Kadushin and Harkness use Chapter 10 to discuss many issues supervisors face, including?the conflict between ongoing supervision requirements within agencies, licensur

300

from intrusion of an observer. What Freud said of analysis is applied to the social work interview: “The dialogue which constitutes the analysis will permit of no audience; the process cannot be demonstrated.”

The group worker’s performance is more open to observation. Miller (1960) pointed out that “what goes on between the worker and the group is directly visible to many people” (72)— to group members, to other workers, and to supervisors. However, “observations of a work- er’s activity take place . . . on an informal, not a deliberately planned, basis” (75).

While the community worker’s activities also seem to be open to observation, this openness is more apparent than real. As Brager and Specht noted:

Community organization practice is at once more visible and more private than casework. Although it takes place in the open forums of the commu- nity, where higher authorities may be present, this is usually only on ceremonial occasions. Surveil- lance of the workers’ informal activities is another matter. The real business of community workers is less likely to occur within the physical domain of higher ranking participants than the activities of other workers. Thus the community worker has ample opportunity, if he wishes, to withhold or distort information. (1973:240)

Many of the community organizer’s activities are highly informal and unstructured:

Whereas casework interviews can be scheduled and group workers conduct meetings on some

Observation of Performance: The Nature of the Problem At different points in the earlier chapters, we have alluded to persistent problems that confront supervisors in social work. Some problems are methodological, related to how supervisors observe and teach social work prac- tice. Others, addressing supervisory goals and environments, are more basic. The first series of problems is primarily technical in nature. The second series deals with professional pol- icy issues. The intent in this chapter is to pull together and make explicit the different sets of problems and to review the innovative methods and procedures that have been proposed to deal with them.

The supervisor faces a technical problem related to access to the supervisee’s perfor- mance. If the supervisor is to be administratively accountable for the worker’s performance in order to help the worker learn to perform his or her work more effectively, the supervisor needs to have clear knowledge of what the worker is doing (Beddoe et al. 2011). However, the super- visor most often cannot directly observe the worker’s performance. This is particularly true in direct practice. The worker-client contact is a private performance, deliberately screened from public viewing. Concealment of what takes place in the physically isolated encounter is reinforced and justified by dictates of “good” practice and professional ethics. Protecting the privacy of the encounter guarantees the client his or her right to confidentiality and guards against the disturbances to the worker-client relationship which, it is thought, would result

c h a P T e r 1 0

Problems and Innovations

C o p y r i g h t 2 0 1 4 . C o l u m b i a U n i v e r s i t y P r e s s .

A l l r i g h t s r e s e r v e d . M a y n o t b e r e p r o d u c e d i n a n y f o r m w i t h o u t p e r m i s s i o n f r o m t h e p u b l i s h e r , e x c e p t f a i r u s e s p e r m i t t e d u n d e r U . S . o r a p p l i c a b l e c o p y r i g h t l a w .

EBSCO Publishing : eBook Collection (EBSCOhost) – printed on 6/7/2023 10:36 PM via CAPELLA UNIVERSITY AN: 707141 ; Alfred Kadushin, Daniel Harkness.; Supervision in Social Work Account: s9174717.main.ehost

P r O b l e m s a N d I N N O v a T I O N s ] 301

do next would probably last no more than one season. Yet this is the way much of the teaching in psychotherapy is done. What is lacking is a systematic critique of actual performances as ob- served by peers or supervisors. (1977:37)

The problem of access to worker performance is compounded by the fact that not only is the performance itself “invisible,” but the outcomes of the performance are vague and ambiguous. The fact that automobiles come off an auto- mobile assembly line is assurance of the com- petence of the worker’s performance without actually having to see the performance. Social work supervisors never see a clear visible prod- uct of performance.

The traditional, and current, heavy depen- dence on record material and verbal reports for information regarding workers’ performance necessitates some evaluation of these sources. Studies by social workers (Everett et al. 2011; Maidment 2000) as well as other profession- als (Del-Ben et al. 2005; Farber 2006; Farnan et al. 2012; Mehr, Ladany, and Caskie 2010) indi- cate that case records and self-reports present a selective and often distorted view of worker performance.

