This study re-investigates the often-found higher probability of enrolling in academic tracks of upper secondary education among immigrant compared to ethnic majority students, often referred to as the “positive ethnic choice effect”, from a causal inference perspective. The difference in enrollment is often attributed to “immigrant optimism”, a particularly high degree of ambition in students whose parents immigrated. In this study, we first theoretically disentangle multiple facets of immigrant optimism that can be found in the literature: (parental) educational aspirations, a drive for status occupational upward mobility, a high value attached to education and high general ambition. We then discuss the estimation strategies of previous studies in light of a causal mediation analysis framework. While we acknowledge that the explanation of ethnic enrollment gaps is largely a descriptive research question, much can be gained by using knowledge from causal inference. First, a causal inference framework forces us to clearly define our estimand of interest (in our case, a descriptive version of the controlled direct effect). Second, using estimation techniques from causal mediation analysis avoids potential difficulties in interpreting quantities that result from traditional regression modeling. We use marginal structural models and covariate balancing propensity scores to estimate the controlled direct effect of immigration background on choosing the academic upper secondary school track in the German educational system over vocational alternatives, holding different facets of immigrant optimism constant. Our results suggest that parental educational aspirations drive ethnic choice effects, while other facets of immigrant optimism play no role.