Group identification of youth in residential care: evidences of measurement and dimensionality

Abstract

This study aims to adapt the Tarrant's scale of group identification for youth in residential care and to provide evidence of the validity and reliability of this scale. This study is part of a larger project including 59 Portuguese institutions and 400 adolescents (55 % males) from 11 to 18 years old (M = 14.68; SD = 1.83). The adolescents that agreed to participate completed a structured questionnaire that was analysed regarding face, construct and concurrent validity, reliability as well as the relationship between group identification, individual characteristics (age, gender) and placement length. Face validity was obtained with feedback from a panel experts and youth in care. The three components structure obtained explains 70.28 % of the variance: Self-categorization, Group Evaluation and Commitment. The CFA revealed good values of fit: CFI = .95, GFI = .91, and RMSEA = .073. Concurrent validity was documented by the positive and significant correlation found between the positive perception of in-group and all dimensions of group identification, as well by a negative and significant correlation between deviant behaviour and all dimensions of group identification. Finally, significant and positive correlations were found between Self-Categorization and the length of placement in the residential setting. Gender differences were found on Group Evaluation with females reporting higher scores than males. The satisfactory psychometric characteristics of this scale strengthened its trustworthiness in the measurement of group identification and emphasize the relevance and validity of this instrument in the evaluation of group identification in residential care populations.