Comprehensive Guide to Psychological Research Methods, Statistics, and Ethics

The Scientific Method and Goals of Psychological Research

• Psychology as an Objective Science: Psychology operates as an empirical science. Researchers utilize the scientific method to maintain objectivity, eliminate cognitive and experimental biases, and guard against subjectivity that could otherwise skew research outcomes.

• Historical Consequences of Unvalidated Research: Flawed research methodologies lead to inaccurate conclusions and harmful practices.

  • Prefrontal Lobotomy: Developed in the early 1900s as a surgical treatment for schizophrenia and severe psychiatric conditions. The procedure involved severing neural connections within the brain's frontal lobes. Controlled scientific studies eventually demonstrated that lobotomies were ineffective and caused severe, permanent neurological and psychological damage.

• Assumption of Lawfulness: The scientific approach rests on the fundamental assumption that biological, physical, and behavioral events are governed by an underlying lawful order or set of overarching principles.

• Three Primary Goals of Scientific Research:

  1. Measurement and Description: Defining, quantifying, and categorizing behavioral or physiological phenomena.

  2. Prediction: Identifying relationships between variables to forecast future behaviors or outcomes.

  3. Control and Application: Applying empirical findings to alter, manage, or improve real-world conditions and clinical outcomes.

  • Note: No single goal is superior to the others; comprehensive scientific understanding requires all three approaches.

The Scientific Process and Operational Definitions

• General Flow of Psychological Research:

  • Theory: A systematic set of concepts, laws, or explanations derived to explain observed phenomena.

    • Example: Biological theories of aggression propose that aggressive behavior stems from elevated hormones (such as testosterone), chromosomal abnormalities, or specific neurotransmitters (brain chemicals).

  • Hypothesis: A specific, testable prediction or research question derived directly from a theory.

    • Example: "Elevated levels of testosterone will cause human males between the ages of $18$ and $30$ to display higher rates of physical aggression."

  • Research Design & Data Collection: Empirical testing of the hypothesis.

  • Evaluation: Empirical findings either support the theory (adding to the cumulative body of knowledge) or fail to support it (requiring theory revision, replacement, or methodological redesign).

• Operational Definitions:

  • An operational definition precisely outlines how a specific variable is defined, manipulated, and measured within a study.

  • Conceptualizing Aggression: Aggression can be conceptualized as physical violence, verbal abuse, social manipulation, or accidental physical contact. Defining aggression specifically (e.g., as physical violence) dictates the exact measurement strategy used in the research.

• Participants:

  • The living organisms (human or non-human animal) observed or evaluated within a study. Formerly referred to as "subjects" in scientific literature; current American Psychological Association (APA) standards mandate the term "participants."

• Five Basic Steps in a Scientific Investigation:

  1. Formulate a testable hypothesis and establish operational definitions.

  2. Select the research design and method of data collection.

  3. Collect empirical data.

  4. Analyze the data using statistical procedures and draw conclusions.

  5. Report findings by writing a research report published in peer-reviewed scientific journals.

Descriptive and Observational Research Designs

• Naturalistic Observation:

  • Systematic observation and recording of behavior in real-world environments without direct intervention or experimental manipulation by the researcher.

  • Examples: Jane Goodall’s field studies observing wild primate populations (apes and chimpanzees); observing childhood social interactions in a classroom setting.

  • External Validity: High. Findings generalize effectively to real-world populations because behaviors occur naturally in unmanipulated settings.

  • Internal Validity: Low. The lack of environmental control prevents researchers from manipulating variables or establishing definitive cause-and-effect relationships (e.g., an observer cannot paint classroom walls or rearrange furniture to evaluate if overcrowding causes aggression).

• Case Studies:

  • An in-depth, detailed investigation of a single individual, small group, or rare occurrence.

  • Applications: Infrequent neurological disorders, rare brain damage, or specific psychiatric illnesses.

  • Example: Chronic Traumatic Encephalopathy (CTE): Research investigating concussions and severe brain damage in professional football players using post-mortem brain donations from families (as depicted in the film Concussion).

  • Limitations: High risk of misleading conclusions; unique single-case dynamics cannot be reliably generalized to the broader public population.

• Self-Report Measures (Surveys and Questionnaires):

  • Instruments used to collect self-reported opinions, abilities, personality traits, diagnostic symptoms, or behavioral tendencies.

  • Random Selection: A sampling technique where every individual in the target population has an equal probability of selection. Ensures a representative and diverse sample.

