| Univariate strategy | A data analysis strategy in which multiple dependent measures are analyzed independently with separate statistical tests.
|
 |
|
|
| Multivariate strategy | A data analysis strategy in which multiple dependent measures are analyzed with a single, multivariate statistical test.
|
|
|
|
| Multivariate design | A research design in which multiple dependent or predictor variables are included.
|
|
|
|
| Multiple regression | Multivariate linear regression analysis used when you have a single criterion variable and multiple predictor variables.
|
|
|
|
| Discriminant analysis | Multivariate statistical technique used when you have multiple predictor variables and a categorical criterion variable.
|
|
|
|
| Canonical correlation | Multivariate statistical techniques used to correlate two sets of variables.
|
|
|
|
| Factor analysis | Multivariate statistical technique that uses correlations between variables to determine the underlying dimensions (factors) represented by the variables.
|
|
|
|
| Multivariate analysis of variance (MANOVA) | Multivariate analog to the analysis of variance used to analyze data from an experimental design with multiple dependent variables.
|
|
|
|
| Partial correlation | Multivariate correlational statistic used to examine the relationship between two variables with the effect of a third variable removed from both of them.
|
|
|
|
| Part correlation | Multivariate correlational statistic used to examine the relationship between two variables with the effects of a third variable removed from only one of them.
|
|
|
|
| Multiple R | The correlation between the best linear combination of predictor variables entered into a multiple regression analysis and the dependent variable.
|
|
|
|
| R-square | The square of the multiple R in a multiple regression analysis. Provides a measure of the amount of variability in the dependent measure accounted for by the best linear combination of predictor variables.
|
|
|
|
| Beta weight | Standardized regression weight used to interpret the results of a linear regression analysis. A beta weight can be interpreted as a partial correlation coefficient.
|
|
|
|
| Multiway frequency analysis | A class of alternatives to ANOVA, MANOVA, or regression analysis for use when you want to measure or manipulate categorical variables.
|
|
|
|
| Loglinear analysis | A nonparametric, multivariate statistical technique used primarily to evaluate data from multifactor research with a nominal dependent variable. It can also be used on interval or ratio data that violate the assumptions of the analysis of variance.
|
|
|
|
| Path analysis | An application of multiple regression used to develop and test causal models using correlational data.
|