Friday, September 12, 2014

How To... Calculate Student's t Statistic for Dependent (Paired) Samples in Excel 2010

In preparing a new set of notes and resources for the Business Data Analysis module (Statistics) on two upcoming undergraduate and one postgraduate course I am creating a new set of videos to support student learning. One of the statistical methods covered is Student's t-test. Learners can sometimes find it awkward as there are two types of t-Test with differing names. We have the t Statistic for Dependent (Paired) Samples, and the t Statistic for Independent (Unpaired) Samples.

In the video below I use scores from a pre and post test to test the difference between the means (Salkind, 2014). My Null and Alternate Hypotheses are:

        H0: µpost-test =  µpre-test
                 H1post-test >  pre-test
                 
The difference between the student's scores on the pre-test and on the post-test is the focus. Participants are being tested mor than once. There are two groups of scores. Therefore the appropriate test statistic is t-Test for dependent (paired) means (Salkind, 2014).



Salkind, N. (2014), Statistics for People Who (Think They) Hate Statistics (5th Edition). SAGE Publications,

No comments:

Post a Comment