crucial in supervisor development process: Context within which supervisor learning occurs; making sense of the supervisor role; acquiring an approach to teaching supervisees; and developing a method or style of evaluating | Small sample size; all supervisors affiliated with single institution |
Lyon, Heppler, Leavitt, & Fisher(2008) | 233 psychology pre-doctoral interns (160 f, 71 m; 177 European-American; mean age=32.45 years) from clinical (N=151), counseling (67), school/counseling (10), and other (N=5) psychology graduate programs; internship sites widely varied | Demographic questions; Psychotherapy Supervisor Development Scale; Supervision questions about selected variables (e.g., supervision training, experience, interest in being a supervisor) | Frequencies, percentages, point-biserial correlations, multiple regression analysis | Survey packets mailed to training directors of American Psychological Association accredited internship sites; follow-up email sent as reminder; participating directors then gave survey packets to interns to compete | Total number of supervision training activities and number of hours of supervised supervision predicted interns’ level of supervisor development (more training plus more supervised supervision=higher development); most interns had provided supervision to a trainee during or prior to internship, but most had not had a graduate course in supervision; interns expressed interest in becoming supervisors and receiving supervision training | 33% return rate; self-report measures used only |
Pelling (2008) | 175 clinical supervisors (76 f, 98 m; 85% European-American; 85% doctoral degree holders; mean age=52.5 years; all members of Association for Counselor Education and Supervision | Psychotherapy Supervisor Development Scale; Three predictor variable questions (# of years supervising others, # of years treating others, and # of years of supervision training experience) | Frequencies, percentages, forward stepwise multiple regression analyses | Introductory postcard, survey package, reminder postcard, and follow-up survey package mailed out over five week period | Experience providing supervision and training in supervision were the best predictors of supervisor development (8.3% and 4.4%, respectively) | 58.3% return rate; selfreport measures used only; highly experienced supervision group |
Ronnestad, Orlinsky, Parks, Davis, & Society for Psychotherapy Research Collaborative Research Network (CRN) (1997) | 1639 psychotherapists (975 f, 458 m; mean age=40) from around the world (two-thirds from Europe); professions represented included medicine (N=395), psychology (N=867), social work (N=174), and other (N=177); theoretical orientations included analytic/dynamic (50.8%), humanistic (33.8%), cognitive (23.4%), systemic (22.9%), and behavioral (17.7%) | Development of Psychotherapists Common Core Questionnaire | Frequencies, percentages, step-wise multiple regression analyses | Large international survey of psychotherapists, psychoanalysts, and counselors conducted by CRN | Noticeable increases in supervisor confidence observed after initial supervisory experiences; supervisor confidence not predicted by amount of therapy supervision received, but was predicted to some extent by amount of therapy supervision experience; becoming a supervisor found to be “part of the normative development of practicing psychotherapists” (Ronnestad et al., 1997, p. 195) | Self-report measure; crosssectional study; highly experienced sample; supervision training or supervisor supervision received not included as variables |
Stevens, Goodyear, & Robertson(1997) | 60 practicing mental health professionals (39 f, 21 m; 46 European-American; mean age=41.6 years; 30 eclectic, 12 psychodynamic, 7 cognitive-behavioral, 3 person-centered, 3 family systems, 5 other) | Demographic survey; Supervisory Emphasis Report Form-Revised (SERF-R); Thought listing; Video-tape stimulus questions | Factor analysis, factorial analysis of variance; multivariate analysis of variance | 12 sites for clinical, counseling, or professional psychology training used; participants asked to complete demographic survey and SERFR, watch stimulus videotape, and then do thought listing and stimulus questions | Limited support found for predicted, linear movement of supervisors as they gain experience; experience identified as necessary but not sufficient for supervisor development; need for supervisory training underscored | Self-report measures; crosssectional study; limited to southern California |
Vidlak (2002) | 99 clinical or counseling psychologists (53% f; 89% European-American; mean age=45.6 years) working at approved (American Psychological Association) pre-doctoral internship sites; all clinical supervision providers | Demographic questionnaire; Psychotherapy Supervisor Development Scale; Supervisory Styles Inventory; Working Alliance Inventory-Supervisor Form | Correlations ± tests Analysis of variance Regression analyses | Online survey packet completed by 99 participants | Supervision training and supervisor supervision highly related to level of supervisor development, but supervisor experience was not; supervisory style served as mediator in relation between supervisor development and supervisory working alliance | Self-report survey data; noncausal design; limited sample size |
Vieceli (2006) | 37 supervisors (in academic and clinical settings; no specifics with regard to gender or ethnicity provided) | Psychotherapy Supervisor Development Scale; Supervisory Emphasis Report Form-Revised; Selected supervision variables (e.g., supervision experience, training) | Pearson productmoment correlations Non-paired ±tests Multivariate analysis of covariance | Survey packets distributed to supervisor group | No significant multivariate effect found for impact of supervisors’ years of experience, supervision training, and perceived workplace support on supervisor development; positive correlation found between supervisor development and emphasis on conceptualization | Small sample size; low return rate; selfreport survey data |
