In: Statistics and Probability
A sample of 52 Elementary Statistics students includes 13 women. Assuming the sample is random:
(a) Estimate the percentage of women taking Elementary Statistics by computing an appropriate confidence interval with 98% confidence.
(b) At 10% significance, test whether less than 40% of the enrollment in all Elementary Statistics classes consists of women.
Solution :
Given that,
n = 52
x = 13
= x / n = 13 / 52 = 0.25
1 - = 1 - 0.25 = 0.75
(a)
At 98% confidence level the z is ,
= 1 - 98% = 1 - 0.98 = 0.02
/ 2 = 0.02 / 2 = 0.01
Z/2 = Z0.01 = 2.326
Margin of error = E = Z / 2 * (( * (1 - )) / n)
= 2.326 * (((0.0.250 * 0.750) / 52)
= 0.140
A 95% confidence interval for population proportion p is ,
- E < P < + E
0.250 - 0.140 < p < 0.250 + 0.140
0.110 < p < 0.390
(11.0% , 39.0%)
(b)
This is the left tailed test .
The null and alternative hypothesis is
H0 : p = 0.40
Ha : p < 0.40
P0 = 0.40
1 - P0 = 1 - 0.40
z = - P0 / [P0 * (1 - P0 ) / n]
= 0.25 - 0.40 / [(0.250 * 0.750) / 52]
= -2.208
This is the left tailed test .
P(z < -2.208) = 0.0136
P-value = 0.0136
= 10% = 0.10
P-value <
Reject the null hypothesis .