The p-value of a test is not the probability that the null hypothesis is true or false
Abstract. One of the most common errors made by physicians in all developed countries is to say that the p-value of a test is the probability that the null hypothesis considered in the test is true or false. Eighty percent of those polled in many surveys make this mistake. The p-value of a test is the probability of obtaining a result like the one obtained in the investigation if the null hypothesis is true. The probability of a pregnancy involving three embryos is very small, at 0.00008 (8 in 100,000). In pregnancies with triplets, the probability of a caesarean section being performed is very high, at 0.98 (98%). These are two very different values, and two very different concepts. Saying that 0.98 is the probability of a pregnancy involving triplets would be a serious mistake. We make the same mistake when we say that the p-value of the test is the probability that the null hypothesis is true, or the probability that it is false.
Key words. A priori probability. Medical research. Null hypothesis. p-value. Probability. Statistical inference.
|