Psychological Research Chapter 2

Why Research?

• Research is used to validate claims with evidence rather than rely on intuition or untested beliefs.
• Historical examples show that beliefs can be wrong (e.g., earth was thought to be flat; mental illness attributed to possession).
• Scientific research is empirical: grounded in objective, tangible evidence that can be observed repeatedly by different observers.
• Psychology is a science; research provides verification and support for findings, not just exploration for exploration’s sake.
• Research is needed to move from groundless assumptions to proven ideas through study and testing.

The Research Process

• The Process (as outlined):
  • Identify the Research Problem
  • Review Existing Literature
  • Formulate a Hypothesis or Research Question
  • Choose a Research Design
  • Select Participants and Sampling Method
  • Data Collection
  • Data Analysis
  • Interpret Findings
  • Draw Conclusions
• Before beginning: many courses emphasize preparing and planning steps; in-class activities (e.g., inductive/deductive reasoning practice) support understanding of the process.
• Note: Some slide numbering is non-sequential, but the essential steps are listed above.

Approaches to Research

• CASE STUDY (Clinical or Case Studies)
  • Focus on one individual or a small group in an extreme or unique circumstance.
  • Pros: rich, detailed insight; cons: limited generalizability to the larger population.
• Naturalistic Observation
  • Observation of behavior in a natural setting.
  • Naturalistic behavior tends to be more genuine when not observed directly; avoids performance biases.
  • Observer bias: observations may be skewed by observer expectations.
  • Remedy: establish clear observation criteria to reduce bias.
  • Notable example: Jane Goodall’s work with chimpanzees (naturalistic observation).
• Surveys
  • Use questions delivered in writing, electronically, or verbally to gather data from a large sample.
  • Flexible administration methods (e.g., paper, online, in-person).
• Archival Research
  • Examine existing records (hardcopy or electronic) to answer research questions.
• Longitudinal and Cross-Sectional Research
  • Cross-Sectional: compares different segments of a population at a single time (e.g., age groups).
  • Longitudinal: follows the same group over an extended period; attrition (dropout) is a common issue; researchers often recruit many participants initially to account for this.

Population, Sample, and Inferential Statistics

• Population vs Sample
  • Population: the entire group of interest to the researcher.
  • Sample: a subset of the population used to represent it.
• Inferential Statistics
  • Using sample statistics to draw conclusions about population parameters.
  • Process: data from a sample infer about the broader population.
• Sampling and Representation
  • Random Sample: every member of the population has an equal chance of being selected.
  • Random sampling helps ensure representativeness across variables such as sex, ethnicity, and socio-economic status.

The Scientific Method and Reasoning

• Goals of Scientific Study (Describe, Identify, Classify, Explain, Propose reasons, Predict, Hypothesize, Influence/Use)
• Inductive vs Deductive Reasoning
  • Inductive reasoning: start with observations and generalize to broader theories.
  • Deductive reasoning: start with a theory/premise and derive specific predictions.
  • The scientific process often moves: Inductive reasoning → Theory → Hypotheses (deduction from theory) → Empirical testing → Refined theory.
• Inductive vs Deductive Reasoning (Quick reminders)
  • Inductive: observations lead to general conclusions; e.g., many observed fruits grow on trees → conclude all fruit grows on trees (illustrative, simplified).
  • Deductive: general rule applied to a specific case; e.g., All living things require energy; a duck is a living thing; therefore a duck requires energy.
• The Process of Scientific Research: Inductive vs Deductive Reasoning
  • Inductive reasoning forms theories; deductive reasoning tests hypotheses derived from theories.
  • Conclusions can lead to new theories or broader generalizations.
• The Scientific Method overview (cycle)
  • Observation → Theory/Idea generation → Hypothesis → Design a study → Data collection → Data analysis → Interpretation → Theory refinement.
  • The method is not necessarily linear; it often loops as new evidence emerges.

The Research Design and Measurement

• Step 1: Identify the Research Problem
• Step 2: Review Existing Literature
• Step 3: Formulate a Hypothesis or Research Question
• Step 4: Choose a Research Design
• Step 5: Select Participants and Sampling Method
• Step 6: Data Collection
• Step 7: Data Analysis
• Step 8: Interpret Findings
• Step 9: Draw Conclusions

Defining Variables and Measurement

• Operational Definition
  • Description of the actions/operations used to measure the dependent variable and manipulate the independent variable.
• Independent Variable (IV)
  • The variable that is manipulated or controlled by the experimenter; ideally the only important difference between experimental and control groups.
• Dependent Variable (DV)
  • The variable measured to assess the effect of the IV.
• Examples (from slides):
  • Meditation on stress reduction; Sleep on memory performance; Social media use on well-being; Positive affirmations on self-esteem; Classroom seating on participation.

The Experimental Method and Biases

• Experimental Design Concepts
  • Experimental Group: participants exposed to the manipulated variable.
  • Control Group: participants not exposed to the manipulated variable.
  • Random Assignment: equal chance of being assigned to either group; helps prevent preexisting differences.
  • Cause-and-effect: achieved when random assignment, manipulation of IV, and control of extraneous variables are in place.
• Bias and Placebo Effects
  • Experimenter bias: researcher's expectations can skew results.
  • Participant bias: participant expectations can skew results.
  • Single-blind: participants unaware of group assignment; experimenter may know.
  • Double-blind: neither participants nor researchers know group assignments.
  • Placebo effect: participants' expectations influence outcomes; to test effects, use a placebo analog.
• Ethics in Experimentation
  • Some questions cannot be tested with experiments due to ethical concerns (e.g., abuse exposure).
  • In such cases, other methods like case studies or surveys may be used.

