Question 9 of my final exam for Design and Analysis of Sample Surveys

9. Out of a population of 100 medical records, 40 are randomly sampled and then audited. 10 out of the 40 audits reveal fraud. From this information, give an estimate, standard error, and 95% confidence interval for the proportion of audits in the population with fraud.

Solution to question 8

From yesterday:

8. Which of the following statements accurately characterize the National Election Studies? (Indicate all that apply.)

(a) The NES began in 1960.

(b) Since 1980, the NES has mostly relied on telephone interviews.

(c) The NES typically has a sample size of about 1000–2000 people.

(d) The NES uses a sampling design that ensures they get respondents from all fifty states and D.C.

Solution: c. This is a purely factual question, not much to say here.

1. awm says:

I assume the key idea here is the inclusion of the finite population correction.
p = 0.25
std err = sqrt(p*(1-p)/n * (N-n) / (N-1)) = 0.053

So the confidence interval is roughly 0.2 to 0.3

• awm says:

Um, wow, bad with the addition there. CI is 0.15 to 0.35.

• zbicyclist says:

95% — should be twice as wide.

2. […] yesterday: 9. Out of a population of 100 medical records, 40 are randomly sampled and then audited. 10 out of […]