Please enable JavaScript.
Coggle requires JavaScript to display documents.
Correlation and Regression - Coggle Diagram
Correlation and Regression
Pearson Correlation
Pearson Correlation: Measures the strength and direction of a linear relationship between two quantitative variables
Sum of Products (SP): Measures how two variables vary together using paired deviations
SP Definitional Formula: Calculates SP by multiplying deviations from each variable's mean and adding them
SP Computational Formula: Calculates SP using raw scores and totals rather than individual deviations
Outliers: Extreme scores that can greatly change a correlation
Restricted Range: Limited variation in scores that can weaken or distort a correlation
Correlation Basics
Correlation: Measures the direction and strength of a relationship between two variables
Scatter Plot: A graph of paired scores showing how two variables relate
Positive Correlation: Both variables tend to increase or decrease together
Negative Correlation: As one variable increases, the other tends to decrease
Perfect Correlation: All points fall exactly on a straight line (r = +1 or −1)
Envelope: An imaginary boundary around the points in a scatter plot that helps show the relationship's shape
Linear Relationship: A relationship that follows a straight-line pattern
Other Correlations
Coefficient of Determination: The squared correlation (r²), representing the proportion of variability shared by two variables
Correlation Matrix: A table showing correlations among multiple variables
Spearman Correlation: Measures relationships using ranked scores rather than original values
Monotonic Relationship: A relationship where one variable consistently increases or decreases as the other increases, but not necessarily in a straight line
Point-Biserial Correlation: Measures the relationship between one quantitative variable and one naturally dichotomous variable
Dichotomous (Binomial) Variable: A variable with only two categories or possible values
Regression
Y-Intercept: The predicted Y value when X equals zero
Regression: Uses one or more variables to predict another variable
Regression Line: The straight line that best predicts Y values from X values
Least-Squared-Error Solution: Finds the line that minimizes the sum of squared prediction errors
Regression Equation for Y: A formula that predicts Y from X using the slope and intercept
Standardized Form of the Regression Equation: Predicts Y in z-scores using the correlation and X's z-score
Linear Equations
Phi-Coefficient: Measures the relationship between two dichotomous variable.
Linear Equation: Describes a straight line using a slope and intercept
Slope: The amount Y changes for every one-unit increase in X
Regression Accuracy
Standard Error of Estimate: Measures the typical distance between actual Y scores and predicted Y scores
Predicted Variability: The portion of Y's variability explained by the regression model
Unpredicted Variability: The portion of Y's variability not explained by the regression model
Analysis of Regression: Separates total variability into predicted and unpredicted portions to evaluate the regression model
Regression Equation: The mathematical formula used to generate predicted scores