Monday, November 16, 2015

How To... Test for Equal and Unequal Variance (F Test) by Hand

I am finding my new videos showing statistics calculations by hand useful for my students - reaction so far has been positive. When comparing the means of two samples, we use a t-test to determine if there is a significant difference between the means. Students often find the decision as to which t-test formula to use a confusing one. First, they have to decide if the two samples are paired (dependent) or unpaired (independent). If  the samples are unpaired, you then have to determine if the variances of the two samples are equal or unequal. Quite often researchers assume that the variances are equal, but we need to conduct what is called an F test first to be sure. The F statistic is easy to calculate, it is simply the larger sample variance divided by the smaller sample variance. In the video below I show how this is done by hand (includes calculations of variance):

No comments:

Post a comment