wiki content export 2026-04-22

This commit is contained in:
Wiki Bot 2026-04-22 04:00:06 +00:00
parent 16aef35682
commit 656c86e6a7
99 changed files with 2277 additions and 3 deletions

View file

@ -0,0 +1,50 @@
---
title: Glossary
collection: Hub User Guide
path: Hub User Guide/Reference/Glossary
parentDocument: Reference
outlineId: 6f6f07f0-86e2-44f6-95f0-60f3273e30c0
createdBy: Jennie R.F.
---
Key terms used throughout Cohort-OS and this documentation.
---
**Blind review** -- A stage setting that hides reviewer identities from other reviewers when viewing consensus scores. Reviewers appear as "Reviewer 1", "Reviewer 2", etc. Admins always see full names regardless of this setting. Enabled by default on all stages.
**Cohort** -- A group of applicants going through a selection process together. Each cohort has its own application form, review stages, reviewers, and decisions. Cohorts move through lifecycle statuses: `draft`, `review`, `decide`, `onboard`, `active`, `closed`.
**Cohort admin** -- A user with the `cohortadmin` role. Can manage applications, assign reviewers, configure stages, make decisions, and send notifications for cohorts they have access to. Cannot manage organization-level settings or users.
**Consensus** -- The aggregated view of all completed reviews for a single application within a stage. Shows average scores per criterion, overall averages, score spread, and a tally of reviewer recommendations. Used by admins and committees to inform decisions.
**Criterion** -- A single dimension of evaluation within a rubric (e.g., "Mission Alignment", "Governance Readiness"). Each criterion has a name, description, maximum point value, and optional scoring thresholds that describe what each performance level looks like.
**Decision** -- The final outcome recorded for an application: `accepted`, `waitlist`, or `declined`. Decisions include a rationale, rationale code, and optional committee notes. Recorded in the **Decisions** tab.
**Flywheel** -- The external assessment engine that Cohort-OS integrates with. Flywheel provides rubrics (scoring criteria and thresholds), surveys (application form question sets), and reporting capabilities. Rubrics are authored in Flywheel and linked to review stages in Cohort-OS.
**Ghostie** -- The avatar system in Cohort-OS. Each user chooses a ghostie -- a small ghost illustration defined by an expression (sweet, mild, exasperated, wtf, disbelieving, double-take) and a body color (from presets or a custom color picker). Ghosties appear on user profile circles throughout the app.
**Guided scoring** -- The primary review interface where reviewers score applications criterion by criterion. In focused mode, one criterion is shown at a time with the applicant's answers alongside. In "show all" mode, every criterion is visible at once. After scoring all criteria, reviewers proceed to the recommendation step.
**Org** -- Short for organization. The top-level entity that owns cohorts, users, and settings. Users can belong to multiple orgs via org memberships. All data is scoped to an org.
**Orgadmin** -- A user with the `orgadmin` role for an organization. Full management access: can manage users, cohorts, settings, and view the activity feed. Can do everything a cohort admin can, plus organization-level administration.
**Orphaned interview** -- A Cal.com booking that could not be automatically matched to an application. This happens when the email on the booking does not match any applicant's contact email in the cohort. Orphaned interviews appear in a modal accessible from the **Applications** tab, where an admin can manually match them to the correct application or dismiss them.
**Permission (view/review)** -- The access level assigned to a reviewer for a specific application. `review` permission allows the reviewer to score the application and submit a review. `view` permission gives read-only access to the application data without the ability to score. View-only reviews are excluded from consensus calculations.
**Rationale code** -- A short code attached to a decision that categorizes the reasoning (e.g., `manual_accept`). Used alongside the free-text rationale to provide structured data about why a decision was made.
**Recommendation** -- A reviewer's suggested outcome for an application, selected after scoring. Four options: **advance** (move to the next stage), **reject** (decline the application), **hold** (flag for further committee discussion), **undecided** (no recommendation yet). Recommendations inform but do not determine the final decision. Reviewers can change their recommendation after submission by providing a reason.
**Reviewer** -- A user assigned to evaluate applications. Reviewers are first added to a cohort's reviewer pool, then assigned to specific applications with either `view` or `review` permission. Reviewers access their assignments through the review dashboard and score applications using the guided scoring interface.
**Rubric** -- A structured scoring framework defined in Flywheel and linked to a review stage. Contains multiple criteria, each with descriptive thresholds/levels. Rubrics standardize evaluation so all reviewers assess applications against the same dimensions.
**Self-assessment** -- A survey sent to members of an applicant's team (e.g., studio members). Each team member receives a unique token-based link that does not require login. Admins can track completion status, resend emails, and view responses. Self-assessment data is available to second-round reviewers during scoring.
**Stage** -- A phase within a cohort's review process. Common stage types are `application-reviews` (initial paper review), `first-round` (first interview round), and `second-round` (second interview round). Each stage has its own rubric, reviewer settings (minimum/maximum reviewers, blind review, self-review), and status. Stages are ordered sequentially and applications advance through them.