The use of network analysis to handle semantic differential data

Abstract

The aim of this paper is to propose a method to transform semantic differential data into a network whose graph representation is interpreted as an empirical network of adjectives. The graph is constituted by the adjectives of the semantic differential task. Two adjectives are linked depending on the scoring assigned by a set of respondents. The proposed approach aims at using concepts and methods of Social Network Analysis to explore the network structure and study roles and positions of dominant adjectives. A simulation design has been realized to assess the stability of results under different conditions, i.e. in order to set the optimal threshold in presence of different data generator processes. A case study carried out on real data shows how the emerging network of adjectives can be effectively used to define the concept arising from a semantic differential task.