Data from the NEPS (Starting Cohort 4) were used for the analyses (H.-P. Blossfeld & Roßbach, 2019). Within Starting Cohort 4, a subsample of school leavers with a university entrance qualification can be identified and tracked over a longer period of time. The data from Starting Cohort 4 are particularly well suited for the analysis of this question because, on the one hand, different immigrant groups can be identified with the data (Olczyk et al., 2016). On the other hand, the data contain a number of indicators that can be used to operationalize educational goals (idealistic aspirations and realistic expectations) as well as migration-related explanations of elevated educational goals (immigrant optimism, relative status preservation, anticipated discrimination, and information deficits).
In a first step, previous work on secondary effects of migration background was replicated. For this purpose, nonlinear regression models were estimated taking into account social origin (education and socioeconomic status of parents) and primary effects (grade of HZB, type of HZB). Different operationalizations of the dependent variable (with and without a third category: "no transition to postsecondary education") and migration background (country of origin, generation, and combinations of both indicators) were tested to determine more precisely for which contrasts of education alternatives and for which migrant groups secondary effects exist.
In a second step, we examined the extent to which the positive secondary effects of migration background can be attributed to increased educational goals (educational aspirations and expectations) or to specific migration-related mechanisms (e.g., immigrant optimism). To this end, further nonlinear regression models and decomposition analyses (Kohler et al., 2011) were used.