Gaia Ghirardi

Gaia Ghirardi

Sociology PhD researcher at the EUI

Selected ongoing projects

Sociogenomics

Ghirardi, G. & Bernardi, F. Compensating or Boosting Genetic Propensities? Gene-Family Socioeconomic Status Interactions by Educational Outcome Selectivity Here the pre-print (R&R)

Abstract This study investigates the extent to which the genetic propensity for education - measured using the polygenic index (PGI) for educational attainment – matters more for the final educational attainment of high or low socio-economic status (SES) students. We propose a model integrating social stratification theories, such as the compensatory and boosting advantage models, into sociogenomics, highlighting the role of educational outcome selectivity. Our model predicts that for low selective educational outcomes (e.g., high school completion), the PGI for education matters more for low- SES individuals, while for highly selective outcomes (e.g., graduate school completion), it matters more for high-SES individuals. We test our model using the National Longitudinal Study of Adolescent to Adult Health, the Health and Retirement Study, and the Wisconsin Longitudinal Study. The results corroborate our predictions and are robust to alternative models’ specifications. Our theoretical model explains previous heterogeneous findings and can be generalized to develop testable hypotheses for other cohorts in the US and other countries.


Bernardi, F., & Ghirardi, G., Socioeconomic Status, Genes, and the Differential Effects of Parental Separation on Educational Attainment. Here the pre-print (R&R)

Abstract Prior research has consistently documented a more pronounced negative impact of parental separation on educational attainment among children from high socio-economic status (SES) backgrounds. This study leverages molecular data to investigate how the parental separation penalty on educational attainment varies by SES and children’s genetic propensity for education. We replicate the analysis on two distinct datasets, the National Longitudinal Study of Adolescent to Adult Health (Add Health) and the Health and Retirement Study (HRS), using two different dependent variables, the probability of college attainment and years of education, parametric (OLS and logit) and non-parametric models (LOWESS) and mother’s education as an indicator of family SES. Our results show that the parental separation penalty clusters among high-SES students with a low genetic propensity for education. While for high-SES students with non-separated parents, the probability of college attainment and completing more years of education is largely independent of their genetic propensity for education, it notably reduces if they have a low genetic propensity for education and their parents separate. These findings suggest that when high-SES parents separate, they experience a reduced capacity to compensate for their children's low genetic propensity for education on college attainment and years of education.


Ghirardi, G., Wolfram, T., Ruks, M., & Tropf, F., Family Socioeconomic Status Compensates and Triggers Genetic Propensity for Externalizing Behavior on Adverse High School Outcomes

Abstract Previous research has investigated how genetic propensities for educational attainment and family socioeconomic status (SES) interact in shaping educational outcomes. Theoretical considerations in this line of research are leading to the Scarr-Rowe hypothesis, that genetic effects are stronger in higher SES compared to lower SES families, and genetic propensities for education are interpreted as encoding genetic potential for cognitive ability. However, various mechanisms link the genome and educational outcomes and require tailor-made theoretical consideration. This study specifically focuses on the children's genetic propensity for externalising behaviour to investigate whether this genetic propensity affects children’s likelihood of experiencing adverse educational outcomes in high school and whether this relationship varies by family SES. Drawing upon behavioural genetics and social stratification theories— the compensatory advantage or social control and diathesis-stress models—we test the predicted stronger association between the genetic propensity for externalising behaviours and adverse school outcomes among low-SES children compared to their high-SES peers. Using data from the National Longitudinal Study of Adolescent to Adult Health (Add Health) from the US, we construct a polygenic index (PGI) for externalising behaviours, and we look at different adverse high-school educational outcomes, such as school failure, suspension, expulsion, absence, and misbehaviours. The findings reveal two key insights. First, a higher genetic propensity for externalising behaviours increases the likelihood of experiencing adverse high-school outcomes. Second, the genetic propensity for externalising behaviours is more predictive of adverse high-school educational outcomes among low-SES children compared to high-SES children. The robustness of these findings is confirmed through several checks, including replication with the British National Child Development Study (NCDS), alternative modelling, correcting for biases in polygenic scores, and using various socioeconomic status measures. Our results suggest that high-SES families compensate for their children's high genetic propensity for externalizing behaviors, whereas low-SES families trigger these genetic propensities in their children.


