due on 04/09/2016- advance statistics - 10 questions quiz - multiple choice

profileallisky
chapter14_quiz.docx
Home>Social Science homework help>due on 04/09/2016- advance statistics - 10 questions quiz - multiple choice

Question 1 (5 points)

 Question 1 Unsaved

Match each term with the correct definition:

Question 1 options:

graph of the bivariate data to examine the relationship

change in the dependent variable divided by the change in the independent variable

also called the ordinary least-squares equation

the variable used to make the prediction or forecast

the variable being predicted or forecast by the regression equation

1.

regression equation

2.

slope

3.

constant

4.

independent variable

5.

scatterplot

6.

dependent variable

Save

Question 2 (1 point)

 Question 2 Unsaved

Which of the following is not a step to derive a regression equation?

Question 2 options:

A) 

determining if the relationship between the variables can be described by a straight line

B) 

obtaining the bivariate data

C) 

determining criteria for null and perfect relationship

D) 

plotting the data on graph

E) 

calculating the regression equation

Save

Question 3 (1 point)

 Question 3 Unsaved

The residual is defined as:

Question 3 options:

A) 

model - data

B) 

Yi - Y ˆ  

C) 

Y ˆ    - Yi

D) 

predicted - actual

Save

Question 4 (1 point)

 Question 4 Unsaved

Which of the following terms is not used to identify the correlation coefficient?

Question 4 options:

A) 

zero-order correlation coefficient

B) 

R2

C) 

Pearon's r

D) 

r

Save

Question 5 (1 point)

 Question 5 Unsaved

If r = 1.0, the regression indicates

Question 5 options:

A) 

a perfect inverse relationship

B) 

a null inverse relationship

C) 

a perfect direct relationship

D) 

a null relationship

Save

Question 6 (1 point)

 Question 6 Unsaved

To study factors associated with frequency of managers' use of report information, a researcher examined five (5) regression models. The independent variables consisted of:

Frequency of report: 1 = annual; 12 = monthly; 52 = weekly; 260 = daily

Medium: 1 = printed; 0 = online

Technique: factor score representing objective information quality

Accessibility: factor score representing ability to obtain and interpret information

Value: factor score representing value of information to specific user

The five models are summarized below:

https://mycourses.spcollege.edu/content/enforced/81939-OFR_PUP3046_3947_0510/Content/Table%2014.9.JPG?_&d2lSessionVal=4bW6Upm6x6nVCmiBHMifAEdJQ

Which model best explains the variations in the managers' information use? Explain your answer.

Question 6 options:

A) 

Model 1 is best since R2 is lowest

B) 

Model 5 is best since R2 is highest

C) 

Model 4 is best since it has the highest R2 with the least number of variables

D) 

Model 3 is best since its R2 is between the lowest and highest R2.

Save

Question 7 (1 point)

 Question 7 Unsaved

To study factors associated with frequency of managers' use of report information, a researcher examined five (5) regression models. The independent variables consisted of:

Frequency of report: 1 = annual; 12 = monthly; 52 = weekly; 260 = daily

Medium: 1 = printed; 0 = online

Technique: factor score representing objective information quality

Accessibility: factor score representing ability to obtain and interpret information

Value: factor score representing value of information to specific user

The five models are summarized below:

https://mycourses.spcollege.edu/content/enforced/81939-OFR_PUP3046_3947_0510/Content/Table%2014.9.JPG?_&d2lSessionVal=4bW6Upm6x6nVCmiBHMifAEdJQ

Based on the information, should an agency concentrate on continually improving the quality of its information? Identify the independent variable you used and explain your answer.

Question 7 options:

A) 

Value; the beta weights for models 4 and 5 indicate that increased value is associated with increased quality so the agency should concentrate on increasing the quality of the information.

B) 

Value; the beta weights for models 4 and 5 indicate that increased value is associated with increased usage so the agency should concentrate on increasing the quality of the information.

C) 

Technique; the beta weights for models 4 and 5 indicate that decreased quality is associated with increased usage so the agency should not concentrate on improving the quality of the information.

