Bootstrapping is a resampling analytic technique that simulates the sampling distribution of any statistic by treating the observed data as if they were the entire statistical population under study. The starting point is an empirical sample of N observations, from which we draw a new random sample of (also) N observations using sampling with replacement, and we repeat this operation a large number of times, while calculating the statistic (e.g., a mean) from each of these bootstrap samples. The resulting histogram of the bootstrap statistics will estimate of the expected distribution of the statistic.