The central limit theorem says that this sampling distribution is approximately normal—commonly known as a bell curve. This approximation improves as we increase the size of the simple random samples that are used to produce the sampling distribution.