Pete317 wrote:
I agree that those two TRL reports are invalid, but for different reasons. They both attempt, by the dubious use of statistical methods, to establish a correlation between speed and accident numbers, and to imply causation.
Causation isn't implied - it's wilfully falsely assumed.
And I'd rate the statistical method as fraudulent rather than dubious.
I'm actually really angry about TRL421 and TRL511. How dare the TRL (of all organisations) undermine UK road safety with wilfully false "scientific research"?
Pete317 wrote:
But, as far as I recall, neither of them attempt to establish any mechanism - although they do allude to mechanism from other research, such as the infamous Kloeden et al papers.
These do attempt to establish mechanism, but their false assumptions render this completely invalid.
So I stand by my version.
Don't you rate Kloeden as invalid? (not that I consider the positioning of the word "valid" as worth an argument

)