Apps

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

Citation Bias

Also Known As: Selective citation Citational favoritism
Statistical Error ID: citation_bias

Definition

Citation bias occurs when studies with statistically significant or positive results are cited more frequently than studies with null or negative results. This creates a distorted impression of the state of evidence: highly cited studies appear more authoritative and important, while uncited null studies become invisible. Over time, a false consensus can emerge because researchers, reviewers, and policymakers encounter the positive evidence repeatedly while the contradicting evidence accumulates citations too slowly to influence the discourse.

Examples

A landmark study claiming a link between a certain food additive and hyperactivity in children is cited over 1,000 times. Three subsequent studies finding no link are cited fewer than 50 times each. A policy review heavily influenced by citation counts concludes the evidence strongly supports the link.

A widely shared social media post links to a 2015 study claiming a specific gut bacteria supplement boosts memory, which has since accumulated 3,400 citations. Two rigorous randomized controlled trials published in 2017 and 2019 finding no cognitive benefit have been cited 31 and 44 times respectively, and are rarely mentioned in popular science coverage.

In climate policy discussions, a controversial paper suggesting a pause in global warming attracted thousands of citations and extensive media debate. Four methodologically stronger studies published concurrently that reaffirmed the warming trend each received fewer than 200 citations, making the scientific consensus appear more contested than it actually was.

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

    Are studies with significant results cited more frequently than those with null results?

    Type: binary
  2. 2

    Does the citing literature selectively reference supportive studies while ignoring contradictory evidence?

    Type: binary
  3. 3

    Could high citation counts be creating a false impression of the strength of evidence?

    Type: binary
  4. 4

    Has the citing paper conducted a balanced review of both supportive and contradictory studies?

    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.

Hierarchical Context