II. Research Methods
A. Elements of Science
1. Broad Overview
a. Observation
b. Classification
c. Explanation
2. Elements of the Scientific Enterprise
a. Theory (e.g., Hullian Learning Theory)
b. Principles
c. Hypotheses/Predictions
d. Models
e. Experimental Methodologies
f. Data
3. Approaches
a. The Empirical Method
b. The Hypothetico-Deductive Method
c. Why Both Are Needed
4. Theory-Testing & Revision
a. Confirmability & Falsifiability
b. Competitive Hypothesis Testing
c. Follow-Up to Negative Results
i. Faulty Procedures/Artifacts?
ii. Adequacy of the Model?
iii. Theory Revision versus Theory Disconfirmation?
d. Peer Review: Science as a Social Activity
i. Review at Conferences & Journals
ii. Two (or 2000) Heads are better than One
iii. Alternative Theories, New Findings, New Perspectives
B. Data Collection Methods
1. Naturalistic Observation
a. When & Why (Ecological Validity Claims)
b. Give & Take With Experimental Method
c. Potential Pitfalls
i. Observer Effect (e.g., The Real World)
ii. Observer Bias & Anthropomorphic Fallacy (e.g., gaze)
iii. Retrospective versus Contemporary Reports
2. Correlational Method
a. When & Why
b. The Correlation Coefficient: -1 .. r .. +1
c. Potential Pitfalls
i. Attributing Causality
ii. Attributing a Direct Relationship
iii. Missing Non-Linear
Trends
3. Experimental Method
a. When & Why (Searching for Causes & Effects)
b. Types of Variables
c. Within-S & Between-S Experiments
d. Potential Pitfalls
i. Hidden Variables
ii. Misinterpreting the Null Result
iii. Inappropriate Statistics & Biases (see below)
4. Clinical Method & Case Studies
a. When & Why
b. ABA Design
c. Potential Pitfalls
i. Retrospective Memory
ii. Multiple Potential Causes
iii. Lack of Clear Control
5. Survey Method
a. When & Why
b. Potential Pitfalls
i. Non-representative samples
ii. Demand characteristics
iii. Subject impression management
C. Statistical Analyses
1. Descriptive Statistics (means; frequencies)
2. Inferential Statistics (t-tests; anovas; etc.)
a. Parametric Statistics (sample t-test for correlation)
b. Non-Parametric Stats (sample sign test)
c. Meta-Analyses
D. Some Specific Threats/Artifacts
1. Experimenter Bias
2. Subject Bias