It's funny, perspicuous, and I'm sure it'll have a very useful impact. It's also a great example of talk design. Anyone who takes science seriously should find the patience to sit through the talk, and perhaps go back once or twice to figure out any examples that they don't get the first time.
Last year I gave a talk to the British Society of Audiology, which was subsequently recorded & made available online. The title is “Science: How not to do it” and the talk covers some common logical and statistical errors in scientific research. There are also some good cartoons, and the opportunity to win points by spotting errors. I realise that it may not be sophisticated enough for many subscribers to this list, but some people have found it useful, so I thought I’d post the link.
MRC Cognition & Brain Sciences Unit