Define variation to a beginner

[mkir]

Define "variation" to a beginner

How do you define the word "variation" "variability" to a complete beginner, like myself.
For Example, I understand 'R' explains how strong the linear relationship between the change in X and Y. But sometimes, let's Say R=.91, people will say the linear trend model explains more than 91% of the variation in sales.


Another Case is the Mean squared error (MSE) is a measure of variability in the forecast errors.

Hope someone can use 'SIMPLE' words to explain it to me.


Thanks a lot!

 

[EnumaElish]

Let's say sales you are looking to explain "sales" with "effort." Suppose effort changes over time, but sales are constant. Then obviously effort cannot "explain" sales, because there is no correlation between the two (sales stay constant as the level of effort changes). There is no correlation unless there is some variation in both sales and effort.