Another Good Reason Not to Trust Medics

[Archibald]

Yes, a series of traffic lights in such a short commute would really roger the stats and null the test.

51 sets, including pedestrian crossings (not including 4 zebra crossings)...

 

[Magnus]

Even using your rough numbers, a 9% rise would be highly significant statistically.

I guess you have very little understanding of statistics?

 

[stainlessguy1]

LoL

I know this is an older thread , but this was funny .
I have to shake my head . oh boy .

 

[datalore]

Poor experimental method, design, and conclusions, typical of medics.

They just included it in the BMJ as Christmas stuffing. Fluff.

Let's have some more time comparisons, Chaps!

+1

This is a classic example of subject and investigator bias. Sadly, it is typical of research by physicians.