G Seven tips for advanced students to improve your experiments yet further

Why not:

use more than one DV?

manipulate variables instead of holding them constant?

use practice trials and pretests?

measure variables that have been randomized?

include manipulation checks?

measure process variables?

If possible, also try to ensure that the experimenter is blind to condition. If this is not possible, leave it as late as you can in the experiment before the experimenter discovers the condition.

Remember, your task is to design a reasonable, meaningful experiment, free of confounds, not to find the “correct” way to design it (see Section 13.9.1 of the book).