Integrating different factorial solutions of a psychometric tool via social network analysis: The case of the mood disorder questionnaire

Evaluating the factorial structure of a psychometric test is crucial to capture the complexity of a psychological phenomenon. Indeed, for the same test, several studies may find different factorial solutions which, in turn, may be explained by within and/or between sample variability. In this paper we introduce a novel quantitative approach to combine different factorial solutions of the same test. We propose to use a method based on Social Network Analysis to create and statistically evaluate an integrated factorial structure based on the information provided by previous researches. We present an application to the Mood Disorder Questionnaire by considering different factorial structures reported in the literature. The integrated factorial solution indicates the presence of three factors supporting the multidimensionality of the test. The role of single items in the composition of factors is also evaluated and discussed in terms of differences and similarities between the five original studies and the new integrated model. From an applied perspective, our approach may be useful to assist researchers in summarizing different factorial solutions for the same test efficiently. Furthermore, the resulting integrated factor solution could serve as baseline model to validate the structure of the test by applying confirmatory factor techniques to new data.