When you construct a confidence interval for the difference between two proportions, what do you use as an unbiased estimate of the difference between the two proportions?

Thus difference in sample proportions would be the unbiased estimate of the difference between the two proportions

Step-by-step explanation:

Given that you construct a confidence interval for the difference between two proportions

Let the proportions be p1 and p2

Sample sizes be n1 and n2

For samples of larger size randomly drawn we know that

p1 follows Normal with mean = sample proportion = p1 and

p2 is N with mean = sample proportion p2

Assuming these two proportions are independent we have

p1-p2 will follow a normal distribution with mean = p1-p2

Thus difference in sample proportions would be the unbiased estimate of the difference between the two proportions