Powerline risks real: expert
It is interesting to contrast the claim made in the headline of this article with the debate included within it. Although there are no figures associated with the statistical claims of correlation in this story, it illustrates an important feature of public debates associated with the beliefs of "experts". The same correlation, statistically significant because of the large data set upon which it was based, may in real terms be so small, that is, explaining such a low percent of the variability, that it is of no practical importance in a real-world environment.
Notice who sponsored Dr. Luben's trip to Australia. Is this important in making an inference about his evidence?
See also the Hydro cancer-link study article.
Where to next?
Student Questions for this article
Index - Related articles
Index - Inference
Numeracy in the News - Main Index