  • Sampling Flaws: Administering a self-esteem questionnaire exclusively to undergraduate university students aged $17$ to $20$ creates a restricted demographic that fails to represent the broader population.

  • Validity: The degree to which an instrument accurately measures the construct it claims to evaluate. Historical child shyness questionnaires measured social wariness (fear-based social avoidance) but mislabeled socially disinterested children (children who prefer solitary play without social anxiety) as shy, undermining construct validity.

  • Reliability: The stability or consistency of a measurement over repeated applications. A weight scale displaying $120\,\text{lbs}$, $170\,\text{lbs}$, and $115\,\text{lbs}$ on consecutive mornings lacks reliability.

Experimental Research Designs

• Establishing Causality: Experiments represent the definitive research design for identifying direct cause-and-effect relationships through variable manipulation and strict environmental control.

• Experimental Variables:

  • Independent Variable (IV / $X$): The factor or condition deliberately manipulated by the researcher (e.g., drug dosage, room temperature, wall color, background music).

  • Dependent Variable (DV / $Y$): The outcome variable measured to assess the effect of the IV (e.g., aggression scores, depression severity, test performance).

• Group Classifications:

  • Experimental Group: Receives the active treatment or experimental manipulation.

  • Control Group: Does not receive the active treatment (receives a placebo or baseline condition).

  • Random Assignment: Assigning participants to experimental or control groups purely by chance (e.g., drawing names from a hat) to ensure baseline equivalence between groups.

• Uncontrolled Variables:

  • Extraneous Variables: Any external variable that might systematically influence the dependent variable (e.g., testing the experimental group in the morning and the control group in the late afternoon creates a time-of-day extraneous variable).

  • Confounding Variables: Hidden or internal participant factors that vary alongside the IV, obscuring true cause-and-effect conclusions (e.g., unmeasured differences in participant motivation or attitude).

• Complex Experimental Variations:

  • Single-Group Repeated Measures: Exposing the same group of participants to multiple IV conditions (e.g., testing one group of participants in a yellow room and subsequently testing the same group in a red room).

  • Multiple Independent Variables: Evaluating the simultaneous effects of two or more IVs (e.g., testing the effects of music type [distracting vs. no distracting music] combined with room temperature [normal vs. hot] on study performance).

  • Multiple Dependent Variables: Measuring the impact of an IV on multiple outcome metrics (e.g., testing if an antidepressant reduces both depression and generalized anxiety scores).

• Methodological Trade-offs:

  • Internal Validity: High (strict experimental control).

  • External Validity: Low (contrived laboratory settings may fail to reflect natural real-world behavior).

• Field Experiments: Experiments conducted in natural settings when laboratory setups are impractical or ethically questionable (e.g., studying behavioral patterns in active illegal drug users without administering illicit substances like heroin in a laboratory).

Statistical Analysis in Psychological Research

• Definition: Mathematics used to organize, summarize, and draw logical inferences from empirical datasets.

• Descriptive Statistics: Mathematical techniques used to organize and summarize the key characteristics of a dataset.

• Measures of Central Tendency: Indicators of the central or typical score in a data distribution:

  • Mean: The mathematical average, calculated by summing all scores ($\sum X$) and dividing by the total sample size ($N$). Highly sensitive to extreme outliers.

    • Calculation Example: In a dataset containing scores $20000$, $20000$, $30000$, $30000$, and $200000$ ($N = 5$), the sum is $300000$, yielding a mean of $60000$. The extreme outlier ($200000$) distorts the mean, making it an unrepresentative indicator of the typical score.

  • Median: The exact middle score position separating the upper $50\%$ of data from the lower $50\%$ of data.

  • Mode: The single score that occurs with the highest frequency in a distribution.

• Measures of Variability: Quantifying the degree to which scores spread out or cluster together:

  • Range: The linear distance between the highest and lowest scores.

  • Standard Deviation: A numerical index reflecting the average distance of individual scores from the distribution mean. High standard deviation indicates wide dispersion; low standard deviation indicates scores cluster tightly around the mean.

• Normal Distribution (Bell Curve): A symmetrical curve where values cluster around the central mean. Exactly $68.26\%$ of all scores fall within one standard deviation above or below the mean ($\pm 1\,\sigma$).

• Correlation: A descriptive statistic evaluating the direction and strength of the linear relationship between two variables ($X$ and $Y$).

  • Correlation Coefficient ($r$): A statistic ranging from $-1.0$ to $+1.0$.