D) 

Technique; the beta weights for models 4 and 5 indicate that increased quality is associated with increased usage so the agency should concentrate on increasing the quality of the information.

Save

Question 8 (1 point)

 Question 8 Unsaved

You are the director of the county welfare office and want to know if a relationship exists between the length of time that people receive welfare (in months) and several variables you believe might affect the length of time on welfare. You run a regression analysis on the county's welfare data.

The results are shown below:

https://mycourses.spcollege.edu/content/enforced/81939-OFR_PUP3046_3947_0510/Content/welfare%20regression.JPG?_&d2lSessionVal=4bW6Upm6x6nVCmiBHMifAEdJQ

Would you remove any variables from the model? If yes, which ones. Explain your answer.

Question 8 options:

A) 

Yes, marital status should be removed since its beta weight is so small.

B) 

Yes, number of dependents is the only variable that should be removed since its beta weight is so small.

C) 

Yes, number of dependents and education should be removed since their t-ratios have p-values greater than 0.05.

D) 

No, all variables are making significant contributions to the regression model.

Save

Question 9 (1 point)

 Question 9 Unsaved

Question 9 options:

Consider this regression equation and correlation:

Y = 6 + 0.25X

r = 0.60

where Y = % of African-American police officers in a particular community in Georgia and X = % of the African-Americans in that particular community in Georgia.

What is the coefficient of determination for this regression line? Enter your answer as a decimal rounded to two decimal places.

r2 =

Spell check

Save

Question 10 (1 point)

 Question 10 Unsaved

Question 10 options:

Consider this regression equation and correlation:

Y = 6 + 0.25X

r = 0.60

where Y = % of African-American police officers in a particular community in Georgia and X = % of the African-Americans in that particular community in Georgia.

In the town of Milton, Georgia, the population is 30% African-American. Estimate the percentage of the police force in Milton that is African-American. Enter your answers with one decimal place. Do not type in % with your answer.

NOTE: To use this regression model, enter 30% as x = 30 and determine the predicted value Y of the % of African-American police officers in this community.

The estimate for the percentage of African American police officers in Milton is

Spell check

%.

Save

Question 11 (1 point)

 Question 11 Unsaved

Question 11 options:

Education level, in average number of years, was used to predict the mortality rate, in deaths per 100,000 people, for 58 U.S. cities.

The regression output is shown below:

https://mycourses.spcollege.edu/content/enforced/81939-OFR_PUP3046_3947_0510/Content/education%20mortality.JPG?_&d2lSessionVal=4bW6Upm6x6nVCmiBHMifAEdJQ

If we know that the correlation coefficient r = -0.64, what is the coefficient of determination? Enter your answer as a decimal with two decimal places.

r2 =

Spell check

Save

Question 12 (1 point)

 Question 12 Unsaved

A sample of 84 2004-model cars from an online information service was examined to see how fuel efficiency (as highway mpg) relates to the cost (Manufacturer's Suggested Retail Price in dollars) of cars.

The analysis of regression model is shown below along with the coefficient of determination.

https://mycourses.spcollege.edu/content/enforced/81939-OFR_PUP3046_3947_0510/Content/car%20efficiency.JPG?_&d2lSessionVal=4bW6Upm6x6nVCmiBHMifAEdJQ

Using this information, what is the correlation cofficient? Explain how you calculated your answer.

Question 12 options:

A) 

Since R2= 30.1%, we find the square root and report as a decimal. The sign of the correlation coefficient matches the sign of the slope, which is positive for this dataset. Therefore r = 0.549.

B) 

Since R2= 30.1%, we find the square root and report as a decimal. The sign of the correlation coefficient matches the sign of the slope, which is negative for this dataset. Therefore r = -0.549.

C) 

We square the coefficient of determination to find the correlation coefficient and express as a decimal. For this dataset, r = 0.091.

D) 

Unable to determine with the given information.

Save

Question 13 (1 point)

 Question 13 Unsaved

Infant mortality is often used as a general measure of the quality of healthcare for children and mothers. It is reported as the rate of deaths of newborns per 1000 live births. Data are recorded for each of the 50 states of the United States and those data are used to build regression models to help understand or predict infant mortality.

