Social Mobilities : Multidimensionality, Operationalization, and Subgroup Heterogeneity

Sammanfattning: Social mobility is a key concern for societies, as it reflects levels of inequality, life chances, and fairness. A higher mobility rate implies that individuals are less constrained by their family origins. In this dissertation, I explore which parental resources matter, whether transfer patterns are shaped by the parent–child configuration of socio-economic status (SES) resources, how such resources should be operationalized, and whether subgroups differ in social mobility rates. The aim is to go beyond a unidimensional view of social mobility and more systematically explore the existence of social mobilities.    Study I addresses how different dimensions of parents’ SES in education, occupation, income, and wealth are associated with educational attainment in secondary tracks and tertiary fields of study. Our results show that there is strong segregation by parents’ SES. However, the association of social background dimensions to educational attainment is not uniform but differs by the combination of dimension and track or field. This selection process by social origin is also linked to future inequality in chances of tertiary graduation and expected earnings.     Study II focuses on how to best combine information on parental SES in studies of intergenerational inequality. We assess how much of the sibling correlations in continuous measures of education, occupation, and earnings are accounted for by parents’ SES in the same dimensions using different operationalizations. In conclusion, parental averages are an attractive and parsimonious one-variable alternative that is preferred over the conventional dominance approach, although the highest explanatory power is attributed to models using two parental measures and an interaction term.   Study III highlights how multiple parental SES resources are transmitted over corresponding child outcomes. The findings suggest, first, that transmission is particular to given parental and child resource configurations (resource specificity). Second, within-resource transmission implies that the same parental resource as the child outcome matters most in the transmission of advantage. Third, resource transmission follows an SES proximity pattern, where parental education is least correlated with child income and parental income is least correlated with children’s education—with parental occupation in between. The bias resulting from ignoring multidimensionality is estimated to an upper bound of 31 percent, with considerable confounding bias found as well.      Study IV is centered on subgroup heterogeneity in social mobility. I propose a three-stage evaluation process that goes beyond a conventional examination of group mean and effect differences using prediction methods. Prediction allows for gauging subgroup-specific changes in explanatory power and comparing the level of uncertainty that each subgroup faces. Although the conventional interpretation of interactive models shows some support for subgroup heterogeneity, predictive results indicate minimal improvements in fit. However, subgroup differences in idiosyncratic error, or the subgroup level of uncertainty in outcomes, suggest substantial heterogeneity. In sum, the interpretation of coefficients and graphical analysis of interactions can be contextualized by expanding the criteria for assessing subgroup heterogeneity. Ultimately this leads to a better understanding of subgroup particularities, thus guiding future research toward richer answers.