By using this site, you agree to our Privacy Policy and our Terms of Use. Close
ioi said:
theprof00 said:

I agree. I want to know why everyone is saying NPD is a statistical tie when they don't know the margin of error.

There is no way to calculate a margin of error.

Game X sells 200k at Walmart/TRU and 250k everywhere else.

Game Y sells 100k at Walmart/TRU and 350k everywhere else.

Both games have actually sold 450k. According to data that NPD track, one has sold around 40% more than the other which unless they have some way of finding out actual sales at Walmart/TRU will reflect in their final data. For example, if they assumed their market coverage was 60% (which it probably is on average), the final sales estimates would be:

Game X = 250/0.6 = 417k
Game Y = 350/0.6 = 583k

You can't really put a margin of error on it when such a large and independent sector of what you are estimating is totally unknown. They could be 5% out, they could be 20% out in some cases.

It doesn't matter established, how accurate and how reputable you think NPD is - the simple fact is that they don't have 100% market coverage so all of these figures are estimates...

Yes I understand that, and I agree. However, it is always possible to find the margin of error, it's just that the margin of error may be wrong since it's only assumed based on a sample size and some other factors. Do they do random sampling?