Unit 0 Research Methods Notes

Hindsight Bias

  • Definition: The mind updates interpretation of the past based on what we know now; phrase like “I knew it all along.”

  • Postdiction: When predicting in hindsight after seeing results, people say outcomes were obvious.

Overconfidence

  • Two types:

    • Performance: overestimate speed, effort, or skill.

    • Accuracy: overestimate the correctness of knowledge.

  • Consequences: eyewitness errors; tests.

  • Check: try explaining the concept to someone else to test understanding.

  • Quick probe: estimate how long a task takes to complete.

Coincidence / Pattern Perception

  • Perceiving order in random events (gambler’s fallacy).

  • Danger: mistaking coincidence for predictive power; test via replication and controlled experiments.

Scientific Attitude: Curiosity, Skepticism, Humility

  • Curiosity: ask new questions; pursue evidence while avoiding harm.

  • Skepticism: challenge claims; consider alternative explanations; seek disconfirming evidence.

  • Humility: value truth over being right; willing to revise beliefs.

Critical Thinking

  • Analyzing information to see if it makes sense.

  • Look for hidden assumptions, bias, and political or personal influences.

  • Evaluate evidence and consider multiple explanations.

The Scientific Method: Theory, Hypothesis, Operational Definitions, Replication

  • Theory: big-picture explanation that can generate predictions.

  • Hypothesis: testable prediction aligned with the theory.

  • Operational Definitions: concrete, observable measures (e.g., impulsivity = # of times a person calls out without raising hand).

  • Replication: repeating studies to verify findings.

  • Data types: description, correlation, causation (via experimentation).

  • If data conflict with ideas, revise and retest.

Theory and Hypotheses

  • Theory: explains and predicts phenomena (e.g., ADHD symptoms as a reaction to sugar).

  • Hypothesis: specific, testable prediction consistent with the theory.

Describing Data and Data Types

  • Descriptive research: description of behavior; cases, surveys, naturalistic observation.

  • Correlational research: examines relationships; can predict but not prove causation.

  • Experimental research: tests causation by manipulating IVs and measuring DVs; includes random assignment.

  • Confounding variables: other factors that may affect DV.

Research Methods: Descriptive, Correlational, Experimental

  • Descriptive: case studies, naturalistic observation, surveys; no control of variables.

  • Correlational: assesses relationships; does not establish cause.

  • Experimental: manipulates IV, uses random assignment; allows causal inference.

Random Sampling vs. Random Assignment

  • Random Sampling: representative subset of population; each member has an equal chance.

  • Random Assignment: participants allocated to groups by chance; controls for confounds.

Placebo Effect and Blinding

  • Placebo effect: improvement due to expectations about treatment.

  • Control group may receive a placebo; studies can be single- or double-blind (neither participants nor staff know group assignment).

Variables in Experiments

  • Independent Variable (IV): the manipulated factor.

  • Dependent Variable (DV): the measured outcome.

  • Confounding variables: other factors that could influence DV.

Ethics and Values in Research

  • Animal vs. human subjects: safety, dignity, and scientific value.

  • Human subjects: informed consent, confidentiality, debriefing, minimization of harm.

  • Values influence topic choice, interpretation, and conclusions; complete value-free research is not achievable.

Statistics in Psychology (Describing Data and Inference)

  • Central tendency: Mode, Mean, Median

    • Mode = most frequent value

    • Mean = arithmetic average

    • Median = middle value

  • Measures of variation: Range, Standard Deviation

  • Distribution shapes: Normal (bell curve) vs. Skewed

  • Statistical significance concept:

    • Common threshold: p < 0.05

    • Significance ≠ truth; sample size and design affect interpretation.

  • Correlation basics:

    • Correlation coefficient rr ranges from −1 ≤ r≤ 1-1 \, \le \ r \le \ 1; direction indicates positive or negative relationship.

    • Correlation does not imply causation.

Data to Insight: Reliability and Generalization

  • Reliability: consistency of measurements across time and observers.

  • Validity: whether a measure actually assesses what it claims to.

  • Generalizability depends on representative sampling and well-controlled methods.

The Role of Replication and Confidence in Inference

  • Replication confirms whether findings hold across samples and settings.

  • Confidence grows with converging evidence from multiple methods.

Quick Reference: Key Concepts to Remember

  • Hindsight Bias: I knew it all along after the fact.

  • Overconfidence: too sure about performance or accuracy.

  • Correlation ≠ Causation: relationships do not prove cause.

  • IV and DV: what you manipulate and what you measure.

  • Random Sampling vs. Random Assignment: represent population vs. control for confounds.

  • Placebo and Blinding: controls for expectation effects.

  • Operational Definitions: concrete measures for abstract concepts.

  • Ethics and Values: protect participants; recognize subjectivity in science.

  • Statistical Significance: practical threshold, not absolute truth; consider effect sizes and reliability.

  • Descriptive, Correlational, Experimental: three pillars of research design

−1≤r≤1-1 \le r \le 1
p < 0.05