Breinholt, A., & Ghirardi, G., The Interaction between Socioeconomic Status and Genetic Propensity for Education: A Systematic Review and Meta-Analysis of Molecular Studies Here the pre-registration

Abstract A growing body of research investigates whether genetic influences on educational outcomes can be suppressed or boosted by the environment in which an individual lives. Indeed, several studies have examined the interaction between the genetic propensity for education and the socioeconomic status (SES) of the family of origin, the neighborhood, or school - so-called GxSES studies. However, results are mixed. Some studies find stronger associations between genetic propensities and educational outcomes in high-SES environments, while other studies find stronger associations in low-SES environments. In this work, we conduct a systematic review and meta-analysis of studies that investigated the interaction between genetic propensity for education and SES on measures of educational attainment and achievement. We focus on studies relying on molecular genetic information in the form of polygenic indices (PGI). This study aims to answer the following questions: (1) Does the genetic propensity for education matter more for educational outcomes in high or low-SES environments? (2) Which factors may explain the inconsistent findings of the previous literature? To address these questions, we collect results from both published and working papers investigating the interaction of interest and using molecular data.


Ghirardi, G., The Development of Body Mass Index from Adolescence to Adulthood: A Gene-Family Socioeconomic Status Interaction Study

Abstract Body weight in adolescence and adulthood may result from the interplay between individuals' genetic characteristics and the social context in which they grow up, such as family socioeconomic status (SES). However, evidence on the interaction between genetic propensity for high body mass index (BMI) and family SES remains inconclusive. This study investigates whether the genetic propensity for high BMI impacts BMI differently among individuals from high-SES versus low-SES families and whether this effect varies with age. Three theoretical frameworks—compensatory advantage or diathesis-stress model and social push model —are tested to predict the moderating effect of family SES on the genetic association with BMI. Drawing on data from the National Longitudinal Study of Adolescent to Adult Health (Add Health), I use the polygenic index for BMI (BMI PGI) to measure genetic propensity for body weight and analyze it at various life stages: adolescence (≃ 16 years old), early adulthood (≃ 22 years old), adulthood (≃ 28 years old), and later adulthood (≃ 37 years old). Results indicate that the BMI PGI is more predictive of BMI in individuals from low-SES families than those from high-SES families. This interaction is observed across both younger and older age groups, suggesting that the interaction between BMI PGI and family SES persists throughout the life course, especially among those who show high levels of BMI – as revealed through unconditional quantile regression. These findings underscore the importance of adopting a life-course perspective in gene-environment interaction studies and highlight the need to account for potential heterogeneity in the gene-environment effect across different levels of BMI.


Early childhood and care

Ghirardi, G., & Gioachin, F. The Role of Different Childcare Arrangements on Social Inequalities in Children’s Early Cognitive Competencies: A Research Note. Here the pre-print

Abstract This research note investigates how childcare arrangements under age three affect early social inequalities in children’s competencies, relying on a novel decomposition approach (Yu and Elwert 2023). This approach allows us to provide a comprehensive evaluation of different childcare arrangements within a potential outcome framework, revealing not only the overall relevance of each childcare arrangement but also the importance of specific channels impacting social disparities such as prevalence (unequal exposure), effect (heterogeneous effect), and selection (within-group propensity to treatment). Using the newborn cohort of the German National Educational Panel Study (NEPS - SC1), we assess the impact of early childhood education and care (ECEC), family day-care, grandparental care, and exclusive parental care on children’s competencies in mathematics and vocabulary at ages 4 and 5. Findings highlight ECEC as the most effective approach in reducing social disparities, positively influencing children from lower socio-economic backgrounds. Conversely, exclusive parental care and family day-care increase social disparities, benefiting children from higher socio-economic backgrounds while negatively affecting those from lower backgrounds. This comprehensive evaluation identifies channels impacting social inequalities, contributing to understanding the nuanced role of childcare arrangements in shaping early social inequalities in children’s competencies.


Berger, L., Ghirardi, G., Panico L., & Solaz, A., The Impact of Early Formal Childcare Attendance and Exposure on Children’s Developmental Outcomes in France

Abstract The importance of early childhood education and care (ECEC) for fostering children’s development and reducing early inequalities is widely recognized by previous studies. However, much of this evidence refers to pre-school (i.e., to 3-to 5-year-olds), and mainly stems from randomized evaluations of small-scale intensive programs based in the United States and other English-speaking countries. This study thus examines the impact of ECEC under the age of 3 on children’s development and its role in tackling the early social inequality in such outcomes in France. We ask whether attending ECEC at age 2 has a positive (or negative) impact on children’s developmental outcomes and if the role of ECEC on child development is different according to the family’s socio-economic status (SES). We address these questions, using the Étude Longitudinale Français depuis l’Enfance (Elfe), and ordinary least squares (OLS) models with inverse probability weight (IPW) to account for confounding bias possibly arising from selection into care arrangements. Our findings show that ECEC attendance positively affects children’s development and that disadvantaged children benefit more in children’s global development from attending ECEC. These results suggest that ECEC institutions may be an effective tool to reduce social inequalities in some, but not all, early child developmental outcomes.