2021/04/23 by André Beauducel, Beauducel, André, Norbert Hilger +1 · 1 citation
Decision Sciences · Computer Science · #Psychometric Methodologies and Testing #Advanced Statistical Modeling Techniques
paper · pdf · doi:10.48550/arxiv.2104.11526
The paper is devoted to the consequences of blind random selection of items\nfrom different item populations that might be based on completely uncorrelated\nfactors for item inter-correlations and corresponding factor loadings. Based on\nthe model of essentially parallel measurements, we explore the consequences of\npresenting items from different populations across individuals and items from\nidentical populations within each individual for the factor model and item\ninter-correlations in the total population of individuals. Moreover, we explore\nthe consequences of presenting items from different as well as identical item\npopulations across and within individuals. We show that correlations can be\nsubstantial in the total population of individuals even when -- in\nsubpopulations of individuals -- items are drawn from populations with\nuncorrelated factors. In order to address this challenge for the validity of a\nscale, we propose a method that helps to detect whether item inter-correlations\nresult from different item populations in different subpopulations of\nindividuals and evaluate the method by means of a simulation study. Based on\nthe analytical results and on results from a simulation study, we provide\nrecommendations for the detection of subpopulations of individuals responding\nto items from different item populations.\n