Part A:
1. In a scatter diagram, the:
A) dependent variable is scaled along the horizontal axis.
B) independent variable is scaled along the vertical axis.
C) graph shows the relationship between two variables.
D) probabilities are plotted.
2. Correlation analysis does the following:
A) It reports the relationship between two nominal scale variables.
B) It evaluates the strength of the association between two interval or ratio variables.
C) It compares two variances.
D) It compares two means.
3. The sample coefficient of correlation:
A) has the same sign as the slope of the regression equation.
B) can range only from 0 to 1.00.
C) is also called Student's t.
D) has "n" degrees of freedom.
4. The coefficient of determination is:
A) the square root of the correlation coefficient.
B) between -1.00 and 1.00.
C) inversely related to (SSE/ SS Total).
D) based on the treatment mean square.
5. Suppose we developed the following least squares regression equation: Y^ = 3.5 + 2.1 X. What can we conclude?
A) The dependent variable increases 3.5 for each unit increase in X.
B) The equation crosses the Y-axis at 2.1.
C) If X = 5, then Y^ is 14.
D) There is a significant positive relationship between the dependent and independent variables.
6. The standard error of estimate:
A) is based on squared deviations from the regression line.
B) may assume negative values.
C) is in squared units of the independent variable.
D) is the regression mean square error.
7. Which of the following is NOT a necessary condition for regression analysis?
A) The standard deviation of each of the conditional distributions must be the same.
B) The Y values are independent.
C) For each X value, there is a group of Y values and these Y values are normally distributed.
D) The slope of the regression line is positive (increasing).
8. Which of the following is not based on squared deviations from the regression line?
A) Correlation coefficient
B) Coefficient of determination
C) Standard deviation
D) Standard error of estimate
9. In an ANOVA table for simple linear regression, the degrees of freedom for the regression mean square is equal to:
A) 1.
B) n - 1.
C) n - 2.
D) n
10. The term (SSR/SS total) is also called the:
A) sum of squares due to regression.
B) coefficient of determination.
C) standard error of estimate.
D) coefficient of correlation.
11. When comparing the 90% confidence interval and the 90% prediction interval for a given regression analysis,
A) the confidence interval is wider than a prediction interval.
B) the confidence interval is narrower than a prediction interval.
C) there is no difference between the confidence and prediction intervals.
D) None of the above.
12. The hypothesis, H0: ρ = 0, is tested using a:
A) z statistic.
B) F statistic.
C) t statistic with n - 1 degrees of freedom.
D) t statistic with n - 2 degrees of freedom.
13. If the coefficient of determination is 95%, then:
A) 95% of the variation in the dependent variable is explained by the variation in the independent variable.
B) 95% of the variation in the independent variable is explained by the dependent variable.
C) the regression equation is 95% correct.
D) the confidence interval is a good estimate 95% of the time.
Part B:
1. In a multiple regression equation, there are two or more:
A) dependent variables.
B) independent variables.
C) intercept values.
D) coefficients of determination
2. Which of the following statements is true about a dummy variable or indicator variable?
A) It may assume only a value of 0 or 1.
B) It is another term for the dependent variable.
C) It is found by (Y - ?).
D) It is equal to ?.
3. The multiple standard error of estimate is:
A) found by taking the square root of SSR/SS total.
B) negative when one of the net regression coefficients is 0.
C) based on the term, (Y - ?)2.
D) the error of estimating the regression coefficients.
4. In the ANOVA table, the value of k is the:
A) sum of squares total.
B) total number of observations
C) number of degrees of freedom.
D) number of independent variables.
5. A correlation matrix shows:
A) all simple correlation coefficients.
B) all possible net regression coefficients.
C) only positive correlation coefficients.
D) the multiple regression equation.
6. Which of the following statements is true for a multiple regression equation?
A) There is only one dependent variable.
B) The R2 term must be at least .50.
C) All the regression coefficients must be between -1.00 and 1.00.
D) There is only one (Y - ?).
7. Multicollinearity occurs when which of the following conditions exists?
A) The residuals are correlated.
B) Time is involved in the analysis.
C) The independent variables are correlated.
D) The residuals are not constant for all values.
8. For a multiple regression analysis, the global hypothesis test determines:
A) which regression coefficients do not equal 0.
B) whether any of the regression coefficients differ from 0.
C) whether any of the correlation coefficients differ from 0.
D) whether the intercept differs from 0.
9. In testing the significance of individual regression coefficients,
A) the test statistic is the t distribution.
B) we test for differences between each pair of regression coefficients.
C) we usually delete the variables where the null hypothesis is rejected.
D) regression coefficients with a negative coefficient are deleted from the equation.
10. Which of the following statements is true about a residual?
A) It has the same degrees of freedom as the MSE term.
B) It cannot assume a negative value.
C) It is summarized in a correlation matrix.
D) It is the difference between the actual and the predicted value of a dependent variable.
11. The multiple coefficient of determination:
A) reports the percent of the variation in the dependent variable explained by the variation in the set of independent variables.
B) can range from -1.00 up to 1.00.
C) is the ratio of the SSE to the SS total.
D) is based on the interaction mean square.
12. The adjusted coefficient of determination:
A) is the ratio of SSR to SS total.
B) can range from -1.00 up to 1.00.
C) is the ratio of the SSE to the SS total.
D) corrects for the number of independent variables in a regression equation.
13. An example of an interaction term in a multiple regression model is:
A) (X1) + (X2).
B) (X1)(X2).
C) (b1)(b2).
D) (b1)(b2) + (X1)(X2).
14. Stepwise regression is a method to:
A) eliminate insignificant variables from a multiple regression equation.
B) decrease the multiple coefficient of determination by selecting variables to include in a multiple regression equation.
C) increase the total sum of squares.
D) use the best regression equation with a single independent variable.