<snip>
>The researchers discovered that the worse a
>state's schooling system was - as measured by
>teacher salaries, spending per pupil and national
>test scores - the greater the interest in the
>Mozart effect.
Here we have a good example of recursiveness. While technically correct, the newspaper summary of the Stanford researchers' results implies a degree of both spontaneous inquiry and finality that isn't borne out in the study -- which is not to say that the study isn't interesting. What Bangerter and Heath did is to try to empirically test some well-established and widely accepted hypotheses in social psych for which there was little empirical evidence. So the anxiety thing wasn't a 'discovery' in the sense, almost, of 'surprise' that non-scientists might assume. Also, contrary to what may be inferred, they didn't measure "what people thought" about the Mozart research or their anxiety levels or even people's interest in the Mozart Effect story. What they measured was the number of articles on the ME appearing in a state's newspapers as a proportion of all articles in that state's newspapers, which they used as a proxy for interest in the story. Then they ran a few regressions and got a statistically significant result -- which means only that the result was very unlikely to be random, not that it definitely confirms the hypothesis.
There is another very possible interpretation of the results that the authors don't mention, which is that the newspapers in some states simply produce fewer articles based on actual local reporting and thus rely, as a proportion of their content, more on syndicated soft news stories. One might also predict a correlation between poor quality news coverage of local and state-wide issues and poor performance of the state's schools. There are probably quite a few alternative explanations of the results. But I like the soft-news-is-no-news story.
In a way, the mass media account of any scientific study contributes to the gestation of a new scientific legend because the 'gist' of the science seems so much more certain than the science itself.
The Sandwichman