195k views
5 votes
A study1 conducted in July 2015 examines smartphone ownership by US adults. A random sample of 2001 people were surveyed, and the study shows that 688 of the 989 men own a smartphone and 671 of the 1012 women own a smartphone. We want to test whether the survey results provide evidence of a difference in the proportion owning a smartphone between men and women. Let group 1 be US men and let group 2 be US women.

(a) State the null and alternative hypotheses.
(b) Give the notation for the sample statistic.
(c) Give the value for the sample statistic.
(d) In the sample, which group has higher samartphone ownership: men or women?

User Mario A
by
7.2k points

1 Answer

6 votes

Answer:

a) Null hypothesis:
p_(1) = p_(2)

Alternative hypothesis:
p_(1) \\eq p_(2)

b)
z=\frac{p_(1)-p_(2)}{\sqrt{\hat p (1-\hat p)((1)/(n_(1))+(1)/(n_(2)))}} (1)

Where
\hat p=(X_(1)+X_(2))/(n_(1)+n_(2))=(688+671)/(989+1012)=0.679

c)
z=\frac{0.696-0.663}{\sqrt{0.679(1-0.679)((1)/(989)+(1)/(1012))}}=1.58

d) For this case we see that
\hat p_1 > \hat p_2 so then the answer for this cae would men

Explanation:

Information given


X_(1)=688 represent the number of men with smartphone


X_(2)=671 represent the number of women with smartphone


n_(1)=989 sample of men selected


n_(2)=1012 sample of women selected


p_(1)=(688)/(989)=0.696 represent the proportion of men with smartphone


p_(2)=(671)/(1012)=0.663 represent the proportion of women with smartphone


\hat p represent the pooled estimate of p

z would represent the statistic


p_v represent the value

Part a

We want to test if we have difference in the proportion owning a smartphone between men and women, the system of hypothesis would be:

Null hypothesis:
p_(1) = p_(2)

Alternative hypothesis:
p_(1) \\eq p_(2)

Part b

The statistic for this case is given by:


z=\frac{p_(1)-p_(2)}{\sqrt{\hat p (1-\hat p)((1)/(n_(1))+(1)/(n_(2)))}} (1)

Where
\hat p=(X_(1)+X_(2))/(n_(1)+n_(2))=(688+671)/(989+1012)=0.679

Part c

Replacing the info given we got:


z=\frac{0.696-0.663}{\sqrt{0.679(1-0.679)((1)/(989)+(1)/(1012))}}=1.58

Part d

For this case we see that
\hat p_1 > \hat p_2 so then the answer for this cae would men

User Michael Sofaer
by
7.3k points