Coaches are not interested in making inferences to a population, statistical significance, nor mean values (P<0.01). What they are interested are individual athletes, their reactions to training and what are practically significant and meaningful changes.
Before proceeding reading this how-to blog post, I urge readers to read the following two presentations from Will Hopkins
- How to Interpret Changes in an Athletic Performance Test
- Client Assessment and Other New Uses of Reliability
If you have read those must read presentations then we can continue. If you haven’t please do.
To evaluate change (is it practically meaningful) we are going to use SWC (Smallest Worthwhile Change) and TE (Typical Error). If you don’t know what they are and how they are used then you haven’t read the mentioned Will Hopkin’s presentations.
Knowing SWC and TE for a test and/or individual (in this case test) we can use NOVEL way of visualizing test change score for the reports so that coaches can understand and see individual players reaction and whether it practically meaningful
Calcuating Typical Error for the test/estimate
I beleive in in-house research, at least doing something simple as this. Knowing typical error (TE) for the test/estimate is of great importance to judge the real change.
What does this mean in real world? It means that if we test a player in vertical jump one day when he was fully rested, and then we re-test him couple of days after in same conditionins (with the assumption that there are no learning effects nor improvement happening; that’s why it is important to do familiarization with the test first) the scores might differ a little. This might be due measurement error and/or biological variability.
Then if we test that same player couple of months later to see the effects of training, how certain are we that the change is real? Here comes TE to help out as you will see later in the visualization and calculus of harmful/trivial/beneficial changes.