When you perform a hypothesis test in statistics, a p-value helps you determine the significance of your results [D.J. Rumsey]. Hypothesis tests are used to test the validity of a claim that is made about a population. This claim that’s on trial, in essence, is called the null hypothesis.
The alternative hypothesis is the one you would believe if the null hypothesis is concluded to be untrue. The evidence in the trial is your data and the statistics that go along with it. All hypothesis tests ultimately use a p-value to weigh the strength of the evidence (what the data are telling you about the population). The p-value is a number between 0 and 1 and interpreted in the following way:
- A small p-value (typically ≤ 0.05) indicates strong evidence against the null hypothesis, so you reject the null hypothesis and state confidently that the alternative hypothesis is true.
- A large p-value (> 0.05) indicates weak evidence against the null hypothesis, so you fail to reject the null hypothesis. p-value bigger than the threshold does not mean that alternative hypothesis is wrong, it just shows that we cannot be sure about the conclusions we draw from the data.
- p-values very close to the cutoff (0.05) are considered to be marginal (could go either way).
For example, let we have 2 analogues similar to our chemical of interest and both of them show negative effect. We make the conclusion that all chemicals from this group (including our target) have some properties which explain the negative effect, so we infer that the target chemical is also negative. The null hypothesis states that our assumption is wrong and there is no relationship between the negative effect and the properties of chemicals in our group.
P-value is the probability for having 2 chemicals both with negative values due to pure chance. In this case it is 0.25 which is bigger than the typical threshold and because of this we cannot be sure if the picture we see is due to some relationship between the effect and the properties of chemicals in the group.
Now let’s consider another example, in which we have 8 analogues all of them having negative effect. In this case p-value is approx. 0.004, which is far below the threshold, so we are confident that the negative effect of chemicals is due to some property they possess, but not due to chance.