Ethics and Bias
Apply ethical guidelines for human and animal research and evaluate bias, validity, reliability, and generalizability.
A lesson of AP Psychology, in Scientific Foundations of Psychology, about 10 minutes.
Ethics and Bias in Psychological Research
By the end of this lesson, you will be able to describe APA ethical guidelines, identify common research biases, and evaluate historical cases of ethical violations.
APA Ethical Guidelines
The American Psychological Association sets standards for ethical research:
Informed consent: participants must be told what the study involves and agree to participate. Exceptions: when deception is necessary and fully justified.
Voluntary participation: participants must be able to withdraw at any time without penalty.
Confidentiality: participant data is protected; not shared without consent.
Debriefing: after studies using deception, participants must be fully informed of the true nature and purpose of the study.
Avoiding harm: researchers must protect participants from physical and psychological harm; benefits must outweigh risks.
Animal research: must minimize pain and distress; justified by scientific benefit.
Historical Ethical Violations
Milgram Obedience Study (1963):
Participants believed they were administering increasingly powerful electric shocks to another person when they answered questions incorrectly. 65% went to the maximum level (450 volts). The "victim" was a confederate; no one was actually shocked.
Ethical issues: deception; potential psychological harm (participants were distressed); lack of easy withdrawal (experimenter insisted they continue).
Zimbardo's Stanford Prison Experiment (1971):
Normal college students randomly assigned as "guards" or "prisoners" in a simulated prison. Guards became abusive; the study was shut down after 6 days.
Ethical issues: harm to participants; Zimbardo himself took on the role of prison superintendent, compromising objectivity; participants weren't truly free to withdraw.
Sources of Bias in Research
Sampling bias: the sample is not representative of the population.
- WEIRD bias: most psychology research uses Western, Educated, Industrialized, Rich, Democratic samples — may not generalize to the rest of humanity
Observer bias (experimenter effect): researchers unconsciously influence results in the direction they expect.
- Blind/double-blind designs control for this
Confirmation bias: tendency to notice and interpret evidence that confirms existing beliefs.