> ## Documentation Index > Fetch the complete documentation index at: https://docs.getrivet.ca/llms.txt > Use this file to discover all available pages before exploring further. # WAI-SR — tracking the therapeutic alliance > The Working Alliance Inventory, Short Revised — 12 client-rated items measuring task agreement, goal agreement, and the therapeutic bond. The WAI-SR is how you find out whether the working relationship is strong before it goes sideways. Twelve client-rated items across three subscales — task agreement, goal agreement, and the therapeutic bond. Sustained low scores are a signal to slow down and repair; a sudden drop session-over-session is a signal to ask what changed. ## What it measures Bordin's (1979) three-component model of the therapeutic alliance: * **Task agreement** — does the client agree with what you're doing in session? * **Goal agreement** — does the client agree on what you're working toward? * **Bond** — does the client experience the relationship as one of mutual respect and care? Strong alliance is the single most reliable common-factor predictor of therapy outcome across modalities — roughly 30 meta-analyses put the alliance-outcome correlation around r = 0.25-0.30. ## When to use it * After session 3-5 to establish an early-alliance baseline * Every 4-6 sessions to track the trend * Any time you sense a rupture * Before a termination decision — sustained low scores plus stalled outcome measures is a referral signal The WAI-SR is **client-rated by default** — the client is the source of truth on whether the alliance is working. ## How clients fill it out About three minutes self-administered. The wording uses an underlined blank where the therapist's name goes ("I believe \_\_ likes me") — the client mentally substitutes your name. Rivet's renderer presents the blank verbatim per the canonical instrument. ## How Rivet scores it Sum of 12 items, each rated 1-5 (Seldom / Sometimes / Fairly often / Very often / Always). Range 12-60. Higher = stronger alliance. | Total | Band | | ----- | --------------------------------------------- | | 12-35 | Weak alliance — explore rupture / repair work | | 36-47 | Moderate alliance | | 48-60 | Strong alliance | ### Three subscales (each 4-20) | Subscale | Items | | -------------- | ------------ | | Task agreement | 1, 2, 10, 12 | | Bond | 3, 5, 7, 9 | | Goal agreement | 4, 6, 8, 11 | The per-subscale scores are exposed so you can see whether a low total is driven by misalignment on tasks, on goals, or by the bond — three very different conversations to have. ## How to read it The WAI-SR is a **continuous tracking variable**, not a diagnostic cutoff. Trend across sessions matters more than any single absolute score. A drop from session N to N+1 is the actionable signal — that's when to ask "is something we're doing not working for you?" ## When NOT to use it The WAI-SR asks the client to evaluate the relationship — which is itself a relationship event. Some clients with interpersonal trauma or strong people-pleasing patterns will inflate scores to keep you happy. Pair the measure with your own clinical read. ## Is it free to use? **Yes.** Free academic and clinical use, per Hatcher. ## Citation Hatcher, R. L., & Gillaspy, J. A. (2006). Development and validation of a revised short version of the Working Alliance Inventory. *Psychotherapy Research,* 16(1), 12-25. Based on Horvath, A. O., & Greenberg, L. S. (1989). Development and validation of the Working Alliance Inventory. *Journal of Counseling Psychology,* 36(2), 223-233. Free academic / clinical use per Hatcher. ## Related articles Reading a WAI-SR trend the way you'd read a PHQ-9 trend — direction over time, not single-point absolutes. What counts as a meaningful WAI-SR shift between sessions. Patterns for routine post-session WAI-SR administration without it feeling clinical or interruptive.