paulo
Erm no actually, the person I asked hasn't answered at all, which was why I asked him again. Somebody who doesn't work for cakewalk posting a link to some gibberish isn't an answer. Thanks for the arrogance.
I don't consider the science of statistics gibberish, so I didn't consider referencing his response as arrogance. And frankly, I didn't look at your post count so thought you might not be aware of subscribing to threads. Besides, people commonly provide answers to questions regardless of whom they're directed to if they feel they have an answer, and I thought Stickman's answer was valid.
What his post is referencing is that to obtain a statistically accurate sampling of opinions, there are formulas for how many people you need to ask to have a representative answer. There is a certain inherent margin of error due to limiting the sample; this is why you hear things like "Candidate X is trailing candidate Z by 2%, which is within the margin of error." However, it has been proven time and time again that you can obtain accurate analyses with a relatively compact sample of people. I think Neilsen's ratings involve only a few thousand households, yet advertisers consider them sufficiently accurate that they're willing to commit to spending millions of dollars on advertising based on the results.
The advantage of limiting the samples is that the more samples you take, the more time it takes to analyze those samples, especially with surveys that have comment fields (don't know if Cakewalk's did or not...I wasn't asked, either). So at least based on statistical sciences, asking more than the needed number of people is a waste of time - which is likely why Cakewalk didn't ask their entire user base.