[ Research Studies ] [ Research Instruments ] [ Research Methodology ] [ Research-Based Therapy Protocols ] [ Promising Leads for Research ]

Proposed Criteria for Demonstrating Empirical Support for Humanistic and Other Therapies

Download humanist.doc

(Working Draft, ©R. Elliott, 9/2000)

It is essential that humanistic and experiential therapists become more involved in research on the process and outcome of their therapies. These proposed criteria are presented as an alternative to restrictive criteria published by the Society for Clinical Psychology (Division 12 of the American Psychologistic Association) and other similar approaches. These approaches emphasize randomized clinical trials over all other forms of research on therapy effectiveness; we believe that this is unduly restrictive condition and prejudices conclusions against humanistic, psychodynamic and strategic therapies.

To this end, I have drafted the following proposed criteria for demonstrating empirical support for humanistic and other therapies. This is an ongoing effort which has been taken up by a task force of the Division of Humanistic Psychology (Division 32) of the American Psychological Association. We are therefore making this draft version available at this time, in order to invite comments and suggestions.

General Approach. These draft criteria propose a self-reflective attempt to test one’s assumptions regarding the efficacy of a favored treatment. This systematic/critical strategy requires that the conditions listed below be met in order to designate a therapeutic approach as empirically supported. At the same time, empirical support is not seen as an absolute or dichotomous judgement, but rather a matter of the degree of confidence with which one can generalize efficacy conclusions developed with one client or group of clients to other, similar clients. Humanistic and other therapists realize that each client is unique and that generalizations must always be taken as tentative hypotheses rather than foregone conclusions.

1. Client Change: Client(s) show reliable, clinically-meaningful change over the course of therapy. (Exception: maintenance of functioning where deterioration is expected.) This demonstration of change can take the form of either or both of the following:

• Quantitative/prospective: repeated measurement across time on standard quantitative measures.
• Qualitative/retrospective: client written or interview accounts after all or some of therapy.

In assessing client change, the demonstration will be more convincing is it includes some questions or measures that address:

1. Clients’ initial problems or goals, and
2. Theory-relevant processes (e.g., openness to experience).

In addition, demonstrations which offer corroboration from multiple perspectives or using multiple methods will be more convincing.

2. Client Specification: Client(s) are described in relevant, useful terms (e.g., gender, age, ethnicity, presenting problems, goals or diagnoses, status on theory-relevant variables such as experiencing). Specifying who the client(s) are makes it possible to generalize conclusions about efficacy, on the basis of important case characteristics. Demonstrations which use some form of systematic description of important client characteristics will be more convincing. Note that pre-DSM-III research can be included (e.g., "neurosis" can be translated as mixed anxiety/depression).

3. Therapy Specification: The researcher provides evidence that the desired therapy was in fact carried out. This may involve use of:

• General, nonrestrictive treatment manual; or
• Adherence or competence checks by raters, supervisor or therapist (e.g., ratings of warmth, empathy and genuineness for Client-Centered therapy)

Evidence on the nature of the therapy carried out can be either quantitative or qualitative, or both. In addition, relevant characteristics of the therapy process are provided, including basic descriptive characteristics of the therapist(s) (e.g., gender, ethnicity, age, suborientation, experience, supervision). This descriptive information facilitates generalization of efficacy conclusions.

4. Efficacy Demonstrations: The researcher provides efficacy demonstrations using one or more of the following group or systematic single case designs:

a. Group designs:

• Single-group pre-post or time series design, especially in naturalistic settings.
• Process-outcome correlational study (e.g., level of therapist empathy predicts change in client depression)
• Attitude X treatment interaction design (e.g., depressed clients with primarilly interpersonal issues do better with empty chair work than depressed clients with primarily self-critical issues).
• Randomized clinical trials, with control groups consisting of either no treatment, wait-list or alternative therapy conditions.

As noted earlier, such group designs may use either quantitative outcome measures or systematic posttreatment client reports or interviews, or both.

b. Systematic single case designs: Efficacy demonstrations can include one or more single cases in which there is systematic (not necessarily statistical) data collection and analysis. These are new and emerging approaches include the pragmatic case study (Fishman, 1999); hermeneutic single case efficacy designs (which use "thick description" to rule out alternative explanations) (Elliott, in press); adjudicational (legal) model case studies (Bohart, 2000); and multiple case depth research (K. Schneider, 1999).

5. Replication: Some kind of replication, either within or across studies, is present. At a minimum, this involves consistent efficacy demonstrations using multiple measures of client change; multiple instances of the same design; or multiple designs. Demonstrations will be more convincing if they involve multiple clients; they will be more convincing yet is they involve multiple studies; more convincing yet if they involved multiple research teams or multiple designs. Systematic data cumulation also contributes to the credibility of the demonstration; this may involve meta-analyses of group designs or application of case-based reasoning strategies using quasi-judicial approaches such as peer-reviewed case databases (Fishman, 1999). The more the replications, and the more varied the replication, they more convincing the demonstration of efficacy.

[ Home ] [ Up ] [ What's New ] [ Site Map ] [ Feedback ] [ Search NREP ]

Send mail to experiential-researchers@focusing.org with questions or comments about this web site.
Copyright © 2000 Network for Research on Experiential Psychotherapies
Last modified: March 20, 2001