This meta-analytic review of 232 effect sizes showed that, across five categories, attitudes were more negative toward older than younger adults. Perceived age differences were largest for age stereotypes and smallest for evaluations. As predicted by social role theory (Eagly, 1987), effect sizes were reduced when detailed information was provided about the person being rated. The double standard of aging emerged for evaluations and behavior/behavioral intentions, but was reversed for the competence category. Perceptions depended on respondent age also. Results demonstrated both the multi-dimensionality and the complexity of attitudes toward older adults (Hummert, 1999; Kite & Wagner, 2002).For over half a century, gerontologists have puzzled over North Americans' fascination with youth and their reluctance to accept aging gracefully. Initially, research on these issues was based on the assumption that negative attitudes toward older adults were widespread (Butler, 1969). Subsequent reviewers have reached different, and sometimes opposite conclusions. At the same time that Green (1981) 242 Kite, Stockdale, Whitley, and Johnson determined that negative age-related stereotypes were the norm, Lutsky (1980) concluded that age, in and of itself, was less important in determining attitudes toward older adults than were other types of information. More recent reviewers (Hummert, 1999; Kite & Wagner, 2002) echoed McTavish's (1971 conclusion; the answer to the seemingly simply question "Does ageism exist?" is not an unequivocal yes. Instead, people's views about aging are multidimensional, with both positive and negative elements.This article examines the complexities of age-related attitudes and stereotypes through a meta-analytic review of literature on ageism. Specifically, we update and expand upon Kite and Johnson's (1988) previous meta-analysis. Kite and Johnson reviewed the literature on this topic prior to December 1985, examining a total of 43 independent effect sizes. Thirty of those effect sizes indicated that people were more negative toward older than toward younger people, eleven indicated more negative attitudes toward younger than toward older people, and two indicated exactly no difference in attitudes toward the two groups. The overall effect size, as indexed by the d statistic was 0.39, indicating that attitudes toward older persons were more negative than attitudes toward younger persons by approximately onethird of a standard deviation.Although this finding indicated a bias against older adults, these effect sizes were not homogeneous. That is, the difference in attitudes expressed about older and younger adults varied widely across studies. Therefore, potential moderators of the effect sizes were explored to account for this variation. Results supported Lutsky's (1980) contention that people do not rely solely on information about age in judging older adults. Specifically, Kite and Johnson (1988) found that the context provided for the ratings moderated the results. That is, when specific info...