In The Free Press, Jonathan Haidt notes the inflection point at which young liberal women started to become depressed at a much higher rate than the rest of the population — a trend that has continued for over a decade:
In September 2020, Zach Goldberg, who was then a graduate student at Georgia State University, discovered something interesting in a dataset made public by Pew Research. Pew surveyed about 12,000 people in March 2020, during the first month of the COVID shutdowns. The survey included this item: “Has a doctor or other healthcare provider EVER told you that you have a mental health condition?” Goldberg graphed the percentage of respondents who said “yes” to that item as a function of their self-placement on the liberal-conservative 5-point scale and found that white liberals were much more likely to say yes than white moderates and conservatives. (His analyses for non-white groups generally found small or inconsistent relationships with politics.)
I wrote to Goldberg and asked him to redo it for men and women separately, and for young vs. old separately. He did, and he found that the relationship to politics was much stronger for young (white) women. You can see Goldberg’s graph here, but I find it hard to interpret a three-way interaction using bar charts, so I downloaded the Pew dataset and created line graphs, which make it easier to interpret.
Here’s the same data, showing three main effects: gender (women higher), age (youngest groups higher), and politics (liberals higher). The graphs also show three two-way interactions (young women higher, liberal women higher, young liberals higher). And there’s an important three-way interaction: it is the young liberal women who are highest. They are so high that a majority of them said yes, they had been told that they have a mental health condition.
Data from Pew Research, American Trends Panel Wave 64. The survey was fielded March 19–24, 2020.
Graphed by Jon Haidt.In recent weeks — since the publication of the CDC’s report on the high and rising rates of depression and anxiety among teens — there has been a lot of attention to a different study that shows the gender-by-politics interaction — Gimbrone, Bates, Prins & Keyes (2022), titled: “The politics of depression: Diverging trends in internalizing symptoms among US adolescents by political beliefs”. Gimbrone et al. examined trends in the Monitoring the Future dataset, which is the only major U.S. survey of adolescents that asks high school students (seniors) to self-identify as liberal or conservative (using a 5-point scale). The survey asks four items about mood/depression. Gimbrone et al. found that prior to 2012 there were no sex differences and only a small difference between liberals and conservatives. But beginning in 2012, the liberal girls began to rise, and they rose the most. The other three groups followed suit, although none rose as much, in absolute terms, as did the liberal girls (who rose .73 points since 2010, on a 5-point scale where the standard deviation is .89).
Data from Monitoring the Future, graphed by Gimbrone et al. (2022). The scale runs from 1 (minimum) to 5 (maximum).
The authors of the study try to explain the fact that liberals rise first and most in terms of the terrible things that conservatives were doing during Obama’s second term, e.g.,
Liberal adolescents may have therefore experienced alienation within a growing conservative political climate such that their mental health suffered in comparison to that of their conservative peers whose hegemonic views were flourishing.
The progressive New York Times columnist Michelle Goldberg took up the question and wrote a superb essay making the argument that teen mental health is not and must not become a partisan issue. She dismissed Gimbrone et al.’s explanation as having a poor fit with their own data:
Barack Obama was re-elected in 2012. In 2013, the Supreme Court extended gay marriage rights. It was hard to draw a direct link between that period’s political events and teenage depression, which in 2012 started an increase that has continued, unabated, until today.
After examining the evidence, including the fact that the same trends happened at the same time in Britain, Canada, and Australia, Goldberg concluded that “Technology, not politics, was what changed in all these countries around 2012. That was the year that Facebook bought Instagram and the word ‘selfie’ entered the popular lexicon.”
