With a massive sample size, even a tiny, useless difference will be "statistically significant."
Interpret high p-values as "inconclusive" rather than "proof of zero effect." 4. Contextualize with Confidence Intervals Wise Use of Null Hypothesis Tests: A Practition...
It measures the probability of seeing your data if the world is actually boring ( H0cap H sub 0 With a massive sample size, even a tiny,
If you run twenty different tests on the same data, one will likely be significant just by chance. With a massive sample size
A p-value tells you if an effect is , not if it is large .
Did I check my data for outliers and normality before testing? Is my sample size justified by a power calculation? Am I reporting the effect size alongside the p-value?