Is the Polythetic Approach Efficient in Identifying Potentially Addicted to Work Individuals? Comparison of the Polythetic Approach With the Item Response Theory Framework

This study included investigation of efficiency of the threshold used to classify symptoms as present, investigation of efficiency of the cut-off point used to identify potentially addicted to work individuals, investigation of magnitude of the problem of class overlap, and investigation of effects of dichotomization of polytomous items on the estimates of the latent trait level. The sample comprised 16,426 working Norwegians (Mage = 37.31; SD = 11.36) who filled out the Bergen Work Addiction Scale (BWAS). The results showed that the difficulty/third threshold parameters corresponding to the threshold used to classify symptoms as present were lower than 1.5 for the items corresponding to tolerance and conflict and higher than or equal to 1.5 for the items corresponding to salience, mood modification, relapse, withdrawal, and problems. The cut-off point used to identify individuals as potentially addicted to work identified 411 individuals (31.9% of all individuals classified by the polythetic approach as potentially addicted to work) whose estimates of the latent trait level were lower than 1.5 as potentially addicted to work. The problem of class overlap (being classified by the polythetic approach into different class despite almost the same level of the latent trait) affected 4,686 individuals (28.5% of the whole sample). The dichotomization of polytomous items had a substantial effect on the estimates of the latent trait level. The findings show that the polythetic approach is not efficient in identifying potentially addicted to work individuals and that the prevalence rates of work addiction based on the polythetic approach are not trustworthy.