How the UK gov't spun 136 survey respondants into 7m infringers
The Advisory Board claimed it commissioned the research from a team of academics at University College London, who it transpires got the 7m figure from a paper published by Forrester Research.How UK Government spun 136 people into 7m illegal file sharers (Thanks, Glyn!)The More or Less team hunted down the relevant Forrester paper, but could find no mention of the 7m figure, so they contacted the report's author Mark Mulligan.
Mulligan claimed the figure actually came from a report he wrote about music industry losses for Forrester subsidiary Jupiter Research. That report was privately commissioned by none other than the music trade body, the BPI...
The 7m figure had actually been rounded up from an actual figure of 6.7m. That 6.7m was gleaned from a 2008 survey of 1,176 net-connected households, 11.6% of which admitted to having used file-sharing software - in other words, only 136 people.
It gets worse. That 11.6% of respondents who admitted to file sharing was adjusted upwards to 16.3% "to reflect the assumption that fewer people admit to file sharing than actually do it." The report's author told the BBC that the adjustment "wasn't just pulled out of thin air" but based on unspecified evidence.
the latest
latest episodes
This story isn't entirely as sensational as people are making it sound. 1176 people is a decent enough sample size to draw conclusions from at a reasonable confidence level. It's been a while since I ran the numbers, but I think it takes less than 2500 people to draw an accurate conclusion (19 times out of 20) on the American population. (You're welcome to check my math, of course.)
The one major grey area is the extra percentage (4.7%) they added on to account for potential liars. So, in other words, the grey area ... is a grey area.
So, I've heard the claim that the only way to stop file sharing (legal or not) is to literally "break" the Internet. Now it looks like that's exactly what their trying to do.
Are we still working on access to the web being a basic human right?
Some of those who acknowledged using file-sharing software may in fact have been lying just because they could. What's more, using file-sharing software in and of itself is not necessarily illegal; it's only illegal when you use it to share copyrighted files.
You could make the argument that 1176 people is a reasonable sample if the demographics of that sample represented, proportionally, the demographics of the (online) nation as a whole.
But basing the study on research sponsored by the music industry, which has an obvious financial interest in a specific result, eliminates any reasonable expectation of objectivity.
As suggested above, the number of illegal file sharers is likely in the millions based purely upon statistical inference.
This however does not imply that more laws need to be enforced, but rather that some laws need to be changed to adapt for current cultural use of information. If the laws are changed, then even though the downloads will continue, they will no longer be illegal.
Change the laws, not the flow of the internet.
@1: Making up random numbers to add to the survey results (and not disclosing them). It's that the government published a stat as having originated with its own original research, when really it was a third-hand rehashing of a study commissioned by an extremely partisan industry group.
If the government proposed to make everyone eat five servings of carrots a day because they'd "discovered" that 7m British centenarians ate carrots 5x a day, and then it came out that this "discovery" was in fact an extrapolation of a summary of a press-release issued by the Carrot Farmers of England; then the underlying research, and the conclusions drawn from it, would be ridiculed just as thoroughly.
I agree with Beryllium that 1176 people might make up a decent enough sample size to draw conclusions from.
But the 4.7% adjustment in the number of file-sharers is actually a 40.5% adjustment in the number of file sharers. They claim that this adjustment was based upon unspecified evidence and was not pulled out of thin air, but if they cannot specify their evidence then we have no way to verify that their adjustment was not pulled out of thin air.
So we are looking at 4.8 million people having used file-sharing software at one time in their life, another 2.9 million users pulled out of thin air and the last 0.3 million was a rounding error.
If you want to use smallish sample groups in your statistical analysis, decide upon arbitrary modifications to apply to the raw data which bias the results, come to the exact sort of conclusion that the entity who pays for the study wants to hear, and then publish the results as an excuse to push forth litigation against the public interest... you do really need a very concrete reason why your research is as good or better as any other nonpartisan research.
Given the facts of the matter it would be foolish to accept anything but the highest standards of professionalism in the people conducting this study. Loose standards which are "sort of as good as others' " in a manner of speaking are just another way of bending over and painting a target on your ass. Most statistics cited in public are bad and biased. The very method it's being stated should be an obvious clue as to what they're trying to force people to believe.
7 million infringers, based upon a small number of people who only admitted to using a file-sharing program? (uh, gee, I guess downloading World of Warcraft patches or freely distributed game models is piracy too?) How about being more honest and scientific, and phrasing that in terms of a percentage of the population? The answer is because the very intent is to throw a large number in people's faces and try to steer their perception of the implications in the direction that gets the lobby what it wants.
This is strange.
In a democracy, the claim for such a huge of infringers should be ground for making file-sharing legal.
7m people downloading, means that 7m people do support file-sharing.
How many votes are they?
What adjustment did they use to account for the non file sharers who answered affirmatively just to be cool?
Independent research which has proved the validity of its sampling methods may use those techniques to extrapolate to the population.
From partisan trash like this, I demand overwhelming empirical data.
If there hadn't been a 40% upward adjustment based on 'unspecified evidence', this would have been a neat trick. Distract everybody with a statistical red herring, when the real ploy, as Francesco points out, is to use "7 million people do it" as evidence that "it must be stopped".
What if the answer had been 51% of the population? I'm sure at least 51% of people lie sometimes, should we cut off their water supply perhaps?
I'm not surprised at all. The home education community are fighting this sort of "research", included in the Badman report, which is being used to justify draconian new rules allow the authorities to enter homes without any reason, and to interview children alone, without an adult or guardian present. If it is passed, it will be the first time in the UK that the authorities will have a right of access to the homes of people who are not suspected of having done anything wrong.
