Prenatal Sex Selection after Migration: Evidences from Italy

Elena Ambrosetti, Sapienza Università di Roma
Livia Elisa Ortensi, University of Milan - Bicocca
Cinzia Castagnaro, Istituto Nazionale di Statistica (Istat)
Marina Attili, Istituto Nazionale di Statistica (Istat)

This paper aims to explore sex ratio at birth (SRB) of migrants in Italy in order to shed light on the possible phenomenon of sex selection at birth. Recent studies addressed the same issue for migrants of Indian and Chinese origin living in Italy (Meldolesi, 2012; Blangiardo and Rimoldi, 2012). The study of Meldolesi was limited to the period of 2006-2009 and used data on birth records, while the latter study uses data from a 2011 survey of 700 women of Chinese and Indian origin in the Lombardy region.

Our objective is to go beyond these studies, analysing births from mothers with a foreign background, from countries where sex selection at birth is widespread and that are among the largest immigrant communities in Italy. The paper aims at assessing 1) if a skewed sex ratio at birth is observed among overseas communities; and 2) the possible factors affecting skewed SRB in the migratory context. When studying the phenomenon of sex selection before birth, it is important to stress that the SRB increases with birth order, as prenatal discrimination with first births is generally infrequent (Guilmoto, 2015). Prenatal sex selection is practiced for higher order births, while first and second births are often left to chance.

Preliminary results show that the sex ratio at birth is above the biological constant for the period 2005-2015 for births of Albanian, Chinese and Tunisian couples from the third child and Indian couples from the second child with a confidence interval of 95% above 105 (Ambrosetti, et al. 2017).

The study will proceed with a logistic regression model only for citizenship with imbalanced sex ratio at birth and with significant presence in Italian territory using different data sources linked together. The dependent variable is the sex of the new-born child.


  1. Aim of the study

This paper aims to explore sex ratio at birth (SRB) of migrants in Italy in order to shed light on the possible phenomenon of sex selection at birth. Recent studies addressed the same issue for migrants of Indian and Chinese origin living in Italy (Meldolesi, 2012; Blangiardo and Rimoldi, 2012). The study of Meldolesi was limited to the period of 2006-2009 and used data on birth records, while the latter study uses data from a 2011 survey of 700 women of Chinese and Indian origin in the Lombardy region.

Our objective is to go beyond these studies, analysing births from mothers with a foreign background, from countries where sex selection at birth is widespread and that are among the largest immigrant communities in Italy. The paper aims at assessing 1) if a skewed sex ratio at birth is observed among overseas communities; and 2) the possible factors affecting skewed SRB in the migratory context. When studying the phenomenon of sex selection before birth, it is important to stress that the SRB increases with birth order, as prenatal discrimination with first births is generally infrequent (Guilmoto, 2015). Prenatal sex selection is practiced for higher order births, while first and second births are often left to chance.

  1. Data and methods

Data used to elaborate descriptive statistics on births at the national level stems from the survey on births from the Resident Population Registers. Live births registered in the Resident Population Registers provide a descriptive analysis of the phenomenon. The last data available refer to 2015 statistics. Average SRB is calculated for the period of 2005-2015 for same-citizenship couples.

As the precision of SRB calculated on the small annual number of births is limited - being subjected to the binomial law - we provided the information on the statistical significance based on a 95% confidence interval for each year and for the whole period of analysis.

A preliminary study was conducted for children born to foreign couples to analyze the sex ratio at birth by order. Preliminary results show that the sex ratio at birth is above the biological constant for the period 2005-2015 for births of Albanian, Chinese and Tunisian couples from the third child and Indian couples from the second child with a confidence interval of 95% above 105 (Ambrosetti et al. 2017).

To improve our analysis taking into consideration the birth’s order, once completed the database containing the offspring of all the Italian and foreign women living in Italy who have had at least one child registered in the Italian Registry from 1999 to 2015 (database under construction by Istat), it’d be possible to calculate the sex ratio of follow-up orders as a function of the sex of the first order births.

This new database, containing a huge and valuable information asset for reading transversal and longitudinal data, allows to perform demographic and social analyses.

To study the determinants of the phenomenon the information from the new database (with precise estimates of the order of birth) will be integrated with those from Delivery Certificate Survey, with the aim of outlining, in the country of destination, the factors related to this discriminatory behaviour towards girls. CEDAP is a continuative data source since 2002 on annual deliveries, collected in hospitals’ birth departments. The dataset contains information on parents’ socio-demographic background, mothers’ reproductive history, pregnancy characteristics, delivery and neonatal characteristics.

The study will proceed with explorative multivariate analyses on different data sources linked together; preliminary studies were applied for the region of Lombardy because it is one of the Italian regions most affected by immigration. Although referring only to a region, in the two-year span of analysis, foreign children born in Lombardy accounted for 27% of the total number of foreign children born in Italy (Istat, 2015).

To carry out multivariate analysis on data referred to Italy, we’ll implement a logistic regression model only for citizenship with imbalanced sex ratio at birth and with significant presence in Italian territory. The dependent variable is the sex of the newborn child.

Among the possible covariates: the parents'' age, educational level, profession, previous abortion, the presence of prenatal diagnosis (villocentesis, amniocentesis, number of ultrasound examinations) in order to assess the possibility that these medical examinations are used as a prenatal sex selection tool and the birth order (presence of one, two or more previous children), a key variable for the study of birth selection (Guilmoto, 2015); once completed the new database containing the offspring of all the Italian and foreign women living in Italy, it’d be possible to consider, between covariates, the birth order by sex.

Presented in Session 1087: International Migration and Migrant Populations