[babydriver]

Science isn't PROVED, science is a process. You can never PROVE a theory, you can only disprove it. The rest of it is always subject to revision.

This is science:


A. Observe a natural phenomenon that doesn't fit in with the present understanding of how things work. Verify that the observation is valid and not the result of a measurement error or some other confounding factor.

B. Form a hypothesis to attempt to explain the observation.

C. Devise a test for the hypothesis that includes the ability to predict the expected outcome.

D. Attempt to prove the null hypothesis: is there a way to explain the resulting data as the result of random chance? Does the outcome of the experiment still support the hypothesis? Are there any possible alternative explanations?

E. If step D does not reject the null, repeat steps B through D with a new hypothesis.

F. If step D rejects the null, test and reject possible alternative explanations.

G. If step F does not lead to rejection of the alternative explanations, redesign the experiment to test for these alternatives (steps B through D again for each one.)

H. Keep in mind that your hypothesis may be proved false at any time based on new evidence. Go back to step B.

and most important,

I. Don't ever fall in love with your own hypothesis. The divorce can get really messy.