America’s Reading Habits

There was a recent tweet I saw saying "I’m willing to bet my entire net worth on neither the median nor the average American reading 11 books a year." There's a lot of people pushing back on it, but all citing different reasons why the data could be wrong. Rather than do that, we can just dig into the underlying data! ¹ A community note has since been added addressing this.

The quote tweet in question is pushing back on a tweet from The Economist. The article links to an earlier December 2023 article about how many books you'll read before you die. The article is paywalled, but I have a subscription ² If you don't, here's an easy hole in the wall from our friends over at archive.is. .

The question was "Have you read or listened to any books in 2023?" and only 54% said yes. Of the people who said yes, the average was 11 books. This already partially explains why the number is so high, but let's dig into the actual data!

The Data

Despite sponsoring the survey, The Economist doesn't link to it. YouGov does ³ And their writeup has fun charts! though, and they also have links to the toplines and crosstabs. As a quick aside, it's always worth checking the sampling, which is on the last page of the toplines. It flags that the data comes from a web-survey of U.S. citizens, age 18 and over from their opt-in panel, and was weighted according to gender, age, race, education, 2020 election turnout and Presidential vote, baseline party identification, and current voter registration status. All numbers are also only reported as rounded percentages. This is pretty typical (the books question were only a few among the reported 97 questions, much of which was politically focused) but it's worth noting that it's a web-survey and it's sampled from an opt-in panel, which may bias results. Now back to the data!

Checking the toplines doc shows that a followup question was "About how many books have you read or listened to so far in 2023?" with breakdowns from "0" to "50 or more". We can multiply out by 1500 survey respondents to answer questions about the mean/median. Ignoring the 1% of responders who said "not sure", and treating "50 or more" as exactly fifty ⁴ 4% of users chose this, and how you handle this has a fair amount of impact for the average. If you remove this it goes down to an average of 4 books, but if you treat it as 60 it goes up to 6.3. Unfortunately the raw data isn't available so we just have to choose something. we get an average of 5.9 books and a median of 1. If you exclude 0 (what The Economist did in the article that kicked this off) you do get an average of 11.1 books and a median of 5. ⁵ Google Sheet with my calculations because you should always link the raw data (*cough* The Economist *cough*)

Breakdowns

I saw some discussion about the "read or listened" distinction. There was a followup of "Were any of the books that you read or listened to in 2023...? Select all that apply" with 42% saying physical, 21% saying digital, and 19% saying audiobooks.

Again the lack of access raw data makes breaking this down tricky, but YouGov did share some aggregated data that digital book readers tended to read more ⁶ This is certainly true for me. While I love reading physical books, the volume is digital because I always have my phone on me, but don't always have a book (also Books is one of my few non-locked apps). (with audiobooks and print roughly evenly split).

Some of the breakdowns were a little surprising. The percentage of men and women saying they read at least one book was the same, 54%, which goes counter to my prior thought that women read more than men (this may be explained by volume — 50 or more was only 2% for males but 5% for females). As another one, liberals were more likely to consume digital books or audiobooks than their conservative counterparts.

Corroborating the Data

But this is just one survey. Does 54% of Americans reading at least one book match what other surveys have shown?

• A 2024 YouGov poll of 3385 people sourced in a similar way found that 63% of people read. It's a little surprising that this changed so much from one year to the next—as the Washington Post article ⁷ Archive link. that motivated this survey also pointed out:

  Other sources don’t point to a seismic shift in reading habits, so this may be a refined estimate rather than a trend.

• The National Endowments for the Arts did a 2022 survey ⁸ The data comes from the NEA survey questionnaire which, since 1982, has been administered roughly every five years by the U.S. Census Bureau—the 2022 data comes from 40,718 respondents. that showed that 54.6% of US adults read any book in 2012, dropping to 52.7% in 2017 and 48.5% in 2022 (51.9% if you include audiobooks, as the previous surveys did).

• An open Google Poll of 1,621 from TestPrepInsight in 2025 showed that only 51.5% of respondents had read a book in the last 12 months, though this may undercount audiobooks/digital books and result in the difference.

• The Arts Basic Survey from 2020 reported that only 39.9% of people reported "[Reading] novels, short stories, poetry, or plays" ⁹ These numbers are likely lower by implicitly/explicitly omitting other forms of books, including academic, nonfiction, and history which especially skews male. , down from 43.1% in 2015. ¹⁰ Recorded from Feb '19 to Feb '20, all pre-COVID. See the raw data here.

• Pew Research has been conducting a survey of ~1,500 people since 2011, and the 2021 numbers have 77% of surveyed Americans reading a book in the last year.

Conclusion

These surveys span a pretty large range, which likely stems from differences in how they handle digital/audiobooks, as well as sampling and self-reporting bias. Somewhere in the range of 50%-75% of Americans having read a book sounds right, but as to @Xilo_K's original tweet, their net worth isn't at risk. I look forward to seeing the 2025 data from YouGov, but from all the data we have, 6 books feels closer to right.