Implications for Counselor Education
The following sections suggest the importance of differentiating training from supervision and the promise of developing student profiles, of conducting effectiveness studies, and of responding to the trend in the field to manualize treatments.
Differentiate Training and Supervision
Counselor educators would be well-served to more clearly differentiate the function of training from that of supervision. One reason for doing this is that the optimal conditions for training might be somewhat different than those for supervision.
In their review of the literature, Holloway and Neufeldt (1995) noted that skill delivery is learned rather quickly. Rather, it is timing and judgment that takes much longer to acquire. Although Holloway and Neufeldt referred to timing and judgment as "advanced skills," it may be more helpful to view them as reflecting advanced integration of an earlier learned skill and requiring a supervisory rather than training context. For example, the call for structure for novice trainees and for trainees of lower conceptual level may in fact be a call for training. Henry, Schacht, Strupp, Butler, and Binder (1993) found that more directive supervisor behaviors, such as interrupting videotapes of sessions frequently and focusing on specific therapist behaviors, accelerated the learning of therapy approaches. Conversely, supervision may be the more reflective, introspective part of trainee development--learning that is decelerated, so to speak. Evidence of training's effects may be easy to recognize on a videotape; the evidence of supervision's effects will be more subtle.
It may be that most counselor educators working with master's level students find a considerable part of their work to be training. Because much of the research in "supervision" has used these students as subjects, it is no wonder that differentiation of training and supervision might become "muddied."
Develop Profiles of Students
We have sufficient information about particular trainee characteristics for counselor education programs to be encouraged to develop profiles of students at the onset and throughout graduate training. We will not know more about training or supervision until we understand better how our students' personal characteristics affect these interventions. Assessment might occur, for example, with respect to trainees' reactance and conceptual levels as well as to their gender and racial identity development. Counselor educators in programs need to more consciously use the data that exist and observe the impact of these variables throughout training and supervision to (a) start to appreciate intraprogram effects, (b) start to appreciate interactions among multiple trainee factors, and (c) begin to form a more informed research agenda for training and supervision.
Conduct Effectiveness Studies
We have already addressed the problems in conducting efficacy studies. Effectiveness studies, however, are not only possible but overdue.
Because of the proliferation of supervision models in the past 20 years, the researcher can choose models that are distinct enough to warrant investigation. For example, live supervision and interpersonal process recall certainly must differ in their effect across a host of conditions. Studies that begin to assess their differences could make important contributions to counselor education programs and postdegree supervisors.
This particular issue speaks most directly to researchers. But it affects practitioners as well, because they certainly have vested interest in the outcomes of supervision effectiveness studies. More-over, they could play an essential role in conducting and even designing the interventions.
To manualize or not. Elsewhere (Bernard & Goodyear, 1998) we discussed the increasing attention given to treatments that are manualized. There can be no question that these will affect supervision practices to one extent or another.
We assert that supervision of manualized treatment is actually a combination of both advanced training and supervision. The critical issue is one of identifying the point at which training ends and supervision begins. Holloway and Neufeldt (1995) described the supervision of manualized treatment as follows:
Technical adherence is improved with manualized training, but the quality of the therapeutic relationship appears to suffer. In addition, it appears easier to teach people technical adherence than the selection of problem and timing of delivery. This suggests that the actual timing and delivery of skills must be exposed to expert feedback until the trainee is capable of assimilating the larger principles that govern the timing, judgment, and appropriateness of delivering therapeutic treatment within the context of an ambiguous and changing social interactional medium. (p. 21l)
Holloway and Neufeldt certainly are describing a situation that calls for supervision. For the trainee to "assimilat[e] the larger principles that govern" referred to calls not only for "expert feedback" but for an invitation to reflect and to consider and reconsider the context and process of change.
But given the trend to manualize therapy, why not then also manualize supervision? We have already noted that Neufeldt et al. (1995) developed such a manual. Moreover, the descriptions in some supervision models (e.g., Kagan, Holmes, & Kagan's, 1995, description of Interpersonal Process Recall) are actually described in such sufficient detail that they constitute manuals, although they have not been thought of in that way. It may be, then, that manualized supervision is newer as a concept than as a practice.
The typical supervision practitioner functions in an integrative manner (see e.g., Bernard, 1997) without manuals. However, there are two important virtues for developing supervision manuals: (a) to serve as training tools (see Neufeldt, 1994), and (b) to allow studies of supervision efficacy and effectiveness. Each of these is sufficiently important to prompt additional work on developing manuals.
Evaluation and Measurement
Most evaluation instruments to date have blended training and supervision experiences and have been used to evaluate trainees who are either in training or transitioning between training and supervision. Differentiating training and supervision should lead to differentiating the criteria around which evaluation tools are created.
Again, we must point to the importance of the supervisory relationship and revisit the fact that so many evaluations of supervision rely on measures of trainee satisfaction. These data are important as one marker of success; at the same time, they are insufficient. Counselor education programs might consider expanding the evaluation process in ways similar to admissions procedures. Most programs would not rely on a single faculty member evaluating the potential of a candidate for graduate study. And yet, the same person who conducts training and supervision of a student is also the sole evaluator of that student's clinical competence. Perhaps some of the relationship variables so important to supervision could be screened out of the evaluation process by asking for independent evaluations based on samples of the trainee's thinking and intervention skill. The challenge, of course, is for faculty to determine criteria and develop measures that would reflect these criteria. This is more likely to be done, however, if evaluation becomes a less solitary act.
Earlier we alluded to Paul's (1967) set of clinical questions (What works, with whom, under what circumstances, etc.?). Recent critiques of the supervision literature have suggested that this is perhaps the direction supervision research should take. From a totally different vantage point, Salvadore Minuchin has been quoted as saying that there is no excellent therapist under the age of 50. This is consistent with Mowrer (1951) when he asserted years ago, "I do not say that a graduate student in his [sic] early twenties cannot have or acquire the purely personal qualifications which are prerequisites for therapeutic proficiency; the odds are against him" (p. 276). There is an Eriksonian (Erik) slant that progress in the field of counseling is slow, subfie, and perhaps resistant to empirical investigation. Somewhere between Paul's (1967) question and the provocation of both Minuchin and Mowrer is a viable posture to pursue in the coming years.
In addition to such a classic dichotomy of expectations, counselor educators are now engaging in Internet supervision (a new millennium indeed!) before we even have any evidence that particular models are more salient than others in a person-to-person format. The implications of technology on what we know (and don't know) about supervision simply boggles the mind of anyone who has spent time reflecting current practice. While we attempt to catch up with a plethora of new training and supervision possibilities, we may need to step back and paraphrase Paul's (1967) question by asking "What is training, what is supervision, and what does each do best using what model(s)?"
- Goodyear, Rodney, & Janine Bernard, Clinical Supervision: Lessons From the Literature, Counselor Education & Supervision, Sep 1998, Vol. 38, Issue 1.
The article above contains foundational information. Articles below contain optional updates.
Reflection Exercise #5
The preceding section contained information
about lessons from the literature on clinical supervision. Write three case study examples
regarding how you might use the content of this section in your practice.
What insufficient data do many supervision evaluations focus on to measure success?
Record the letter of the correct answer the