Multidimensional scaling of the eight state questionnaire and the differential emotions scale

Abstract

While the interrelationships between multivariate measures of mood states have been analysed using various statistical procedures including exploratory factor analysis, discriminant function analysis, multiple regression analysis, and canonical-redundancy analysis, the techniques of Multidimensional Scaling (MDS) however, have been used less extensively in psychometric research. MDS provides a readily interpretable representation of empirical relationships between different sets of data. The present study attempted to replicate Boyle's (1986a; Psychological Reports, 59, 503–510) factor analytic findings on two self-report multidimensional mood-state measures, namely the Eight State Questionnaire (8SQ), and the Differential Emotions Scale (DES-IV), to demonstrate the utility of MDS to uncover the underlying structure of each instrument, and to investigate further the correspondence across the two scales. Results supported previous findings that relatively little overlap exists between the two instruments, despite the apparent similarity of the subscales. Most of the DES-IV subscales were found to lie in close proximity, suggesting inadequate separation. Hence, MDS was found to be a viable procedure for investigating such interbattery relationships.