In 1954 Darrell Huff wrote a book called "How to lie with statistics". It is well-known and has often been used as an introductory college textbook. The point of the book was of course to alert people to statistical skullduggery so that they were not deceived by it.
But even though it sold a lot of copies the book has been an almost complete failure. In most academic fields where statistics are used (e.g. medical research, psychological research, climate research) statistics are still routinely misused. I spent 20 years getting papers published in the academic journals of the social sciences pointing out the defective reasoning in other articles in my field and, more recently, my FOOD & HEALTH SKEPTIC blog has tackled the outlandish conclusions that prevail in much of medical research. It's all a very sad tale. A lot of so-called "science" is basically corrupt.
So I would very much like readers to take the time to listen to a breezy video below by statistican W.M. Briggs in which he gives examples of corrupt statistical reasoning from psychological, medical and environmental research.
And the lesson from the above? Don't accept ANY scientific statement until you have seen what people say who don't agree with that statement. Corrupt science is so common that the odds are that the critics will be right.
So why is statistically-based science so corrupt? Briggs gives you some answers but I will give you another one that you may not see elsewhere: Most users of statistics are Left-leaning academics and for them "There is no such thing as truth". There is however a desperate need for them to defend their ideology. Their egos depend on it.
Even medical science has become heavily politicized with the "war on obesity" and the general elitist view of the Left that anything popular is either wrong or bad for you (cue cellphones).