Experimentation process

Question	Answer
Hypothesis	A suggested explanation of a phenomenon. Must be falsifiable. Should be specific, which makes it easier to test.
Independent variables	Variable that are manipulated by experimenter.
Dependent variables	Variable that are measured.
Variables relationship	Manipulating independent variables is assumed to affect dependent variables results.
Null hypothesis	When there is no relationship between manipulating independent variables and the result of dependent variables. When testing hypothesis, we are seeking to disproof it.
Control condition	No experimental manipulations performed.
Experimental condition	Experimental variable(s) are manipulated.
Validity	What we say is being measured, is really what is being measured?
Reliability	Running the experiment several times, do we get the same results every time?
Confounds	Are there uncontrollable variables that might be influencing the results?
Subject pool	Should represent the intended user population. Should not be biased.
Between-subjects	Each subject experience only one condition. Needs more subjects. Do not suffer form "learn effect".
Within-subjects	Every subject experiences every condition. Need fewer subjects. Order of conditions should be counterbalanced to minimize learning effects.
Subjects influence on results	More subjects increase confidence of findings. Larger effects need less subjects to make it statically relevant. Smaller effects require subjects.

Resource summary

	Created by dcardosods almost 11 years ago