The p-value is the probability that you would observe an effect of this size if the hypothesis were not correct. So you want them to be small. In most fields, you want to get at least p < 0.05 or 95% certainty, which is about two standard deviations from the mean in a normal distribution. That means that your result is far enough away from what you would get by random chance. https://en.wikipedia.org/wiki/Standard_deviation#mediaviewer...