an experiment should be replicated for one: it my give more reason to people who did not do it them selves two: during the experiment one variable could have been wrong like a gust of wind or a draft that's why you should conduct the experiment 3 to 4 times then get the medium or the average of your results giving you a more accurate answer.