For many Americans, polarization is a major concern. They worry that diverging worldviews threaten to fracture a common American identity into competing subgroups, making it harder to interact, communicate, and trust one another.

And while there are some signs of increasing division, it’s unclear how much American culture is really diverging.

In a paper in the American Economic Journal: Applied Economics, authors Marianne Bertrand and Emir Kamenica developed a new measure of cultural distance and tracked it over a recent 60-year period.

Using machine learning techniques, the researchers were able to predict group membership based on media diet, consumer behavior, time use, social attitudes, and newborn names. They argue that the likelihood of correctly guessing an individual’s group membership based on those variables serves as a good indicator of cultural distance—where a higher chance of guessing group membership correctly suggests a larger cultural gap between groups. 

Figure 1 in the authors’ paper shows how these indicators changed in the United States from the 1960s to the 2010s. Income, education, gender, race, and political ideology correspond to high or low income, high or low education, male or female, White or non-White, and liberal or conservative groups, respectively.

Figure 1 from Bertrand and Kamenica (2023)

The chart shows only a few trends pointing toward larger cultural divides. The most prominent divergence was seen when predicting political ideology based on social attitudes (orange line in Panel E). But some divergence is also seen when guessing income based on social attitudes (orange line in Panel A) and race based on consumer behavior (red line in Panel D).

The rest of the indicators showed no clear trends toward greater polarization over the last half century. In fact, between 1965 and 1995, men and women’s time use became more similar (green line in Panel C), but there was no change over the last 20 years of the sample. 

The authors say that their results do not support the view that US cultural divides are growing. Their findings suggest that cultural distance between many groups has been roughly constant since the 1960s.