Non experimental research (1)
Non-Experimental Research Overview
Focus on Non-Experimental Research: The document discusses various aspects of non-experimental research, emphasizing its significance in scenarios where experimental design is not feasible.
When Non-Experimental Research is Preferred
Single Variable Focus: Research questions relating to a single variable rather than relationships between two (e.g., accuracy of first impressions).
Non-Causal Statistical Relationships: Examining correlations between variables without assuming causation (e.g., verbal vs. mathematical intelligence correlation).
Unmanipulable Independent Variables: Situations where independent variables cannot be manipulated due to practical or ethical constraints (e.g., effects of hippocampus damage on memory).
Broad Exploratory Questions: Research about specific experiences (e.g., experiences of working mothers with depression).
Types of Non-Experimental Research
Observational Studies:
Focus on qualitative research and ethnography studies.
Correlational Research:
Types include survey research, archival data, and naturalistic observation.
Quasi-Experiments:
Include case-control studies, cohort studies, and natural experiments.
Correlational Research Explained
Definition: Investigates relationships between variables without manipulation by the researcher.
Correlation Strength: Indicates both the direction and strength of relationships between variables.
Understanding Correlation Strength (Cohen, 1988)
Correlation Size Indicators:
Negative Correlation:
Small: -0.10 to -0.29
Medium: -0.30 to -0.49
Large: -0.50 to -1.00
Positive Correlation:
Small: 0.10 to 0.29
Medium: 0.30 to 0.49
Large: 0.50 to 1.00
Practical Example of Correlation
Example: Exploring the correlation between height and basketball dunking ability illustrates application in sports research.
Types of Correlational Research
Naturalistic Observation: Study behavior in natural environments without interference.
Survey Method: Collect data through questionnaires.
Archival Research: Analyze existing data or documents.
Archival Research Insights
Case Study: In Ancient Rome, a positive correlation noted between urban treasury coins and femur lengths of adult men.
Survey Research Insights
Vaccine Hesitancy Case: High correlation between vaccine hesitancy and political affiliation, identifying trends across various states.
Economic Freedom and Living Standards
Context: High economic freedom generally correlates with higher GDP per capita, showcasing the interrelations between economy and governance.
Longitudinal Research and IQ
The Flynn Effect: Documenting increases in mean IQ scores across diverse global regions over time.
Mathematically Precocious Youth Study Cohorts
Cohorts: Different cohorts identified based on SAT scores showcasing the talent pool in mathematically gifted students.
Limitations of Case-Control Studies
Bias Types: Includes selection bias, survival bias, and recall bias which can impact the study's validity.
Control Group Matching: Difficulty in ensuring both control and experimental groups are comparable.
Prospective Cohort Study Limitations
Challenges: Include difficulties in studying uncommon diseases, diseases with long latency periods, and multiple exposures.
Natural Experiments and Endogeneity
Endogeneity Problem: This occurs when predictor variables correlate with dependent variables due to unmeasured factors, potentially leading to biased results.
Omitted Variable Bias: Leaving out relevant variables which can misattribute effects in research.
Differences-in-Differences Analysis
Observing Trends: Evaluating changes in outcome trends for groups over time to determine intervention effects.