An "evidentiary standards" approach was used to examine how standards for trait diagnosis may become more lenient or stringent depending on category membership of the target being judged and the type of standard-minimum or confirmatory-being referenced. Three studies demonstrated that minimal standards are lower but confirmatory standards are higher for members of groups assumed to be deficient on a trait relative to those stereotyped as having the trait (e.g., men on emotionality, women on competitiveness and aggression), though this was more typical among judges who were themselves members of the group stereotyped as having the trait in question. It was also the case that for targets who belonged to groups stereotyped as possessing the relevant trait, minimal, and confirmatory standards did not differ: When one expects to see evidence of a trait, suspicion that a target possesses it readily becomes certainty.