Chapter 1-7: Introduction to Research Methods and Bystander Effect
Chapter 1: Introduction
Psychology as a science means we use the scientific method to study behavior and mental processes.
Central idea: aim to gather, analyze, and interpret information in a way that reduces error and allows dependable generalization.
Practical illustration: a real-life scenario to motivate empirical questions about helping behavior under threat.
You’re followed late at night; you face a fork: one crowded street vs. a sparsely populated street.
Question: Are you more likely to get help on the crowded street, or on the sparsely populated street where fewer people may be around?
Class poll example: a student suggests going to the crowded street to maximize chances of being helped; the instructor reframes toward empirical testing of this idea.
Empirical question: a question we can answer through observation and experimentation as scientists.
Real-world case motivating study: Kitty Genovese case.
Dozens of neighbors reportedly heard screams for help but did not intervene.
This case spurred formal investigation into why people don’t help in the presence of others.
Viv Latney and John Darley (two social psychologists) used the scientific method to study bystander intervention.
Definitions and goals of the scientific method (direct book quote summarized):
A set of procedures used in science to gather, analyze, and interpret information in a way that reduces error and leads to dependable generalization.
Goals include reducing error (to avoid incorrect conclusions) and enabling dependable generalizations across contexts.
Research process (project lifecycle):
Topic selection and literature review: what has been studied before? In the Genovese case, initial literature on bystander behavior was limited.
Develop theory and formulate hypotheses.
Select and apply a scientific method (including ethical evaluation in modern practice).
Collect and analyze data.
Report results.
Hypothesis: a tentative statement about the relationship between two or more variables (tentative means not yet confirmed).
Example hypothesis related to bystander intervention (Latane & Darley): generally, the more people who witness an emergency, the fewer people intervene; i.e., more witnesses predict less helping.
What is a variable?
A variable is a measurable condition, event, characteristic, or behavior that is controlled or observed in a study.
Examples in Latane & Darley study:
Independent variable (IV): the number of people present (the factor thought to influence another variable).
Dependent variable (DV): the amount or likelihood of helping (the outcome that changes based on the IV).
Operationalization: turning abstract concepts into concrete, observable operations.
How is "helping" defined in the study? Could include direct action, signaling, calling for help, or other observable interventions.
Before data collection, define what counts as helping and what counts as no helping or partial helping.
Research methods overview (four broad approaches):
Case studies
Surveys
Correlational research
Experiments
Key idea: different methods have different strengths and limits; choice depends on the research question and ethical considerations.
Chapter 2: The Scientific Method
Core components of the method (as described in the chapter):
Topic selection and literature review
Theory development and hypothesis formation
Method selection (and ethical review where applicable)
Data collection and analysis
Reporting results
Hypothesis formalization (reiterated): a tentative statement about the relationship between two or more variables.
Operationalization recap: define how variables will be measured or manipulated.
Example of a counterintuitive hypothesis in the bystander literature:
The more people who witness an emergency, the fewer people intervene (the bystander effect).
Ethical evaluation reminder:
Modern studies typically undergo ethical review by an independent board to ensure participant safety and ethical treatment.
Chapter 3: Bunch Of People
Case study vs. survey vs. other methods:
Case study: an in-depth description of a single instance or person.
Pros: rich, detailed data; potential insights not accessible by other methods.
Cons: vulnerable to bias; limited generalizability to broader populations or contexts.
Example discussed: brain case of a person with severe epilepsy where removing a hippocampal region helped control seizures.
Result: the patient (often cited as HM) could not form new memories after hippocampal damage but retained older memories.
Significance: helped infer that the hippocampus plays a critical role in the formation of new memories.
Generalizability concern: findings from a single case may not apply to others or to different settings.
The hippocampus example illustrates the leverage case studies provide for understanding brain function, especially before modern imaging.
When is a case study appropriate? When a phenomenon is difficult to study at scale or when detailed, contextual understanding is needed.
Generalizability question: to what extent can we generalize from one case to others or to broader populations?
Case studies remain valuable in cognitive neuroscience and clinical psychology as exploratory, hypothesis-generating evidence.
Surveys: questions posed to many participants to gather broad data about experiences or attitudes.
Pros: higher generalizability than a single case; can reach diverse populations; efficient for large samples.
Cons: potential biases and measurement limits; data are often correlational, not causal.
