Statistics And Probability In Scientific Research Is Immensely Complicated, And Is Almost Always Misinterpreted By The Media And Especially By Juries

I am not sufficiently knowledgeable about glyphosate (Round-Up) or statistical analysis to speak intelligently about this study, but it makes some interesting points, and illuminates some of the things about scientific inquiry that many people don't understand. The analysis and interpretation of the data is just as important as the collection of the data. And; What questions are the researchers asking? Are they starting from a neutral premise, or are they tipping their hats, like...oh...for instance..."The world is warming because of man: How much and why?"
Glyphosate Vs. Caffeine: Acute and Chronic Toxicity Assessments Explained

The current brouhaha over Round-Up is a colossal failure of our legal system and, perhaps even more catastrophic, our scientific education in America.

1. Observe some phenomenon.
2. Guess how it works.
3. Look at others' work and fill in whatever gaps you can.
4. Formulate a hypothesis, with specific predictions that can be falsified.
5. Design experiments to test that hypothesis.
6. If your predictions are borne out by experiment, you get a cookie.
7. If your predictions are not borne out by experiment, shitcan or modify the hypothesis.

Notice there is no step that includes massaging the data to fit the hypothesis. Notice too that there is no step that works backwards from a conclusion.

The Scientific Method is the bedrock of Western Culture and success. When we tamper with it (AGW, Alar, Asbestos, etc) we risk diving into an irrational culture in which homeopathy and copper wristbands and dousing and clever plaintiffs' attorneys are considered just as valid as scientifically proven concepts.

It doesn't matter how beautiful your theory is, it doesn't matter how smart you are. If it doesn't agree with experiment, it's wrong.

[Hat Tip and input: Jay Guevara]

posted by CBD at 12:15 PM

| Access Comments