P-Value

A p-value is a number between 0 and 1 that tells how surprising a study's result would be if there were really no effect. A small p-value, such as 0.01, means the result would rarely occur by chance alone; by convention, a p-value below 0.05 is called statistically significant. It is not the probability that the finding is true.

Origin

P-values grew out of significance testing developed by statisticians including Karl Pearson and Ronald Fisher in the early 20th century. In 2016 the American Statistical Association issued a formal statement warning against common misuses of p-values.

Significance in Asbestos Context

P-values appear in nearly every study cited in asbestos cases, and they are easily misread. A p-value of 0.08 for an increased lung cancer risk does not mean the risk is absent; it means the study could not rule out chance at the usual 5 percent level, often because it was small. A p-value of 0.001 does not mean the effect is large or that the study was free of bias or confounding. The American Statistical Association has cautioned that scientific conclusions should not rest only on whether a p-value passes a threshold. In court, defense experts may describe non-significant p-values as showing no association, and plaintiffs' experts may answer with confidence intervals, the size of the effect, and the consistency of many studies. Judges deciding whether expert testimony is reliable may hear arguments over these numbers. For families reading reports, a p-value answers a narrow question about chance and needs to be read with the rest of the evidence.

Example Sentence

The increase in mesothelioma among the plant's workers had a p-value of 0.03, below the conventional 0.05 cutoff.

Related Terms

Statistical Significance, confidence interval, Relative Risk, Odds Ratio, Epidemiology, Confounding

Notes

• A p-value says nothing about how big a risk is; look at the risk ratio and confidence interval too.
• A p-value just above 0.05 is not evidence of no effect.