draw the flat (i.e. no stereochemistry defined) structure, just so you know what it looks like.
Start with one of the stereocenters. Arbitrarily draw it in the 'up' position. Go to the next stereocenter - there are only two choices for that one. Up or down. draw both possibilities (keeping the first one 'up'). Then switch the first stereocenter to 'down'. go to the next stereocenter - again, only two possibilities. Up and down. draw them both.
You'll have drawn all 4 maximum possibilities (remember the 2n rule).
The last thing to do is see if any of them are the same - that is if you've drawn two identical compounds (that can be drawn the same through rotations and flipping). If two of the structures are the same, then because they're the same there is one fewer total stereoisomer.
Based on that, what do you think for 1,3-cyclopentanediol?