  • Positive Correlation: Variables move in the same direction (as $X$ increases, $Y$ increases; e.g., higher class attendance percentages correlate with higher exam grades).

  • Negative Correlation: Variables move in opposite directions (as $X$ increases, $Y$ decreases; e.g., increased alcohol/beer consumption correlates with lower exam scores).

  • Zero Correlation: Complete absence of a linear relationship between variables.

  • Correlation Magnitude: Absolute value determines strength regardless of sign:

    • Low Correlation: $0.0$ to $0.3$ (or $-0.1$ to $-0.3$).

    • Moderate Correlation: $0.4$ to $0.6$ (or $-0.4$ to $-0.6$) (e.g., SAT/ACT scores correlating $0.4$ to $0.6$ with first-year university GPA, functioning as moderate predictive tools).

    • High Correlation: $0.7$ to $1.0$ (or $-0.7$ to $-1.0$).

  • Correlation vs. Causation & The Third-Variable Problem: Correlation indicates predictive utility but does NOT demonstrate cause-and-effect.

    • Foot Size and Vocabulary: Positively correlated due to the third variable of age/child development.

    • Ice Cream Consumption and Drownings: Positively correlated due to the third variable of ambient outdoor temperature/summer heat.

    • Household Electrical Appliances and Academic Success: Positively correlated due to socioeconomic status and family education level.

    • Smoking and Depression: Correlated due to the third variable of the personality trait neuroticism.

• Inferential Statistics: Mathematical methods used to interpret data, draw generalizations from sample data to populations, and determine whether observed results are statistically significant or due to random chance.

  • Statistical Significance: An outcome indicating that the probability of observed findings occurring by chance is extremely low.

  • Alpha Significance Levels: Standard thresholds set at $p < 0.05$ ($5\%$ probability due to chance, $95\%$ confidence in true research findings) or $p < 0.01$ ($1\%$ chance, $99\%$ confidence).

Evaluating Research Quality, Biases, and Pitfalls

• Meta-Analysis: A statistical procedure combining empirical data from numerous independent scientific studies addressing the same research question to calculate an overall effect size.

• Sampling Bias: Systematic distortion occurring when a sample does not accurately reflect the broader target population.

  • WEIRD Samples Bias: Psychological research predominantly relies on participants from Western, Educated, Industrialized, Rich, and Democratic societies, severely limiting cross-cultural generalizability.

• Placebo Effect: Physical or psychological improvement observed when participants receive an inactive or sham treatment (e.g., sugar pill in clinical drug trials) driven purely by subjective expectations.

• Distortions in Self-Report Data:

  • Social Desirability Bias: Tendency of participants to answer questions in a manner viewed favorably by others, skewing self-reports on sensitive or taboo topics like illegal drug usage.

  • Response Sets: Unreflective answering strategies where participants repeatedly select the same score option (e.g., circling "3" continuously down a questionnaire).

• Experimenter Bias and Blinding Strategies:

  • Experimenter Bias: Unintentional influence exerted by researchers on study outcomes or data interpretation to align with their hypotheses.

  • The Clever Hans Case Study: A German horse alleged to perform mental arithmetic by stamping his hoof. Investigation revealed Hans was reacting to subtle, nonverbal facial and body tension cues exhibited by questioners as he approached the correct answer. When Hans was blinded from seeing the questioner's nonverbal expressions, his mathematical ability vanished.

  • Single-Blind Design: Study setup where participants remain unaware of whether they are in the experimental or control group, controlling for participant expectations.

  • Double-Blind Design: Study setup where neither the participants nor the primary experimenters know which individuals belong to the treatment or placebo groups, simultaneously eliminating experimenter bias and participant placebo effects.

Ethics in Psychological Research

• Historical Context and Governance: Escalated global concern regarding research ethics emerged post-World War II following revelations of unethical Nazi medical experiments, alongside controversial longitudinal studies.

  • The Tuskegee Syphilis Study: Conducted from the 1930s to the 1970s, tracking impoverished African American males infected with syphilis to study untreated disease progression. Researchers failed to obtain informed consent, reportedly injected healthy subjects, and actively withheld standard curative treatment (penicillin) after it was discovered.

  • Stanley Milgram’s Obedience Study: Conducted using a modified "shock box" setup where participants were deceived into believing they were delivering painful, potentially lethal electric shocks to another individual.

• Regulatory Bodies: All institutional research involving human or animal subjects must be pre-approved and monitored by a Research Ethics Board (REB) or Institutional Review Board (IRB).

