In statistics, bootstrapping is a procedure done by resampling a single data set to create a multitude of simulated samples. Those samples can be used to calculate confidence intervals and to test a hypothesis. This approach allows you to generate a more accurate sample from a smaller data set than the traditional method.

Why bootstrapping?

Most of the time, when you’re conducting research about a population, it’s not possible to obtain the whole population. Additionally, it is not realistic or practical to collect data from the entire population. So, in using bootstrap, we can use a subset of the population and use it to gain some insight to learn more about the population. However, you should note that many factors can affect how well a sample reflects the population, and therefore, the validity and reliability of the conclusions.

“The advantages of bootstrapping are that it is a straightforward way to derive the estimates of standard errors and confidence intervals, and it is convenient since it avoids the cost of repeating the experiment to get other groups of sampled data. Although it is impossible to know the true confidence interval for most problems, bootstrapping is asymptotically consistent and more accurate than using the standard intervals obtained using sample variance and the assumption of normality.” – Graysen Cline^[1]

The bootstrapping method takes the original sample data and then resamples it multiple times to create many bootstrap samples. In bootstrapping, we do not need to worry about the assumptions and conditions of the sampling distribution. Additionally, we are no longer constrained to test only the means and proportions of the data set. Bootstrapping allows us to find accurate estimates of statistics without access to the whole population.