To examine the nature, extent, and impor- tance of what supervisees withhold from their supervisors, Ladany et al. (1996) surveyed twenty counseling and clinical psychology training programs and received 108 usable responses from graduate students. Therapists- in-training acknowledged withholding negative aspects of their performance, such as clinical mistakes, evaluation concerns, impressions of their clients, negative reactions to clients, coun- tertransference reactions, and client-counselor attraction. The reasons provided for withhold- ing information were that supervisees per- ceived the information as unimportant or too negative or personal to reveal to the supervi- sor; that information was withheld to manage the supervisor’s impression of the supervisee; and that revealing information was a form of political suicide. In most cases, nondisclosures

scheduled basis, the activities of community workers defy regulation and schedule. Much time is absorbed with informal telephone conversa- tions, attending meetings in which they may have no formal role, talking to other professionals and other difficult-to-specify activities. (Brager and Specht 1973:242)

By far the most common source of informa- tion used by supervisors in learning about a worker’s performance is the written case-record material supplemented by a verbal report pre- pared and presented by the supervisee (Ameri- kaner and Rose 2012; Everett et al. 2011; Hicks 2009). Thus, in most instances, the supervisor observes the work of the supervisee at second hand, mediated through the supervisee’s per- ception and written description of it. A medical social worker supervisee wrote:

The only information my supervisor receives about my clients is through my filter, what I tell her at patient staffing and what she reads when I do all my charting. Her thoughts and recommen- dations are manipulated by what information I chart and choose to share.

A worker in an adolescent treatment facility added:

It’s pretty scary. In reality my supervisor has no clear idea of what I am doing. She knows only what I choose to tell her. Sometimes I feel pleased that she trusts my judgments. But sometimes I feel we might be placing my supervisor, as well as my clients, at risk and she could end up on the hot seat on the “Oprah Show.”

Discussing the supervision of psychiatric residents, Wolberg said:

A professional coach who sends his players out to complete a number of practice games with in- structions on what to do and who asks them to provide him at intervals with a verbal description of how they had played and what they intend to

EBSCOhost – printed on 6/7/2023 10:36 PM via CAPELLA UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use

302 [ P r O b l e m s a N d I N N O v a T I O N s

Comparisons of process recordings with audio or video recordings of the same contacts indicate that workers fail to hear and remember significant, recurrent patterns of interaction— discrepancies that can lead to important les- sons in supervision (Bransford 2009). Workers do not perceive and report important failings in their approach to the client. This omission is not necessarily intentional falsification of the record in order to make the worker look good, although that does happen. It is, rather, the result of selective perception in the service of the ego’s attempt to maintain self-esteem (Yerushalmi 1992). In 1934, Elon Moore wrote an article entitled, “How Accurate Are Case Records?” The question, which he answered negatively, is still pertinent today (Gillingham and Humphreys 2010). Supervision based on written records supplemented by verbal reports is supervision based on “retrospective recon- structions which are subject to serious dis- tortions” on the part of the supervisee (Ward 1962:1128). A supervisee wrote candidly:

The tendency in writing process notes is to sort of gloss over things that you found embarrassing or that you found difficult. I think there are times when I’ve made super boo-boos that I’ve left out purposely. And my reason for leaving out super mistakes is that I don’t feel like being embar- rassed. And if I know it’s a mistake, why I have to present it to the supervisor? (Nash 1975:67)

Seeking approval through selective reporting, another supervisee said:

If I’m concerned about my supervisor writing down in his evaluation that I tend to ignore trans- ference phenomena, then, even if I don’t believe it, I’ll make sure to include material that shows transference phenomena, because he happens to be interested in that. (Nash 1975:68)

There is some inevitable distortion of the reali- ties of the encounter in the intellectualized reconstruction for the purpose of verbal and/

remained secret because neither the supervisee nor the supervisor brought them up. Some find- ings from the study suggested that withholding information from the supervisor was related to perceptions of supervisory styles and rela- tionships; supervisees often revealed the same information withheld from their supervisors to their peers or friends in the field, particularly if the information was perceived as important. Studies by Farber (2006), Hess et al. (2008), and Mehr et al. 2010 have reported similar patterns of findings.

