Freedom to Tinker has an excellent post on journalistic skepticism. I agree wholeheartedly that numbers can take on a life of their own — it concerns me even more because it is a real downside of estimates and measurements, of which I am a huge advocate.

I felt a similar way regarding (the book) Geekonomics' approach to estimating the $180 billion costs of insecure software, and dissected the number to understand the details.

I worry quite a bit that only the end results get reported without showing the work. But I don't really know any way around it, except to be fair and open throughout the process.