This study examined a methodological approach to research on social representations. The approach provides an empirical validation of the possibility that a representation has a consensus among a certain social group. This methodological approach is illustrated in a study that aimed at revealing the structure of the social representation of fear among children from seven to ten-years old, with different social and cultural backgrounds (private schools and orphanages). The main aim was to examine the role of social background in establishing this kind of representation. First, with a group of thirty children, the elements of the representation were collected through a free association method, which consisted of asking the children to express what they thought of when the word 'fear' was evoked. From this information, fifteen kinds of fears were selected from those that had been evoked most frequently by the two groups of children. The social representation of fear was examined using a non-verbal sorting technique with a second group of fifty-eight children. They were asked to think of about fifteen words that were presented on cards and to place them in order according to their higher or lower level of association with the sensation of fear. The data were analysed using non-metric multidimensional scaling techniques (SSA, MSA) and pointed to the existence of similarities and differences in consensus levels between the different child groups compared with respect to the representation of fear in society. In this way it was possible to establish the meaning and the general attitude of the children in relation to fear. The results are discussed emphasising the methodological problems associated with the study of social representations, referring to the objective establishment of consensus level.
Fear; emotion; social representation; culture; non-verbal sorting technique
