Data Collectionhard
0:00.0
A survey of 1000 respondents has sample distributions deviating from known population margins:
Sex: Sample is 40% Male, 60% Female; Target is 50% Male, 50% Female
Age: Sample is 35% <30, 50% 30–60, 15% >60; Target is 30% <30, 50% 30–60, 20% >60
Raking (iterative proportional fitting) adjusts weights to match both margins simultaneously. Which statements about raking are correct?
A) After raking, the marginal distributions of sex and age exactly match population targets
B) Raking ensures the joint distribution (sex × age) exactly matches the population's unknown joint distribution
C) Raked estimates have lower variance than unraked estimates because they use population information
D) Raking assumes no association between sex and age
A survey of 1000 respondents has sample distributions deviating from known population margins:
Sex: Sample is 40% Male, 60% Female; Target is 50% Male, 50% Female
Age: Sample is 35% <30, 50% 30–60, 15% >60; Target is 30% <30, 50% 30–60, 20% >60
Raking (iterative proportional fitting) adjusts weights to match both margins simultaneously. Which statements about raking are correct?
A) After raking, the marginal distributions of sex and age exactly match population targets
B) Raking ensures the joint distribution (sex × age) exactly matches the population's unknown joint distribution
C) Raked estimates have lower variance than unraked estimates because they use population information
D) Raking assumes no association between sex and age