Chapter 2: Psychology's Scientific Method
Scientific Method
- Science is a method, not just what you study; it's how you study it.
- Scientists propose theories to explain the world. A theory is a system of ideas that attempts to explain observations and make predictions about future observations.
- Scientific theories are tested and either rejected, supported, or refined using the Scientific Method.
Five-Step Scientific Method (as presented)
- Step 1: Observe
- Observe some phenomenon
- Curiosity & critical thinking
- Formulate or challenge a theory
- Step 2: Hypothesize
- Formulate hypotheses and predictions
- A testable prediction derived from theory
- Step 3: Test
- Test through empirical research
- Use operational definitions of variables
- Analyze data using statistical procedures
- Step 4: Conclusions
- Draw conclusions
- Replication of results
- Reliability
- Step 5: Evaluate the Theory
- Evaluate the theory
- Change the theory if needed
- Peer review and publication
- Meta-analysis
Descriptive, Correlational, and Experimental Research
Descriptive Research
- Goal: Describing a phenomenon
- Methods: Observation, surveys/interviews, case studies
- Limitation: Does not answer why things are the way they are; can identify issues nonetheless
Correlational Research
- Goal: Identify relationships between variables
- Key statistic: correlation coefficient r with -1 \,\le\, r \,\le\, 1
- Characteristics:
- Magnitude indicates strength of the relationship
- Sign indicates direction (positive or negative)
- Common values (approximate):
- 1.00: Perfect
- .75: Very Strong
- .50: Moderate
- .25: Weak
- 0: None
- Scatter plots illustrate relationships (positive vs negative correlations)
- Examples mentioned: longer lecture associated with more yawns; longer lecture associated with lower attentiveness; various combinations of lecture length and attention
Correlation and Causation
- Important principle: Correlation does not equal causation.
- Possible explanations for observed correlations:
- Direct causation (A causes B)
- Reverse causation (B causes A)
- Third-variable problem (a third factor causes both A and B)
- Examples from the material:
- Parental harshness and child rebellion: multiple plausible explanations; correlation does not settle why behavior occurs
- Happy mood and sociability: multiple plausible explanations; correlation does not reveal causation
- Implication: Be cautious about causal inferences from correlational data; seek converging evidence and consider potential confounds
- Formula example (conceptual):
- r = \frac{\sum{i=1}^{N} (xi - \bar{x})(yi - \bar{y})}{\sqrt{\sum{i=1}^{N} (xi - \bar{x})^2}\; \sqrt{\sum{i=1}^{N} (y_i - \bar{y})^2}}
- r ranges from -1 to +1
Experimental Research
Purpose and Key Features
- Goal: Determine causation
- Design features:
- Random assignment to groups
- Experimental group (receives manipulation)
- Control group (no manipulation or standard treatment)
- Independent Variable (IV): the manipulated variable
- Dependent Variable (DV): the measured outcome
- How it tests causation: differences between groups on the DV attributable to manipulation of the IV
Experimental Procedure Details
- Random assignment to groups helps ensure equivalence at start
- Experimental Group: where the hypothesized cause is manipulated
- Control Group: treated equally except for the manipulation
- Observed/Measured Effect: difference between groups on the DV
- Key variables:
- Independent Variable (IV)
- Dependent Variable (DV)
Application Question Practice (Sample from transcript)
- Scenario: A counseling psychologist tests a self-help book's PTSD-reducing exercises vs ordinary journaling
- 3a) Independent variable: type of journaling (refunctional writing vs ordinary entries)
- 3b) Dependent variable: level of PTSD symptoms
- 3c) Control group: participants writing ordinary journal entries
- 3d) Experimental group: participants writing refunctional entries
Sampling, Populations, and Settings
Important Concepts in Sampling
- Population: Entire group about whom conclusions are to be drawn
- Sample: Portion of the population actually observed
- Representative Sample: characteristics similar to population
- Random Sample: Each individual in the population has an equal chance of being selected
- Biased vs representative samples: Bias compromises external validity
Validity
- External Validity: Do results generalize to the real world? (Sample size and representativeness matter)
- Internal Validity: Are changes in the DV due to manipulation of the IV? Are there biases, confounds?
Bias and Expectations
- Experimenter Bias: Researchers' expectations influence outcomes
- Research Participant Bias:
- Demand Characteristics
- Socially desirable responding
- Placebo Effect
- Solution: Double-blind experiment (neither participants nor experimenters know group assignments)
Research Settings
- Laboratory Setting (Artificial): controlled environment; advantages include control over confounds; disadvantages include low external validity
- Natural Setting (Real World): naturalistic observation; advantages include real-world relevance; disadvantages include less control over variables
Analyzing and Interpreting Data
Descriptive vs Inferential Statistics
- Descriptive statistics: summarize data (mean, median, mode; measures of dispersion like range and standard deviation)
- Inferential statistics: draw conclusions about the population from samples; bridge between sample and population; assess whether data confirm the hypothesis
- Core concepts:
- Mean: \bar{x} = \frac{1}{N}\sum{i=1}^{N} xi
- Median: middle value
- Mode: most frequent value
- Range: max − min
- Standard Deviation (sample): s = \sqrt{\frac{1}{N-1}\sum{i=1}^{N} (xi - \bar{x})^2}
- Significance testing concept:
- Alpha level: \alpha = 0.05 (common threshold)
- Statistical significance: when p < 0.05, the observed pattern is unlikely due to chance
Inference and Evidence
- Inferential step asks whether data confirm the hypothesis beyond chance
- Acknowledges that statistical significance does not prove a theory; it supports or challenges it
Ethics and Responsible Research
Research Ethics (APA Guidelines)
- Informed consent: participants understand the study and agree to participate
- Confidentiality: protect participants' information
- Debriefing: explain the study afterward, especially if deception was used
- Deception: allowed only if justified and followed by debriefing
- Institutional Review Board (IRB): oversees ethical compliance and risk assessment
Animal Research
- Benefits to humans; used by about 5% of researchers
- Species used: rats and mice account for ~90% of animal research
- Standards of care include housing, feeding, and ensuring psychological and physical well-being
Critical Consumer: Evaluating Psychological Research
- A Wise Consumer: skeptical yet open-minded
- Cautions:
- Avoid overgeneralizing results to individuals
- Be cautious in applying group trends to individual experience
- Question causal inferences
- Look for converging evidence
- Consider the source
Chapter 2 Objectives (Summary of what you should be able to do)
- Explain the scientific method
- Describe the three types of research used in psychology and the settings: Descriptive, Correlational, Experimental; and the conclusions that can be drawn from each
- Explain research samples and settings
- Distinguish between descriptive statistics and inferential statistics
- Explain the need to think critically about psychological research
- Correlation coefficient: r = \frac{\sum (xi - \bar{x})(yi - \bar{y})}{\sqrt{\sum (xi - \bar{x})^2}\; \sqrt{\sum (yi - \bar{y})^2}}
- Mean: \bar{x} = \frac{1}{N}\sum{i=1}^{N} xi
- Standard deviation (sample): s = \sqrt{\frac{1}{N-1}\sum{i=1}^{N} (xi - \bar{x})^2}
- Population parameter concepts cited: population vs sample; representative and random sampling concepts
- Significance level: \alpha = 0.05 ; significance criterion: p < 0.05
Additional Notes
- The lecture-style figures and examples (e.g., longer lectures vs yawns, mood vs sociability) illustrate correlational reasoning and the care needed before claiming causation.
- The material covers ethical safeguards (informed consent, confidentiality, debriefing, deception, IRB) and practical considerations (external vs internal validity) that affect how studies are designed and interpreted.