The root mean square of a data set is exactly what it sounds like: it is the result of the square root of the mean of each item in a data set squared. There are specific applications for this, such as when trying to find the mean of an alternating data set, where the raw mean would always be 0. Because each value has an equivalent negative counterpart in such sets, they would each cancel out and render no usable information from the mean. By squaring the values each one becomes positive, so the mean can be found. This is compensated for by taking the root of the answer once the square mean has been calculated.