Preliminary investigation of the Italian version of The Tromsø Social Intelligence Scale (TSIS) for children

Abstract

Introduction: Social intelligence (SI) is a construct that has shown promising practical applications, but its use in research and applied settings has been limited by both definitional and operational problems. Overall, SI is currently defined as a set of cognitive, emotional, and behavioral abilities that individuals can use to interpret events, plan their life, achieve personal and social goals, solving personal and interpersonal problems. These abilities could be fundamental in primary school children because classrooms represent the context for first experiences with social acceptance and rejection, as well as for the development of relationships between classmates. However, there is a lack of studies on the measurement of SI in children. Methods: An adapted version of the Tromsø Social Intelligence Scale (TSIS-IV) was administered to 228 Italian primary school children (115 females; mean age: 9.57; SD: .62). The original questionnaire consists in 21 items on a 7- point Likert scale and provides scores to three subscales: Social information Processing (SP), Social Skills (SS), and Social Awareness (SA). Results: Internal reliability was evaluated using Cronbach’s alpha coefficient. The adapted version for children showed acceptable levels of reliability (α=.75). Explorative factor analysis with varimax rotation showed three factors that correspond to the subscales of the original version of the TSIS: SP (eigenvalue= 3.04), SS (eigenvalue= 2.59), and SA (eigenvalue= 2.43). However, while all items of SP load onto one factor, three items are switched between SS and SA. Nevertheless, these three factors explain 14.47%, 12.34%, and 11.57% of the variance, respectively. Each item showed a loading of almost .30, except for item 5. Conclusion: These preliminary results showed that the factorial structure of the TSIS-IV for children is somehow different to the adult form, with SS and SA subscales seeming to be structured differently. This could be interpreted by the fact that a number of psychological variables measured by the questionnaire,such as those included in the subscale SA, may not be already developed by children and may overlap with more concrete dimensions which are included into the subscale SS.