research methods study guide2 for exam 3

Chapter 1: Introduction to Research

• Empiricism: Knowledge based on observation.

• Basic vs. Applied Research:

  • Basic = understanding fundamental principles.

  • Applied = solving real-world problems.

Chapter 2: Research Articles

• Sections of an empirical article:

  • Abstract, Introduction, Method, Results, Discussion, References.

• Research > Personal Experience: Controlled, systematic, less biased.

Chapter 3: Types of Claims

• Frequency Claim: Describes rate or level.

• Association Claim: Relationship between variables.

• Causal Claim: One variable causes another.

• Conceptual vs. Operational Definitions:

  • Conceptual = idea or theory.

  • Operational = how it’s measured/tested.

Chapter 4: Belmont Report

• Three ethical principles:

  • Respect for persons: Informed consent.

  • Beneficence: Do no harm.

  • Justice: Fairness in participant selection.

Chapter 5: Reliability & Validity

• Reliability:

  • Test-retest, Interrater, Internal.

• Validity:

  • Face, Content, Criterion, Convergent, Discriminant.

Chapter 6: Biases & Effects

• Response sets: Tendency to respond in patterned ways.

• Observer bias: Researcher's expectations skew results.

• Reactivity: Participant behavior changes when observed.

Chapter 7: Sampling

• Probability Sampling: Random, representative.

• Non-Probability Sampling: Convenience, quota.

• Examples: Cluster, Stratified, Snowball, etc.

Chapter 8: Correlational Research

• Problems with causal conclusions:

  • Directionality Problem

  • Third-variable Problem

• Statistical Validity Concerns: Sample size, effect size, outliers, restriction of range.

📘 New Material for Exam 3

Chapter 9: Multivariate Research

🧩 Types of Multivariate Correlations

• Cross-sectional: Different variables, same time.

• Autocorrelations: Same variable, different time.

• Cross-lag: Different variables, different times (→ helps determine temporal precedence).

🧠 Regression Analysis

• Regression equation: Y = a + bX

  • Y = outcome variable

  • X = predictor variable

  • a = intercept, b = slope

• Purpose: Controls for third variables.

• Interpretation:

  • Look at Beta coefficients (β), p-values, and confidence intervals.

• Validity Supported: Internal validity, if third variables are ruled out.

📌 Takeaway:

Regression helps with covariance and third variables, but random assignment in experiments is still superior for causal claims.

Chapter 10: Experiments and Design

🔁 Variables & Causal Criteria

• Independent Variable (IV): Manipulated.

• Dependent Variable (DV): Measured.

• Causal Criteria:

  • Covariance

  • Temporal Precedence

  • Internal Validity

🧪 Design Types

• Independent-groups:

  • Posttest-only, Pretest/Posttest.

• Within-groups:

  • Repeated-measures, Concurrent measures.

⚠ Threats to Internal Validity

• Design Confounds

• Selection Effects: Use random assignment.

• Order Effects (within-groups): Use counterbalancing.

🔄 Counterbalancing

• Full: All possible sequences.

• Partial: Some sequences.

• Latin Square: Ensures each condition appears in each position.

🔍 Other Concepts

• Demand Characteristics

• Systematic vs. Unsystematic Variability

• Control Variable: Holds constant to isolate IV effect.

• Cohen’s d: Effect size.

• Confidence Interval:

  • If CI includes 0 → Not significant.

  • If CI excludes 0 → Significant.

Chapter 11: More Validity Threats & Null Effects

🔐 Threats to Internal Validity (Any Design)

• Observer bias: Use masked/blind design.

• Demand characteristics: Participants guess study purpose.

• Placebo effects: Use double-blind placebo control group.

⏳ Threats (Pretest/Posttest)

• Regression to the Mean

• Attrition: Participant drop-out.

• Testing Effects: Familiarity with test.

• Instrumentation

• History

• Maturation

➡ Solutions: Comparison groups, counterbalancing, posttest-only design.

❌ Null Effects: Why No Effect Found?

• Too little between-groups variability:

  • Ceiling/floor effects

  • Poor manipulation/measurement

• Too much within-groups variability:

  • Measurement error

  • Individual differences

  • Situation noise

➡ Solutions:

• Improve reliability of measures

• Increase sample size

• Use within-groups design

💡 Statistical Power:

• Power = Probability of detecting an effect if one exists.

• Increased by minimizing within-group variability and increasing sample size.

📎 Bonus Practice & Exam Tips

• Match study designs to examples (e.g., pretest/posttest, repeated measures).

• Understand how to read regression tables.

• Know which validity is being addressed in examples.

• Be able to identify confounds in experimental scenarios.

• Recognize types of correlations in longitudinal studies.

• Memorize key definitions AND examples.

• Prepare your 3x5 notecard with formulas, definitions, and diagrams!