Calculation--when to do it?

In my teaching days, I often told my students that "the only stupid question is the one you don't ask."

With that in mind, I ask the following question:

I have a bad chess habit that I want to break; namely, the habit of not consistently calculating variations.

Just when do you calculate, and under what circumstances?

(p. 50)

   In a complicated game this tree of analysis usually stays within a

depth of four or five moves--that is, four or five moves for each player,

(p. 51)

or eight to ten total moves. (We call these half moves: one move for

white and one for black equals one full move.) Unless there are special

circumstances--a particularly dangerous position or a key moment in a

game--I know from years of experience that's a safe, practical amount of 

calculation.

Kasparov, Garry, and Mig Greengard. 2007. How Life Imitates Chess: Making the Right Moves, From the Board to the Boardroom. New York, NY: Bloomsbury USA.

I start calculating heavily when I see an idea and I want to see if the math checks out.  

Thank you, but I found the answer to my question here:
http://www.swfloridachessclub.com/Chess_Calculation_Technique.pdf