White (1997) | 4 clinical psychology doctoral students (3 f, 1 m) followed over 1-year period, in which they provided supervision to therapy trainees during the last four months of year; participants’ supervision training experiences varied (e.g., didactic/experiential practicum, “learning by doing” or “trial by fire”) | Psychotherapy Supervisor Development Scale (PSDS); Supervisory Styles Inventory (SSI); Working Alliance Inventory (WAI); 2-hour semistructured interview | Visual inspection of data graphs Correlation of coefficients Atlas ti qualitative data analysis program | In June, participants asked to complete PSDS and SSI every 2 to 3 weeks until September (beginning of supervisor training); participants completed semistructured interview first week of September; SSI and PSDS completed at each month’s end until study finished; participants filled out WAI after first meeting with their supervisees and at each month’s end until study finished; supervisor trainees completed semistructured interview 3 weeks after beginning actual therapy supervision; completed another interview 3 weeks before study’s end | Developmental model studied (Supervisor Complexity Model; SCM) “does describe the experience of these supervisees fairly well, with exceptions” (p. 199); issues considered fundamental in SCM “to supervisor development do appear to have been central to those supervisors’ experiences” (p. 200); speculated that supervisor development may occur more quickly than theory suggests and that structure and support may diminish or eliminate “impostor phenomenon” and other highly negative experiences associated with early supervisor development; PSDS viewed as being potentially useful in tracking supervisor development longitudinally | Intensive, longitudinal mixedmethods design, small number of participants, supervision of therapist experience only 4 months in duration |
Ybrandt & Armelius(2009) | 9 supervisor trainees (6 f, 3 m; all Caucasian; mean age=48 years; 7 psychologists, 2 social workers) participating in a three-semester postgraduate supervisor and teacher training program at Swedish University; supervisor group consisted of 12 highly experienced supervisors (7 f, 5 m; mean age=49 years), all clinical psychologists | Swedish Structural Analysis of Social Behavior (SSASB) | Three self-image ratings (before, after, and follow-up) compared on two levels of aggregation, clusters, and vectors with paired sample ± tests | SSASB self-image questionnaire mailed to supervisor trainee group participants at three points: Before training, after training, and at four-month follow-up | Supervisor trainees self-image found to change over the course of supervisor training, reflecting more autonomy and positive selfacceptance; gains also maintained at follow-up | Self-report measure used; small sample size; developmental changes studied over three semesters plus fourmonth follow-up |
Authors | Setting/Sample | Measures | Analyses | Procedure | Findings/Conclusions | Limitations/Strengths |
---|---|---|---|---|---|---|
Majcher ± Daniluk (2009) | 6 doctoral students (4f, 2m; all European-American; age range late 20s to mid 40s) participating in 8 month supervision training course (with didactic and experiential components) | Three in-depth, audiotaped interviews, 1 to 2 hours in duration, conducted with each supervisor trainee | Interpretive phenomenological analysis and thematic review | In-depth audiotaped interview conducted with each supervisor trainee at three different points: Beginning of supervision course, midway through, and at course’s end. Main interview question was: “How do you experience and make sense of your process of becoming a clinical supervisor? Please describe your experience of becoming a supervisor in as much detail as possible” (Majcher ± Daniluk, 2009, pp. 64-65). | Relationship dimension emerged as core common element for all participants across all identified themes across all interviews; interview 1 common themes—role ambiguity/uncertainty, competence/incompetence, excitement, and accomplishment; interview 2 common themes—transformation, developing competence/confidence, supervisory role identification, and role/boundary clarity; interview 3 common themes—supervisory commitment, congruence, and having experienced a significant growth process; findings reflected a process similar to supervisor development models | Small sample, all European-American, drawn from single program; 8 month course |
Nelson, Oliver, ± Capps (2006) | Study 1 sample: 13 doctoral students (10 f, 3 m; 7 European-American, 6 Hispanic; taking 3 semesters of practicum/internship) who were supervisors in training Follow-up sample: 5 doctoral students (4 f, 1 m; 2 European-American, 1 Hispanic, 2 African-American; taking practicum/internship) used to explore and confirm/disconfirm thematic results from Study 1 sample; followed sample 1 by 1 ½ years | 8 primary research questions (e.g., “How did this process help you develop supervisory skills?”) asked of each participant | Constant comparative methods used to analyze individual/focus group interview transcripts, student interview notes, and researcher memos/reflections; open coding, selective coding, and theoretical sampling employed | Individual interviews used to gather responses to 7 primary research questions; at each semester’s end, identified themes presented to participants’ focus group