The Ambivalent Ageism Scale: Developing and Validating a Scale to Measure Benevolent and Hostile Ageism.

Purpose: Much like sexism, ageism is a multifaceted prejudice; it involves benevolent and hostile attitudes toward older adults. There are many scales designed to measure hostile ageism, yet none dedicated to measuring benevolent ageism. In the current studies, we developed and validated a 13-item measure: the Ambivalent Ageism Scale (AAS). Design and Methods: We employed four stages of scale development and validation. In Stage 1, we created 41 benevolent ageist items adapted from existing ageism measures. In Stages 2 and 3, we further refined the pool of items through additional testing and factor analysis and retained nine items loading strongly on two factors related to benevolent ageism: cognitive assistance/physical protection and unwanted help. In order to enable researchers to contrast benevolent and hostile attitudes, we then added four hostile ageist items. In Stage 4, we assessed the test-retest reliability of the 13-item scale. Results: The AAS had good test-retest reliability (r = .80) and good internal consistency (α = .91). As predicted, the benevolent and hostile ageism subscales differentially predicted attitudes toward older adults: higher scores on the hostile subscale predicted lower competence and warmth ratings, whereas higher scores on the benevolent subscale predicted higher warmth perceptions. Implications: The AAS is a useful tool for researchers to assess hostile and benevolent ageism. This measure serves as an important first step in designing interventions to reduce the harmful effects of both hostile and benevolent ageism.

Comparing young and older adults' perceptions of conflicting stereotypes and multiply-categorizable individuals.

Individuals can be simultaneously categorized into multiple social groups (e.g., racial, gender, age), and stereotypes about one social group may conflict with another. Two such conflicting stereotype sets are those associated with older adults (e.g., frail, kind) and with Black people (e.g., violent, hostile). Recent research shows that young adult perceivers evaluate elderly Black men more positively than young Black men, suggesting that components of the elderly stereotype moderate the influence of conflicting Black stereotypes (Kang & Chasteen, 2009). The current research begins to examine whether this pattern of perceiving multiply-categorizable individuals is maintained among older adults or altered, perhaps due to aging-related cognitive and motivational changes. In three studies using different targets and evaluative tasks, both young and older participants showed evidence of an interplay between Black and elderly stereotypes, such that they perceived elderly Black targets more positively than young Black targets. A similar pattern was observed when assessing emotion change (Study 1), making ratings of warmth and power in the past, present, and future (Study 2), and when directly comparing young and old Black and White targets on traits related to warmth and power (Study 3). The absence of age differences suggests that evaluation of multiply-categorizable targets follows comparable underlying patterns of stereotype activation and inhibition in younger and older adults.

Does age group identification differentially influence younger and older adults' intergenerational perceptions?

People's knowledge that they will change age groups over time likely makes relations between age groups function differently from relations between racial or gender groups. We tested this possibility by examining whether people's identification with their current age group influenced their responses to threats to the status of their current age group. We found that age group identification was a good predictor of responses to status threat, particularly among young adults. Specifically, young adults who were strongly identified with their age group showed patterns of bias typically found in strongly identified members of other groups, whereas older adults did not. These results provide further evidence that age groups function differently from other group memberships and that models of intergroup conflict need to be tested in groups for whom memberships change over time.