# Information gain or regression for measuring to what extent a continuous quantity detemines a discrete one?

I have a small population S and two quantities P and Q associated to its members. P is integer valued (say, age taking integer values between 1 and 20) and Q real valued (say, height in meters). I can measure them for the whole population S. I want to see to what extent Q determines P. I think information gain is a good measure for this, better than regression, but am not sure.

• Welcome to our site. This post may be a little too abstract to be answerable. Could you say more about what $Q$ and $P$ represent? Could $P$ be ordinal? How many possible values might it have? Is it accurately measured or known? How to you sample the population? – whuber Jan 11 at 21:42
• Hi, I did so and clarified. – Reza Jan 15 at 20:33