Changing Effects of Birth Order on Education Over Time: Evidence From Cambodia

Abstract

Despite the vast literature on birth-order effects in developing countries, there has been little research on their long-term changes. This study uses three population censuses of Cambodia to investigate how birth order affects educational attainments and how its effects changed over time. We estimate regressions with family fixed effects and cohort fixed effects, and present four findings. First, later-born children are less likely to enroll in primary school and gain literacy skills, and end up with fewer years of schooling than first-born children. Second, controlling for gender differences in education, we find smaller birth-order effects for girls. Third, there is a strong relationship between the gender composition of siblings and birth-order effects. Finally, not only have these birth-order effects attenuated over time but the relationship between the gender composition of siblings and birth-order effects has also changed. In the 1998 census, if the firstborn child is male, his sisters’ educational outcomes are influenced more by their birth order compared with his brothers. In contrast, in the 2019 census, if the firstborn child is male, his sisters' educational outcomes are less susceptible to birth-order effects compared with his brothers. Our findings indicate gendered economic disparities within households and their long-term changes.

Keywords:

• birth order
• education
• Cambodia

Acknowledgements

We thank Tien Manh Vu for his invaluable comments and suggestions. All remaining errors are our own. Although we are not allowed to share the census data with other researchers, we explain how to construct our sample in detail in the main text and the Supplementary Material.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 As explained by Song and Chea (2023), the length of vocational training differs by qualification (specified by the Cambodia Qualifications Framework). However, the details on respondents’ vocational training are not available in the GPCC, and the percentage of people who have received formal vocational training is very small in Cambodia (0.49% of our sample in 2008, and 0.54% of our sample in 2019; there is no information for 1998). Accordingly, we will set pre-secondary vocational qualifications equal to 12 and post-secondary vocational qualifications equal to 14.

2 The estimated coefficients are visualized with 95% confidence intervals for the sake of comparison across the three waves of GPCC in Section S3 in the Supplementary Material.

3 The definition of “employment” in the GPCC includes own-account workers (70% of the employed) and unpaid family workers (9% of the employed) among male respondents aged 24 – 64. Among female respondents in the same age range, own-account workers and unpaid family workers account for 35% and 51% of the employed, respectively.

Additional information

Funding

This work was supported by JSPS (Japan Society for the Promotion of Science) KAKENHI [Grant Number 22K18534].