Current Views on Social Class, Status, and Mobility

measure social position. These options reflect differences at the conceptual level but also in the empirical approach taken to construct measures of social position. Is it better to operationalize social position as a single dimension or multiple dimensions (Alwin & Wray, 2005), is it a continuous measure, a collection of continuous measures, or is it a discrete variable (Weeden & Grusky, 2005)? If it is a discrete variable, how many class categories are sufficient for capturing groups that are sufficiently different from one another, but homogenous within? Some schemes use three classes, others use seven or eleven, and still others (the micro-class approach) use more than one hundred groupings (Erikson & Goldthorpe, 1992; Weeden & Grusky, 2005). Are we talking about groups that are clearly delimited from one another, or, rather, about fuzzy groups with flexible boundaries? Is one measure enough to capture education? The same question applies for other measures such as occupation or social origins. Recent recommendations in the literature and the use of models with latent variables in social stratification research have suggested that it is better to use multiple indicators to measure each of these, in order to account for measurement errors (Ganzeboom, 2009; Ganzeboom & Treiman, 2003; Kerckhoff, 1984). For researchers adopting a model in which social origins and social status are latent variables with multiple indicators, a further question is: should the model be a reflective indicators model, or a formative indicator model (Alwin, 1988; Blau & Duncan, 1978; Hauser, 1972; Hauser, Tsai, & Sewell, 1983; Heise, 1972; Tufiș, 2010)? There is no best way to approach the conceptualization and measurement of social position, and each of the above strategies has strengths and limitations depending on the research question being tackled. In the landscape of available options for conceptualizing and measuring social positions, the multidimensional approach (Alwin & Wray, 2005) and the micro-class approach (Weeden & Grusky, 2005) seem to be the most complex and to have a better conceptual grounding. The former would be better suited to capture social status in conceptual frameworks that take into account various capitals (cultural, material, social, DOI 10.1515/irsr-2015-0001