Behavior

The DNA of a Student's Peers Might Help Them Stay in School

A study of the "social genome" of 12 to 18-year-olds found that the genetics of a student's classmates influenced their educational attainment.

A mathematics teacher (in blue) works with high school students in the Manor New Tech High School, Manor, Texas. | Robert Daemmrich Photography Inc/Corbis via Getty Images
A mathematics teacher (in blue) works with high school students in the Manor New Tech High School, Manor, Texas. | Robert Daemmrich Photography Inc/Corbis via Getty Images

School-aged friends might have more in common than a favorite color or toy. They might share some genes, as well, thanks in part to their shared environment, according to a new study. The research builds on previous observations that people in close relationships, including spouses and adult friends, tend to be genetically similar. Now, scientists are asking how and why this happens.

The study, published in the journal Proceedings of the National Academy of Sciences, focused on the idea of a “social genome,” where one person’s genetic makeup may influence another person’s behavior. This could have implications for a range of things, including how much school a child eventually completes.

"It seems that in some cases the genetics of other people are going to have implications for us,” Ben Domingue, lead author of the study and an assistant professor at Stanford Graduate School of Education, told Seeker.

The researchers started with a large dataset from the National Longitudinal Study of Adolescent to Adult Health, known as Add Health, which surveyed 90,118 students between 12 and 18 years old during the 1994-1995 school year. The study asked participants to list their friends, which was then used to identify social groups in different schools. Researchers then took 20,745 students and conducted in-home interviews with them and their parents. Follow-up interviews were conducted four times over 14 years. During the last interview in 2008-2009, roughly 12,000 subjects provided DNA samples. The researchers focused on around 5,500 of these students for their study, chosen for their similar ancestry.

The researchers then linked the DNA collected in 2008 with data collected in the original survey on friend groups, as well as information from the four other surveys that included a broad range of characteristics including height, weight, and grade point average. They then asked: Are friends more similar genetically then when compared to other random peers? How does school assignment affect friends’ genetic similarity? And what impact does a social genome have on an individual, such as their height, body mass index, and educational attainment.

“As an education researcher I’m sensitive to the fact that educational attainment is a fairly socially contextualized outcome, so this is kind of curious,” said Domingue, when talking about the rising number of links being made between genes and different factors, including how far a student will go in school.

RELATED: Chimps and Six-Year-Olds Want to Mete Out Punishment — Even If It Comes at a Cost

For the study, the scientists computed genetic kinships — the metric of how much genetic material you share with others — of all potential pairs in the data. Comparing the kinships of friends with the kinships of random people from the same school, the authors found, based on scale where identical twins would be a 1, that there was more genetic similarity between friends: 0.031 amongst friends, or about the same level of genetic similarity between third or fourth cousins, compared to 0.019 amongst schoolmates. To put that into context, a previous study by Domingue found that spouses have a 0.045 similarity.

These results suggested that the genetic likeness might be a result of two things: social homophily, where individuals are drawn to those that are like themselves (“Birds of a feather flock together.”), and social structuring, where individuals form bonds with people sharing the same environment. The genetic similarity between classmates suggests that a shared environment might be responsible for a large chunk of the genetic similarity among friends.

“What’s happening is you’re already in schools that are similar genetically, so it’s larger structural forces that are placing you in an environment where your choice of friends is already restricted,” said Domingue.

The study showed that the educational attainment of a student is largely associated with the genetics of his or her friends and classmates. This wasn’t true for height — important, because height is not related to the school environment like educational attainment.

These genetics could be a proxy for many things, including school quality, said Domingue. But, “the basic idea is your social environment has a lot to say about how far you go in school,” he said.

The Add Health study is funded by the NIH–Eunice Kennedy Shriver National Institute of Child Health and Human Development, as well as 23 other federal agencies and foundations.

RELATED: To Friend or Unfriend? How Your Relationship Style Plays Out in Social Networks

The idea of studying shared genes and how they might predict certain traits isn’t new. Scientists have explored the genetic similarities between adult friends, and others have tried to understand which genes can predict someone’s height or if they have a predisposition for certain diseases.

But, Domingue stressed that we can’t treat all phenotypes — observable properties of a person like appearance and behavior — the same. Genes that may predict someone’s physical characteristics may not be completely comparable to genes that are impacted by social forces, including those that may predict how much education someone will receive.

“You can’t do genetic analysis of educational attainment in the exact same fashion that you did with height and BMI,” said Domingue. “That will introduce biases because those phenotypes weren’t socially in context.”

Moving forward, Domingue hopes that more genetic studies will consider the way an environment might influence certain genetic factors.

“Outcomes that are socially contextualized,” she said, “may require special handling.

WATCH: How to Change Our Own Genes