theprof00 said:
r505Matt said:
De85 said:
Galaki said:
Yeah, a survey of 1000 is very... researchy.
|
Um, yes actually, it can be. During elections most polls are taken with similar sample sizes. What matters is getting a sample that is statistically representative of the population.
@topic: That's more than I would have guessed, so I'd like to see more about the survey methodology. If true though the PS3 could be shaping up to have a monster fall.
|
Exactly, they didn't really describe their sample, and I have a feeling that it may not be a good representation. I really wish they explained the research better, as it stands, it just looks like they could have pulled the number out of a hat.
I would expect a more scientific approach just like on Sciencedaily.com where they pretty much show a lot more information for studies. Here's a decent example of how to report findings of a study:
http://www.sciencedaily.com/releases/2010/04/100405174942.htm
They don't explain every detail in the study, but thoroughly explain the study's set-up and conclusions/trends. The "39%" study just shows conclusions, not very professional at all.
|
We don't publish how we gather data either. I guess you should stop coming to this site for numbers then. FYI it's called protecting the business property.
|
What does sales tracking have to do with publishing a study? The difference is for sales tracking, it's the norm for data gathering methods to be withheld. For studies though, it's the norm to explain/show your work so to speak. You don't need to reveal everything (though the actual published piece will usually detail everything) but the less you show upfront, the less professional it seems.