• Core Ethical Principles for Human Research:

  1. Informed Consent: Written, voluntary agreement obtained after full disclosure of study parameters. Guardians must provide consent for minor children.

  2. Risk versus Benefit Ratio: Scientific and societal benefits must outweigh potential risks to participants.

  3. Protection from Harm: Absolute requirement to safeguard participants against physical and psychological harm.

     • Real-World REB Example: An REB prohibited a researcher studying pregnancy anxiety from administering questionnaires containing explicit questions about stillbirths or fetal deformities to pregnant mothers due to the potential for causing severe psychological distress.

  4. Justification of Deception: Deception is permitted only when essential to study validity, must be formally justified to the REB, and must not cause distress.

  5. Debriefing: Comprehensive post-study explanation provided to participants detailing the study's purpose, methodologies, handling of anonymous data, and results access.

• Ethics in Animal Research: Evaluates whether medical/scientific benefits (e.g., discovering insulin to treat diabetes) justify animal testing, while strictly forbidding harm for non-essential applications (e.g., blinding animals during cosmetic/mascara formula testing).

1. The Scientific Method and Goals of Psychological Research

• Psychology as an Objective Science:

  • Psychology relies on empirical scientific methods to collect real-world evidence.

  • Why it matters: Using structured scientific steps eliminates personal cognitive biases, subjective opinions, and experimental errors that would otherwise distort research findings.

• Historical Consequences of Flawed Research:

  • Bad or unvalidated research leads to harmful real-world practices.

  • Example - Prefrontal Lobotomy: Developed in the early $1900$s to treat schizophrenia and severe mental illness by severing neural pathways in the frontal lobe. It was widely performed until controlled scientific studies proved it was completely ineffective and caused permanent neurological and psychological damage.

• Assumption of Lawfulness:

  • Science assumes that physical, biological, and behavioral events are not random; they follow predictable, lawful rules and principles.

• Three Primary Goals of Research:

  1. Measurement and Description: Precisely defining, quantifying, and categorizing behaviors or physiological processes.

  2. Prediction: Identifying patterns and relationships between variables to forecast future outcomes.

  3. Control and Application: Applying research findings to solve practical problems, improve clinical treatments, and manage real-world conditions.

  • Key Principle: No single goal is superior to the others; complete scientific understanding requires all three.

2. The Scientific Process and Operational Definitions

• General Research Process:

  • Theory: A broad, systematic model or set of laws explaining why certain phenomena occur.

  • Lecture Example: Biological theories of aggression propose that violent behavior is caused by internal biological factors, such as elevated testosterone levels, extra chromosomes, or specific brain chemicals (neurotransmitters).

  • Hypothesis: A specific, testable prediction derived directly from a theory.

  • Lecture Example: "Elevated testosterone levels will cause human males between the ages of $18$ and $30$ to show higher rates of physical aggression."

  • Research Design & Data Collection: Choosing a method to test the hypothesis empirical setup.

  • Evaluation: Statistical results either support the theory (building scientific knowledge) or fail to support it (requiring the theory to be revised or discarded).

• Operational Definitions:

  • An operational definition explicitly specifies how a variable is defined, manipulated, and measured in a specific study.

  • Lecture Example (Conceptualizing Aggression): Aggression could mean physical hitting, verbal shouting, social manipulation, or accidental bumping. Defining aggression specifically as "physical violence" dictates the exact measuring tool used in the experiment.

• Participants:

  • The living human or animal organisms studied in research. Scientific guidelines (APA) require calling human subjects "participants."

• Five Steps in Scientific Investigation:

  1. Formulate a testable hypothesis and set operational definitions.

  2. Select the research design and data collection method.

  3. Collect empirical data.

  4. Analyze the data using statistics and draw logical conclusions.

  5. Report findings in peer-reviewed scientific journals.

3. Descriptive and Observational Research Designs

• Naturalistic Observation:

  • Observing and recording behavior in real-world settings without manipulating or interfering with the participants.

  • Lecture Examples: Jane Goodall's field research observing wild chimpanzees in nature; watching children interact in a classroom setting.

  • External Validity (High): Findings apply well to real-world populations because behavior happens naturally.

  • Internal Validity (Low): Lacks control over variables; researchers cannot alter conditions to prove direct cause-and-effect (e.g., an observer cannot paint classroom walls or rearrange furniture to see if it changes aggression levels).

• Case Studies:

  • An detailed, in-depth investigation of a single individual, small group, or rare event.

  • Application: Used for rare brain injuries, unique neurological conditions, or unusual psychiatric disorders.

