Experimental errors are normally classified into the random error and the systematic error.

The method is subject to greater experimental errors than the freezing-point-depression method .

For fractional factorial designs several different approaches to estimate this variance have been suggested.