In this review we aimed to clarify the impact of symptom and sample selection on the results of an analysis of dimensionality versus taxonicity, and to show how prior knowledge about the data can guide the choice of method. The paper is structured as follows. The next section covers some general methodological considerations that can influence the inference of dimensionality or taxonicity that are independent of the method used. This is followed by a section providing an overview of the essential features of (1) taxometric methods (e.g. Meehl, 1995, 2004; Waller & Meehl, 1998; Gordon et al. 2007; McGrath & Walters, 2012), (2) model-based clustering (e.g. Banfield & Raftery, 1993; Fraley & Raftery, 2002; Frühwirth-Schnatter, 2006), and (3) LVMM (e.g. Dolan & van der Maas, 1998; Muthén & Shedden, 1999; Lubke & Muthén, 2005; Tueller & Lubke, 2010). Although not inclusive, a discussion of these three methods illustrates the complexities involved in deciding whether a psychiatric disorder is dimensional or categorical. The description of the three methods focuses specifically on those aspects and assumptions that can influence the outcome of
