In: Statistics and Probability
Independent random samples of
n1 = 800
and
n2 = 670
observations were selected from binomial populations 1 and 2, and
x1 = 336
and
x2 = 378
successes were observed.
(a) Find a 90% confidence interval for the difference
(p1 − p2) in the two
population proportions. (Round your answers to three decimal
places.)
to
(b) What assumptions must you make for the confidence interval to
be valid? (Select all that apply.)nq̂ > 5 for samples
from both populationssymmetrical distributions for both
populationsnp̂ > 5 for samples from both
populationsindependent random samples
n1 + n2 > 1,000
(c) Can you conclude that there is a difference in the population
proportions based on the confidence interval found in part (a)?
No. Since zero is not contained in the confidence interval, the two population proportions are likely to be equal.Yes. Since zero is contained in the confidence interval, the two population proportions are likely to be different. No. Since zero is contained in the confidence interval, the two population proportions are likely to be equal.Nothing can be determined about the difference between the two population proportions.Yes. Since zero is not contained in the confidence interval, the two population proportions are likely to be different.
a)
Here, , n1 = 800 , n2 = 670
p1cap = 0.42 , p2cap = 0.5642
Standard Error, sigma(p1cap - p2cap),
SE = sqrt(p1cap * (1-p1cap)/n1 + p2cap * (1-p2cap)/n2)
SE = sqrt(0.42 * (1-0.42)/800 + 0.5642*(1-0.5642)/670)
SE = 0.0259
For 0.9 CI, z-value = 1.64
Confidence Interval,
CI = (p1cap - p2cap - z*SE, p1cap - p2cap + z*SE)
CI = (0.42 - 0.5642 - 1.64*0.0259, 0.42 - 0.5642 +
1.64*0.0259)
CI = (-0.187 , -0.102)
b)
nq̂ > 5 for samples from both populationssymmetrical distributions for both populationsnp̂ > 5 for samples from both populationsindependent random samples
c)
.Yes. Since zero is not contained in the confidence interval, the two population proportions are likely to be different.