  • Lecture Example (Chronic Traumatic Encephalopathy - CTE): Studying brain damage in deceased professional football players using post-mortem brain donations from families.

  • Limitation: High risk of misleading conclusions because unique findings from one person cannot be generalized to the entire public.

• Self-Report Measures (Surveys and Questionnaires):

  • Tools used to gather self-reported opinions, abilities, personality traits, or symptoms.

  • Random Selection: A sampling method where every person in the target population has an equal chance of being picked, creating a diverse and representative sample.

  • Sampling Flaws: Giving a survey only to undergraduate university students aged $17$ to $20$ creates a narrow, unrepresentative sample that fails to represent the general population.

  • Validity: Whether a test actually measures what it claims to measure. (e.g., early child shyness questionnaires measured fear/social wariness, but incorrectly labeled socially disinterested children who prefer playing alone as shy, lowering validity).

  • Reliability: Consistency of measurement over time. A bathroom scale showing $120\,\text{lbs}$, $170\,\text{lbs}$, and $115\,\text{lbs}$ on consecutive mornings is unreliable.

4. Experimental Research Designs

• Establishing Causality:

  • Experiments are the gold standard for proving direct cause-and-effect relationships because the researcher actively manipulates conditions.

• Experimental Variables:

  • Independent Variable (IV / $X$): The factor deliberately changed or manipulated by the experimenter (e.g., wall color, drug dosage, room temperature, background music).

  • Dependent Variable (DV / $Y$): The outcome variable measured to see the effect of the IV (e.g., aggression level, depression score, test grade).

• Group Classifications:

  • Experimental Group: Receives the active treatment or experimental manipulation.

  • Control Group: Does not receive the active treatment (receives baseline conditions or a placebo).

  • Random Assignment: Assigning participants to the experimental or control group entirely by chance (e.g., drawing names out of a hat) to ensure both groups are equal at the start.

• Uncontrolled Variables:

  • Extraneous Variables: External factors that might unintentionally influence the DV (e.g., testing the experimental group in the morning and the control group in the late afternoon makes time-of-day an extraneous variable).

  • Confounding Variables: Hidden internal participant differences that vary along with the IV, confusing the cause-and-effect results (e.g., unmeasured differences in participant motivation or attitude).

• Complex Experimental Variations:

  • Single-Group Repeated Measures: Testing the exact same group of participants under multiple conditions (e.g., placing one group in a yellow room and then testing the same group in a red room).

  • Multiple Independent Variables: Testing two IVs at once (e.g., testing background music type [distracting vs. no music] combined with room temperature [normal vs. hot] on study performance).

  • Multiple Dependent Variables: Measuring how one IV affects several outcomes (e.g., testing if a drug reduces both depression and anxiety scores).

• Methodological Trade-offs:

  • Internal Validity: High (strict laboratory control over variables).

  • External Validity: Low (artificial lab settings may not reflect real-world behavior).

• Field Experiments:

  • Experiments conducted in real-world environments when laboratory settings are impractical or unethical (e.g., observing active illegal drug users in real-world settings rather than giving illicit drugs in a lab).

5. Statistical Analysis in Psychological Research

• Definition: Mathematical procedures used to organize, summarize, and draw conclusions from research data.

• Descriptive Statistics: Math used to summarize and organize key features of a dataset.

• Measures of Central Tendency (Typical Scores):

  • Mean: Mathematical average (sum of all scores divided by sample size $N$). Highly sensitive to extreme outlier scores.

  • Lecture Example: For scores $20000$, $20000$, $30000$, $30000$, and $200000$ ($N = 5$), the sum is $300000$, giving a mean of $60000$. The single outlier ($200000$) inflates the mean, making it a poor representation of the typical score.

  • Median: The exact middle score that splits the distribution in half ($50\%$ above, $50\%$ below).

  • Mode: The most frequently occurring score in a dataset.

• Measures of Variability (Score Spread):

  • Range: The distance between the highest and lowest scores.

  • Standard Deviation: Index showing how much individual scores spread out from the mean. High standard deviation means scores are widely scattered; low standard deviation means scores cluster tightly near the mean.

• Normal Distribution (Bell Curve):

  • A symmetrical bell-shaped curve where $68.26\%$ of all scores fall within one standard deviation above or below the mean ($\pm 1\,\sigma$).

• Correlation:

  • A statistic measuring the strength and direction of the linear relationship between two variables ($X$ and $Y$).

