Overview of today's topic covering:
Research questions
Levels of measurement
Types of research approaches
Differentiate between types:
Descriptive: Describes population/sample
Correlational: Examines relationships between variables
Causal: Investigates cause-effect relationships
Types:
Nominal: Categorical data (e.g., animal preference)
Ordinal: Order data (e.g., marathon rankings)
Interval: Equal differences between values (e.g., Likert scales)
Ratio: Contains a true zero value
Experiments
Often yield causal claims
Qualitative Research
Explores themes and patterns in data
Quantitative Research
Relies on numerical data and statistical analysis
Includes exploratory designs (correlational, longitudinal, etc.)
Mean: Average score (sum of scores / number of scores)
Median: Middle value in ordered data
Mode: Most frequently occurring score
Variance: Measures score spread
Standard Deviation: Average distance of scores from the mean
Parametric: Assumes underlying statistical distributions
Non-parametric: Does not assume distribution; useful for categorical data
Examples include t-tests and Chi-squared tests
Explores subjective experiences; more descriptive and thematic
Based on numerical data, statistical models, and hypotheses testing
Correlational Studies: Measures relationships without manipulation
Cross-Sectional Studies: Observations at a single point in time
Longitudinal Studies: Observations over multiple time points
Cohort Studies: Focus on groups sharing a characteristic
Observations in natural contexts without manipulation from researchers
Involves deliberate manipulation to observe effects
Randomized control trials are key type of experimental design
Objective observation of phenomena
Control of variables, isolation of the independent variable (IV)
Systematic measurement of outcomes (dependent variable, DV)
Relationship Condition: Variables must be associated
Temporal Order Condition: Changes must occur in the correct order
No Alternative Explanation Condition: Rule out other variable explanations
Nisbett and DeCamp Wilson's (1977) study on the halo effect:
Variables manipulated and measured provide strong basis for causal claims
Discuss weak designs prone to internal validity issues
Importance of random assignment to strengthen causal claims.
Strengths:
Experimental designs support strong causal inferences
Weaknesses:
Ethical or practical limitations in manipulating certain variables
Understanding these concepts aids in evaluating psychological research and its effectiveness.