Yes, that’s what I quoted too. I’m drawing a distinction between what is “statistically” significant and what is pragmatically significant. An example is the p

Yes, that’s what I quoted too. I’m drawing a distinction between what is “statistically” significant and what is pragmatically significant. An example is the processed foods study from a few years back that showed a statistically significant increase in cancer risk. But if you look harder at the data, it increases your risk from 1.0% to 1.18%. A relative difference of 18% sounds huge, but the absolute difference is very small. That’s not really much difference in the grand scheme of mortality risk. You can measure very small statistically significant differences with large data sets. It doesn’t necessarily mean it should drive people to change their behavior for every “significant” scientific finding.