Age-Related Stereotypes and the Linguistic Intergroup Bias 1 Running head: AGE-RELATED STEREOTYPES AND THE LINGUISTIC INTERGROUP BIAS An Examination of Age-Related Stereotypes and the Linguistic Intergroup Bias Using Two Measures

The linguistic intergroup bias is a phenomenon where people use more abstract language to talk positively about in-groups and negatively about out-groups (Maass, Salvi, Arcuri, & Semin, 1989). This has been established for many in-groups, but has not been extended to age-related stereotypes. This study extended the linguistic intergroup bias to attitudes towards older adults. It was predicted that statements about what participants liked about their peers and disliked about older adults would be more abstract than statements about what participants disliked about their peers and liked about older adults. Results supported these predictions. Also, a new measure of linguistic abstractness was tested, but was found not to be useful in this context. Age-Related Stereotypes and the Linguistic Intergroup Bias 3 An Examination of Age-Related Stereotypes and the Linguistic Intergroup Bias Using Two Measures Stereotypes and prejudice have been a major concern of social psychology for several decades, and research in the area is continuing. In the last 20 years, stereotype research has been extended to linguistics, and the role of language in creating and transmitting stereotypes. One area of relevant research is in the area of linguistic abstractness, specifically in the Linguistic Category Model (LCM; Semin & Fiedler, 1988). The research concerns the abstractness of language that is used to express stereotypes, and the descriptions of behaviors relative to stereotypes. Linguistic abstractness is an important area of study because it can give insight into how thoughts are converted to linguistic statements. Also, the nature of stereotypes and biases can be better understood. The other major reason that the language of stereotypes is being studied is because stereotypes are largely a social phenomenon and language is a primary means of communicating a stereotype. Stereotypical language also may play an important role in stereotype formation and retention. Also, the language of stereotypes gives some insight into the cognitive processes that underly stereotypes. There are two primary goals of this research. The first is to Age-Related Stereotypes and the Linguistic Intergroup Bias 4 develop and test a new measure of linguistic abstractness. This measure is far easier to use and may be more sensitive than the current measure of linguistic abstractness. Also, the proposed measure addresses some of the shortcomings of the current measure, both from a measurement standpoint and a linguistic standpoint. The second goal is to extend the Linguistic Intergroup Bias (LIB; Maass, Salvi, Arcuri, & Semin, 1989) to age-related stereotypes. The Linguistic Intergroup Bias model is a robust linguistic effect present in the language of stereotypes and has been generalized to many different groups and stereotypes. It has not been extended to agerelated stereotypes, which have been extensively researched and are consistently present across the population. Linguistic Abstractness and the Linguistic Intergroup Bias Linguistic abstractness is one dimension that is currently examined to study how language is used in a social context. Words used in language lie somewhere between being concrete, where a word describes a specific item or action, and abstract, where a word conveys a large amount of inferred information. This distinction is important because more abstract language implies dispositional attribution, whereas concrete language implies situational attribution. For example, an accountant could overlook some crucial information Age-Related Stereotypes and the Linguistic Intergroup Bias 5 on an account, causing a mistake. A concrete explanation of the situation could be that the accountant “rushed” through that account, implying that the mistake was due to situational factors. On the other hand, the situation could be described as the accountant “was careless,” implying that the mistake was caused by the accountant's disposition and is therefore likely to occur again. The second example uses more abstract language than the first example. The measure of linguistic abstractness can be used to find differences between descriptions of different social targets. Traditionally, linguistic abstractness is measured using the LCM, developed by Semin and Fiedler (1988). According to Semin and Fiedler, the definition of abstractness in the LCM is the degree of subjectivity and permanence that a word conveys. This method generates a score for a stretch of discourse by counting the number of adjectives and different categories of verbs. The most concrete verb is a descriptive action verb (DAV), which describes an action without making any inferences about that action. An example would be “run.” Interpretive action verbs (IAV), which describe an action that requires an inference, are more abstract. This includes verbs like “swindle.” State verbs (SV) describe a relationship between two things, such as “like.” These are the most abstract verbs. DAVs describe an Age-Related Stereotypes and the Linguistic Intergroup Bias 6 objective, impermanent action, while SVs and adjectives describe subjective states of being that have some degree of permanence. IAVs fall somewhere in between, but have a level of subjective inference greater than that of DAVs. Finally, adjectives are considered the most abstract words in the LCM. Using these categories, a composite score can be constructed for a written transcript by assigning a score of one to four based on the relative abstractness of the category, with DAVs being 1 and adjectives being 4. These scores are then summed and divided by the number of scored words in a transcript, yielding a mean abstractness score ranging between 1, meaning extremely concrete, and 4, meaning extremely abstract. Linguistic Category Model Research using the Linguistic Intergroup