POS3330 BB4
CHAPTER 4: RESEARCH DESIGN
Outline of Research Design
The Purposes of Research
The Logic of Nomothetic Explanation
Necessary and Sufficient Causes
Units of Analysis
The Time Dimension
How to Design a Research Project
The Research Proposal
The Ethics of Research Design
Quick Quiz
The Purposes of Research
1. Exploration
Aims to satisfy the researcher’s curiosity.
Tests feasibility of conducting a more extensive study.
Develops methodologies for subsequent studies.
Examples needed for clarification.
2. Description
Describes situations and events through scientific observation.
not necessarily looking for an explanation, observing to observe is useful too
Examples needed for clarification.
3. (Re-) Categorization
4. Explanation
Descriptive Studies: Address what, where, when, how.
Explanatory Studies: Address why.
Examples needed for clarification.
Prediction and Prescription
Involves making careful short-term forecasts for policy-making.
The Logic of Nomothetic Explanation
The goal is to identify a few factors that account for variations in a phenomenon.
Example: Legalization of Marijuana
Idiographic Approach: Focus on individual experiences (parents, teachers, clergy).
Nomothetic Approach: General factors influencing attitudes (political orientation, etc.).
Criteria for Nomothetic Causality
Correlation: An empirical relationship between two variables where changes in one relate to changes in the other.
Temporal Order: The cause occurs before the effect.
Non-Spuriousness: Eliminates false correlations caused by a third variable.
Logical Explanation: A non-standard, yet reasoned causal relationship.
Example of Spurious Correlation
Observed Correlation: Bigger shoe size may seem linked to greater math skills.
Actual Causation: Both shoe size and math skills increase with age.
Nomothetic Causal Analysis and Hypothesis Testing
Hypotheses are not mandatory in nomothetic research.
To test a hypothesis:
Specify related variables.
Measure variables.
Hypothesize correlation and statistical significance.
Specify tests for spuriousness.
False Criteria for Nomothetic Causality
Complete Causation can lead to misinterpretations.
Exceptional Cases may not represent majority trends.
Necessary and Sufficient Causes
Necessary Cause: Must be present for an effect to occur.
Sufficient Cause: Guarantees the effect when present.
Most research outcomes benefit from identifying both types of causes.
Units of Analysis
Aggregates vs. Individuals
Individuals: Common unit in social research.
Groups, Organizations, Social Interactions:
Individuals: Students, voters, etc.
Groups: Family units, friendship groups.
Organizations: Corporations, colleges.
Social Interactions: Online chats, phone calls.
Social Artifacts
Defined as products of social beings or their behavior.
Faulty Reasoning about Units of Analysis
Ecological Fallacy: Incorrectly deducing individual characteristics from group data.
Reductionism: Restricting relevant concepts to genetic factors only (as in sociobiology).
The Time Dimension
Cross-Sectional Studies
Observations represent a single point in time.
Longitudinal Studies
Collect data at various points in time:
Trend Study: Monitors a characteristic over time.
Cohort Study: Studies a specific subpopulation over time.
Panel Study: Repeatedly collects data from the same people over time.
Examples of Research Strategies
Choose between Exploration, Description, or Explanation.
Identify data sources.
Define your unit of analysis.
Consider relevant time dimensions.
How to Design a Research Project
Define project purpose (exploratory, descriptive, explanatory).
Specify meanings of concepts.
Choose a research method.
Measure your results.
Determine whom or what to study.
Collect empirical data.
Process the data.
Analyze the data.
Report findings.
The Research Proposal
Key Elements
Problem or Objective
Literature Review
Subjects for Study
Measurement
Data Collection Methods
Analysis
Schedule
Budget
Institutional Review Board