A Penn State-led study says a majority of U.S. adults report weak understanding of basic statistics, even as many say they would use statistical evidence more often if they understood it better. According to Penn State’s research news release, the study was published in PLOS One and analyzed nationally representative survey data from 1,000 people. The headline finding: 62% of respondents said they had no or limited statistical knowledge. Only 11% said they regularly use statistics in daily life. The reported breakdown is stark. Penn State says one-quarter of respondents said they had no understanding of statistics, while 37% reported limited familiarity. Slightly more than one-quarter said they had learned some statistics in school. The survey also asked whether people understood p-values, a common statistical measure used in research. Penn State’s Mark Ramos, an assistant research professor of health policy and administration, described p-values as a tool for interpreting study results; in simplified terms, the smaller the p-value, the stronger the statistical evidence, according to Ramos. The questions came from the 2025 Joint Statistical Meetings in Nashville, Tennessee, where a market-research and survey company solicited survey questions from attendees for a nationwide poll. Ramos and Samuel Anyaso-Samuel, a postdoctoral fellow at the National Institutes of Health’s National Cancer Institute, submitted two questions that were included: how much respondents understood about statistics and p-values, and how often they would base decisions on reported statistics if they understood the subject better. Penn State says the findings do not show indifference to statistical reasoning. Nine out of 10 participants said they would rely on statistics more for decision-making if they had a better grasp of the topic. For technology and AI readers, the study is a reminder that the public’s ability to interpret evidence may lag behind the statistical framing now embedded in product claims, research announcements, policy debates and health guidance. The source material does not test AI-specific literacy, but it does point to a broader constraint: statistical claims are only useful to decision-makers if their audience can interpret uncertainty, evidence strength and limits. Who benefits: Educators, research communicators and organizations that translate technical evidence into plain language have a clear opening. Products or training programs focused on practical statistical literacy may also find demand, given that most respondents said they would use statistics more if they understood them better. Who's exposed: Institutions that rely on statistical claims to persuade non-specialist audiences are exposed to misunderstanding. The source material is too thin to quantify business impact, but it suggests a communications gap around evidence and uncertainty.