Assuming that one member's traits represent the whole group, or that a group decision reflects every individual's view.
Imagine you meet one kid from another school who is really loud. Then you think, 'All kids from that school must be loud!' Or imagine a class votes to skip recess and you think every single kid wanted to skip, even though maybe only half of them did and the others just lost the vote. That's the group attribution error—treating one person or one decision as if it speaks for everybody in the group.
The Group Attribution Error operates in two distinct forms. First, people observe the behavior or traits of one group member and generalize those characteristics to the entire group, even when told the individual is atypical. Second, people assume that a group's collective decision—such as a vote or policy outcome—reflects the private attitudes of every member of that group, ignoring the decision rules, dissent, or structural constraints that shaped the outcome. The error intensifies when the target group is perceived as an outgroup, adversarial, or homogeneous, and it tends to disappear when people evaluate their own groups, where they more readily appreciate internal diversity and situational pressures.
The same glitch looks different depending on the terrain. Finance, medicine, a relationship, a team — same mechanism, different costume.
Investors may judge an entire company's workforce as incompetent or brilliant based on the actions of a single executive, leading to disproportionate stock reactions to individual scandals or successes that do not reflect the organization's overall capability.
A clinician encountering one non-compliant patient from a particular demographic group may begin to assume patients from that background are generally non-compliant, adjusting communication or treatment intensity based on group membership rather than individual assessment.
Scott T. Allison and David M. Messick, 1985. Formalized in their paper 'The Group Attribution Error' published in the Journal of Experimental Social Psychology. Earlier related work on generalizing from atypical cases was done by Hamill, Nisbett, and Wilson in 1980.
In ancestral environments, rapidly categorizing other tribal groups as uniformly friendly or hostile based on limited encounters with individual members would have been a survival-relevant heuristic. Treating an outgroup as monolithic allowed faster threat assessment when inter-tribal encounters were rare and the cost of misreading a hostile group was potentially fatal.
Machine learning models trained on group-level labeled data can inherit group attribution errors by learning to associate individual predictions with aggregate group characteristics. For example, a hiring algorithm trained on department-level performance scores may penalize all applicants from a low-performing department, ignoring individual variation. LLMs can reproduce stereotypical group attributions found in training data, generating text that treats groups as homogeneous agents.
This page taught you the name. The deck turns the name into reflex. 1,100+ swipeable scenarios, 1,100+ defenses, 650+ detection prompts — spaced-repetition Swipe Deck, unlimited Spot-the-Bias Quiz, Defense Playbook, Pre-Flight, My Blindspots, Cheat Sheets, Field Guide e-book. $39.53$59.