
Index indicating the proportion of the variance in the dependent variable that can be explained by the independent variables.
Coefficient of determination

Technique used to analyze relationships between variables and make predictions about the value of variables.
Linear regression

Equation for the bestfitting straight line to characterize the relationship between independent and dependent variables.
Regression equation

Point at which the line passes through the Y axis and the value of X is zero.
Intercept constant

Rate at which a line rises across a horizontal distance; the steepness of a regression line.
Slope

Error term or unexplained variance.
Residuals

Method of estimating parameters in a model such that the squared error term is minimized.
Least squares

Standard deviation of the errors from the regression line; used to indicate the accuracy of the predictions.
Standard error of the estimates

Procedure for understanding the effects of two or more independent variables on dependent variable using least squares estimation.
Multiple regression

Index that summarizes the magnitude of the relationship between two or more independent variables and a dependent variable.
Multiple correlation coefficient

Squared multiple correlation coefficient adjusted for sample size and number of predictors to give a more accurate estimate of the relationships in the population.
Adjusted R2

Multiple regression analysis in which all predictor variables are entered into the equation simultaneously.
Simultaneous regression model

Multiple regression analysis in which the predictor variables are entered into the model in steps prespecified by the analyst.
Hierarchical regression model

Multiple regression analysis in which predictor variables are entered into the model in steps, in the order in which the increment to R is greatest.
Stepwise regression model

A method of coding categorical variables into dichotomous variables using codes such as 1 and 0 to represent the presence and absence of an attribute (e.g., 1=male, 0=female).
Dummy coding

A way of coding categorical variables in multiple variate analysis that uses 1, 0, and 1 to designate categories.
Effect coding

A method of coding categorical variables in a regression analysis to make specific planned comparisons.
Orthogonal coding

When two or more independent variables in a matrix are highly correlated with each other.
Multicollinearity

A statistical method of used to test mean group differences on a dependent variable, while controlling for one or more extraneous variables (covariates).
Analysis of covariance (ANCOVA)

A variable that is statistically controlled (held constant) in ANCOVA; typically an extraneous, confounding influence.
Covariate

Statistical procedure used to test the significance of differences between the means of two or more groups on two or more dependent variables, after controlling for one or more covariates.
Multivariate ANCOVA (MANCOVA)

Multivariate statistical procedure for examining the relationship between two or more independent variables and two or more dependent variables.
Canonical analysis

