Weighting Quiz

Question 1

1. What does a weighting efficiency of 27.6% suggest about the sample?

Accepted Answer

C. The sample is highly skewed and may not represent the population well

Question 2

5. Which situation indicates multicollinearity in a linear regression model?

Accepted Answer

A. VIF > 5 or Tolerance values close to 0

Question 3

6. Why can a predictor's coefficient turn negative in multiple regression after being positive in bivariate analysis?

Accepted Answer

B. A suppression effect caused by multicollinearity with other predictors

Question 4

7. Which correlation method would be most appropriate for two dichotomized variables (e.g., 1 = Yes, 0 = No)?

Accepted Answer

D. Tetrachoric correlation

Question 5

8. What is a limitation of using Exp(B) from a logistic regression model to rank variable importance?

Accepted Answer

C. The logit function distorts linear interpretation; use Wald test instead

Question 6

9. What does the Shapley Value Regression help resolve that traditional regression may not?

Accepted Answer

B. Fairly distributing R² among predictors, esp when they are correlated

Question 7

4. What is a key limitation of using Pearson correlation in survey analysis?

Accepted Answer

It assumes linearity & normality, which may not reflect real data.

Question 8

2. Why might Rim Weighting be preferred over Cell Weighting in certain situations?

Accepted Answer

B. It adjusts margins even when full cell-level data isn't available

Question 9

3. In what scenario would the Spearman correlation be more appropriate than the Pearson correlation?

Accepted Answer

D. When the relationship is monotonic but not linear

Questions 9