Base Rate
How often a candidate cause, such as a software update or a schedule change, occurs in general without producing the failure under investigation. A correlation in time with a rare event is stronger evidence than a correlation with something that happens constantly and usually causes no harm; checking the base rate before treating a correlated change as a likely branch guards against overweighting the first coincidence noticed.
Defined in 6 GAGE programs, which carry 10 distinct definitions of it. The wording above is taught in Engineering Judgment and Professional Formation.
How each discipline defines it
The same term does different work depending on who is using it. These are the definitions as each program teaches them, unedited.
How often a candidate cause, such as a software update or a schedule change, occurs in general without producing the failure under investigation. A correlation in time with a rare event is stronger evidence than a correlation with something that happens constantly and usually causes no harm; checking the base rate before treating a correlated change as a likely branch guards against overweighting the first coincidence noticed.
How common an outcome is in a dataset (for example, the fraction of cases that are actually fraud). A high accuracy figure can be meaningless if the base rate is very low, because a model can score well simply by always predicting the common outcome, which is why the base rate is a required companion to any accuracy claim.
What usually happens in situations of a given type, used as the anchor against which a specific prediction is judged. A prediction that departs sharply from the base rate must justify the departure; ignoring base rates for a vivid specific claim is a classic judgment error.
The underlying frequency of a condition in a population. Ignoring it is what makes a red flag screening system produce a flagged list that is almost entirely innocent, however accurate the indicators look in isolation.
The historical frequency or cost of an event (a fraud loss, an error, a stockout) used to anchor a risk-reduction claim. Without a base rate, a claimed reduction is a guess; with one, it is a defensible number.
Where it is taught
The exact lessons this term appears in. The first module of every program is free with a free account.
- Interconnected Literacies: How Digital, Data, and AI Literacy Connect · Digital Foundations, Certified AI Practitioner: Workplace Foundations
- AI for Innovation: Predictive Analytics Applications · Advanced AI Literacy, Certified AI Practitioner: Workplace Foundations
- Train a model with your own hands and watch what it actually learns · Build Before You Govern, Certified AI Governance Professional (CAIGP)
- Where the bias came from: tracing a bad prediction to its data · Build Before You Govern, Certified AI Governance Professional (CAIGP)
- Fixing the model, breaking it again: why fixes are never free · Build Before You Govern, Certified AI Governance Professional (CAIGP)
- Labor Market and Innovation Economics: Displacement, Creation, Productivity · AI Economics and Impact Modeling, The AI Lobbyist: Certified AI Policy Strategist
- Value Creation and ROI Modeling: Your Three-Scenario Value Model · The Economics and the Business Case, Certified AI Transformation Professional (CATP)
- Fraud psychology, assessment, and detection · Financial Risk, Fraud, and Financial Crime, Certified Enterprise Governance, Risk and Compliance Professional (CEGP)
- Root cause analysis that finds the real cause · Failure Analysis, Learning From How Robots Break, Engineering Judgment and Professional Formation
- Proving a robot saves or makes money · The Economics of Robots, the Business Case, Engineering Judgment and Professional Formation
Where it shows up at work
A definition is where Base Rate starts. The lessons above teach it inside the competencies below, which are what job descriptions ask for and what the GAGE credential records. Every link here is drawn from the same careers map that builds a role's readiness check.
Builds an honest value model: baseline, measured change, cost, risk, and the projects that should be stopped.
2 lessons on this page teach it as a primary outcome.
Decides where AI should be used, where it should not, and which use cases create measurable value, tied to the organization's priorities.
One lesson on this page teach it as a primary outcome.
Uses generative AI tools daily, knows what a model can and cannot do, and can say where an output should not be trusted.
One lesson on this page teach it as a primary outcome.
Roles that hire for it
Each role page shows the full competency set, what the seat decides, and a readiness check against your own evidence.
- Chief AI Officerneeds it at depth
- Chief Information Officer, technology leadership editionneeds it at depth
- AI Adoption and Enablement Leadneeds it at depth
- Operations Manager, AI-readyneeds it at depth
- AI Program Managerneeds working knowledge · 11 open postings this week
- Chief Technology Officer, AI engineering editionneeds working knowledge
How job descriptions say it
Postings rarely use the textbook word. These are the phrases the careers map reads as asking for the competencies above.
Where the frameworks place it: EU AI Act Article 4; DOL AI Literacy Framework.
One interview question about it
How would you measure the return on an AI deployment honestly, including the costs people prefer to leave out?
Asked of candidates for roi and value measurement for ai. Answering it in your own words is exactly how a GAGE topic is passed, and the credential records the pass where an employer can check it.
If you come from another field
Some backgrounds already carry this competency, and the readiness check gives credit for it.
- A business, MBA or technology management degree carries it as a claim to back up: Value and return measurement are a course you passed.
- Recent graduate in law, policy, business or data carries it as a claim to back up: Most graduates already use the tools daily.
- Teachers, trainers and instructional designers carries it as a claim to back up: Most educators are already using AI in preparation.
Terms it appears with
Not an alphabetical neighbourhood: these are the terms taught in the same lessons, ranked by how often they appear together.