Distorted reports and withheld information are self-protective measures against the possi- bility of criticism and rejection by the supervi- sor (Doherty 2005; Hahn 2001;Yourman 2003)). They are also an effort to obtain approval and approbation for work seemingly well done, per- haps even an integral part of the social worker’s professional development (Lazar and Itzhaky 2000; Noelle 2002). It needs to be remembered that approval and criticism are intensified by inevitable transference elements in the relation- ship with the supervisor, and that autonomous decision-making is a cornerstone of profes- sional practice.

Pithouse (1987), studying the use of records by agency workers, noted that, among other things, they were sometimes constructed as a gloss—a protective device vindicating the worker’s practice. He noted that supervisors, having themselves once been workers, recog- nized records as presenting the appearance of expected practice, not necessarily a record of actual practice. As Gillingham and Humphreys (2010:2602) observed, “There may be an unof- ficial version of practice, which is quite differ- ent from the official version, as represented in formal procedures and practice guides.”

Case records are used not only to collect and store information for use, but they are used to justify a worker’s decisions, reconcile conflict- ing impressions, document events for worker protection, and present an understandable pic- ture of a confusing situation to communicate an impression of success (Bush 1984; Munro 1999).

EBSCOhost – printed on 6/7/2023 10:36 PM via CAPELLA UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use

P r O b l e m s a N d I N N O v a T I O N s ] 303

what workers are aware of may not be recalled; if they are aware of it and do recall it, they may not report it. The comment by two supervisors of child psychotherapy trainees is applicable to social workers:

In supervision of child psychotherapists over the year the authors have become impressed with . . . the unexpected degree to which direct observa- tion of the trainee’s psychotherapeutic hours re- veals important and often flagrant errors in the trainee’s functioning—errors which are somehow missed during supervision which is not supple- mented by direct observation. This seems to be the case despite the trainee’s attempt to be as hon- est as possible in talking with his supervisors, his use of the most detailed and complete process notes or his attempt to associate freely about the case without looking at his notes. (Ables and Aug 1972:340)

This clinical observation is confirmed in an empirical study by Muslin, Thurnblad, and Meschel (1981). They systematically compared the actual interview material as recorded on videotape with an audiotape of the reports of these interviews to supervisors. The interviews were conducted by medical students during their psychiatric clerkship. They found that less than half of the material was actually reported to the supervisor, and some degree of distor- tion was present in 54 percent of the interview reports. The four clinicians who indepen- dently studied the actual videotaped inter- views and the reports to supervisors also made a judgment on the significance of the interview material that was not reported to supervisors. Forty-four percent of the material that the judges felt would totally alter the evaluation of the patient was omitted and 9 percent of such material was distorted. “These results indicate that to proceed in supervision as if [emphasis in original] an adequate data base were pres- ent is misleading” (Muslin et al. 1981:824; see also Wolfson and Sampson 1976). Some addi- tional studies on the problem of assessment of

or written communication. Imposing order, sequence, and structure on the interview in reporting the typically discontinuous, redun- dant, haphazard interactions presents it differ- ently from the way it actually happened.

Stein et al. (1975) compared the psychiat- ric evaluations of patients made under two conditions. In one condition, the psychiatric resident described the patient in a supervisory conference and the supervisor completed an evaluation statement. In the second condition, supervisors directly observed the interview between the psychiatric resident and the same patient and on the basis of this observation completed an evaluation. “The results of the study supported the hypothesis that a supervi- sor who does not see the patient is handicapped in his evaluation of the patient’s psychopathol- ogy” (267), indicating that “indirect supervi- sion results in decreased accuracy” (268).

