In: Statistics and Probability
A 2016 poll sampled 523 adults who were planning a vacation during the next six months and found that 141 were expected to travel by airplane. A similar survey question in 2019 found that of 477 adults who were planning a vacation in the next six months, 81 were expecting to travel by airplane.
(i) Express in words for this study, what is meant by the probability of making a Type I and a Type II error.
(ii) Use the p-value method to test for a significant difference. Assume a significance level of 1%. Write the null and alternative hypotheses.
(iii) Repeat part (ii) using the critical value method.
(iv) Repeat part (ii) using the confidence interval approach.
1)
Type 1 error:
A type I error is the probability rejection of a true null hypothesis.
Type2 error:
A type II error is the probability of non-rejection of a false null hypothesis.
2)
Ho: p1 - p2 = 0
Ha: p1 - p2 ╪ 0
sample #1 ----->
experimental
first sample size, n1=
523
number of successes, sample 1 = x1=
141
proportion success of sample 1 , p̂1=
x1/n1= 0.2696
sample #2 -----> standard
second sample size, n2 =
477
number of successes, sample 2 = x2 =
81
proportion success of sample 1 , p̂ 2= x2/n2 =
0.170
difference in sample proportions, p̂1 - p̂2 =
0.2696 - 0.1698 =
0.0998
pooled proportion , p = (x1+x2)/(n1+n2)=
0.2220
std error ,SE = =SQRT(p*(1-p)*(1/n1+
1/n2)= 0.0263
Z-statistic = (p̂1 - p̂2)/SE = ( 0.100
/ 0.0263 ) = 3.7924
p-value = 0.000149176
[excel formula =2*NORMSDIST(z)]
decision : p-value<α,Reject null hypothesis
3)
Ho: p1 - p2 = 0
Ha: p1 - p2 ╪ 0
sample #1 ----->
first sample size, n1=
523
number of successes, sample 1 = x1=
141
proportion success of sample 1 , p̂1=
x1/n1= 0.2696
sample #2 ----->
second sample size, n2 =
477
number of successes, sample 2 = x2 =
81
proportion success of sample 1 , p̂ 2= x2/n2 =
0.170
difference in sample proportions, p̂1 - p̂2 =
0.2696 - 0.1698 =
0.0998
pooled proportion , p = (x1+x2)/(n1+n2)=
0.2220
std error ,SE = =SQRT(p*(1-p)*(1/n1+
1/n2)= 0.0263
Z-statistic = (p̂1 - p̂2)/SE = ( 0.100
/ 0.0263 ) = 3.7924
z-critical value , Z* =
2.5758 [excel formula =NORMSINV(α/2)]
decision : z stats>Z critical,Reject null
hypothesis
4)
level of significance, α = 0.01
Z critical value = Z α/2 =
2.576 [excel function: =normsinv(α/2)
Std error , SE = SQRT(p̂1 * (1 - p̂1)/n1 + p̂2 *
(1-p̂2)/n2) = 0.0259
margin of error , E = Z*SE = 2.576
* 0.0259 = 0.0668
confidence interval is
lower limit = (p̂1 - p̂2) - E = 0.100
- 0.0668 = 0.0330
upper limit = (p̂1 - p̂2) + E = 0.100
+ 0.0668 = 0.1666
so, confidence interval is ( 0.0330 < p1
- p2 < 0.1666 )
Since it does not contain zero value.
Thanks in advance!
revert back for doubt
Please upvote