The extrapolation from 1176 is reasonable (I think the error is roughly +-2% for n=1000) if the sample is not biased..
But the real issues is how the question was asked. From the article:
...11.6% of which admitted to having used file-sharing software
So what question were they asked? If I use Instant Messenger to send a photo to a friend, have I used 'file-sharing software'? Is Skype file sharing software? I used to write surveys that tried to get at issues like this (sw piracy), and I believe it is nigh-on impossible to get good data by asking a direct question in this way. You either make it very pointed (e.g. ".. used file-sharing software to illegal share files"), and then few people say yes, or you make is less specific, and people accidentally say yes because they don't understand it. That's the real problem.
A study can be designed to account for embarrassing and/or incriminating questions.
Say the question is about "illegal file-sharing" and Yes means the subject has participated in this behavior in the past and No means the subject has not. Under private conditions, have the subject perform the following:
1. Flip a coin.
2. Have the subject answer Yes if they have participated in the illegal sharing of files.
3. If the subject has not done this activity, have them answer No only if they flipped 'tails' in step #1. Otherwise, if the coin came up 'heads', they answer Yes.
The true Yes proportion in the survey population of size n can then be determined by (Y_count - N_count) / n.
While that "works", what fraction of survey participants will actually do that?
I suspect that many/most folks who are reluctant to admit that they've done something won't admit it even if you tell them that other people "will" falsely admit to doing said thing so based on a coin flip. That's completely rational because I suspect that many people who haven't done said thing won't say that they have just because the coin tells them to.
I think people would assume that it was things like Limewire, Azureus, Napster, etc., ie bittorrent[-like] clients - interpreting broadly would encompass things like email clients, browsers, ftp clients, etc..
I've used several torrent applications .. for downloading linux distros. So hopefully the question was not just about file-sharing but actually about sharing music files that the owner didn't have rights to use.
Like you say, this is problematic, it requires the respondent to make a legal decision on an issue which may not be determined yet in caselaw and so may be an open question : can I dl music files that I've bought already on CD for example. The /a priori/ answer is no; but it's not at all cut-and-dried.
Title is silly. The wonderful thing about statistics is that if you have a truly random sample of 1176 people you can extrapolate from 136 people to seven million (plus or minus a certain error bar which I'm too lazy to figure out right now).
The other points are somewhat valid, but by the usual standards of political misuse of statistics this is pretty small beer.
I agree a small truly random sample can often be reliably extrapolated - but in this case the 136>7m figure was based on research paid for by the BMI (who lobbied heavily against file-sharing) in a report still not fully published, based on two numbers of which one differs from the official ONS number without explanation. The fact that the lobbyist organisation paid for the original report was obscured by misattributing the source, which also seems dubious to me.
I don't doubt that political misuse of statistics is common, but it should always be called out.
Yeah, I was immediately screaming the same thing. Arguing that "only" a tiny subset of a population is involved with a survey as a basis for rejecting its results is just plain ignorance.
The other stuff in the article does seem dodgy though, like arbitrarily tacking on 50% to reflect an assumed-but-unmeasured bias in the input sample. Ridiculous.
Still, the title is innumerate, sensationalist junk, and the poster should be ashamed.
The title was actually copied directly from the article. I completely agree that the title appears to be linkbait, but the blame should be placed on the editor and/or author of the post, not the poster.
The editor/author created the title to be sensationlist on purpose. HN is, I hope, not here to offer sensational headlines merely to grab extra viewers but as a way to share knowledge, information and wisdom. We should have higher standards for titles here IMO.
What I'd like to see is a subtitle (or tagging) system with a descriptive subtitle that can't be applied by the poster only by someone else. Poor subtitles would be downmodded too.
But the real issues is how the question was asked. From the article:
...11.6% of which admitted to having used file-sharing software
So what question were they asked? If I use Instant Messenger to send a photo to a friend, have I used 'file-sharing software'? Is Skype file sharing software? I used to write surveys that tried to get at issues like this (sw piracy), and I believe it is nigh-on impossible to get good data by asking a direct question in this way. You either make it very pointed (e.g. ".. used file-sharing software to illegal share files"), and then few people say yes, or you make is less specific, and people accidentally say yes because they don't understand it. That's the real problem.