Search

This thesis explores empirically the links between parental migration and educational performance of children, as well as bilingualism and family status and cognitive development of children. The first three chapters focus on a scenario in which households send one member, usually a parent, for temporary employment abroad. I firstly examine the implications of such a family structure on educational performance of teenagers. I then investigate whether this impact may spill over through peer interactions at school. I have designed and collected a survey for the purpose of the analysis. I elaborate on the process in the first chapter. The gathered data contain information about over 2800 16- year -old pupils, including their socio- economic background, performance at school and migration experiences within a family over a period of three years. Parental migration is common in the studied population and is mostly characterised by relatively short, repeated spells of legal employment of fathers in other European countries. The nature of the migration experience sets it apart from cases considered so far. In the second chapter I utilise the data to investigate the relationship between parental absence due to emigration and the child's performance at school in that period. I find that, on average, children's grades improve when they have a parent abroad. A negative impact may, however, emerge in cases of prolonged separation. Meanwhile, sibling emigration exerts a strong positive effect on educational attainment which accumulates over time. The results are plausible if parental emigration significantly increases household income, whilst not disproportionately burdening children by means of increased responsibility. The third chapter extends the analysis by looking at the influence children with parents working abroad may exert on their classmates. I find that pupils in classes with a high proportion of children of migrant parents perform better. The impact is greater for those who experienced family migration themselves. I consider various possible explanations of the result and conclude that the positive individual effect found in the second chapter may spill over through the peer interactions. Increased teachers' involvement in classes with many migrant children may play an additional role. Parental education is key to the positive effects found in both chapters. The children of parents, who have themselves graduated from high school, benefit most from their parents' emigration experience. They also are the influential group among their peers. In the final chapter I consider a different scenario where families use two languages at home to investigate whether it affects the development of cognitive and non -cognitive skills of their children. Importantly, I notice that bilingualism may be an insufficient element to explain any differences, as bilingual families are a heterogeneous group. Therefore, I also differentiate between families with two native, one native, one foreign and two foreign parents. Using the data for families with children under the age of 6 in Scotland, I find that overall children's cognitive and non -cognitive skills are similar. The performance in the English Vocabulary Naming exercise is an exception. On average, bilingual children do not perform worse than monolinguals in the task. There is heterogeneity within the group, however. Bilingual mixed -nationality children lag behind the monolingual native children at the age of 3 but they catch up by the age of 5. However, there is some evidence that bilingual children who have two foreign born parents may perform worse than the monolingual native children and not improve with age.

The overarching aim of this research is twofold. First, I explore the disparities in academic attainment and progression between ethnic minority undergraduates and their White peers in the UK, in terms of the probability of graduating with a "good" class of degree and the dropout likelihood. Second, I estimate the size of ethnic pay inequalities in the British labour market among UK-born graduates and quantify the contribution of specific characteristics to the wage differential. Regarding the likelihood of degree non-completion, for the first time in the ethnicity context, I distinguish between compulsory dismissal (because of academic failure) and voluntary dropout (because of personal, financial, or other reasons), acknowledging that an effective policy response to student withdrawal ought to be closely associated with the dropout causes. I provide firm evidence that all ethnic minority groups have, on average, a higher probability of failing their degrees than White students, even after allowing for differences in a wide range of socio-demographic and university characteristics. Most worryingly, Black African students are twice as likely (7.1%) as their White peers (3.8%) to fail. On the contrary, White students are more inclined towards quitting voluntarily than ethnic minorities, although the differences are smaller than those relating to academic dismissal. The probability of attaining a good degree class stands at 76.2% for White students, and the difference relative to ethnic minorities ranges from 4.3 percentage points (for students from a Mixed ethnic background) to 15.1 percentage points (for Black African students). These performance gaps remain striking across all types of institutions and social class levels, suggesting that ethnic minorities' under-attainment is a pervasive issue. With respect to the second target of this thesis, a key conclusion of this work is that having graduated from university does not eliminate ethnic pay inequalities. For Black men, the labour market experience is more disappointing, as their wage penalties stand at higher levels (16.7%) than those of Black women (4.5%) and the other ethnic groups. The occupational segregation and the shorter job tenure of ethnic minorities relative to White people make up the primary determinants of pay gaps relating to observed characteristics. Although it is difficult to disentangle the extent of racial discrimination from other unobserved factors that also shape the wage levels, by implementing a novel method that partially addresses the selection on unobservable determinants of earnings, I find supportive evidence of ethnic discrimination, particularly against Black and Asian men. I reveal that ethnic penalties worsen with age across both genders, implying that some mechanisms endogenous to the labour market (such as discrepancies in promotion procedures, access to company-specific training, and the time required to find a job, which, in turn, affects the accumulative experience) should, at least partially, explain these results. These research findings should help policymakers better comprehend the mechanisms that produce diverse academic and earnings outcomes and identify targeted strategies for each ethnic group. In this context, I propose certain policymaking actions that should alleviate the barriers faced by ethnic minorities in higher education and the UK labour market.