91Ó°ÊÓ

When conducting a hypothesis test, the null hypothesis (often denoted as \( H_0 \)) serves as the starting point. It is a statement that there is no effect or no difference, and it is assumed to be true until evidence suggests otherwise. In simpler terms, it's like saying, "We believe things are as usual until we have enough proof to state otherwise."
In the given exercise, the null hypothesis proposes that the proportion of successful hits or "kills," with the new radar system is at most 0.8. This translates to \( H_0: p \leq 0.8 \), where \( p \) is the unknown true proportion of hits. Here, the null hypothesis essentially claims that the new system is not any better than the existing one in terms of effectiveness.
Importantly, the null hypothesis can never be proven true. Instead, we might fail to reject it if the evidence is insufficient.

The alternative hypothesis (\( H_a \)) contradicts the null hypothesis. It is what researchers aim to support with evidence. Think of it as the statement that changes or differences do exist and need to be recognized. In this case, the alternative hypothesis claims the proportion of kills with the new system is greater than 0.8: \( H_a: p > 0.8 \).
This hypothesis represents the assertive claim that the new radar system is actually an improvement. Here, your hypothesis test will attempt to gather evidence in support of this claim. If the alternative hypothesis is accepted, it hints at potential enhancements in the system's performance.
Rejecting the null hypothesis in favor of the alternative indicates statistical significance in your results, pointing towards an improvement with the new radar device.

The significance level is a crucial component in hypothesis testing that helps determine the threshold for rejecting the null hypothesis. It is denoted by \( \alpha \) and represents the probability of making a Type I error — rejecting the null hypothesis when it is actually true.
In simpler terms, it sets the bar for how extreme the sample data must be before we reject \( H_0 \). In the given exercise, the significance level is 0.04, meaning there is a 4% risk of mistakenly declaring that the new system is superior when it might not be.
Lesser the \( \alpha \), higher the confidence needed to reject \( H_0 \). While common \( \alpha \) levels are 0.05 or 0.01, the chosen 0.04 level suggests a balanced approach between risk and caution. This value helps you decide alongside the test statistic whether to support \( H_a \) or not.

The Z-test is a statistical method used to determine if there is a significant difference between the observed sample statistic and the claimed population parameter. It is particularly useful for large sample sizes, typically those above 30.
In this testing scenario, we apply the Z-test for proportions. This particular test is ideal when you want to compare the sample proportion to the population proportion. For our radar system exercise, it's used to check if the sample proportion of 0.833 is significantly greater than the assumed population proportion of 0.8.
The calculated Z-statistic in this scenario was 1.79, as derived from the formula:

• \( z = \frac{\hat{p} - p_0}{\sqrt{\frac{p_0(1-p_0)}{n}}} \)

Where \( \hat{p} \) is the sample proportion, \( p_0 \) is the population proportion, and \( n \) is the sample size.
This test statistic is then compared with the critical value derived from the significance level, and in this case, it exceeded the critical value of 1.75 — leading to the rejection of the null hypothesis.