Posts Tagged 'lying with statistics'

Having your way with statistics

The British government released the official figures on the level of illegal file sharing in the UK. According to them, 7m people are engaged in illegal file sharing in UK alone.

How did they come up with this number you say ? PcPro traced the number and came up with a nice story.

It all started when the music industry trade body (BPI) commisioned a private survey to Jupiter Research. Jupiter research interviewed 1,176 house holds with internet access and, 136 of them admitted that they have been engaged to illegal file sharing. This 11.6% figure was then changed to 16.3% “to reflect the assumption that fewer people admit to file sharing than actually do it.” (Just to be on the safe side).

The official figure of english people with internet access is 33.9 million but Jupiter Research chose the number 40 Million, for no reason at all (or maybe to compensate for the households that use their neighbours’ wifi connection).

So they got the number 6.52 million (40*0.163) which was rounded up to 7 million propably because their software couldn’t handle floating points.

A few months later, the Strategic Advisory Board for Intellectual Property -a Government advisory body- commissioned the research from a team of academics at University College London, who apparently got the 7m figure from this interesting study.

nice huh ?

The normal result of this study, taking the real numbers into account would be 33.9*0.116 = 3.93 million. But this is still based on a 1000-answers telephone survey, and should be further rounded *down* , for example to compensate for all the house holds that do not have a telephone line. But, hey, what can you do, it is an official number now.