In: Math
Test whether the proportion of iphone owners is more than the proportion of android owners. Take two samples of at least 20 each. In the first sample ask "Do you own an iphone?" in the second sample ask "do you own an android?" Use a significance level of 0.01.
a. survey results: iphone -> 13 yes, 7 no android -> 11 yes, 9 no
b. state your claim
c. null hypothesis and alternative hypothesis
d. which type of test are you running? show calculator input.
e. P-value
f. Decision: reject null or fail to reject null
g. conclusion
h. calculate a 98% confidence interval to estimate the difference in proportions of iphone and android users. write it in sentence format.
i. does the confidence interval agree or contradict your hypothesis test conclusion? why?
b)
claim: proportion of iphone owners is more than the proportion of android owners or p1 > p2
c)
Ho: p1 - p2 = 0
Ha: p1 - p2 > 0
d)
Hypothesis Test: Difference of two Proportions
e)
sample #1 -----> iphone
first sample size, n1=
20
number of successes, sample 1 = x1=
13
proportion success of sample 1 , p̂1=
x1/n1= 0.6500
sample #2 -----> android
second sample size, n2 =
20
number of successes, sample 2 = x2 =
11
proportion success of sample 1 , p̂ 2= x2/n2 =
0.550
difference in sample proportions, p̂1 - p̂2 =
0.6500 - 0.5500 =
0.1000
pooled proportion , p = (x1+x2)/(n1+n2)=
0.6000
std error ,SE = =SQRT(p*(1-p)*(1/n1+
1/n2)= 0.1549
Z-statistic = (p̂1 - p̂2)/SE = ( 0.100
/ 0.1549 ) = 0.6455
z-critical value , Z* =
2.3263 [excel function =NORMSINV(α)]
p-value = 0.2593 [excel function
=NORMSDIST(-z)]
f)
decision : p-value>α,Don't reject null
hypothesis
g)Conclusion: There is not enough evidence to conclude that the proportion of iphone owners is more than the proportion of android owner
h)
level of significance, α = 0.02
Z critical value = Z α/2 =
2.326 [excel function: =normsinv(α/2)
Std error , SE = SQRT(p̂1 * (1 - p̂1)/n1 + p̂2 *
(1-p̂2)/n2) = 0.1541
margin of error , E = Z*SE = 2.326
* 0.1541 = 0.3585
confidence interval is
lower limit = (p̂1 - p̂2) - E = 0.100
- 0.3585 = -0.2585
upper limit = (p̂1 - p̂2) + E = 0.100
+ 0.3585 = 0.4585
so, confidence interval is ( -0.2585 <
p1 - p2 < 0.4585 )
we are 95% confident that the difference in proportions of iphone and android users lies within confidence interval
i)
since, confidence interval contains zero, null hypothesis is
not rejected.so, confidence interval agree
your hypothesis test conclusion