QUESTION 1 : Correlation coefficient r quantifies linear relationship with a number between

a. -1 and 1.

b. Negative to positive infinity

c. 0 to 1

d. 0 to infinity

QUESTION 2 : When all points fall on a line with an upward slope, r = ?

a. 1.

b. 0.

c. -1

d. Indeterminate

QUESTION 3 : When all data points fall on a line with a downward slope, r = ?

a. 1.

b. 0.

c. -1

d. Indeterminate

QUESTION 4 : When data points trend upward, r is ?

a. Positive

b. Negative

c. Neutral

d. Indeterminate

QUESTION 5 : When data points trend downward, r is ?

a. Positive

b. Negative

c. Neutral

d. Indeterminate

QUESTION 6 : The closer r is to 1 or -1, the ______the correlation.

a. Weaker

b. Stronger

c. Neutral

d. Non significant

QUESTION 7 : In regression the quantitative response variable Y is the __

a. "dependent variable"

b. "independent variable"

c. "individual variable"

d. "composite variable"

QUESTION 8 : In regression the quantitative explanatory variable X is the __

a. "dependent variable"

b. "independent variable"

c. "individual variable"

d. "composite variable"

QUESTION 9 : Which of the following correctly describes the interpretation of an r2 statistic of 0.673 in a model relating BMI (kg/m2) (X) to frequency of hospital visits (Y)?

a. There is a high degree of correlation between BMI and frequency of hospital visits.

b. There is a strong, positive relationship between BMI and frequency of hospital visits

c. 67.3% of hospital visits are due to elevated levels of BMI

d. 67.3% of the variation in frequency of hospital visits is explained by BMI

QUESTION 10 : Which of the following most accurately describes a correlation coefficient of 0.83 between height (inch X) and weight (lbs Y)?

a. 83% of the variation in weight can be explained by height

b. There is a strong, positive association between height and weight

c. There is an association between height and weight

d. 83% of the variation in height can be explained by weight

QUESTION 11 : Which of the following is not a formal condition for inference about a regression estimate?

a. Sampling independence of bivariate observations

b. Linearity between variables X and Y

c. Normality in the distribution of residuals

d. Heteroscedasticity between X and Y variables

QUESTION 12 : Which of the following is the correct set of null and alternative hypotheses for a t-test regarding the regression equation between variables X and Y?

a. Ho: r=0, Ha: r≠0

b. Ho: ρ=0, Ha: ρ≠0

c. Ho: b=0, Ha: b≠0

d. Ho: β=0, Ha: β≠0

QUESTION 13 : Which of the following best describes the difference between the interpretation of the correlation coefficient and the beta coefficient for a regression equation expressing the relationship between variables X and Y?

a. The beta coefficient represents the linear relationship between X and Y, whereas the correlation coefficient represents the percent variability in Y explained by X.

b. The beta coefficient represents the difference between observed and expected values of Y, whereas the correlation coefficient represents the linear relationship between X and Y.

c. The beta coefficient predicts increases or decreases in Y with increases or decreases in X, whereas the correlation coefficient provides a unit-free measure of the strength of the relationship.

d. The correlation coefficient predicts increases or decreases in Y with increases or decreases in X, whereas the beta coefficient provides a unit-free measure of the strength of the relationship.

QUESTION 14 : Which of the following is false concerning the interpretation of a regression equation expressing the relationship between variables X and Y?

a. The beta coefficient predicts a per unit increase in Y based on X

b. The regression equation is typically adapted to populations beyond the observed range for X

c. Regression equations are strongly influenced by the presence of outliers

d. Regression equations only describe linear relationships between two variables

QUESTION 15 : Which of the following is not determined in the case of multiple linear regression?

a. Potential confounding effects from "lurking" variables included in the analysis

b. Observed relationship between predictor and outcome variables adjusting for confounding relationships

c. Causal relationships between predictor and outcome variables adjusting for confounding relationships

d. Potential observed interaction effects between variables.