The Alliance, Graded: A Free Entry Point to the Relationship Evidence
- A free, open web summary from the Society for the Advancement of Psychotherapy (APA Division 29) presenting the conclusions and recommendations of the Interdivisional (APA Divisions 12 and 29) Task Force on Evidence-Based Therapy Relationships, compiled by John C. Norcross; written for clinicians, supervisors and trainers who want the graded evidence rather than slogans.
- Free to read online, no login or purchase; citable as Norcross (2014); it is the second task force's synthesis, since carried forward into the expanded Oxford University Press reference of the same name.
- It sorts each relationship element by strength of evidence, from demonstrably effective (the alliance in individual, youth and family therapy; group cohesion; empathy; collecting client feedback) down to promising but insufficient research to judge (congruence/genuineness, repairing alliance ruptures, managing countertransference).
- Read it at societyforpsychotherapy.org under evidence-based therapy relationships.
"The alliance predicts outcome" has become one of the most repeated sentences in our field, and lately one of the most abused, invoked to sell apps, chatbots and weekend certifications. Before the phrase can mean anything at a case conference, a clinician needs to see what the evidence actually grants the relationship, element by element, and where the confidence runs out. This free, open summary from the Society for the Advancement of Psychotherapy is the most compact place to get exactly that.
What it is
The page reproduces the conclusions and recommendations of the Interdivisional (APA Divisions 12 and 29) Task Force on Evidence-Based Therapy Relationships, compiled by John C. Norcross, Distinguished Professor of Psychology at the University of Scranton and a board-certified clinical psychologist. It is not a blog post or a marketing landing page: it is the citable summary (Norcross, 2014) of a panel that reviewed the meta-analytic evidence for each component of the therapy relationship and then rated it. The fuller argument is developed at book length; this page is the free door to its bottom line.
The grading, and why it matters
The task force sorts relationship elements into three tiers. Demonstrably effective are the alliance in individual, youth and family therapy, cohesion in group therapy, empathy, and collecting client feedback. A middle tier is judged probably effective. A third tier, congruence/genuineness, repairing alliance ruptures and managing countertransference, is labelled promising but insufficient research to judge. That last column is the corrective the current hype omits: several relational moves clinicians treat as established are, on this reading, not yet proven. The panel is candid about the ceiling, too. Its conclusions "do not by themselves constitute a set of practice standards," and the underlying evidence base is largely correlational.
What the task force does defend, it defends firmly. The relationship "makes substantial and consistent contributions to psychotherapy outcome independent of the specific type of treatment," and the panel warns that efforts to promote evidence-based practices "without including the relationship are seriously incomplete and potentially misleading." It pairs that with concrete recommendations: make cultivating the relationship a primary aim, adapt it to the individual patient rather than to the diagnosis alone, and routinely monitor the patient's response so ruptures can be caught and repaired.
How to use it
Fifteen minutes with this page is worth having before you next teach, supervise or argue about "common factors." Trainers can lift the three-tier structure straight into a seminar; supervisors can use the feedback-monitoring recommendation to justify routine outcome measurement; and any clinician offered a product that promises "a therapeutic alliance" can check the promise against the column it actually belongs in. Treat the page as the index, not the whole book: it points to the fuller treatment, the Oxford University Press reference of the same name, Psychotherapy Relationships That Work.
The relationship earns demonstrably effective, but several moves we treat as established sit, honestly labelled, under insufficient research to judge.
The summary is free but abbreviated, and it reflects the second task force (Norcross, 2014); the fuller, current synthesis is the paid Oxford University Press reference. The underlying evidence is largely correlational, and the panel stresses that its conclusions are not practice standards.