One of the assumptions in regression analysis is that

1. the errors have a mean of 1.

2. the errors have a mean of 0.

3. the observations (Y) have a mean of 1.

4. the observations (Y) have a mean of 0.

A graph of the sample points that will be used to develop a regression line is called

1. a sample graph.

2. a regression diagram.

3. a scatter diagram.

4. a regression plot.

When using regression, an error is also called

1. an intercept.

2. a prediction.

3. a coefficient.

4. a residual.

In a regression model, Y is called

1. the independent variable.

2. the dependent variable.

3. the regression variable.

4. the predictor variable.

A quantity that provides a measure of how far each sample point is from the regression line is

1. the SSR.

2. the SSE.

3. the SST.

4. the MSR.

The percentage of the variation in the dependent variable that is explained by a regression equation is measured by

1. the coefficient of correlation.

2. the MSE.

3. the coefficient of determination.

4. the slope.

In a regression model, if every sample point is on the regression line (all errors are 0), then

1. the correlation coefficient would be 0.

2. the correlation coefficient would be $-1$ or 1.

3. the coefficient of determination would be $-1.$

4. the coefficient of determination would be 0.

When using dummy variables in a regression equation to model a qualitative or categorical variable, the number of dummy variables should equal

1. the number of categories.

2. 1 more than the number of categories.

3. 1 less than the number of categories.

4. the number of other independent variables in the model.

A multiple regression model differs from a simple linear regression model because the multiple regression model has more than one

1. independent variable.

2. dependent variable.

3. intercept.

4. error.

The overall significance of a regression model is tested using an F test. The model is significant if

1. the F value is low.

2. the significance level of the F value is low.

3. the $r^2$ value is low.

4. the slope is lower than the intercept.

A new variable should not be added to a multiple regression model if that variable causes

1. $r^2$ to decrease.

2. the adjusted $r^2$ to decrease.

3. the SST to decrease.

4. the intercept to decrease.

A good regression model should have

1. a low $r^2$ and a low significance level for the F test.

2. a high $r^2$ and a high significance level for the F test.

3. a high $r^2$ and a low significance level for the F test.

4. a low $r^2$ and a high significance level for the F test.