The null hypothesis concept began to be used in the first agricultural and medical applications of the statistics, and is responsible for making statements about a certain parameter of the population. What does this mean? That the null hypothesis is a statement that is never rejected ... unless the data of the sample used show otherwise. Therefore, if the results of our sample do not support the null hypothesis, we are no longer talking about the null hypothesis, but about the alternative hypothesis. We are going to learn more about these concepts:
What is the difference between the null hypothesis and the alternative hypothesis?
The null hypothesis is an untested general idea that the researcher tries to reject or nullify; while the alternative hypothesis is the cause or the real and verified consequence of a phenomenon (be it medical, economic, financial or demographic, among others). As we see, one term cannot be understood without the other.
How to formulate a null and alternative hypothesis?
To better understand the correlation between null and alternative hypotheses, nothing better than to resort to an illustrative example.
Suppose that a null hypothesis indicates that there is no cause-and-consequence relationship between a new treatment being tested and a reduction in symptoms of a disease. What does this mean? That, according to the null hypothesis, the new drug will not offer any improvement over the drug that was being used so far, so any improvement would be the result of chance.
To test and, if applicable, reject this null hypothesis, we would have to develop a scientific study on this new treatment. If we discover that there is indeed a cause-and-consequence relationship between the new medicine and the improvement of the patient suffering from the disease, we would be showing that the null hypothesis is false. Therefore, if this were to occur, we could adopt an alternative hypothesis: the new drug is better than the old drug because significant advances have been demonstrated in the patient.