Professional Documents
Culture Documents
(𝑝1 − 𝑝2 ) − (𝑃1 − 𝑃2 )
𝑧=
1 1
√𝑝𝑞 (
𝑛1 + 𝑛2 )
Where;
𝑝1 – proportion of the first sample
𝑝2 – proportion of the second sample
𝑃1 – proportion of the first population
𝑃2 – proportion of the second population
𝑛1 – size of the first sample
𝑛2 – size of the second sample
Example 5: A random sample of 120 females, where 108 passed Statistics on their first exam, and
another random sample of 80 males, where 60 passed Statistics on their first exam. Perform a test
of hypothesis that the proportions who passed Statistics on their first take differ for males and
females. Used the 0.05 level of significance.
Solution: Let 𝑝1 and 𝑝2 be the proportions of females and males, respectively who passed Statistics
on their firs exam.
Steps:
1. State the null and alternative hypothesis. Mathematically,
Ho: 𝑝1 = 𝑝2
𝐻𝑎 : 𝑝1 ≠ 𝑝2
2. Level of significance α = 0.05.
3. Select an appropriate test statistic.
The test statistic is the z test for proportions, and the formula is
(𝑝1 − 𝑝2 ) − (𝑃1 − 𝑃2 )
𝑧=
1 1
√𝑝𝑞 (
𝑛1 + 𝑛2 )
4. Determine the critical value and critical region
Since the level of significance is 0.05 and the critical value (𝑍α/2 ) = ±1.96.
Reject Ho, if t computed is less than -1.96 or greater than 1.96.
5. Compute the value of the test statistics: Let x – be the number of sample in the first group
possessing the characteristics of interest; y – be the number of sample in the second group
𝑥+𝑦 108+60 168
First, we compute the pooled estimate p = 𝑛 = 120+80 = 200 , the z test value is
1 +𝑛2
108 60
(120 − 80) − 0
𝑧= = 2.83.
√ 168 168 1 1
200 (1 − 200) (120 + 80)
6. Decision: Since the computed Z = 2.83 is in the rejection region, thus, reject Ho at 0.05
level of significance.
7. Conclusion:
Therefore, there is sufficient evidence to conclude that the proportions who passed
Statistics on their first exam differ for males and females.