Two siblings playing i the woods
Photo: Mostphotos/Gabby Baldrocco


Siblings' educational choices are correlated, and associated with factors such as parental education and occupation. Researchers have now been able to isolate a direct effect, and measure how an older sibling's educational experience affects the choices of younger siblings.

“We see the same pattern in Chile and Sweden, despite large differences in everything from education systems to culture and the degree of inequality in society” said Adam Altmejd, PhD in economics at the Swedish Institute for Social Research (SOFI) at Stockholm University and Stockholm School of Economics, and one of the authors of the study.

Younger siblings follow their older, both to better and worse* majors. They follow also to the same college, regardless of specialization.

“We find that the older siblings' experience impact individuals in all types of families. In the US, we see a much stronger effect in families that are unlikely to send their kids to college. We find that in this group, the probability of studying at any four-year college is up to six times higher for individuals with older siblings who do so”, said Adam Altmejd.

The effect exists between siblings with an age difference of more than five years, which suggests that it is not driven by siblings wanting to study together. However, if the older sibling drops out, the effect disappears.

 

Possible explanation for persistent inequality

Social networks might be an important driver of persisting educational inequality, even in societies with low income inequality like Sweden, the researchers believe.

“We want to understand why we still find such large differences in the level of education between communities. Why children of people without a university degree have a much lower probability of studying of continuing to higher education, even in Sweden where university education is free. These research results show the importance of academic role models”, said Adam Altmejd.

Researcher Adam Altmejd, Ph D in Economics at SOFI, Stockholm University
Adam Altmejd, Ph D in Economics at SOFI

From previous research, mainly in sociology, we know that family background is strongly associated with choice of education. With the statistical analysis method now used, researchers have been able to separate causation from correlation and identify the older sibling's direct impact, within different types of families and educational systems.

“What we can take with us when educational issues are debated is that the social context is extremely important for which educational path individuals take. When we plan efforts to reach groups that have difficulties finding higher education, it’s important keep in mind that an older sibling can open up such paths for their younger siblings.

* The program is worse than what the younger sibling would otherwise have applied for in terms of average starting salary among graduates, average grades among enrolled students and the share that continues in the second year of the program.

 

More about the study

The study is published as an "advance article" on the journal's website and will be published in August. The study is peer reviewed.

Data from four countries with completely different education systems were studied, Sweden, USA, Croatia and Chile. In Sweden, data is obtained from the Swedish Council for Higher Education, the Swedish national archives and Statistics Sweden. In total, approximately 240 000 Swedes were included in the sample which consisted of applicants to universities and colleges since 1993 who were just above or below the admission threshold and had at least one younger sibling. In total, 440 000 applicants and their siblings were analyzed across all four countries.

The statistical method used to identify a causal relationship is called regression discontinuity design and is quasi-experimental, which entails trying to get the same features from the data as from a randomized experiment. As the exact admission threshold for an education program is not known in advance, applicants who are close to that cutoff will not know for sure whether they will be admitted or not. The method studies only these applicants. Comparing those who are are just above the admission limit to those slightly below is similar to randomly assigning admission and makes it possible to determine the causal relationship between the older sibling's admission and the younger’s application behavior.

 

Contact

Adam Altmejd, PhD in economics at the Swedish Institute for Social Research (SOFI) at Stockholm University, mobile +46 734 20 01 20 , e-mail adam.altmejd@sofi.su.se