Research institute · Pew Research Center
36% of US Adults Belong to an Online Community
Pew's survey of 9,750 US adults finds 36% belong to an online community, most often on social media or forums, and 42% of members say advice or recommendations is a major reason they visit. For brands, that makes communities a recommendation channel to earn a place in, not one to buy.
Source published Summarised
Key findings at a glance
- 36% of US adults say they are part of at least one online community
- 48% of US adults aged 18 to 29 are part of an online community
- 26% of US adults belong to an online community on social media
- 15% of US adults belong to one on discussion forums such as Reddit, Nextdoor or Discord
- 23% of US adults are part of an online community about a hobby
- 21% of US adults are part of an online community about their local area
- 42% of community members say getting advice or recommendations is a major reason they visit their main community
- 74% of community members post or share in the community they visit most
- 56% of community members visit their most-visited community every day
- 54% of community members say political discussion does not belong in their main community
- 34% of US adults have left an online community at some point
- 85% of those who have left a community cite losing interest as a reason
What the report covers
Pew Research Center's report Online Community and Connection in America, published on 14 October 2026 by its Pew-Knight Initiative, measures who belongs to online communities, where those communities form, what they are about, why people visit and why they leave. Pew defined an online community broadly: people connecting with others online by joining or feeling part of a group, or doing things together, whether on social media, discussion forums, group messaging or online games.
The headline is that 36% of US adults say they are part of at least one online community. That rises to 48% of 18 to 29-year-olds and falls to 26% of those aged 65 and over. Social media is where most of it happens, with 26% of all adults in a community there, followed by discussion forums such as Reddit, Nextdoor and Discord at 15%. By subject, hobbies (23%) and the local area (21%) lead. Members are active: 74% post or share in the community they visit most, 56% go there every day, and 63% say it offers something they could only find online. Among members, 57% say staying informed is a major reason they visit and 42% say the same of getting advice or recommendations.
The report also covers politics and departures. 54% of members say political discussion does not belong in their main community, and 71% see little or none of it there. 34% of adults have left an online community at some point, most often because they lost interest (85% of leavers), because it was not as good as it used to be (71%) or because of too much arguing (57%). It follows our report on what US health and wellness influencers post. The data is US-only; for the UK platform picture, see our report on UK social media use from Pew's global survey.
What this means for UK marketers
Communities form around interests and places. Hobbies and the local area are the most common subjects in Pew's data, and the communities members visit most centre on hobbies, video games, and friends and family. For most UK businesses the job is to be useful inside the spaces customers already use, such as a Mumsnet thread, a local Facebook group or a trade subreddit, rather than to launch a branded community and hope people come.
Recommendations are a core reason people turn up. 42% of members say advice or recommendations is a major reason they visit, and those threads reach well beyond their members: Google surfaces forum content in Search, and AI answer tools can cite the same pages. That is why we treat community presence as part of getting your brand mentioned in ChatGPT and Perplexity and of our generative engine optimisation work.
Take part as yourself. Most members (63%) say their main community offers something they could only find online, and people leave when it stops being worth their time: 85% of leavers cite losing interest and 57% too much arguing. A brand account that only posts promotions, or picks fights, speeds up both. There is a legal line too. The CMA's short guide to unfair commercial practices says businesses must not use hidden advertising on social media, pretend to be a consumer leaving a review, or commission fake reviews. Posting in a forum as a supposedly independent happy customer falls on the wrong side of it.
Keep politics out. 54% of members say political discussion does not belong in their main community, and 44% of leavers cite too much politics as a reason. That is a clear steer for community managers and for brand accounts commenting in other people's spaces.
If you run your own forum, make it readable by search. Google's documentation on discussion forum markup says DiscussionForumPosting structured data helps Search identify online discussions for features such as Discussions and Forums. Moderation, clear rules and a reason to come back each week matter more than the markup, though, because losing interest is the most common reason people leave.
How they got the data
Pew surveyed 9,750 US adults from 26 May to 1 June 2026 as Wave 194 of its American Trends Panel, a nationally representative panel recruited through address-based sampling of US households. SSRS ran the fieldwork online (9,464 interviews) and by live telephone (286), in English and Spanish. Of 11,184 panellists sampled, 9,750 responded, a survey-level response rate of 87% (3% cumulative, allowing for non-response at recruitment and panel attrition). Results are weighted to US population benchmarks, and the margin of error for the full sample is plus or minus 1.4 percentage points. It is a probability sample of real respondents, the kind our report on synthetic survey respondents found AI cannot yet stand in for.
Questions about where communities form, why people visit and how they behave were asked only of the 3,635 respondents who said they belong to an online community, so figures such as the 42% and 74% are shares of members, not of all adults. Some topic questions, including the local-area item, went to half of the sample. The subjects of members' most-visited communities came from open-ended answers, coded with AI-assisted classification that researchers reviewed and corrected. Pew first studied online communities in 2001, but the 2026 report does not compare figures with that study. The findings describe US adults only.
Research institute
Read the full report
From Pew Research Center · published 14/10/2026