Differences between diagnostic reports pro- vided to supervisors and independent assess- ments by observers of the same interaction have been confirmed once again in studies of intake interviews (Del-Ben et al 2005; Spitzer et al. 1982). This is a significant concern in clinical settings (Ponniah et al. 2011), where misdiag- noses or underdiagnoses can have educational, legal and medical consequences for providers and clients (Farnan et al. 2012; Harkness 2010).

Even if worker reports formed a valid basis for educational supervision, using only those reports for evaluating worker performance would be a hazardous procedure. Valid evalua- tion requires that supervisors know what work- ers actually did, not what they think they did or what they say they did. By applying what is known about human behavior to the supervisee reporting on his or her own performance, one can recognize the inadequacies of such a pro- cedure as a basis for either good teaching or valid evaluation. As a consequence of anxiety, self-defense, inattention, and ignorance of what should be looked for, workers may be unaware of much that takes place in the encounters in which they are an active participants. Some of

EBSCOhost – printed on 6/7/2023 10:36 PM via CAPELLA UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use

304 [ P r O b l e m s a N d I N N O v a T I O N s

(Amerikaner and Rose 2012; Reiser and Milne 2012), and social workers (Beddoe et al. 2011; Everett et al. 2011). Direct observation is scarcer still in professional work settings, where “work environments lack needed resources,” “heavy productivity demands from agency adminis- trators have the effect of severely limiting their ability to make use of the supervision strategies they would prefer,” supervisors are often “only paid for the limited time needed for supervision sessions, and thus have little time or incentive for utilizing strategies such as live observation,” and agencies governed by HIPAA regulations have adopted “policies which limit or elimi- nate” the direct observation of clinical work (Amerikaner and Rose 2012:78–79; see also Fromme et al. 2009).

Feedback rooted in direct observation has been described as one of the key elements in developing worker skills in evidence-based practice (Carlson, Rapp, and Eichler 2010), but the evidence base for this claim is not well developed (Champe and Kleist 2003; Farnan et al. 2012). There is some evidence that supervis- ees who receive live supervision form a stron- ger working alliance with their supervisors (O’Dell 2008), if not with their clients (Moore 2004), and that feedback based on direct observation may change supervisee behavior (Craig 2011; Scheeler, McKinnon, and Stout 2012)—although, here, the evidence is mixed (Craig 2011). There is also some evidence that “live supervision” makes “a difference in thera- pists’ ratings of progress on the problem over the course of therapy” (Bartle-Haring et al. 2009:406), and that direct observation leads to better patient outcomes (Farnan et al. 2012) in medical settings, by offering medical stu- dents “the opportunity to see beyond what they know and into what they actually do” (Fromme et al. 2009:265). But, for the most part, the pertinent research has been con- cerned with the experiences and perceptions of students and workers (Champe and Kleist 2003; Haber et al. 2009), not clients (Champe and Kleist 2003).

performance on the basis of supervisee self- reports were cited in the chapter on evaluation.

In response to these difficulties, various innovations have been proposed to give the supervisor more direct access to the worker’s performance. A review of the literature by Goodyear and Nelson (1997) enumerated twenty-two strategies that supervisors have used to observe worker performance; Bernard and Goodyear (2009), Kaslow et al. (2009), and Walker (2010) identified even more. Of these, direct observation and the indirect observa- tion of audio- and video-recorded interviews have been rated highly by clinical supervisors and their trainees (DeRoma et al. 2007; O’Dell 2008), although few social work supervisors appear to use either method (Everett et al. 2011; Knight 2001; Scott et al. 2011). For both super- visors and workers, the postsession observa- tion of video-recorded practice has been the most highly rated supervisory method (Ber- nard and Goodyear 2009; Wetchler, Piercy, and Sprenkle 1989).

direct Observation of Performance “Supervision is a dynamic process that requires the use of a diversity of supervisory styles and approaches” (Graf and Stebnicki 2002:41).The simplest procedure is direct observation of the interview, either by unobtrusively sitting in on the interview or by observing the inter- view through a one-way mirror. The client’s informed consent is needed, of course, for this and any other procedure that opens the client- worker contact to outside observation (NASW 2008). (Whether social work’s involuntary cli- ents can give valid informed consent for direct observation is a controversial question.)

