Questions: Variables: Independent, Dependent, and Confounding
3 questions to test your understanding
Score: 0 / 3
Question 1 Multiple Choice
A study tests whether caffeine improves memory by giving one group 200mg of caffeine and a control group a placebo, then measuring scores on a word-recall test. What is the dependent variable?
ACaffeine dose (200mg vs. placebo)
BWord-recall test scores
CThe participants themselves
DThe amount of sleep participants got the night before
The dependent variable is what is measured as the outcome — word-recall scores. Caffeine dose is the independent variable (what is manipulated). Amount of sleep is a potential confounding variable. The DV is always the outcome you are measuring to see whether the IV had an effect.
Question 2 True / False
In an observational study that finds a correlation between screen time and attention difficulties in children, screen time is the independent variable.
TTrue
FFalse
Answer: False
In an observational (non-experimental) study, neither variable is manipulated — both are simply measured. The labels 'independent variable' and 'dependent variable' are defined by research design, not by correlation. Calling screen time the IV implies it was assigned or manipulated, which it was not. In correlational studies, researchers use terms like 'predictor' and 'outcome' to avoid implying causation.
Question 3 Short Answer
A researcher studying the effect of exercise on mood fails to account for the fact that participants who exercise more also tend to sleep more. What threat does this represent, and how could it be addressed?
Think about your answer, then reveal below.
Model answer: Sleep is a confounding variable — it co-varies with exercise (the IV) and independently affects mood (the DV), making it impossible to attribute mood changes solely to exercise. It could be addressed by measuring and statistically adjusting for sleep, or by holding sleep constant across conditions.
Confounds are threats to internal validity because they provide alternative explanations for the IV-DV relationship. Good experimental design either holds confounds constant, randomizes them across groups, or measures and statistically controls them.