Answer research design prompts using variables, operational definitions, ethics, statistics, validity, and experimental control.
A lesson of AP Psychology, in AP Psych Exam Readiness, about 15 minutes.
AP Psychology: Research Design and the FRQ
By the end of this lesson, you will be able to design a valid psychological experiment, analyze ethical issues in research, interpret statistical results, and write effective AP Psychology FRQ responses.
Presents a scenario; asks you to apply psychological concepts to specific parts of the scenario
Usually tests concepts from multiple units
Question 2: Research Design (typically 4 parts, worth 4 points)
Involves research methods, experimental design, and/or interpreting results
Tests your understanding of how psychological research is conducted and evaluated
Experimental Design: Core Concepts
Independent and Dependent Variables
Independent variable (IV): What the researcher manipulates — the "cause" being tested.
Dependent variable (DV): What the researcher measures — the "effect" being observed.
"Researchers wanted to test whether caffeine improves memory. They gave half the participants 200mg of caffeine and the other half a placebo, then tested recall 30 minutes later."
IV: Whether participants received caffeine or placebo
DV: Memory test score (recall)
FRQ answer format: "The independent variable is [specific IV], and the dependent variable is [specific DV]."
Control and Experimental Groups
Experimental group: Receives the independent variable (the treatment)
Control group: Does NOT receive the IV; provides the baseline for comparison
Placebo group: A type of control group that receives a sham treatment (a pill that looks identical but has no active ingredient)
Why both are needed: Without a control group, you can't determine whether changes in the DV are due to the IV or some other factor.
Random Assignment vs. Random Sampling
Random assignment: Randomly assigning participants to experimental or control groups — ensures pre-existing differences between participants don't systematically affect results. Eliminates confounds.
Random sampling: Randomly selecting participants from a population — ensures the sample represents the population; allows generalization.
Common confusion: Random assignment and random sampling are different procedures with different purposes. An experiment can have one, both, or neither.
Confound/Confounding Variables
A confound is a variable OTHER than the IV that could explain changes in the DV.
"In the caffeine study, if the experimental group participants were also students who had just taken an exam (while the control group had not), stress level would be a confound."
How to control confounds:
Random assignment (distributes confounds equally across groups)
Double-blind procedure: neither participants nor researchers know who is in each group — prevents experimenter bias and demand characteristics
Standardized procedures
Types of Research Designs
Experimental
Manipulates IV; can establish causation; requires control group; random assignment is ideal.
Strength: Can establish cause-and-effect relationships
Weakness: Artificial lab conditions may not generalize; some variables can't be manipulated ethically
Correlational
Measures relationship between variables; cannot establish causation; can be done in natural settings.
Correlation coefficient (r): Ranges from -1 to +1
+1: perfect positive (as X increases, Y increases)
-1: perfect negative (as X increases, Y decreases)
0: no relationship
Strength: Can study things that can't be experimentally manipulated; can detect relationships in natural settings
Weakness: Third variable problem; directionality problem — cannot establish causation
Case Study, Survey, Naturalistic Observation
Case study: In-depth study of one person or group; good for rare conditions; can't generalize
Survey: Self-report questionnaires; large samples; cheap; vulnerable to social desirability bias, sampling bias
Naturalistic observation: Behavior in natural settings; good ecological validity; researcher cannot control variables; observer effect
Statistical Concepts for the FRQ
Measures of Central Tendency
Mean: Average; affected by outliers
Median: Middle value; not affected by outliers
Mode: Most common value
When to use which: If the data is skewed (e.g., a few very high or very low scores), the median is more representative than the mean.
Measures of Variability
Standard deviation (SD): Average distance of scores from the mean; larger SD = more spread in the data
Normal distribution: Bell curve; ~68% of scores within 1 SD; ~95% within 2 SD; ~99.7% within 3 SD
Statistical Significance
p-value: The probability that the observed result occurred by chance. p < .05 means there is less than a 5% chance the results are due to random chance — the result is statistically significant.