Although 51 percent of master’s degree pro- grams and 57 percent of doctoral programs in counselor education are said to use direct observation in training supervision (Champe and Kleist 2003), such methods are used more rarely in the training of physicians (Craig 2011; Fromme, Karani, and Downing 2009), and more rarely still in the training of psychologists

EBSCOhost – printed on 6/7/2023 10:36 PM via CAPELLA UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use

P r O b l e m s a N d I N N O v a T I O N s ] 305

special room, and it has its own hazards. Gruen- berg, Liston, and Wayne (1969) noted that “the physical setting of the one-way mirror arrange- ment has been less than conducive to continu- ous alertness in the supervisor. The darkened room is more often conducive to languor than attentiveness” (Gruenberg et al. 1961:96; see also Adler and Levy 1981).

In a study of ten social work graduate stu- dents on the other side of the one-way mirror, Wong (1997) found that social work students expected and experienced anxiety before and during supervision. In time, however, the trainees begin to relax, and by the end of their training they generally perceived this form of supervision as a valuable experience. As shown in table 10.1, the limited evidence suggests that this mode of observation is rare in social work supervision.

Co-Therapy Supervision The supervisor-observer, by sitting in on an interview (or otherwise intervening in real time), can easily move to a new role—that of co-worker or co-therapist (Tuckman and Fin- kelstein 1999). Co-therapy has also been termed multiple therapy and, in group work, co-lead- ership. If supervision through co-therapy is offered, it is generally provided as a supplement to, rather than a substitute for, individual super- vision. See Beddoe et al. (2011) and Evans (1987) for rich introductions.

Supervision conducted by sitting in with the worker and client has been championed

Sitting In Kadushin (1956a, 1956b; 1957) tested the fea- sibility of sitting in on an interview in both a family-service agency and a public-assistance agency. Very few clients objected to the intro- duction of an observer. Postinterview dis- cussion with both the worker and the client, supplemented by some objective measures of interview contamination attributable to obser- vation, indicated that an unobtrusive observer had little effect on the interview.

Schuster and his colleagues used this proce- dure in the supervision of psychiatric residents: “We decided on a simple direct approach to the matter. We decided to have the supervisor sit with each new patient and the resident, as a third party, relatively inactive and inconspicu- ous but present…. In very few instances did our presence seem to interfere significantly with either the resident or the patient” (Schus- ter, Sandt, and Thaler 1972:155). For the most part, this has also been the conclusion of recent reviews of the research (Bernard and Goodyear 2009; Champe and Kleist 2003; Ellis 2010).

One-Way Mirrors The one-way mirror permits observation with- out the risk, or necessity, of participation and minimizes observer intrusion on the inter- view or group session (Fleischmann 1955). The supervisor can see and hear the interview with- out himself being seen or heard. Peer group observation of the interview or group session is also possible. One-way viewing requires a

T a b l e 1 0 . 1 Observational methods used in intern, student, and trainee supervision

Profession Client records and/ or process recording Self-report

Audio recording

Video recording

One-way mirror Co-therapy

Marriage and family therapistsa 34% 76% 53% 65% 58% Not reported Marriage and family counselorsb 72% 83% 39% 57% 28% Not reported Psychologistsc 58% 73% 3% 11% 3% 13% Social workersd 93% 81% 12% 7% 9% 68%

aNational sample of 330 approved American Association for Marriage and Family Therapy supervisors from Lee et al. (2004:65). bCalifornia sample of 389 supervisees from DeRoma et al. (2007:421). cNational sample of 173 supervisees from Amerikaner and Rose (2012:68–69). dRegional sample of 81 field instructors from Everett et al. (2011:258).

EBSCOhost – printed on 6/7/2023 10:36 PM via CAPELLA UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use

306 [ P r O b l e m s a N d I N N O v a T I O N s

The supervisor intervenes only when the super- visee is experiencing some difficulty, when the supervisee signals a request for intervention, or when the supervisor sees a clear opportunity of modeling behavior that he or she is anxious to teach.

