x.com/DKThomp/status/2111129498963902524
1 correction found
reduced hosting and dinner parties by 70% in 50 years
The 70% figure refers to one narrow measure—hosting friends or family at home at least monthly—not to dinner parties broadly. The cited sources say monthly dinner-party participation fell by about half from 1975 to 1998, while entertaining people at home at least once in the past year fell from 88% to 67% from 1995 to 2026.
Full reasoning
This line conflates different social measures and applies the largest one (about 70%) to both “hosting” and “dinner parties.”
Here’s what Derek Thompson’s own cited sources say:
- In his October 2026 Atlantic piece, Thompson writes that “the share of Americans who say they host friends or family at their home at least monthly has fallen from 42 percent in 1975 to 12 percent in 2026”—that is the statistic he calls a “70 percent collapse.”
- But for dinner parties, the same article says something different: “From 1975 to 1998, the share of Americans who gave or attended a monthly dinner party declined by half.”
- Thompson also cites a new Data for Progress replication showing that “entertaining people in your home” fell from 88% in 1995 to 67% in 2026—a 21-point drop, not 70%.
So the post overstates the evidence by treating hosting and dinner parties as if they both declined 70% over 50 years. The 70% number applies to a specific monthly-hosting measure, not to dinner parties generally.
2 sources
- Death of the Dinner Party - The Atlantic
The article says: “From 1975 to 1998, the share of Americans who gave or attended a monthly dinner party declined by half.” It separately says: “The share of Americans who say they host friends or family at their home at least monthly has fallen from 42 percent in 1975 to 12 percent in 2026—a 70 percent collapse in a half century.”
- Americans' Social Activity Has Steeply Declined in the Past 30 Years
Data for Progress reports that “entertaining people in your home” fell from 88% in 1995 to 67% in 2026, and notes that the 2026 comparison is against 1995 because that is “the earliest year in which full data for all comparison questions was available.”