MikeB said: @ Avinash_Tyagi Actually size of a sample isn't the main judge of accuracy, how representative that sample is is a bigger issue Assume you count 99% of the market and have extrapolate the remaining 1%. You know that this 1% only relates to small shops, however you have data for 80% of similar small shops within the 99% data, so based on this you extrapolate 20% of uncounted sales for small shops and thus 1% of the market as a whole. The end estimation would be very reliable. This method can be expected to be more accurate than just directly extrapolating 1% from the whole market. The sample volume and diversity of sources is directly related to reliability above. Would the above example have been (an alledged) 15% of the market, not knowing what kind of stores they are, dedicated expensive game stores, budget pricing game stores, mass deversity retail stores, toystores, audio-visual stores, etc. |
No tracking service has 99% sample, NPD is 60% sample, so a good chunk of the market is outside their sample, in addition you're making a lot of assumptions about NPD and VGC and their ability or lack of ability to accurately extrapolate for the rest of the market. Unless you have something which you can show to ioi as incontrvertible proof, i'm afraid you'll just have to wait for him to recheck the numbers and decide
Predictions:Sales of Wii Fit will surpass the combined sales of the Grand Theft Auto franchiseLifetime sales of Wii will surpass the combined sales of the entire Playstation family of consoles by 12/31/2015 Wii hardware sales will surpass the total hardware sales of the PS2 by 12/31/2010 Wii will have 50% marketshare or more by the end of 2008 (I was wrong!! It was a little over 48% only)Wii will surpass 45 Million in lifetime sales by the end of 2008 (I was wrong!! Nintendo Financials showed it fell slightly short of 45 million shipped by end of 2008)Wii will surpass 80 Million in lifetime sales by the end of 2009 (I was wrong!! Wii didn't even get to 70 Million)