Chris Sakellariou

This study used Add Health longitudinal data to investigate whether any effect of religiosity on later delinquency (likelihood of arrest and incarceration) is linear - as opposed to high-religiosity individuals driving an inverse relationship. The study assessed potential mediator-moderator interactions involving self-control and social control variables. Finally, the robustness of findings to potential biases due to omitted variables was assessed. Only highly religious participants were associated with a substantially lower likelihood of arrest. Self-control and social environment variables are important predictors of delinquency. There is little evidence that these variables mediate the effect of religiosity on outcomes; they instead moderate the effect of religiosity on outcomes. Finally, conventional regression estimates likely underestimate the true effect of religiosity on the outcome.

This study used the latest waves of Add Health longitudinal data to assess the size of the protective effect of religiosity on later sexual behaviour in young adults (number of opposite sex partners and early sexual initiation). Related research questions were investigated, namely whether the influence of higher religiosity on sexual behaviour operates linearly and whether findings are robust to omitted variables bias. A general finding is that only for highly religious male and female participants is the effect robust. A related finding is that, after accounting for the protective effect of high religiosity, the independent effect of being religiously affiliated is protective for women of Catholic affiliation (compared to having no affiliation) but weakly towards more promiscuous behaviour for men of Catholic affiliation (gender difference statistically significant). This suggests that some religiously affiliated young males may see less of a need to reconcile any conflict between religious teachings and their sexual behaviour, unless religious teachings are internalized. Finally, after investigating the robustness of findings to biases due to omitted variables bias, there is additional support to the main conclusion that a robust effect of religiosity on sexual behaviour can be established only in individuals of high religiosity.

Introduction: This study investigates the bidirectional relationship between earlier sense of school belonging and later academic achievement in schoolchildren at grades 4 and 5 in US schools, using ECLS-K:2011 longitudinal data.Methods: Two alternative estimation methods were used, both addressing biases due to endogenous covariates.Results and discussion: The findings (sample size > 8,000 observations) provide strong evidence that (1) the dominant effect is from sense of school belonging to achievement, where lower bound effect sizes are substantially larger than those reported in correlational studies; and (2) in the opposite direction biases are small, and bias-corrected effect estimates are generally in line with the multiple regression estimates. The findings also provide suggestive evidence of larger effects for girls compared to boys in the direction from sense of school belonging to achievement scores. The study’s findings provide useful insights into the potential impact of school-based interventions.

Research on the bidirectional relationship between social connectedness and health/mental health in adolescents is scarce, with most studies on adults. Some of the existing studies exploited the availability of longitudinal data to provide evidence of the existence of a causal relationship, either from social connectedness to health or establish a bidirectional relationship. There are at least two weaknesses associated with earlier research to assess the size of the effects. As acknowledged in the literature, one relates to attributing causality to empirical findings, due to well-known but inadequately addressed endogeneity biases. The other relates to failure to account for potentially important covariates, sometimes due to data limitations, or because such variables are not frequently used in sociological research. Existing research predominantly finds that the strongest path is from social connectedness to health/mental health, with effect estimates modest in size. I followed a quasi-experimental strategy by modelling adolescent students’ perceptions of social connectedness and mental health perceptions as potentially endogenous variables when estimating bidirectional effects. An instrumental variables (IV) modelling approach was followed, supplemented with a recently developed alternative approach to testing the exclusion restrictions of individual instruments. I exploited the rich information available in the PISA 2018 multi-country dataset, which allows for conditioning for a wide array of information on adolescent students’ personal circumstances, self-reported personality-related attributes, relationships with parents; and school characteristics. I found that (1) accounting for endogeneity biases is important; and (2) as opposed to findings reported in the literature, the dominant effect is from mental health perceptions to social connectedness for both male and female participants. The policy relevance of the findings is that adolescent mental health should be the primary focus of interventions, i.e., identifying and treating mental health symptoms as a primary intervention and as a precursor to improving the social connectedness of adolescents.

