Confusing Statistical Concepts

Mediators, Moderators, and Suppressors: What IS the difference?

March 10th, 2010 by

One of the biggest questions I get is about the difference between mediators, moderators, and how they both differ from control variables.Stage 2

I recently found a fabulous free video tutorial on the difference between mediators, moderators, and suppressor variables, by Jeremy Taylor at Stats Make Me Cry.   The witty example is about the different types of variables–talent, practice, etc.–that explain the relationship between having a guitar and making lots of $$.

 


Parameters and Variables

December 17th, 2008 by

I once had a client from engineering.  This is pretty rare, as I usually work with social scientists and biologists.  And despite the fact that I was an engineering major for my first two semesters in college, I generally don’t understand a thing engineers talk about.

But I digress.  In this consultation, we had gotten about 20 minutes into me not understanding a word he was talking about when I realized he was using “Parameters” when he meant “Variables.”  As in, “I measured four flexibility parameters on the doohickey.”

In statistics, Variables are things you measure that vary from observation to observation.  Height, weight, flexibility, bending strength, % ground cover–these are all Variables if they vary from one observation to another.  (They are constants if they don’t).

Parameters are things you measure about the variables.  Their means, their variances, the size of their effect on another variable.  And parameters specifically refer to the measurements made about the entire population.

I suppose it makes sense that engineers consider variables to be parameters, since to them, parameters are things you measure about doohickeys.  In statistics, variables are the doohickeys getting measured.

So it makes it hard to talk with engineers because I have to translate as they speak.  But I’ve come to accept that they speak a different language although with the same words.

But lately, I’ve seen other people (like ecologists) calling their variables Parameters.  And in the same sentence as using the terms like p-value and adjusted R-squared, so I know they knew statistics well.

What’s going on?