Immigrant background and expected early school leaving in Europe: evidence from PISA

This technical brief analyses the relationship between immigrant status and educational expectations in PISA. Immigrants have become a very policy-relevant issue during the last recent years, in particular with the occurrence of the refugee crisis. Moreover, the freedom of movement in the EU has led to relevant migration flows across EU Member States. At the same time, the EU has set itself the Europe 2020 headline target of reducing the share of early school leavers to 10% within the EU. Early school leavers are generally disadvantaged socially and economically in later stages in life, so that it is important to better understand their motivations and provide adequate policy solutions. The European Commission (2016, p. 3) indicates that early school leavers are more likely to come from immigrant student groups, as their "early school leaving rates are nearly twice as high as for the native population". Yet it also emphasises that there is still a lack of evidence pointing to the underlying reasons. In consequence, this study considers jointly these two groups, immigrant students and early school leavers. More specifically, we analyse the factors that are most strongly related to disparities in the probability to leave school early, putting special attention to immigrant status (by differentiating among first and second generation immigrants and, where possible, among EU and non-EU immigrants). To this end, we use OECD’s PISA data, which are the most widely employed data on international student assessment. However, early school leavers cannot directly be considered with these data, but it is possible to analyse educational expectations, including the expectation to dropout early from school. As the related literature emphasises, these expectations are very closely linked to actually realised educational career patterns. Therefore, we can use these expectations to gain insights on the factors influencing early school leaving. In addition, we also employ data from Eurostat to complement the picture on early school leavers and immigrants. We analyse the issues at stake in various ways. First, we provide a range of descriptive data on immigrants and expected early school leavers. Second, we run a number of two-level logit regression models, including a range of student- and school-level variables. In particular, we consider all (available) EU Member States together, before providing results for each MS individually. Finally, we also distinguish more specifically between EU and non-EU immigrants in our regression models. The results show that immigrant students do mostly not structurally differ in their expected early dropout probability to natives across Europe. In other words, the reasons why students expect to leave school early are the same for both immigrant students and natives. This finding implies that it is more important to focus on the specific factors that lead to expected early school leaving common to all students, than to concentrate only on specific immigrant-related factors to decrease the occurrence of expected early school leaving among immigrant students. In particular, our results suggest that the factors most strongly increasing the probability of early school leaving at the student level are the socio-economic background of students, epistemological beliefs and grade repetition, while we find that the most consistent factor is to be found at the school level, being the school’s expected early school leavers probability. The school-environment thus appears to play a key role in shaping educational expectations. Among the student-related factors, grade repetition is the most amenable by policy, so that grade repetition practices may be reconsidered by national policy makers.