It only takes one counterexample to nullify a hypothesis. That is the epitome of science. The t-test isn't science in the strictest sense. It's a statistical m

It only takes one counterexample to nullify a hypothesis. That is the epitome of science. The t-test isn't science in the strictest sense. It's a statistical method which informs us of probabilities and distributions. If I find that eating fruit makes me fatter while a study concludes that it makes people skinnier, do I believe my own experience or the study?