The variables used in this model are:

Child deaths: deaths per 100,000 children aged 1-14 yrs

HS Drop%: Percent of teens (aged 16-19) who drop out of high school

Low BW%: percent of low-birth weight babies

Teen births: births per 100,000 females aged 14-17 yrs

Teen deaths:deaths by accident, homicide, suicide per 100,000 teens aged 15-19 yrs.

The regression results are shown below:

https://mycourses.spcollege.edu/content/enforced/81939-OFR_PUP3046_3947_0510/Content/infant%20deaths.JPG?_&d2lSessionVal=4bW6Upm6x6nVCmiBHMifAEdJQ

Which variables should be kept in the regression equation? Assume alpha = 0.05. Explain your answer.

Question 13 options:

A) 

Teen births and teen deaths since these variables have the largest p-values

B) 

All variables should be retained since models always predict better with more independent variables.

C) 

Child deaths, and low birth weight since the p-value for each coefficient is less than 0.05.

D) 

Child deaths, low birth weight, teen births and teen deaths since their coefficients are positive.

Save

Question 14 (1 point)

 Question 14 Unsaved

Question 14 options:

Infant mortality is often used as a general measure of the quality of healthcare for children and mothers. It is reported as the rate of deaths of newborns per 1000 live births. Data are recorded for each of the 50 states of the United States and those data are used to build regression models to help understand or predict infant mortality.

The variables used in this model are:

Child deaths: deaths per 100,000 children aged 1-14 yrs

HS Drop%: Percent of teens (aged 16-19) who drop out of high school

Low BW%: percent of low-birth weight babies

Teen births: births per 100,000 females aged 14-17 yrs

Teen deaths: deaths by accident, homicide, suicide per 100,000 teens aged 15-19 yrs.

The regression results are shown below:

https://mycourses.spcollege.edu/content/enforced/81939-OFR_PUP3046_3947_0510/Content/infant%20deaths.JPG?_&d2lSessionVal=4bW6Upm6x6nVCmiBHMifAEdJQ

If we observed a 1% increase in the percentage of live births that are low birth weight, what increase do we predict in infant mortality using this regression model? Enter your answer with two decimal places.

The predicted increase in infant mortality is

Spell check

.

Save

Question 15 (1 point)

 Question 15 Unsaved

True or False:

A value of the correlation coefficient greater than 0.65 is always statistically significant.

Question 15 options:

A) 

False

B) 

True

Save

Question 16 (1 point)

 Question 16 Unsaved

True or false:

In a multiple regression equation, each regression coefficient bi is known as a partial regression coefficient.

Question 16 options:

A) 

False

B) 

True

Save

Question 17 (1 point)

 Question 17 Unsaved

True or false:

R2 is used to indicate a multivariate relationship while r2 is used to indicate a bivariate relationship.

Question 17 options:

A) 

True

B) 

False

Save

Question 18 (1 point)

 Question 18 Unsaved

True or False:

The adjusted R2 "shrinks" the value of R2 by penalizing for each additional independent variable in the model.

Question 18 options:

A) 

False

B) 

True

Save

Question 19 (1 point)

 Question 19 Unsaved

True or False:

A symptom of multicollinearity is that the addition of an independent variable does not change the beta weights or the regression coefficients.

Question 19 options:

A) 

True

B) 

False

Save

Question 20 (1 point)

 Question 20 Unsaved

True or False:

Logistic regression is used when the dependent variable is not dichotomous.

Question 20 options:

A) 

False

B) 

True

Save

Question 21 (1 point)

 Question 21 Unsaved

True or False:

The linear regression model is appropriate for forecasting if over time the variable's values can be reasonably described by a straight line.

Question 21 options:

A) 

False

B) 

True

Save

Question 22 (1 point)

True or False:

An assumption of causality seems defensible if the regression model is statistically significant.

Question 22 options:

A) 

False

B) 

True

Save

18

19

20

21

22

2

3

4

5

6

7

1

8

9

10

11

12

13

14

15

16

17