Key Concepts for PS 215 Exam 4: Correlational Research

Bivariate Correlational Research

Study of relationships between two measured variables.

Correlation Coefficient

Indicates strength and direction of a relationship.

Cohen's Guidelines

Standards for interpreting effect size in correlations.

Scatterplot

Graphical representation of correlation between two variables.

Association Claims

Involve measured variables, not manipulated ones.

Confidence Interval

Range estimating the precision of a correlation.

Statistical Significance

Determined by effect size and sample size.

Effect Size

Magnitude of a relationship's impact, distinct from significance.

Meaningful Correlations

Statistical significance doesn't guarantee real-world relevance.

Large Effect Size

Indicates importance of an effect in real-world contexts.

Small Effect Size

Can be significant when aggregated across many instances.

Outlier

Data point significantly different from others in a dataset.

Outliers and Correlation

Can distort the correlation coefficient's accuracy.

Sample Size Impact

Outliers affect small samples more than large ones.

Curvilinear Relationship

Non-linear relationship between two variables.

Pearson Correlation Limitations

Not suitable for curvilinear relationships.

Restriction of Range

Limits variability, affecting correlation size.

Causal Relationships Criteria

Criteria include covariance, temporal precedence, and ruling out alternatives.

Correlation vs. Causation

Correlation does not imply one variable causes another.

Directionality Problem

Uncertainty about which variable influences the other.

Third Variables Problem

Unaccounted variables may affect observed relationships.

Moderator Variable

Influences strength or direction of a relationship.

External Validity

Generalizability of findings to other contexts.

Construct Validity

Evaluates how well a test measures its intended concept.

Longitudinal Design

Research design studying the same subjects over time.

Cross-Sectional Study

Analyzes data from a population at one point in time.

Cross-Sectional Correlation

Correlation observed at a single point in time.

Autocorrelation

Correlation of a variable with itself over time.

Cross-Lag Association

Relationship between variables measured at different times.

Cross-Lag Importance

Helps determine directionality in longitudinal studies.

Multiple Regression

Statistical method to control for third variables.

Controlling for Variables

Statistically adjusting for third variables' influence.

Criterion Variable

Dependent variable in regression analysis.

Predictor Variable

Independent variable used to predict outcomes.

Regression Betas

Indicate strength and direction of relationships in regression.

Confidence Interval Significance

Negative and positive values indicate non-significant beta.

Variable Differences

Third, moderating, and mediating variables have distinct roles.

Parsimonious Theory

Simplest explanation with least assumptions.

Value of Parsimony

Preferred for its simplicity and clarity in science.