A new survey from the American Assembly, a research center at Columbia University, sheds new insight into consumer attitudes and behaviour regarding music acquisition and and copyright enforcement. One significant finding, among other things, was that people who shared more music on P2P networks were also the larger (in absolute terms) purchasers of music through legitimate means.
What was more interesting though, was this comment from reader ‘Titanium Dragon’, questioning the reliability and validity of the findings:
1) Surveys, while useful for determining opinions (to some extent), are actually nearly useless for monitoring actual behavior, especially behavior percieved to be unethical.
2) Surveys show a very strong trend towards socially accepted behavior in excess of what is actually observed, and against identifying with groups which are negatively presented to the population. For instance, surveys of weekly church attendance will have twice as many respondents claim to show up to church as actually show up to church on a weekly basis. Likewise, surveys of gun owners over-represent the use of firearms in self defense by a full order of magnitude. Finally, while 15-20% of the population of the United States will tell you that they do not believe in any sort of deity, only 2% will tell you they are atheists if they are asked. Numbers regarding what is seen as socially acceptable behavior is likely to be exaggerated, any behavior which is seen as particularly desirable is likely to be vastly exaggerated, and any group which is perceived in a negative manner is likely to be underrepresented.
3) The survey makes no attempt to correlate actual behavior with reported behavior. Indeed, they only claim, based on the result of -another- survey (which was improperly averaged to boot) that piracy is not underreported, which is obviously stupid. This is just a terrible flaw when you’re actually trying to determine real behavioral patterns, and suggests a lack of understanding of proper methodology.
4) The way in which you ask questions, and even the order in which you ask them, can have a very large impact on the answers you receive. This report does NOT list the exact questions asked, nor does it mention how it asked the questions or in what order, or if the order was randomized. This is incredibly bad, and deeply suspect: without that, you have no way of knowing whether the poll was conducted properly, and whether the questions were written correctly.
Unfortunately, this renders the entire survey completely without meaning. Not that it will stop people who are unfamiliar with proper survey techniques and statistics, especially ones with agendas, from picking up and running this story, and tons of ignorant folk citing it as support for their arguments for or against piracy.
It would be nice for someone to actually do a real behavioral study of people’s habits, but I suspect few people engaged in illegal activity would actually acquiesce to it, and without a random sample, its kind of pointless
What is your own take on the survey findings? Does it legitimize, in a way, the existence of P2P sharing networks and challenge myths that it undermines creativity and the music industry?