Question

1) True or False?

1. In multinomial logistic regression analysis, the model fit is given by the value of R-squared.

2. In linear regression analysis, the overall significance test is an F test.

3. In logistic regression analysis, the overall significance test is an F test.

4. When building a model for regression analysis, the type of data for the outcome variable

   guides you to choose linear, logistic, or multinomial logistic.

5. Two variables that covary might not have a causal relationship.

6. Causation implies correlation.

2) Which one should be analyzed by multinomial logistic regression analysis? Select all correct cases.

• 1) Dependent V: Attitude toward immigration policy (strongly agree/agree/disagree/strongly

  disagree). Independent V: Gender (female/male).

• 2) Dependent V: College major (Business/Biology/Economics/Psychology). Independent V:

  High school GPA (points)

• 3) Dependent V: Individual preference about automobile brand (Ford/Kia). Independent V: Age

• 4) Dependent V: A country’s trade policy (low tariff/moderate tariff/high tariff). Independent V: GDP ($)

• 5) None of them is correct.

3) Which one should be analyzed by logistic regression analysis? Select all correct cases. No partial credit is given.

• 1) Dependent V: Attitude toward gun policy (Agree/Disagree). Independent V: Gender (female/male).

• 2) Dependent V: College major (Business/Economics/Psychology). Independent V: Father’s major (Business/Economics/Psychology)

• 3) Dependent V: Individual preference about automobile brand (Ford/Kia). Independent V: Age

• 4) Dependent V: Starting salary (QAR). Independent V: College major (Business/Biology)

• 5) None of them is correct.

Answer 1

1. i) In multinomial logistic regression analysis, the model fit is given by the value of pseudo-R-squared or chi-square test. Thus it is FALSE.

ii) In linear regression analysis, the overall significance test is an F test. TRUE

iii) In logistic regression analysis, the overall significance test can be determined by the chi-square test. Thus False.

iv) True

v) Two variables that covary might not have a causal relationship as there may be a spurious correlation. True

vi) Causation doesn't imply correlation. False.

2. Multinomial Logistic Regression is the regression analysis to conduct when the dependent variable is nominal with more than two levels. Thus the correct options are:

i) Dependent V: Attitude toward immigration policy (strongly agree/agree/disagree/strongly disagree). Independent V: Gender (female/male).

ii) Dependent V: College major (Business/Biology/Economics/Psychology). Independent V: High school GPA (points)

iii) Dependent V: A country’s trade policy (low tariff/moderate tariff/high tariff). Independent V: GDP ($)

3. Logistic Regression is the regression analysis to conduct when the dependent variable is nominal with exactly two levels. Thus the correct options are:

i) Dependent V: Attitude toward gun policy (Agree/Disagree). Independent V: Gender (female/male).

ii) Dependent V: Individual preference about automobile brand (Ford/Kia). Independent V: Age