Chapter 21: Confidence Intervals and Significance Tests with Two Proportions

Question 1

2-PropZInt Conditions

Answer 1

1. Both random samples or data collected from 2 groups that have been randomly assigned treatments
2. n≤10% of pop. for each (not for experiments)
3. n1p̂1≥10; n1(1-p̂1)≥10; n2p̂2≥10; n2(1-p̂2)≥10

Question 2

2-PropZInt On Calculator

Answer 2

x1: # of successes group 1
n1: sample size group 1
x2: # of successes group 2
n2: sample size group 2
C-Level: CI% in decimal

Question 3

2-PropZInt Interpretation

Answer 3

“We are [C%] confident that the interval from [A to B] captures the [value to be estimated].”
(+,+) - convincing evidence that first is greater
(-, -) - convincing evidence that first is less
(-,+) - no convincing evidence of a difference

Question 4

2-PropZTest Conditions

Answer 4

1. Both random samples or data collected from 2 groups that have been randomly assigned treatments
2. n≤10% of pop. for each (not for experiments)
3. n1p̂C≥10; n1(1-p̂C)≥10; n2p̂C≥10; n2(1-p̂C)≥10
   p̂C = combined # of successes/combined # of observations

Question 5

2-PropZTest On Calculator

Answer 5

x1: # of successes group 1
n1: sample size group 1
x2: # of successes group 2
n2: sample size group 2
p1(≠p2; <p2; >p2)

Question 6

2-PropZTest p-value Interpretation

Answer 6

“Assuming [H0 in context] is true, there is a [p-value] probability of getting a difference in proportions of [observed difference] or [greater/less/more different] by chance alone in the [random assignment/random samples].”