Here’s Why Microsoft Are Wrong to Think 0.0005% is a Good Sample Size

Apparently the average Xbox gamer spends £64 a month on games, as reported first by MCV in some weird PR love-in with Microsoft, then by a number of different outlets. And that would all be fine, only the figure is bullshit.

It’s all based on a survey Microsoft say they conducted with 251 UK Xbox gamers (no information on how the data was gathered, how the individuals were selected, or any methods of data analysis are given, we just have to take their word for it). That’s a sample size, assuming total Xbox sales are around 50 million, of less than one thousandth of a percent, roughly 0.0005%.

Yet apprently this sample is enough to come up with statistics for all “360 owners” — like the “fact” that 13% of 360 owners spend more than £100 per month on gaming.

Or, to put it another way, 13% of US 360 owners pump 7.6 billion dollars into the US gaming economy. Which, considering last year the total sales of all gaming software (including all consoles and PC) was roughly 10 billion dollars, means those 13% are doing impressive — some would say fictional — work.

To repeat for effect: Microsoft are saying that 13% of Xbox gamers are buying so many Xbox games that they single handedly account for 75% of all gaming software sales.

[Editor's note: this maths was worked out on the back of a napkin using statistics gathered from vgchartz, use at your own risk]

I read it in MCV and actually made a verbal noise at what’s blatantly just pure PR shite — hence me sharing it with you, the dear reader, now. It’s a combination of two of the things I hate most: the press publishing adverts disguised as articles, and abuse of statistics.

Come back to me once you’ve got statistics of 251 thousand Xbox gamers and we can talk. A little bit. I’ll probably still give you a dirty look and talk about you unfavourably when you’ve gone though.