Darell Huff's "How to Lie with Statistics"...
An almost 60 years old book, still useful and accurate as hell.
The final chapter, giving some giveaway methods, tells that...
"the method may be direct misstatement or it may be ambiguous" -> anonymous studios...
"selection of favourable data and suppression of unfavourable" -> Only two studio, and physical media sales labelled as "not relevant"
"practice of using one year for one comparison and sliding over to a more favorable year for another" -> again, depend of what was out of those 2 studios.
"Use of OK names" : Carnegie Mellon University.
"A correlation given without a measure of reliability (probable error, standard error) is not to be taken very seriously" -> To be confirmed, can't check the paper from my office...
"Given percentages and missing raw figures" -> checked again.
And so on...
A very interesting read if you're into manipulation - even if most examples are a bit outdated.
Perhaps they should also discover a link between megaupload and the unemployment rate, or between abortion and criminality... (yes, it's a deliberate poke)