This post was adapted from articles by Will Dunham for Reuters, Sarah Kuta for the Daily Camera, and Dennis Thompson for Healthday.
School of Education alumnus Ben Domingue, PhD (REM, 2012) recently released a study conducted with CU Boulder Sociology Professor Jason Boardman, that has garnered a great deal of attention in the national news. Both are researchers for CU's Institute of Behavioral Science.
The study, published in the Proceedings of the National Academy of Sciences, compared married pairs to non-coupled pairs and found that spouses are more genetically similar than two randomly selected individuals within the same sample. Domingue and colleagues focused the study on 825 white, non-Hispanic couples in the University of Michigan's Health and Retirement Study, a longitudinal survey of 26,000 Americans older than 50 conducted every two years, controlling for ethnicity and geographic location to the extent possible.
"We know that people tend to marry individuals that are like themselves in terms of education level," said Domingue. "Our question was, do we observe a genetic preference when it comes to spouse selection, and if so, how does that compare and interact with education?"
Educational similarities do trump genetics, with genetic similarity between spouses roughly one-third the magnitude of educational-level similarity.
Domingue and colleagues measured genetic similarity by comparing 1.7 million individual DNA building blocks, known as single nucleotide polymorphisms, in the study participants. They compared the genetic makeup of the married couples to other randomly chosen people of the opposite sex in the same pool of study participants.
Domingue said the actual mechanism for a person being drawn to another person's genetic similarities is probably complicated and multifaceted - "just a whole host of things," he noted.
The researchers noted that people usually pick spouses with similar backgrounds and characteristics in addition to education, including race, religion, age, income and body type. Genetic similarity can be added to the list, they said, which has implications for statistical models used to predict the role of genetics in health and disease in families and populations that traditionally have assumed that mating is random with respect to genetics.
"For example, people clearly care about height in picking partners. To the extent that tall people marry other tall people, that is going to result in genetic similarity among spouses. But it is difficult to know whether height or genes are driving this decision," Domingue said.
Neil Risch, director of the Center for Human Genetics at the University of California, San Francisco, points out that “genes are, in a sense, a bystander. … Historically, in a highly ethnically and geographically structured population, say Chicago, it might have been the case that Eastern Europeans only married other Eastern Europeans, Southern Europeans only married Southern Europeans, and Northern Europeans only married Northern Europeans. This would lead to significant correlations for genetic factors differentiating these ethnic groups, but has nothing to do with any traits or characteristics which underlie mate choice."
Domingue agrees with Risch’s point, and was careful to avoid causal explanations for the tendency for married couples in this sample to share genetic similarities. A natural extension of the study, he said, is to look at the genetic similarities of couples in other cohorts, such as same-sex pairs, cross-race pairs and non-white pairs. Other studies might offer a more nuanced look at which regions of the genome are similar among couples, which are different, and why.
Other study authors were Professor Jason Fletcher of the University of Wisconsin and Professor Dalton Conley of New York University.