Statistics Made Easy 3: Relationships or Differences?

For inferential statistics classes, what I always emphasis to my students from the very first classes is:  "Know what you are trying to find: Relationships or Differences?"  This is especially for continuous data, rather than discrete or categorical data.  Please do a bit of research and read up these types of data (and variables). I will probably write a post about this some other day.

So the big question: "Relationships" or "Differences"?
This question looks at the link between your independent variable (IV) and dependent variable (DV), and also helps you decide what type of tests you are going to use (once you decide on whether you are going to parametric or nonparametric statistics for the different types of variables (to be discussed in another post).

To better understand whether you are going to do a statistical test to evaluate the relationship or difference, we need to first look deeper into the IV and the DV by asking some simple questions:
1.  Are there different groups in the IV and the scores in the DV might differ as a reason of that?  If "Yes", you are looking for a difference (between the groups).  If there are no groups, there is a high chance that you are looking for a relationship between the IV and DV.
2.  Does the scores in the DV fluctuate or vary with the IV, i.e. you are expecting that whenever the IV scores increase, the DV scores should increase/decrease in a fixed pattern?  If that is the case, it would certainly be an analysis towards the relationship between the IV and the DV.   
Hope this clears up a big question mark for you towards understanding statistical analyses!!  If you wish to read the rest of the statistics-related posts, here they are!!!

No comments :

Post a Comment