If your experiment needs statistics, you ought to have done a better experiment.
What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.
There's the wonder of being able to do research from your own living room, of course. I do find that my biggest research issue, though, is how to frame my questions.
In much of society, research means to investigate something you do not know or understand.
The edifice of science not only requires material, but also a plan. Without the material, the plan alone is but a castle in the air-a mere possibility; whilst the material without a plan is but useless matter.
Hypotheses are the scaffolds which are erected in front of a building and removedd when the building is completed. They are indispensable to the worker; but the worker must not mistake the scaffolding for the building.
A researcher has to be able to identify key points and suggest intelligent questions.
Hypotheses are nets: only he who casts will catch.
Quantitative methods are no more synonymous with objectivity than qualitative methods are synonymous with subjectivity.
People doing the kind of sound research that I'm interested in still have a difficult time.
Thus, in general, in the first instance, the direction of interest in empirical fact will be canalised by the logical structure of the theoretical system.
It is important to consider whether the sample size selected by the Environmental Genome Project will provide sufficient power to discover most alleles relevant to gene-environment interactions.
I think that too ended up affecting a lot of the different research questions that I later asked was really was about well to what extent... How do we balance choice as possibility and choice as limitations?
I have never suggested any principled difference between the natural and social sciences.
In performing experiments, it is necessary... that they be simplified as much as possible, and that every circumstance that could complicate the results should be completely removed.