This study investigated the role of endogeneity of variables of interest and associated biases when deriving effect estimates in the relationship between perceptions of social integration and students’ academic achievement. The dataset used is a large sample of 15-year-old students from the 2018 PISA surveys. Following an Instrumental Variable (IV) approach, paying particular attention to testing the suitability of instruments used, I derived bias-corrected effect estimates which are generally larger than conventional regression estimates. This was especially the case for female students in the direction from sense of social integration to reading literacy. Reciprocal effects were also established, with small to medium effect sizes in the direction from reading and mathematics scores to sense of social integration, for both male and female students. Clear policy implications emerge, targeting heterogeneous pairs/groups of students with respect to their social integration in school, as well as their academic performance.

Objective: To investigate the reciprocal relationship between high school students’ academic self-efficacy and achievement in mathematics using US data from the HSLS:2009 and first follow-up longitudinal surveys, while accounting for biases in effect estimates due to unobserved heterogeneity.Methods: Instrumental Variables (IV) regressions were estimated, to derive causal effect estimates of earlier math self-efficacy on later math achievement and vice versa. Particular attention was paid to testing the validity of instruments used. Models were estimated separately by gender, to uncover gender differences in effects.Results: Evidence of robust reciprocal effects between self-efficacy and achievement for male students is presented, with the dominant effect from earlier achievement to later self-efficacy. For girls, evidence of such effects is weak. Generally, IV estimates are higher than OLS estimates for males, but not for females. As opposed to earlier correlational studies which did not find significant gender differences despite theoretical expectations for their existence, the findings support higher effects for male students.

I brought together two strands of literature, one investigating the moderate but persistent underachievement of girls in mathematics in most countries, and the other examining the role of self-efficacy and other self-beliefs in predicting behaviour and achievement. I implemented detailed decompositions of the gender mathematics gap, both at the mean and for low and high performing students, for a large and diverse group of countries. I found considerable heterogeneity and different cross-country patterns in decomposition components and the contribution of self-beliefs. In OECD-Europe and more affluent East Asian countries, most or all of the gap is explained by gender differences in self-beliefs, especially self-efficacy; on the other hand, in Latin America and the Middle East, most of the gap remains unexplained. I also investigated the cross-country relationship between the gender mathematics gap and gender equity and found that a clearly negative association can be established after controlling for cross-country heterogeneity in gender differences in mathematics self-beliefs, which correlate with gender equality.

The literature on the immigrant-native educational achievement gap is suggestive of a better performance of immigrant students in countries practising selective immigration policies. However, the origin of such differences has not been investigated. I considered both important observed characteristics as well as the role of unobservables in the formation of immigrant-native achievement differences in 4 selective immigration countries. I found that, if the comparison is between selective immigration countries as a group and other European/Organisation for Economic Co-operation and Development (OECD) countries, immigrant students in selective immigration countries perform better. However, the results are heterogeneous and different conclusions apply to each of the 4 countries. The findings are discussed in the context of successful immigration policies as screening devices, which induce a self-selection of immigrants with transferable and adaptive skills. Potentially important are also institutional differences with respect to how inclusionary and conducive to the integration of immigrants into the host country policies are.

It is known that in most countries, students in private schools outperform students in public schools in international assessments. However, assessing the true effect of private school attendance requires addressing selection and sorting issues on both observables and unobservables. The existing empirical evidence on the private school effect mostly covers OECD and Latin American countries, with little evidence on other parts of the world. There is recent emerging country specific evidence doubting the existence of a private school advantage. I use PISA 2012 data for Mathematics and two different methodologies to derive baseline and bias-corrected estimates of the private-dependent and independent school effect for 40 countries. A robust private school advantage if found only in a handful of countries. Public schools generally perform equally well as private subsidised schools and outperform independent schools. Accounting for both peer effects and selection is necessary when evaluating school effectiveness, especially in the case of independent schools.