That this should be forced through on the back of very, very dubious statisics is frightening to me: only about 10% of Local Education authorities responded to their survey and they extrapolated figures for the whole country from those respondents; they made claims about the number of home educators known to social services without accounting for the fact that many more home educators are reported to social services simply because neighbours think it is illegal to keep your children away from school; and then extrapolated (from the number of families reported) a level of concern about home educating families that isn't borne out by the facts in any of the education authorities, even those they relied on for the "facts" that they extrapolated from...!
I have learned to treat with a high level of suspicion the information which the government treats as proven fact... and that informs my whole impression of government:
http://ahed.pbworks.com/LiesDamnedLiesStatistics
*This* is what happens to Freedom of Information requests for the information
http://www.whatdotheyknow.com/request/evidence_in_support_of_badmans_r
Did someone mention democracy? Ha ha ha ha ha ha.
They do ever more consultations with communities and seem to have no problem with ignoring the reponses even when overwhelmingly against the course of action they have chosen. Of course, the always say that "after a consultation process, we have..." as though that actually meant something.
The legal aid system, which offers assistance to people who need to defend themselves and don't have the means to do so, has been systematically dismantled to the point where it is becoming quite hard to find lawyers willing to take on legally funded work, and that was predicted by the consultation response in advance, and they ignored that. They can massage figures from anything into justification for anything, which isn't good for democracy... or for the future working of the country on a non-democratic level for that matter... because it leads to poor decision making.
I thought the argument (which I support) was that we were ALL filesharers; that casual copyright infringement - from photocopying knitting patterns to taping stuff off the radio - was the norm - and that copyright was out of balance and unreasonable because of it.
Now we're complaining that 7 million is too high?
Surely it should be every man, woman, child and their cat that engages in rampant copy monopoly abuse, and the reason for that abuse is the bogus, unreasonable copyright cartel authored laws.
I used, before i got bored with the same old leading questions, to answer these official (seeming) surveys, in fact I may be one of the people who admitted to using file sharing services to download music, but I was only ever asked if I used p2p services, like pirate bay, never if I used such services Illegally, I used to often download Nine inch nails cc licensed material and other cc licensed music from pirate bay which I am quite legally entitled to do but was given no opportunity to qualify my answers
So, if the entertainment industry wants the ability to kick 7mil people off the net, shouldn't the ISPs be lobbying against them? That seems like a significant chunk of revenue that could potentially be cut off.
I think the suits in LA best reflect on the nature of people who are fully web-invested. Try to chop off their lives and they may be the kind that show up on your doorstep wearing C-4 underwear. Seriously. How to explain this to a Hollywood vampire? OK: try to imagine someone cutting off your blow, silicone tits, fast cars and hordes of "admirers".
Small samples like this assume a homogenous population, no variance in certain basic key data points. Then to apply on top of it a number of arbitrary adjustments, which now means that your error range can be orders of magnitude.
I've watched this in operation in some of the stuff in our company. Start off with an assumption, do a study that ends up not really showing anything, make a few corrective adjustments, instead of redoing the study with the proper questions being asked, and viola with a silver tongued delivery, assumption proved to be correct!
Then six months down the road, "Well whatta ya know, who could have known we were wrong?"
I don't care what anyone else says, the idea of getting a perfect representation of the population through almost any survey is impossible, and will most likely be affected by the bias of those using the survey's recorded data, in this case, the British government.
nutshell: Surveys are bullshit.
We can however make another assumption, that negates this studies conclusion. The music industry is dishonest, unethical and will do whatever is needed to skew information in their favor, 100% of the time.
In the interest of furthering the crumble of the record industry - here are two sites that keep me entertained:
http://www.kerevizli.com/
http://mixtapemixtape.com/
The Labour government in the UK has been making figures up for years, unemployment, immigration, time it takes for Saddam to unleash mythical WMD on the UK, their expenses, school league tables, exam results, you name it they can make it up. The corrupt evil sethlord Mandleson just wants more power. Meanwhile the ISPs know that if they make any attempt to limit or cut off P2P users, legal or otherwise, then who will need those fat 20MB-50MB internet connections? You can surf, email, play games quite well on a 2MB connection. So the ISP's will do nothing that is going to strangle their revenue stream.
By this point we can all agree that:
a) The music, software, and other industries get whatever they want through skewed research and intensive lobbying.
b) The government will do whatever they want, even if the majority doesn't want it, and that means they'll do whatever the people with money want them to do.
c) The current laws are already out of line with what the majority of people consider acceptable and the new laws being introduced are even further from what the majority considers acceptable.
With all that in mind, what do we do about it? Lobbying the government won't work -- we don't have enough money. The corporations don't give a shit what we say, because even if we threaten to not buy things, now they can just sue us. Clearly, the government is failing us, so we need to work around it. Tough luck for the Brits and us Canucks, but the Yankees still have their guns, right? Isn't this what the right to bear arms was all about?
It is not a government that makes decisions and statements. People do it.
It is crucial that the NAMES of these people are right in the title of an article.
The NAMES! The NAMES!
The most important thing everyone should get from this article is how awesome More or Less is. You should all be listening to it if you aren't already.
@24:
Oddly enough, the government rarely tell us the names.
You would think that they were trying to keep us in the dark or something ... naaah.
Let's look at this another way: How do we deal with a government that arbitrarily decides that 11.6% of its subjects/citizens are criminals?
The fact of the matter is that the current state of copyright laws do not serve the interests of the people, and as such, these laws must be drastically changed.