Noting the developmental importance of observing supervisees directly early in their training, Hunt and Sharpe said:

[We] evaluated the perceptions of 37 interns and 49 patients regarding communication between supervisor and intern during the therapy ses- sion. Most supervisors used such communication infrequently. While some interns and patients viewed call-ins or walk-ins as intrusive and un- helpful, the majority did not. Indeed, some valued the feedback. No association was found between the use of call-ins or walk-ins and patient reports of therapist behaviours or intern reports of super- visory style. It is concluded that within-session communications is not deemed as necessary by most supervisors, who should remain cautious in their use of such communication, unless patient care is compromised. (Hunt and Sharpe 2008:121)

Productive use of co-therapy for purposes of supervision “requires a conscious effort by the supervisor to modify what tendency he might have to take over and be the expert and for the [supervisee] to resist a tendency to sit back and be an observer” (Sidall and Bosma 1976:210). Alternatively, the supervisor may elect to consult with the worker in the presence of the client, in order to “heighten” the client’s “awareness of particular dynamics” (Bernard and Goodyear 2009:265).

The helping professions employ and evaluate co-therapy supervision differentially (Carlozzi, Romans, Boswell, Ferguson, and Whisen- hunt 1997), and this has changed over time. In 1986, McKenzie, Atkinson, Quinn, and Heath reported that 64.9 percent of marriage and fam- ily-therapy supervisors engaged in co-therapy supervision; eleven years later, DeRoma et al. (2007:419) noted that “direct supervision in the

for safeguarding client welfare, immersing the supervisor in the direct practice experience, and allowing clients to observe supervision in action (Bernard and Goodyear 2009). One of the principal advantages of co-therapy is that the supervisor, as an active participant in the supervisee’s performance, is in a position to witness firsthand the behavior of the super- visee (Finkelstein and Tuckman 1997). Having initiated co-therapy, a supervisor notes that he became immediately aware of a supervisee’s problematic approach to the client—a problem that “had not been clear to me during the few months of traditional supervision we had had” (Rosenberg, Rubin, and Finzi 1968:284). In ana- lyzing the experience over a six-month period of co-therapy between a supervising psycho- therapist and students, Rosenberg, Rubin, and Finzi noted that “the direct observation of the student did away with retrospective falsification in the student’s traditional role in reporting his work to the supervisor” (1968:293).

The power and manifestations of counter- transference are more apparent to the supervi- sor as he or she experiences the client firsthand. Co-therapy then makes more information and more valid information available to the super- visor. Consequently, it is an innovation that helps resolve the problem regarding the infor- mation needed by the supervisor for effective supervision.

Munson (2002) ranked the live co-therapy interview as the most effective method of social work supervision, but the use of co-therapy for educational supervision may present problems. If the supervisor becomes overactive in the worker- client meeting, then the dynamics may resemble those of co-therapy conducted by a novice paired with an experienced practitioner (Smith, Mead, and Kinsella 1998), in which “the junior therapist” tends to defer to the senior therapist, who tends “to take over” (Altfeld and Bernard 1997:381). Co- therapy supervision is also expensive.

During the co-therapy session, it is advisable for the supervisor to allow the supervisee to take primary responsibility for the interaction.

EBSCOhost – printed on 6/7/2023 10:36 PM via CAPELLA UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use

P r O b l e m s a N d I N N O v a T I O N s ] 307

that the experience as it actually occurred is lost forever. Similarly, direct interview observation and observation through co-therapy or a one- way mirror leave no record for retrieval, study, and discussion. To correct this deficiency, some social workers make audio or video recordings of their interviews for use in supervision. The use of such procedures is said to be widespread in the supervision of marriage and family counselors (DeRoma et al. 2007), but less so in psychology and social work, as shown in table 10.1. The importance of such technical aids for supervision is that their use enables one to observe performance indirectly and reenact the performance for examination.

Video may be recorded through an unob- trusive port from an adjoini