hello, all

I have had a hard time trying to understand the concepts of fixed effects and random effects. I have googled a lot for help, but still confused. The following is a web talking about those two effects:

http://faculty.ucr.edu/~hanneman/linear_models/c4.html

It says " The key issue between fixed and random effects, statistically, is whether the effects of the levels of a factor are thought of as being a draw from a probability distribution of such effects. If so, the effect is random. If the levels of a factor are not a sample of possible levels, the effects are fixed".

I could not understand those sentences, what does it mean by saying "a probability distribution of such effects"?; can someone use some examples to explain those sentences to me?

Thank you very much for educating me.