Survey specifics and biases:
Self-presentation bias: participants may respond in a way that presents them more favorably rather than honestly (e.g., claiming they helped more than they did).
Wording biases: slight changes in how a question is framed can shift responses dramatically (e.g., assisted suicide wording examples).
Acquiescence bias: tendency to agree with statements or prompts; can distort results when forced-choice or strongly agree options are used.
Examples given: legal/ethical framing of suicide or abortion questions shows how wording shapes responses.
Practical caution: researchers must craft questions carefully to avoid biasing results.
Chapter 4: A Positive Correlation
Correlational data basics:
A correlation measures the direction and strength of the relationship between two variables.
Range: correlations lie between .
Positive correlation: as one variable increases, so does the other (e.g., exposure to violence on TV and aggressive behavior).
Negative correlation: as one variable increases, the other decreases (e.g., optimism and illness: higher optimism associated with less illness).
Perfect correlation: (knowing one variable predicts the other perfectly).
No or weak correlation: values near 0 indicate little or no linear relationship.
Interpreting correlations: they reveal patterns, not causation.
Correlations show that two variables are related but do not indicate that one causes the other.
Example discussion: violence in video games related to violence in real life, but causation is not established by correlation alone.
Common pitfall examples included:
The idea that violent video games cause real-life aggression is a claim that requires causal evidence beyond correlation.
Spurious correlations (e.g., unrelated variables that move together by coincidence) illustrate the danger of misinterpreting correlations.
Chapter 5: Bunch Of People (Causation vs. Correlation and Third Variables)
Central point: correlation does not imply causation.
Three possible causal scenarios with correlated variables:
Variable X causes Variable Y.
Variable Y causes Variable X.
A third variable C causes both X and Y (common cause).
Third-variable problem: a third factor (e.g., parental supervision) might influence both observed variables (e.g., time spent watching TV and aggression).
The example of violent video games and aggression illustrates the third-variable problem and directionality ambiguity.
The literature review suggestion: be cautious about drawing causal conclusions from correlational data alone.
Additional humor/illustration: cartoons and anecdotes emphasize that correlation can be misinterpreted as causation, highlighting the need for controlled experiments to establish causality.
Chapter 6: People To Groups
How to establish causality in research: experiments with controlled manipulation and random assignment.
Core idea: manipulate an independent variable and observe effects on a dependent variable in a controlled setting.
Example design concept (Latane & Darley): manipulate whether a participant is alone or with others and observe helping behavior when a crisis occurs (e.g., smoke in a room).
Random assignment importance:
Random assignment gives each participant an equal chance to be in any condition.
This helps ensure that observed effects are due to the manipulation of the IV, not preexisting differences among participants.
Confounding variables: extraneous factors that could influence the DV and threaten internal validity.
Example confounds include time of day, participant characteristics, or order effects.
The aim is to minimize confounds so that changes in the DV can be attributed to the IV.
Experimental design considerations:
A classic demonstration involved smoke filling a room; observing how long it took participants to seek help under different conditions.
The diffusion of responsibility effect arises in groups: individuals may feel less personal responsibility to act when others are present.
People look to others to gauge whether the situation is an emergency (pluralistic ignorance) and may interpret lack of visible distress in others as non-emergency.
Ethical considerations in experiments:
Informed consent: participants must be informed about the study and can withdraw at any time.
Internal Review Board (IRB): an independent committee that assesses risks and ensures participant safety.
Chapters note: studies should not unduly harm participants; some deception is sometimes used but must be justified and followed by debriefing.
Chapter 7: Conclusion
Summary of the bystander effect findings: people are more likely to help when there are fewer bystanders; in groups, helping is less likely due to diffusion of responsibility and ambiguity.
The social-context explanation: humans are social animals (pack behavior); in unfamiliar or ambiguous situations, people look to others for cues about how to respond.
Observations from demonstrations or media emphasize that even when people feel strongly about helping, situational factors (group size, ambiguity, diffusion of responsibility) can delay or inhibit helping.
Real-world implication: interventions that reduce ambiguity or assign clear responsibility (e.g., bystander training, explicit directives) can increase helping behavior.
Final note from the course: ethical conduct and careful experimental design are essential to uncover reliable patterns in social behavior and to generalize findings beyond specific incidents.