Quote:

Originally Posted by

Well, obviously it depends on the data. But do you honestly think the median isn't going to be lower than the mean in the college-student computer market? I think that's a fairly safe assumption.

I think teckstuds only point is that using the median - rather than the mean - selling price is a superior data point to look at. I think that's absolutely true.

**Flounder**Well, obviously it depends on the data. But do you honestly think the median isn't going to be lower than the mean in the college-student computer market? I think that's a fairly safe assumption.

I think teckstuds only point is that using the median - rather than the mean - selling price is a superior data point to look at. I think that's absolutely true.

I like to work from the data we have. And we have no data, only the average.

Now, we all know that almost everything follows a gaussian distribution, or "normal" curve. That means that the majority of anything will be in the mid portion of the curve. The drop on both sides is about the same, diminishing to zero at some point.

We have no reason to believe this works in any other way. Until we do, we can't make the assumption that it does.

If, one wants to believe that it does, without any data to prove so, then one must figure out a way to get data points on a curve that makes sense, and is likely to be true.

We would need some numbers posted to give us that conclusion, even if those numbers were made up for the sake of the argument.

We could have a very interesting discussion as to what numbers could fit the curve.

Otherwise, you know that the statistically most likely curve is the gaussian distribution, and that curve sits in the middle of the universe of possible curves determined by a gaussian distribution.