Please find attached image for table Estimating Relationships – There are many statistical tools used to estimate relationships between variables. These include covariance, correlation, ANOVA, simple linear regression, and multiple regression. In the table below, enter a Yes (or Y) or No (or N) in each box to indicate the best response. If a box does not apply, enter NA. **Estimating Relationships**There are many statistical tools used to estimate relationships between variables. These include covariance, correlation, ANOVA, simple linear regression, and multiple regression. In the table below, enter a Yes (or Y) or No (or N) in each box to indicate the best response. If a box does not apply, enter NA.| | Covariance | Correlation | ANOVA | Simple Linear Regression | Multiple Regression ||--------------------------------------------|----------------------|-----------------------|--------------------------------------|----------------------------|--------------------------|| Example: Works well with a categorical dependent variable. | NA (no dependent variable) | NA (no dependent variable) | N (should be continuous dependent variable) | N (logistic regression) | N (logistic regression) || Example: Can be used to estimation direction of relation between variables | | | | | || Works well with one or more categorical variables | | | | | || Can be used to measure the magnitude of relationship between variables | | | | | || Works well with mostly continuous variables | | | | | || Estimated relationship depends on scale | | | | | || Can be used with multiple independent variables | | | | | |

Using the association between the variables, one can predict one unknown variable using the known variables. Covariance, correlation, ANOVA, simple linear regression, and multiple regression are different types of statistics that are used to estimate association.The table is obtained as follows: Covariance Correlation ANOVA Simple Linear Regression Multiple Regression Example: Works well with a categorical dependent variable. NA (no dependent variable) NA (no dependent variable) N (should be continuous dependent variable) N (logistic regression) N (logistic regression) Example: Can be used to estimation direction of relation between variables Y Y NA (analyze any statistical differences present between the average of more than two independent variables) Y (can be found from the slope of the regression equation) Y (can be found from the slope of the regression equation) Works well with one or more categorical variables N (should be two quantitative variables) N (should be quantitative variables) N (should be categorical independent variable) Y (if the categorical variables are independent and are converted into dummy variables) Y (if the categorical variables are independent and are converted into dummy variables) Can be used to measure the magnitude of relationship between variables NA (only estimation direction of association between the variables) Y Y Y Y Works well with mostly continuous variables N (works both on discrete and continuous variables) N (works both on discrete and continuous variables) Y (if the continuous variables are independent ) Y Y Estimated relationship depends on scale Y N (Independent on change of scale) N (Independent on change of scale) Y Y Can be used with multiple independent variables N (should be only two variables) Y (if correlation matrix is used) Y N (use one independent variable) Y