Interpreting Findings and Reporting

• Analyzing Findings
  • Statistical analysis determines how likely observed differences are due to chance.
  • Significance: commonly, results are deemed significant if p-value ≤ 0.05, i.e., p \,\le\, 0.05.
• Reporting Findings
  • Results are often published in scientific journals, usually peer-reviewed.
  • Peer-reviewed journal article: reviewed by other scientists to assess quality and replicability.
  • Replication checks reliability of results and can expand or challenge original findings.
• Bad Science and Retraction: Vaccine-Autism example
  • Early studies claimed vaccines caused autism; later larger studies found no link and some original studies were retracted due to financial conflicts of interest.
  • Public health consequences included decreased vaccination rates and measles outbreaks.
  • Important example of why replication and scrutiny are essential in science.

Reliability, Validity, and Measurement Quality

• Reliability
  • Consistency and reproducibility of results.
  • Inter-rater reliability: agreement among observers when recording events.
  • A measure can be reliable without being valid.
• Validity
  • Accuracy of a result in measuring what it is intended to measure.
  • A valid measure is always reliable, but a reliable measure is not necessarily valid.

Correlation, Causation, and Illusory Correlations

• Correlational Research
  • Examines relationships between two or more variables.
  • Correlation coefficient: r\in[-1,1]; indicates strength and direction of the relationship.
  • Positive correlation: variables move in same direction.
  • Negative correlation: variables move in opposite directions.
  • Correlation does not imply causation: only experimental manipulation can establish causality.
  • Confounding variable: an outside factor that influences both studied variables, potentially creating a spurious association.
• Illusory Correlations and Confirmation Bias
  • Illusory correlations: perceiving a relationship where none exists.
  • Confirmation bias: tendency to favor evidence that supports preconceptions and ignore contradictory data.
  • These biases can contribute to prejudicial attitudes and discriminatory behavior.
• Case Study Activity Example (Ice Cream vs Shark Attacks)
  • Demonstrates that a strong correlation can be observed even when no causation exists; potential confounding variable (e.g., warm weather increases both ice cream sales and beach attendance, which could raise shark incidents).
  • Question prompts encourage identifying how to interpret findings and what confounds might be involved.

Ethics: Human and Animal Research

• Institutional Review Board (IRB)
  • A committee that reviews research proposals involving human participants; exists at institutions receiving federal support for human-subjects research.
  • IRB approval is generally required before proceeding.
• Informed Consent
  • Process of informing participants about the study, risks, implications, voluntariness, and confidentiality; participants provide consent.
• Deception and Debriefing
  • Deception is sometimes used to prevent bias, but requires debriefing afterward to inform participants about the true nature of the study.
  • Example: Tuskegee Syphilis Study (1932): unethical withholding of diagnosis and treatment; highlights evolution of ethical guidelines.
• Animal Research Ethics
  • IACUC: Institutional Animal Care and Use Committee reviews proposals for non-human animal research.
  • Most psychology research with animals uses rodents or birds; animals are used when experiments would be unethical in humans.
  • Core aim is to minimize pain or distress.

Additional Concepts and Terms

• The placebo effect
  • The expectation of improvement can cause real changes in experience; testing with a placebo helps isolate true treatment effects.
• Operationalization and measurement design
  • Clear operational definitions ensure that variables are observable and measurable in a consistent way across observers and contexts.
• Reliability vs Validity visual example
  • A chart shows Target A/B/C illustrating how a measurement can be reliable (consistent) but vary in validity (accuracy toward the target).

Quick Reference: Key Formulas and Thresholds

• Statistical significance threshold commonly used: p \le 0.05
• Correlation coefficient range: r \in [-1, 1]
• Conceptual distinction: correlation does not imply causation; causality established via experimental design with random assignment and controlled manipulation

Connections to Foundational Principles and Real-World Relevance

• Emphasizes empirical validation to avoid intuition-based errors in everyday claims (e.g., advertising claims must be scrutinized for evidence).
• Critical thinking framework: assess expertise, potential gains, evidence justification, and peer consensus before accepting claims.
• Ethical considerations are central to research design, ensuring participants’ rights and welfare while enabling scientifically valid conclusions.

Ethical and Practical Implications Highlight

• Research aims to balance scientific rigor with ethical responsibility, especially when involving humans or animals.
• Missteps (e.g., biased results, deceptive practices without proper debriefing, or unethical experiments) can erode trust and have real-world health consequences.
• Replication and transparency are essential for building a reliable body of knowledge.

Summary of Core Concepts

• Research is essential to validate claims with empirical evidence.
• The research process is iterative and multidimensional, encompassing problem definition, literature review, design, sampling, data collection, analysis, interpretation, and reporting.
• Multiple data-collection approaches exist (case studies, naturalistic observation, surveys, archival research, longitudinal/cross-sectional).
• Inductive and deductive reasoning form the backbone of theory development and hypothesis testing.
• Experimental design (IV, DV, random assignment, control groups) is necessary to establish causality.
• Reliability and validity determine measurement quality; both are crucial for credible findings.
• Correlation does not equal causation; beware confounding variables and illusory correlations.
• Ethics govern all stages of research, with IRBs, informed consent, debriefing, and, when applicable, animal welfare considerations.
• Reporting and replication underpin scientific progress; bad science can have harmful real-world consequences.