Apps

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

Undercoverage Bias

Also Known As: Frame bias Coverage error
Aspect ID: undercoverage_bias

Definition

Undercoverage bias occurs when some members of the target population have zero or near-zero probability of being included in the sample because the sampling frame does not cover them. Classic examples include telephone surveys that miss people without phones, or online surveys that miss people without internet access. The excluded groups often differ systematically on the very variables being measured.

Examples

A 1936 Literary Digest poll predicted a landslide victory for Alf Landon based on 2.4 million responses, but only surveyed car owners and telephone subscribers, systematically excluding poorer voters who overwhelmingly supported Roosevelt.

A city conducts a resident satisfaction survey by emailing registered voters. Renters who move frequently, undocumented residents, and people without stable internet access are almost entirely excluded, making satisfaction scores appear higher than they would if the full residential population were captured.

A national health survey is conducted via landline telephone calls during weekday business hours. Working-age adults, people with only mobile phones, and shift workers are systematically underrepresented, causing the survey to overestimate retirement-age demographics and underestimate health issues common in younger working populations.

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 sampling frame exclude identifiable segments of the target population?

    Type: binary
  2. 2

    Are the excluded groups likely to differ from the included groups on the outcome variable?

    Type: binary
  3. 3

    Is the sampling method (e.g., phone surveys, online polls) inherently inaccessible to some subgroups?

    Type: binary
  4. 4

    Is the claim generalized to the full population despite the exclusion?

    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.