Mazrekaj, Deni, Kristof De Witte, and Sofie Cabus. “School Outcomes of Children Raised in Same-Sex Families: Evidence from Administrative Panel Data.” Revise and Resubmit. American Sociological Review.
[2018 Impact Factor: 5.391, 2/148 in Sociology, ABS Ranking: 4*]
Although widely used in policy debates, the literature on children’s outcomes in same-sex families has mostly relied on small selective samples or on samples based on cross-sectional survey data. This led to a lack of statistical power and the inability to separate children born in same-sex families from children of divorced parents. We address these issues by using unique administrative longitudinal data from the Netherlands: the first country to legalize same-sex marriage in the world. The results indicate that children raised in same-sex families perform better than children raised in different-sex families in both primary and secondary education. Our findings are robust to the use of Cousin Fixed Effects, as well as Coarsened Exact Matching to improve covariate balance and to reduce model dependence. Further analyses using a novel bounding estimator suggest that the selection on unobserved characteristics would have to be more than two and a half times higher than the selection on observed characteristics to reduce the positive estimates to zero.
Mazrekaj, Deni, and Sofie Cabus. “Does a High School Diploma Matter? Evidence Using Regression Discontinuity Design.”
The current study explores whether education raises productivity (human capital theory) or just reflects it (sorting theory) by estimating earnings returns to a high school diploma. For this purpose, we compare earnings of students who barely passed and barely failed standardized high school exit exams in the Netherlands. These students have similar levels of human capital but different diploma status. Using a regression discontinuity design on administrative population data, we find that a diploma raises hourly net earnings by about 4%. Thereby, the results indicate that a high school diploma serves as an important signal on the labour market.
Mazrekaj, Deni, Fritz Schiltz, and Vitezslav Titl. “Identifying Politically Connected Firms: A Machine Learning Approach.
This article introduces machine learning techniques to identify politically connected firms. We use a unique dataset of all contracting firms from the Czech Republic. In this dataset, various forms of political connections can be determined from publicly available sources. The results indicate that over 75% of firms with political connections can be accurately identified. The model obtains this high accuracy by using only firm-level financial and industry indicators that are widely available in most countries. Compared to the logistic regression model that is commonly used to predict binary outcome variables, the proposed technique can increase the accuracy of predictions by up to 36% using the same set of variables and the same data.