The numbers are now granular enough to stop debating whether America's social decline is real. A new Data for Progress survey of 1,098 U.S. adults directly replicates the DDB Needham Life Style surveys that underpinned Robert Putnam's "Bowling Alone" thesis — and finds that every major category of in-person communal activity has suffered double-digit percentage-point declines since 1995. Entertaining people at home dropped 21 points (88% to 67%). Church attendance fell 20 points. Club meetings, picnics, and card games each dropped 19. Even going out to breakfast at a restaurant declined 13 points. The activities that grew are revealing by contrast. Internet use surged 90 points — an obvious artifact of web adoption — but the other risers are home video games (+22), exercise classes (+14), pop/rock concerts (+13), and personal airline trips (+5). The pattern is clear: Americans shifted toward solitary or consumption-oriented experiences and away from reciprocal, locally embedded social participation. You attend a concert; you don't co-create it. You play a video game; you don't host a card night. The methodological caveat matters and the researchers are honest about it. The 1995 DDB Needham survey used mail-based stratified quota sampling; the 2026 Data for Progress survey used web panels with post-stratification weighting. Mode effects are real — religiosity and loneliness questions are known to exhibit social desirability bias differences across modes. But the researchers argue, reasonably, that the magnitude and consistency of declines across many distinct activities constitutes a signal too large to attribute to mode alone. This survey adds texture to what was already directionally established. The American Time Use Survey has shown Americans spending more time alone and less time with friends. The American Perspectives Survey found that the share of Americans with five or more close friends dropped 25 points, from 63% to 38%, between 1990 and 2021. The Surgeon General's 2023 advisory declaring loneliness an epidemic was not hyperbole — it was descriptive. What the new data adds is activity-level specificity: this is not an abstract "people feel lonely" finding but a concrete "people stopped doing the things that generate social bonds" finding. The structural implications are severe. Home entertaining, club meetings, church attendance, and volunteer work are not just leisure activities — they are the infrastructure through which communities generate trust, resolve disputes, share information, and build mutual aid networks. Putnam's original argument was that declining social capital would corrode democratic institutions and civic capacity. Twenty-six years later, the data trail confirms the prediction. The question is no longer whether social capital is declining but whether any institution or policy lever exists that can reverse the trajectory. Notably absent from the rising activities is anything that rebuilds what was lost. Exercise classes are the closest analog to a communal social setting, and they grew — but a spin class is a weak substitute for hosting neighbors for dinner or attending a PTA meeting. The activities that replaced communal participation are overwhelmingly passive, individual, or market-mediated. The social fabric didn't just fray; it was replaced by a consumer experience layer that generates revenue but not reciprocity. If this trajectory continues for another 30 years, the consequences extend well beyond loneliness as a public health problem. Communities that cannot self-organize are communities that cannot respond to crises, hold local institutions accountable, or generate the informal networks that economic mobility depends on. The Surgeon General called it an epidemic. The data says it's a structural transformation — and no one is building the replacement infrastructure.
