---
name: 360-feedback-synthesis
description: Synthesizes 360-degree survey responses into privacy-safe competency scores, rater-group themes, self-versus-other gaps, and a focused coaching agenda. Use for multi-rater reviews, leadership development, performance feedback, anonymized comment analysis, and competency calibration.
license: Apache-2.0
metadata:
  adlass.categories: "hr-people/performance-development"
  adlass.industries: ""
  adlass.tags: "360-feedback,multi-rater,competency,anonymization,coaching,performance-review"
  adlass.adaptation: "reference-doc"
  adlass.source: "original"
  adlass.version: "1"
---

# 360 feedback synthesis

## Purpose

Convert completed 360-degree assessments into an anonymized view of competency results. Separate self-ratings from eligible observer ratings, show response denominators, identify repeated behavior themes, and turn evidence-backed gaps into a coaching agenda.

## Scope

Cover one participant, one review cycle, the survey response table, competency rubric, role profile, and a prior-cycle report when its competency IDs and rating scale match. Include self, manager, peer, and direct-report groups. **Excluded:** promotion or compensation decisions, protected-characteristic inference, diagnosis, and comment attribution.

## Data basis

- Survey table with `participant_id`, `rater_group`, `competency_id`, `rating`, `comment`, `response_status`, `cycle_id`, and `response_id`.
- Competency rubric with scale anchors and behavioral indicators.
- Role profile describing expected behaviors.
- Prior-cycle score table with cycle, competency, mean, denominator, and citation.

## Result

Produce a competency comparison sheet with self mean, observer mean, group means, denominators, gap, and citations; a masked theme register with supporting response IDs; and a coaching agenda linking each focus to a competency and observable behavior.

## Quality criteria

- Exclude blank, withdrawn, and draft responses from every denominator.
- Suppress any rater group with fewer than three eligible responses.
- Show the rating scale, denominator, and response count beside every mean.
- Require two independent comments or two rater groups before naming a theme.
- Cite response IDs, rubric sections, role-profile sections, or prior-cycle pages for each finding.

## Instructions

Group by `participant_id`, `cycle_id`, and `competency_id`. Treat `rater_group=self` as the self score and other eligible groups as observer evidence; never mix them into one mean. Calculate the self-versus-observer gap only when both means have denominators. Mask names, email addresses, distinctive incidents, and recognizable writing patterns in comments. Keep observed wording, interpretation, and coaching proposal separate. A prior-cycle comparison is valid only when competency IDs, scale anchors, and participant scope agree.

## Adapt before use

- Add the approved competency rubric and rating-scale anchors.
- Confirm the anonymity minimum and local response-status vocabulary.
- Map rater-group labels and the review-cycle identifier.
- Define the coaching-agenda format and audience.