In: Statistics and Probability
a. According to a study conducted for Gateway Computers, 60% of men and 70% of women say that weight is an extremely/very important factor in purchasing a laptop computer. Suppose this survey was conducted using 374 men and 481 women. Do these data show enough evidence to declare that a significantly higher proportion of women than men believe that weight is an extremely/very important factor in purchasing a laptop computer? Use a 5% level of significance. b. Use these data to construct a 95% confidence interval to estimate the difference in population proportions on this question.
(a)
H0: Null hypothesis: P1 = P2
HA: Alternative hypothesis: P1 < P2
n1 = size of sample 1 = 374
p1 = proportion of sample 1 = 0.6
n2 = size of sample 2 = 481
p2 = proportion of sample 2 = 0.7
Q = 1 - P = 0.3437
Test statistic is:
Z = (p1 - p2)/SE
= (0.6 - 0.7)/0.0328 = - 3.0488
= 0.05
One tail Left side test
From Table, critical value of Z = - 1.64
Since the calculated value of Z = - 3.0488 is less tha critical value of Z = - 1.64, H0 is rejected.
Conclusion: These data show enough evidence to declare that a significantly higher proportion of women than men believe that weight is an extremely/ very important factor in purchasing a laptop.
(b)
From Table, criical value of Z = 1.96
95% Confidence interval:
(p1 - p2) Z SE
= (0.6 - 0.7) 1.96 X 0.0328
= -0.1 0.0643
= ( - 0.1643, - 0.9357)