본 연구에서는 고정관념 내용 모델(stereotype content model)을 활용하여 한국사회의 연령집단에 대한 고정관념 내용을 분석하였다. 각 연령집단에 대한 고정관념을 지적 특성인 유능함(competence)과 사회적 특성인 따뜻함(warmth)의 2차원으로 분석하였는데, 특히 사회정체성 이론(social identity theory)에 근거하여 평가대상이 되는 연령집단이 내집단(in-groups)인 경우 내집단 편향효과(in-group bias)가 나타나는지를 가설화하고 검증하였다. 실증연구 결과, 연령집단에 대한 유능함과 따뜻함을 평정함에 있어 평가자집단과 평가대상집단 간 일치 여부가 중요한 것으로 나타났다. 즉, 평가대상집단이 내집단일 경우 유능함과 따뜻함을 보다 호의적으로 평가하는 내집단 편향현상이 나타났으며, 이러한 편향된 평가경향은 평가대상집단에 대한 동일시(identification) 수준이 높아지기 때문인 것으로 매개효과 분석에서 검증되었다. 종합토론에서는 연구결과 요약 및 해석과 함께 이론적․실무적․정책적 시사점을 논의하고 있다. 본 연구는 연령집단을 연구 대상으로 함으로써 고정관념 내용 모델을 보다 폭넓게 이해할 수 있는 기회를 제공하는 한편, 내․외집단에 대한 고정관념의 차이와 그 심리적 기제를 밝힘으로써 사회정체성 이론을 확장하는 데 도움이 될 것으로 생각된다. 또한, 연령집단에 대한 고정관념 및 내․외집단 간 차이 분석은 특정분야에서의 미시적 소비자 행동을 이해하고 연령집단이 중요한 상황에서의 마케팅 전략을 수립하는 데 이론적 기초를 제공할 것이다. 최근 한국사회에서 연령집단 간 갈등이 고조되고 있는 상황을 감안할 때, 연령집단에 대한 인식과 고정관념을 정확하게 이해하는 작업은 정책적인 차원에서도 매우 의미 있고 시사하는 바가 클 것이다.
In this research, we investigated whether in-group bias would occur in perceiving competence and warmth about age groups in the Korean society. According to the stereotype content model (e.g., Cuddy, Fiske, and Glick 2008; Fiske, Cuddy, and Glick 2007; Fiske, Cuddy, Glick, and Xu 2002), the stereotype contents regarding social groups are systematically and succinctly captured by two dimensions, competence and warmth. To empirically uncover the stereotype contents about age groups in the Korean society, we measured perceived competence and warmth about the young (in their 20s and 30s), the middle-aged (in their 40s and 50s), and the elderly (in the 60s and older). Drawing on social identity theory (Brown 2000; Tajfel 1982; Tajfel and Turner 1986) that proposes self-categorization and its motivational consequences, we hypothesized that the respondents would exhibit more favorable perceptions of competence and warmth about their in-groups than about their out-groups, and that the biased perceptions would be explained by the level of the respondents’ identification with the age groups. We recruited 910 participants in their 20s to 60s from five major cities in Korea and administered a survey that consisted of a series of questions to measure demographic variables (such as age, occupation, education, and income), perceived competence and warmth about the three age groups, the perception of relative status and competition, and finally the level of identification with the groups. Perceived competence and warmth were measured on 5-point scales (1 = not at all; 5 = extremely): how [competent or warm] are members of [the young, the middle-aged, or the elderly]? (Cuddy, Fiske, and Glick 2007; Fiske et al. 2002). The relative status and competition (Fiske et al. 2002) were also measured on 5-point scales (status: how well-educated are members of [the young, the middle-aged, or the elderly]?; competition: if members of [the young, the middle-aged, or the elderly] get special breaks (such as preference in hiring decisions), this is likely to make things more difficult for people who belong to the same age group as me). The level of cognitive identification with the age groups was measured on a 7-point pictorial scale which had two circles varying in the extent of the overlap between the two (Bergami and Bagozzi 2000). The one circle represented the self and the other represented each of the rated age groups. After responding to the demographic variables, the participants answered to which age groups they belonged. We eliminated those participants whose self- reported in-groups were different from their real in-groups based on their age and entered 801 participants to subsequent analyses. The participants provided all ratings about the young first, then the middle-aged, and finally the elderly. First, we found that perceived competence was highest for the middle-aged, followed by the young and the elderly, while perceived warmth was highest for the elderly, followed by the middle-aged and the young. Second, to test hypothesis 1, we conducted 3 (target: the young, the middle-age, or the elderly) × 3 (evaluator: the young, the middle-age, or the elderly) mixed analyses of variance on competence and warmth respectively, treating the former as a within- participants factor and the latter as a between-participants factor. The two-way interaction was significant in both analyses (competence: F(4, 1596) = 2.93, p = 0.0198; warmth: F(4, 1596) = 12.49, p < 0.0001). Overall, the perceptions of competence and warmth were higher when the evaluators provided ratings for their in-groups than for their out-groups. Third, to test hypothesis 2, we conducted the same 3 (target: the young, the middle-age, or the elderly) × 3 (evaluator: the young, the middle-age, or the elderly) mixed analyses of variance on identification. The two-way interaction between the two factors was also significant (F(4, 1596) = 167.30, p < 0.0001), revealing that overall, the level of identification with the in-groups was higher than with the out-groups. Finally, we investigated whether the level of identification with age groups would account for in-group bias observed in perceptions of competence and warmth about the groups. The mediational analyses using a bootstrapping procedure confirmed the statistical significance of the indirect effect of the evaluator groups on perceived competence and warmth via identification. This research is expected to contribute to a deeper understanding of the stereotype content model by expanding its scope to age groups and integrating the phenomenon of in-group bias to the stereotyping of social groups. It also widens the applicability of social identity theory by focusing on age-based social identity. The findings of this research offer important future research directions and implications to consumer research topics such as social exclusion and intergroup helping, to effective marketing strategies of services, media, and brands in relation to age-based segmentation, and to policies to address inter-group conflicts that have recently emerged in many areas in the Korean society.