  • Correlation Coefficient ($r$): Ranges from $-1.0$ to $+1.0$.

  • Positive Correlation: Both variables move in the same direction (e.g., higher class attendance correlates with higher exam grades).

  • Negative Correlation: Variables move in opposite directions (e.g., higher alcohol consumption correlates with lower exam grades).

  • Zero Correlation: No relationship exists between variables.

  • Strength/Magnitude:

  • Low: $0.0$ to $0.3$ (or $-0.1$ to $-0.3$).

  • Moderate: $0.4$ to $0.6$ (e.g., SAT/ACT scores correlating $0.4$ to $0.6$ with college GPA).

  • High: $0.7$ to $1.0$ (or $-0.7$ to $-1.0$).

  • Correlation vs. Causation (The Third-Variable Problem):

  • Correlation proves prediction, NOT cause-and-effect. A third hidden variable often explains the link:

    • Foot Size and Vocabulary: Linked by the third variable of child development/age.

    • Ice Cream Sales and Drownings: Linked by the third variable of warm summer weather/heat.

    • Household Appliances and Academic Success: Linked by family socioeconomic status.

    • Smoking and Depression: Linked by the personality trait of neuroticism.

• Inferential Statistics:

  • Math used to determine if sample findings represent real population differences or are just random chance.

  • Statistical Significance: Results are statistically significant when the probability of them happening by random chance is extremely low.

  • Alpha Thresholds: Standard cutoffs set at $p < 0.05$ (less than $5\%$ chance likelihood) or $p < 0.01$ ($1\%$ cutoff).

6. Evaluating Research Quality, Biases, and Pitfalls

• Meta-Analysis:

  • A statistical procedure combining data from many independent studies on the same topic to calculate an overall effect size.

• Sampling Bias:

  • Distortion that occurs when a sample does not accurately represent the broader population.

  • WEIRD Samples: Research heavily over-relies on participants from Western, Educated, Industrialized, Rich, Democratic nations, limiting global generalizability.

• Placebo Effect:

  • Psychological or physical improvements caused purely by participant expectations when given an inactive treatment (like a sugar pill).

• Distortions in Self-Report Data:

  • Social Desirability Bias: Participants giving fake or acceptable answers to look good on sensitive or taboo topics (e.g., underreporting drug use).

  • Response Sets: Unreflective answering patterns where a participant repeatedly picks the exact same option (e.g., circling "3" for every question).

• Experimenter Bias and Blinding:

  • Experimenter Bias: Unintentional researcher behavior that skews results toward their hypothesis.

  • Clever Hans Case Study: A horse believed to solve math by tapping his hoof. Researchers discovered Hans was actually reading subtle nonverbal facial and body tension cues from the questioner as he approached the correct answer. When Hans was blinded from seeing the questioner, his math ability vanished.

  • Single-Blind Design: Participants do not know if they are in the experimental or control group (prevents participant expectations/placebo effects).

  • Double-Blind Design: Neither the participants nor the active experimenters know group assignments (eliminates both experimenter bias and placebo effects).

7. Ethics in Psychological Research

• Historical Governance:

  • Standard ethics guidelines were created after World War II due to unethical Nazi medical experiments and historical research abuses.

  • Tuskegee Syphilis Study: Tracked poor African American men with syphilis for decades without informed consent, injected healthy individuals, and deliberately withheld penicillin treatment after it was proven to cure the disease.

  • Milgram Obedience Study: Deceived participants into believing they were giving dangerous electric shocks to another person.

• Regulatory Oversight:

  • All institutional research must be reviewed and approved beforehand by a Research Ethics Board (REB) or Institutional Review Board (IRB).

• Core Ethical Principles for Human Research:

  1. Informed Consent: Voluntary, written agreement signed after understanding study details (parents/guardians must consent for minors).

  2. Risk vs. Benefit: Scientific benefits must far outweigh any potential risks to participants.

  3. Protection from Harm: Mandates protecting participants from physical and psychological distress (e.g., an REB blocked a pregnancy anxiety questionnaire because asking graphic questions about stillbirths caused severe emotional distress).

  4. Justification of Deception: Deception is only allowed when essential to the study, fully justified to the REB, and non-harmful.

  5. Debriefing: Providing full explanations to participants after the study ends, explaining procedures, results, and data anonymity.

• Ethics in Animal Research:

  • Requires that significant scientific breakthroughs (e.g., discovering insulin to treat diabetes) outweigh animal suffering, while strictly banning non-essential harm (e.g., blinding animals to test cosmetics or mascara).