How To Lie With Statistics: France Pretends HADOPI Law Is Working
from the doing-the-math dept
The French government was one of the first to push for a three strikes policy to kick people off the internet based on accusations (not convictions) of copyright infringement. So, I guess it should be no surprise that they feel like lying with statistics to make the program sound better. The government and various agencies are running around touting the claim that, according to their survey the HADOPI law has convinced more than 50% of users to stop file sharing. Problem is, that's not what the data really says. The real data shows that of people surveyed only 7% said either they "or someone close" had received a warning letter. Now, of those 7%, 50% claimed that they would stop infringing.Now, if you're playing along with the home game, you should have quickly realized that the actual percentage of people surveyed is more like 3.5% -- and I could argue that it's even lower for a few key reasons:
- The key question asked wasn't whether the individual would stop file sharing, but whether or not they or someone close to them had. Suddenly you have a big statistical problem, because -- to take an extreme example -- let's say that everyone in a town knows the one big file sharer who shares content online, but no one else in the town does. And, that guy knows and makes it clear that if he gets an injunction, he'll stop. Now, since everyone knows this guy, the reports from that town would be that 100% of people receive letters and 100% of those recipients would stop using P2P, even if that wasn't true at all. Including the "or someone close to you" makes the effective data pretty close to useless, because there's no way to separate out the overlap.
- The whole thing is based on a survey, which is notoriously unreliable in getting accurate data. People quite frequently answer what they think others want them to say, rather than what they're really thinking. And, when asking them if they'll stop engaging in illegal activity, many are simply going to say yes, even if they have no intention to follow through.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: france, hadopi, statistics
Reader Comments
Subscribe: RSS
View by: Time | Thread
43% of all statistics are meaningless
What do we expect from a law pronounced Ha Dopey.
[ link to this | view in chronology ]
Re: 43% of all statistics are meaningless
(Shamelessly swiped from WikiQuote: http://en.wikiquote.org/wiki/Statistics)
[ link to this | view in chronology ]
Re: 43% of all statistics are meaningless
everyone knows that 72.3% of all statistics are made up on the spot by people that dont have any real numbers or sources to back up their claim.
[ link to this | view in chronology ]
Re: Re: 43% of all statistics are meaningless
[ link to this | view in chronology ]
But it *is* working!
>:)
@Seriousness:
Why not just let this law stew in its own stupidity? If the article says the survey data is wrong on logic, does it mean anyone in France will actually believe it?
In two years, there will be news another law is needed because HADOPI isn't working.
Hey, this sounds familiar. :|
[ link to this | view in chronology ]
Wait a minute...
Link
They'll be out of commission for a while.
[ link to this | view in chronology ]
[ link to this | view in chronology ]
And the real question should be...
[ link to this | view in chronology ]
Yes, but none of it French.
[ link to this | view in chronology ]
Ah surveys.... Ask the TSA
So why the need for scanners or pat downs?
[ link to this | view in chronology ]
- a quote popularized in the U.S. by Mark Twain
-- from another source.
[ link to this | view in chronology ]
7% had received, or knew someone who had received, letter.
50% who talk about letter say or imply they or the someone they knew would stop.
Yes, 3.5% of 1500, but 100% of the 7%, the 7% being the "target" group.
Of course, a survey is only as good as the sample selected, and depends upon the accuracy of answers.
If sample was random, and if answers relatively accurate, 50% would be correct.
Intuitively, start cutting people off from intenet access and one would expect some curtailment of "sharing" since untoward consequenses can come to fruition.
Numbers can always be massaged, and this cuts both ways...government data and data secured by those regularly cited here as irrefuteable evidence presented by in research papers.
Sounds to me as if the program is starting to make some headway. Will it stop people with a Tenebaum mindset? Maybe. Maybe not. But at least now they are on notice that there are consequences to what they have been doing to date with relative freedom.
[ link to this | view in chronology ]
Re:
I know someone who has had a No 1 hit. I knew someone who won the Nobel prize.
Anyone who posts like you did is either stupid or a liar. Which is it?
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Encryption
[ link to this | view in chronology ]
Re: Encryption
BTW, even with an encrypted file one still has to identify an unencrypted title somewhere. Moreover, how many "sharers" really download and upload multi-GB files, not at all unusual for moview and new generation software. Even music in the MB range are far larger than an ordinary business document, academic paper, lengthy court opinion, etc., most of which are in the order of modest KBs.
[ link to this | view in chronology ]
Re: Re: Encryption
so whats the next strawman to burn? Its kinda fun :)
[ link to this | view in chronology ]
[ link to this | view in chronology ]
True at the moment - because most can get away with it without encryption - but if that situation changes then the ease of use, speed and availability of encrypted sharing systems will increase dramatically and then it will become a mass activity. The security services know this - which is why they lobbied against the DEA/B in the UK. They don't want to have to search for their needles in a bigger haystack!
BTW, even with an encrypted file one still has to identify an unencrypted title somewhere.
Read up about how (for example) Freenet works - and you will realise that you miss the point. It isn't just the file (and the title) that is encrypted - it is also your identity. Someone somewhere does see an unencrypted title - they are the person at the other end o=f the link - and even they don't know who they are sending the file to - (or receiving it from).
[ link to this | view in chronology ]
Thus, I would not be inclined to dismiss it.
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Re:
Yes but
though these is no way or knowing the absolute numbers within this category.
It removes the possibility of drawing conclusions.
Typically file sharer's are students. Students typically exist in multiple overlapping peer groups (accommodation, course, clubs and societies at college - plus their old peer group from home.
Thus one person receiving a letter would probably get to around 100-300 people who could say that they "knew someone".
Plus it could easily be that the 50% who "said they or the person would stop" are responsible for less than 5% of file sharing - once again we don't know.
[ link to this | view in chronology ]
[ link to this | view in chronology ]
Additional big file sharer problem
[ link to this | view in chronology ]