The first requirement in using statistics is that the facts treated shall be reduced to comparable units.

True science teaches us to doubt and, in ignorance, to refrain.

When a physician is called to a patient, he should decide on the diagnosis, then the prognosis, and then the treatment. ... Physicians must know the evolution of the disease, its duration and gravity in order to predict its course and outcome. Here statistics intervene to guide physicians, by teaching them the proportion of mortal cases, and if observation has also shown that the successful and unsuccessful cases can be recognized by certain signs, then the prognosis is more certain.

Experiment is fundamentally only induced observation.

Theories are like a stairway; by climbing, science widens its horizon more and more, because theories embody and necessarily include proportionately more facts as they advance.

But while I accept specialization in the practice, I reject it utterly in the theory of science.

In every enterprise ... the mind is always reasoning, and, even when we seem to act without a motive, an instinctive logic still directs the mind. Only we are not aware of it, because we begin by reasoning before we know or say that we are reasoning, just as we begin by speaking before we observe that we are speaking, and just as we begin by seeing and hearing before we know what we see or what we hear.