By using this site, you agree to our Privacy Policy and our Terms of Use. Close
Epoch on 11 March 2009
Edit Thread
Gnizmo said:

Epoch said:

Fair enough.  But I know statistics fairly well, and did quite well in Biostatistical Analysis at my University (Canadian).  This data set is biased, probably because of the fact that it came from an online survey.

If you can't follow my reasoning then don't worry about it.  Just don't read too much into the percentages that are given.  They are more a representation of total marketshare than of "this % of wii/360/ps3 owners game online".

This survey ascribes to the "If you can't dazzle them with brilliance, baffle them with bullshit" motto.

 Actually i understand statistics quite well. Math minor from the University of New Orleans. No heavy focus on statistics so I am at a loss to explain all of this, but I can follow quite well normally. The problem is we have fuck all for data here. The Arstechnica report was written by someone who doesn't understand statistics and so is untrustworthy as far as original data is concerned (their usage of the term "statistically significant" is a big red flag). The press release from NPD proper has even less information and is even more confusing when you try to decifer it. I'd give an arm and a leg to see what the raw data on this is. Not only to see where the PS2 and Xbox figur into it all (curiosity) but also see what the numbers actually say.

 And yes, it is unlikely that this generalizes perfectly to the general population of the US or the world. They key factor though is if it generalizes to the general gamer population. I can't find where it says this is from an online survey, but I find it relatively suspect. The original pol included 20,000 people and only a fraction of them actually played online.

I would also be interested to see the raw data.  As it stands, from their numbers, there are too many gamers for their sample size.  Probably because of the source (online survey) of their data set.  They state this in their methodology section at the bottom of their official release, which you posted.

The problem arises when the data set becomes too skewed, and is no longer usable as a reference.  Scientific journals and Non-Fiction publishers have a statistical threshold.  If the data doesn't hold up to this standard, it is considered useless.  This data would fall into that category.  It is far too skewed.

I think thats why the report is so vague, because you can't really get alot out of the data they have.  They didn't even tell us what % of the 20,000 own each console.  It is suspect at best.

However, I do agree that it may be slightly representative of gamers in general.  Enough at least to say that there are quite a few Wii gamers going online.  And far more 360 players than seems possible.