In: Statistics and Probability
6. Suppose that the sample unemployment rate for those aged 25–60 is 8% based on a survey of 150 people, while the unemployment rate for those aged 16–24 is 12% based on a survey of 100 people.
(a) Form a 90% confidence interval for the difference between the two population unemployment rates.
(b) Construct a test statistic to test the null hypothesis that the two population rates are the same against the alternative hypothesis that the rate is higher in for those aged 16–24, and report the associated P-value.
a)
level of significance, α = 0.10
Z critical value = Z α/2 =
1.645 [excel function: =normsinv(α/2)
Std error , SE = SQRT(p̂1 * (1 - p̂1)/n1 + p̂2 *
(1-p̂2)/n2) = 0.0393
margin of error , E = Z*SE = 1.645
* 0.0393 = 0.0647
confidence interval is
lower limit = (p̂1 - p̂2) - E = -0.040
- 0.0647 = -0.1047
upper limit = (p̂1 - p̂2) + E = -0.040
+ 0.0647 = 0.0247
so, confidence interval is (
-0.1047 < p1 - p2 <
0.0247 )
b)
Ho: p1 - p2 = 0
Ha: p1 - p2 < 0
sample #1 -----> 25-60
first sample size, n1=
150
number of successes, sample 1 = x1=
12
proportion success of sample 1 , p̂1=
x1/n1= 0.0800
sample #2 -----> 16-24
second sample size, n2 =
100
number of successes, sample 2 = x2 =
12
proportion success of sample 1 , p̂ 2= x2/n2 =
0.120
difference in sample proportions, p̂1 - p̂2 =
0.0800 - 0.1200 =
-0.0400
pooled proportion , p = (x1+x2)/(n1+n2)=
0.0960
std error ,SE = =SQRT(p*(1-p)*(1/n1+
1/n2)= 0.0380
Z-statistic = (p̂1 - p̂2)/SE = ( -0.040
/ 0.0380 ) = -1.0518
p-value = 0.1465 [Excel
function =NORMSDIST(z)
decision : p-value>α,Don't reject null hypothesis
Conclusion: There is not enough evidence that
the rate is higher in for those aged 16–24
thanks
revert back for doubt
please upvote