Researchers find widespread statistical misunderstandings persist even among experts
A new systematic review published in PLOS One shows that experienced researchers and students frequently misinterpret statistical concepts like p-values despite extensive training.
New study analyzes decades of data
Researchers from Uppsala University conducted a systematic review analyzing 55 studies involving over 15,000 participants. The work covers publications spanning from 1963 to 2024 and appears in the journal PLOS One. The group comprised statisticians, teachers, health care workers, researchers and students who responded to the surveys. These individuals judged whether various statements about statistical results were correct or incorrect.
Common errors involve p-values and significance
Per Lytsy from the Department of Public Health and Caring Sciences notes that understanding of inference is inadequate. Ronnie Pingel, a senior lecturer at the Department of Statistics, explains that people often mistake p-values for proof of truth. A common error is viewing statistical significance as evidence that a hypothesis is true or an effect is real. Such misinterpretations can lead researchers to attribute greater certainty to findings than the data warrants.
Practical meaning often ignored by experts
The review finds that a medicine may show statistically significant improvement without practical significance for patients. Statistical concepts are difficult because they do not always match how people intuitively reason about probability. Different theoretical foundations and combined methods make it hard to determine which conclusions are justified. The researchers hope this broader picture will improve education and future scientific decisions.
Reported by one outlet
Only one outlet has published this. Nothing here has been checked against a second report, so read it as that outlet's account and follow the link below for the original.
Reported by
1 independent outlet. Headline as published. Links open the original report.