Statistical significance vs. practical significance: A result can be statistically significant but trivially small in practice (especially with very large samples).
Interpreting Results
"Results showed a positive correlation of r = .62 between hours of sleep and GPA":
Positive correlation: more sleep → higher GPA (on average)
r = .62 indicates a moderate-to-strong relationship
Cannot conclude that sleep CAUSES higher GPA — third variable (e.g., general conscientiousness) could explain both
Ethics in Psychological Research
APA Ethical Guidelines require:
Informed consent: Participants must understand the nature of the study and consent voluntarily
Confidentiality: Participants' identities and data kept private
Debriefing: After the study, participants must be informed of the study's true purpose (especially if deception was used)
Right to withdraw: Participants can leave at any time without penalty
Minimal harm: No unnecessary physical or psychological discomfort; benefits must outweigh risks
Deception: Allowed ONLY when necessary for the study's validity and when participants are fully debriefed afterward. Cannot be used if it would cause participants to refuse to participate (thus violating informed consent).
IRB: Institutional Review Board must approve research before it is conducted.
Historical ethical violations:
Milgram's shock study: deception, psychological distress (though subjects were debriefed)
Watson's Little Albert: no parental consent, no debriefing, participant not deconditioned
Tuskegee Syphilis Study: participants not told they had syphilis; denied treatment — led to major reform of research ethics
Writing FRQ Responses
Format
FRQ answers are not essays — they are direct responses to specific prompts. Each point is earned by answering a specific sub-question correctly.
The AP Reader's question: "Did the student correctly identify [concept] and correctly apply it?"
The Name-Define-Apply Formula
For each part of the FRQ:
Name the concept being asked about
Define it in your own words (brief)
Apply it to the specific scenario
Example FRQ Part: "Identify the independent variable in the study."
Weak answer: "The caffeine."
Correct answer: "The independent variable is whether participants received caffeine (200mg) or a placebo. This is what the researchers manipulated to test its effect on memory."
The additional explanation is not always required, but it demonstrates understanding and helps if the grader thinks you might be confusing the concept.
What Costs Points
Describing the wrong concept (even if what you say is correct, if it doesn't match the question, you don't get the point)
Being correct but vague (naming the concept without applying it to the scenario)
Writing an essay instead of directly answering each sub-question
Contradicting yourself (if you say something correct and then something incorrect that qualifies it, you may lose the point)
FRQ Practice Scenario
Scenario: A researcher wants to test whether listening to classical music while studying improves test performance. She recruits 60 college students. Thirty are assigned to study in silence and thirty study while listening to Bach. She randomly assigns participants to the groups. After studying, all participants take the same memory test.
FRQ Questions:
Identify the IV and DV.
Why was random assignment important in this study?
What is one confound the researcher should try to control?
If the p-value is .03, what does this mean?
Answers:
IV = whether participants listened to classical music or studied in silence. DV = score on the memory test.
Random assignment ensures that pre-existing differences between participants (like music training or attention span) are distributed equally across groups, preventing these differences from explaining the results.
Possible confounds: the volume at which music was played; the difficulty of the material studied; whether participants already had preference for studying with music.
p = .03 means there is only a 3% probability that the observed difference in test scores occurred by chance, indicating the result is statistically significant (p < .05).
Summary
Experimental design: IV (what you manipulate) → DV (what you measure); control group; random assignment eliminates confounds; double-blind controls for bias.
Statistics: Mean, median, mode; standard deviation; normal curve; p < .05 = statistically significant.
Ethics: Informed consent, confidentiality, debriefing, right to withdraw, minimal harm; IRB approval.
FRQ formula: Name → Define → Apply. One correct, specific answer per part; don't be vague.
AP Exam Tip: The research design FRQ is one of the most predictable parts of the AP Psychology exam. Know the vocabulary precisely (IV, DV, confound, random assignment, p-value, correlation coefficient) and practice applying each concept to novel scenarios. Students who know the definitions but can't apply them will struggle; students who can apply them to any scenario are prepared.