Experiments ask one question: if we change one thing, does it cause another thing to change?
Random sampling and random assignment both use chance but serve different purposes: one chooses participants, the other places them into groups.
A control group gives researchers something to compare against, so they can tell whether the treatment, rather than something else, caused a change.
Groups should be treated the same except for the independent variable, because any other difference can become a confounding variable.
Random assignment, experimental and control groups, manipulating the IV, measuring the DV, and controlling confounds all serve to support stronger cause-and-effect claims.
Key terms
experiment
A study designed to test whether changing one thing causes another thing to change.
random assignment
Using chance to place participants into groups, such as experimental and control groups.
experimental group
The group that receives the treatment.
control group
The group used for comparison, which does not receive the treatment.
independent variable
The variable the researcher actively changes.
dependent variable
The outcome being measured; its value may depend on the independent variable.
confounding variable
A factor other than the independent variable that could influence the dependent variable (confound means to confuse).
Watch out
Random sampling vs. random assignment: random sampling chooses who is in the study; random assignment decides which group each participant is in.
Giving only one group an extra activity (like a relaxing book) creates a confounding variable — you can no longer tell what caused the change.