I. Thou shalt not hunt statistical significance with a shotgun.

II. Thou shalt not enter the valley of the methods of inference without an experimental design.

III. Thou shalt not make statistical inference in the absence of a model.

IV. Thou shalt honor the assumptions of the model.

V. Thou shalt not adulterate thy model to obtain significant results.

VI. Thou shalt not covet thy colleague's data.

VII. Thou shalt not bear false witness against thy control-group.

VIII. Thou shalt not worship the 0.05 significance level.

IX. Thou shalt not apply large-sample approximations in vain.

X. Thou shalt not infer causal relationships from statistical significance.

Michael F. Driscoll
The American Mathematical Monthly