for discussion/clarification purposes; sample 1 thematic data used to stimulate further discussion about supervisor development in follow-up sample’s focus group | Six major themes in supervisor development identified: Learning (e.g., through academics, experience), supervisee growth, individual uniqueness, reflection (i.e., pondering on one’s experience as supervisor), connections (e.g., to peers, supervisees, professional associations), and putting it all together (i.e., finding integration where once there was none); “ voices of doctoral students in this study mirror the [supervisor development] literature ” (p. 29); [o]ver the year and a half of study, marked by less need for structure, increased confidence and decreased anxiety about supervision, and a transition from relying on external resources to relying on internal resources” (p. 29). | One doctoral program; limited number of research participants; follow-up sample used as check and balance |
Rapisarda, Desmond & Nelson (2011) | 7 doctoral students(5 f, 2m of varied ethnicities; mean age=37.7 years) taking supervision practicum course | Two overarching questions guided interview process—(1) How do students describe supervisee to supervisor transition? and (2) What challenges/surprises did they find in that transition? | Collective case study design, constant comparative method used, QSRN6 qualitative research software used throughout for data analysis | 2 45-minute interviews conducted with each participant; first interview took place at mid point of semester, second interview at end of semester | Two main themes identified in supervisee to supervisor transition; (1) establishing a supervisory relationship; and (2) developing a supervisory skill set. Two main themes identified as transition challenges/surprises: (1) the challenge of time; and (2) increased appreciation for and valuing of supervision as professional activity. | One doctoral program; limited number of participants, most of whom were Caucasian females; development followed for only 1 semester |
Authors | Setting/Sample | Measures | Analyses | Procedure | Findings/Conclusions | Limitations/Strengths |
---|---|---|---|---|---|---|
Barnes & Moon(2006) | 255 clinical supervisors (58% f; 86% European-American; average number of years providing supervision=8.34) drawn from counseling programs across the nation | Psychotherapy Supervisor Development Scale (PSDS); Demographic form; Counselor Supervisor Self-Efficacy Scale | Confirmatory factor analyses of second-order factor model and four-factor model | Survey packets distributed to participants and, once materials completed, then returned to researchers via mail | Results offered validation for PSDS, supported a slightly modified four-factor model as best fit, and suggested that supervisor competence/effectiveness, supervisor identity development/commitment, supervisor selfawareness, and supervisor sincerity were each key areas in the supervisor development process; “counseling” supervisors and “psychotherapy” supervisors (Watkins et al.,1995) responded to PSDS in virtually identical developmental fashion | Self-report survey data; possible social desirability effect |
Hillman, McPherson, Swank, & Watkins (1998) | 43 clinical supervisors (27 f, 16 m; 28 European-American; 36 practitioners, 7 supervisor trainees); all participants involved in providing psychotherapy supervision, supervisor supervision, or receiving supervisor training | Demographic data sheet; Pychotherapy Supervisor Development Scale (PSDS); Theory description | Pearson productmoment correlations; Guttman splithalf coefficient estimates | 83 survey packets sent to counseling center training directors and supervision course instructors for distribution to interested participants; after completing initial packet, participants were again mailed PSDS four weeks later for completion | Four-week temporal stability coefficient=.85; split-half coefficient=.95; PSDS considered to be highly homogeneous, stable measure of supervisor development | Small sample, self-report survey data; all counseling centers in Southwestern United States; single training program also in Southwest |
Results
General Study Characteristics
Quantitative/Mixed Method Studies
Qualitative Studies
Measurement Studies
Discussion
Thoughts on theory
Thoughts on Research and Measurement
Thoughts on Practice and Education
Unwilling as we might be to accept it, most supervisors simply might not improve with experience. One reason for this might be that supervisors have little training in how to supervise effectively and thus may perpetuate the mistakes of their own supervisors…. Mere experience might be insufficient to enable one to view one’s work objectively or to take different perspectives on one’s work (p. 206).
Discussion Summary
Conclusion
REFERENCES
Information & Authors
Information
Published In
History
Keywords:
Authors
Metrics & Citations
Metrics
Citations
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
There are no citations for this item
View Options
View options
PDF/ePub
View PDF/ePubGet Access
Login options
Already a subscriber? Access your subscription through your login credentials or your institution for full access to this article.
Personal login Institutional Login Open Athens loginNot a subscriber?
PsychiatryOnline subscription options offer access to the DSM-5-TR® library, books, journals, CME, and patient resources. This all-in-one virtual library provides psychiatrists and mental health professionals with key resources for diagnosis, treatment, research, and professional development.
Need more help? PsychiatryOnline Customer Service may be reached by emailing [email protected] or by calling 800-368-5777 (in the U.S.) or 703-907-7322 (outside the U.S.).