Apps

🧪 This platform is in early beta. Features may change and you might encounter bugs. We appreciate your patience!

Flawed Comparison

Also Known As: Apples to oranges Misleading comparison False equivalence (statistical form) Baseline manipulation
Aspect 📰 Media Bias ID: flawed_comparison

Definition

Flawed comparison occurs when media coverage juxtaposes two entities, events, statistics, or claims in a way that implies equivalence or contrast where none meaningfully exists. The comparison may involve different scales (comparing absolute numbers with rates), different time periods (cherry-picked baselines), different populations (demographics that differ on confounding variables), or different definitions of the same term across contexts.

Examples

A business story compares a company's quarterly profit margin to a competitor's annual margin, presenting both as 'profit margin' without clarifying the time period difference — making the company appear more or less profitable than the comparison actually supports.

A political story compares crime rates across cities using absolute numbers rather than per-capita rates. A city of 5 million is shown to have 'more crime' than a city of 500,000 — a comparison that tells us nothing about safety and actively misleads readers about relative risk.

A healthcare story compares mortality rates from a new treatment tested on elderly patients to rates from an old treatment tested on younger patients, attributing the difference entirely to the treatment — without acknowledging that the structurally different populations make the comparison invalid.

Verification Steps
Verification Steps
Binary yes/no questions that an AI must answer to detect a reasoning pattern in a text.
Each of the 452 aspects has verification steps — simple yes/no questions designed to systematically detect whether a pattern appears in a text. For ad hominem: "Does the argument attack a person rather than their claim?" For false dichotomy: "Are only two options presented when more exist?" This ensures consistent, reproducible analysis.

Binary (yes/no) questions an LLM must answer to identify this aspect:

  1. 1

    Does the coverage compare two or more things to make a point about relative size, quality, risk, or value?

    Type: binary
  2. 2

    Are the compared items substantively different in ways that make the comparison misleading — different scales, time periods, populations, definitions, or contexts?

    Type: binary
  3. 3

    Would an accurate comparison require qualifications, context, or additional data that is absent?

    Type: binary
  4. 4

    Does the flawed comparison serve a rhetorical purpose — making one option look better or worse — rather than informing understanding?

    Type: binary
Deep Dive
The expandable detail section on each aspect page with examples, psychology, and counter-strategies.
The Deep Dive section provides in-depth information about each aspect: a real-world example showing the pattern in action, an explanation of why it works psychologically, practical advice on how to counter it, alternative names, and links to related aspects.