Family Size and Domestic Violence in a High-Fertility Society

TTPI - Working Paper 5/2023

40 Pages Posted: 7 Jun 2023

See all articles by Dyah Pritadrajati

Dyah Pritadrajati

Australian National University (ANU)

Date Written: May 31, 2023


Does family size affect the prevalence of domestic violence? Using nationally representative survey data from Samoa, which has among the world’s highest fertility rates, I extend the classic work on child quantity-quality trade-offs to also consider domestic violence. Identification is based on instrumental variable (IV) strategies exploiting three distinct and plausibly exogenous drivers of additional fertility: (1) same-sex sibling pairs in families with two or more children, (2) multiple births (twin), and (3) a female firstborn. I find evidence of a direct causal link between family size and an increased prevalence of intimate partner violence by, on average, 5 percentage points, equivalent to a 13 percent increase from the mean. This significant effect is largely driven by physical or sexual abuse often associated with serious victim injuries. The IV estimates also suggest that larger families tend to have attitudes that condone violent behaviour. The normalisation of violent behaviour in larger families may be linked to a lack of resources available to effectively address and resolve conflicts, ultimately contributing to an increased likelihood of violent incidents. These findings highlight the need for greater awareness of the potential victimisation risks for larger families and the importance of integrated family planning and domestic violence prevention efforts.

Keywords: Family size, domestic violence, instrumental variable

JEL Classification: : D19, J13, J16, O10

Suggested Citation

Pritadrajati, Dyah, Family Size and Domestic Violence in a High-Fertility Society (May 31, 2023). TTPI - Working Paper 5/2023, Available at SSRN: or

Dyah Pritadrajati (Contact Author)

Australian National University (ANU) ( email )

Canberra, Australian Capital Territory 2601

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
PlumX Metrics