| Burek said: So, to sum up your gripe: You believe that ToysRUs sold a incredibly large number of Bayonetta 2, and thus NPD data is 200K less than VGC, which is more accurate because they track less and extrapolate more. |
He wasn't salty about Bayonetta's sales. It's just a general question that came up in the Bayonetta thread and he decided to make a thread about it.
| Burek said: So, to sum up your gripe: You believe that ToysRUs sold a incredibly large number of Bayonetta 2, and thus NPD data is 200K less than VGC, which is more accurate because they track less and extrapolate more. |
i dont care about the bayo2 sales, it was just that thread that sparked my interrest about the methods.
and yes, tracking less and but more represantive could lead to better numbers.
I get OP's point about sample size, to a degree.
For example, you have 2 samples; one, non-representative, comprising 50% of the market. and the other, representative, but only 1%.
The first doubled, would be less accurate, than the second x100. As the first, despite being bigger, isn't an accurate representation of the whole market, whilst the second is.
However we've no reason to believe that not having Toys R Us makes NPD non-representative, they may well have other toy retailers that provide insight into that market segment. And we also have no reason to believe that VGChartz sample is more representative. It could well be less.
we've been here before with you haven't we
everyone takes NPD so it should be good enough for you too
I smell like the dark forest of Dwemereth in the morning mist 
fanboy til i'm bankrupt
NPD gets numbers directly from ~90% of the retail market in the US. They estimate the remaining ~10%
Won bet with t3mporary_126 - I correctly predicted that the Wii U's LTD at the end of 2014 would be closer to 9 million than 10 million. http://gamrconnect.vgchartz.com/post.php?id=6673287
| ToxicJosh said: I get OP's point about sample size, to a degree. For example, you have 2 samples; one, non-representative, comprising 50% of the market. and the other, representative, but only 1%. The first doubled, would be less accurate, than the second x100. As the first, despite being bigger, isn't an accurate representation of the whole market, whilst the second is. However we've no reason to believe that not having Toys R Us makes NPD non-representative, they may well have other toy retailers that provide insight into that market segment. And we also have no reason to believe that VGChartz sample is more representative. It could well be less. |
well they use other math i hope^^
the size of a sample doesnt matter so much, we predict election with 1000 people in germany, and they get a margin of error of 1%.
| Welfare said: NPD gets numbers directly from ~90% of the retail market in the US. They estimate the remaining ~10% |
do you have a source for the 90% so i can copy it in the op?
| Ka-pi96 said: You what? We know a lot more about VGC than nothing. It uses the exact same kind of tracking as NPD just on a much smaller scale. Bigger sample = more accurate data, that can't possibly be argued against. |
well doh, there isnt another way of getting the data... the question is how he gets the sample and how big is it.
generic-user-1 said:
a big sample size needs a lot more math to get to beeing close to represantive. you need a random sample, thats realy hard to get if you track a big none random part. |
Ah, no. Sure, you can do your modeling so wrong, that your better sample turns worse data, but that's a problem of the model not of the sample. Also it is possible that a smaller sample better covers more different groups of (in this case) consumers. But we have no sign to believe the VGC is better. If you don't know more, it is best to assume the bigger sample is also the better sample. You can argue the NPD sample has systematical error, because they don't get some user groups. But for that you can model (having other user groups in your sample still is better). But it's bordering on impossible that VGC has the more diverse sample.
| ToxicJosh said: I get OP's point about sample size, to a degree. For example, you have 2 samples; one, non-representative, comprising 50% of the market. and the other, representative, but only 1%. The first doubled, would be less accurate, than the second x100. As the first, despite being bigger, isn't an accurate representation of the whole market, whilst the second is. However we've no reason to believe that not having Toys R Us makes NPD non-representative, they may well have other toy retailers that provide insight into that market segment. And we also have no reason to believe that VGChartz sample is more representative. It could well be less. |
This explains it better than I could.
About Us |
Terms of Use |
Privacy Policy |
Advertise |
Staff |
Contact
Display As Desktop
Display As Mobile
© 2006-2024 VGChartz Ltd. All rights reserved.
