Which of following statements of T distribution is NOT true?

a. The total under-the-curve area is 0.5.

b. The mean=median=mode.

c. It is a symmetric bell-shaped curve.

d. It looks more and more like the standard normal distribution as the degree of freedom increases.

The histogram is a suitable graph for __________

a. continuous variables.

b. non-orderable discrete variables.

c. dichotomous variables.

d. orderable discrete variables.

The degree of freedom of a four-by-three (m=4; n=3) crosstabulation is:

a. 12

b. 8

c. 6

d. 9

The Pearson's correlation coefficient is a measure of _______ used to determine the existence and strength of the relationship between two _______ variables.

a. causality; discrete

b. causality; continuous

c. association; discrete

d. association; continuous

The intercept (or say constant) of a simple regression y=a+bx+e is _______.

a. the point where the regression line crosses the x-axis

b. the point where the regression line crosses the y-axis

c. the mean value of x when y=0

d. always larger than the slope

Linear relationship is a relationship between two _______ variables in which the observations displayed in a ______ can be approximated with a ________.

a. continuous; histogram; straight line

b. discrete; bar chart; straight line

c. continuous; scatter plot; U-shape curve

d. continuous; scatter plot; straight line

Regression sum of squares (SS_Regression) _______.

a. is the unexplained portion of total variation in the dependent variable

b. is always larger than the sum of square total (SS_Total)

c. reflects the improvement in the prediction error due to using the regression line

d. is always larger than the sum of square error (SS_Error)

For a Pearson's correlation coefficient (r_(x,y)), which statement is NOT TRUE?

a. It is the square root of coefficient of determination of a simple regression.

b. A positive r_(x,y) implies b >0.

c. It can be zero if two variables are not linearly related.

d. It is possible to have a negative product of r_(x,y)Cov(x,y).

For a simple regression y=a+bx+e, which statement is NOT TRUE?

a. R2 is the squared correlation coefficient between y and its predicted value y .

b. It is possible to have r_(x,y)b <0.

c. The mean square regression equals the sum of square regression.

d. Point (x ,y ) is right on the regression line.

For a simple regression y=a+bx+e of a given valid sample size N, which statement is TRUE? [1 bonus point]

a. If R^2=SS_Regression/SS_Total , then 1-R^2=SS_Error/SS_Total .

b. The error term (e) is assumed to follow an F distribution.

c. As the |r_(x,y) | approaching +1, the MS_Error will be getting larger and larger.

d. The standard error of slope is the variance of the sampling distribution of slopes.