91Ó°ÊÓ

The P-value is a fundamental concept in statistics used during hypotheses testing. It helps us understand how likely it is to obtain a test statistic as extreme as the one observed, assuming that the null hypothesis is true.
A small P-value indicates that the observed data is unlikely if the null hypothesis is true. Thus, it may lead us to reject the null hypothesis. Conversely, a large P-value suggests that the observed data is not unusual, supporting the null hypothesis.

• If the P-value is less than the significance level (commonly 0.05), we reject the null hypothesis.
• If the P-value is greater than the significance level, we do not reject the null hypothesis.

In the exercise given, the one-sided test for the alternative hypothesis of \(H_{a} : \mu > 5\) provides a P-value of approximately 0.044. This suggests the evidence against the null hypothesis is strong enough at a 5% significance level, but not at a 1% significance level.

The significance level is pre-determined and helps to decide whether to reject the null hypothesis. It's denoted by \( \alpha \) and often set at 0.05 or 0.01, depending on the desired level of confidence.
When the P-value is below this threshold, we conclude there is statistically significant evidence against the null hypothesis. Here's how it works:

• Significance level of 5% (\( \alpha = 0.05 \)) implies there is a 5% risk of concluding that a difference exists when there is none.
• Significance level of 1% (\( \alpha = 0.01 \)) implies a stricter criterion, requiring stronger evidence to reject the null hypothesis.

In the exercise, the conclusions drawn were dependent on the chosen significance level. At \( \alpha = 0.05 \), we rejected the null hypothesis for the one-sided test as the P-value (0.044) was less than 0.05. However, at \( \alpha = 0.01 \), the P-value was greater than 0.01, and hence we retained the null hypothesis.

Hypotheses testing is a statistical process used to determine if there is enough evidence in a sample to support a particular claim about a population parameter. The typical framework involves:

• The null hypothesis \( H_0 \): A statement of no effect or no difference, which we aim to test against.
• The alternative hypothesis \( H_a \): A statement we may accept if the evidence strongly contradicts \( H_0 \).

Two types of tests are often discussed: one-sided and two-sided tests. A one-sided test, as seen in the exercise when testing \( H_a : \mu > 5 \), focuses on an effect in one direction. A two-sided test looks for differences in both directions, such as in part (b) where \( H_a : \mu eq 5 \) was tested.
Conclusions drawn from these tests depend on the comparison of the P-value with the pre-set significance level. Rejecting or not rejecting the null hypothesis leads us to infer statistical significance or the lack thereof.

The T-distribution is a probability distribution that is used when estimating population parameters when the sample size is small and the population variance is unknown. It is similar to the normal distribution but has thicker tails. This implies that it accounts for higher variability in small samples.
Key properties include:

• Dependent on degrees of freedom, which is calculated as sample size minus one \( n-1 \).
• Approaches a normal distribution as sample size increases.

In the problem provided, with a sample size of 20, the degrees of freedom were 19. Using the T-distribution allows for finding critical values of the test statistics and the P-value accurately. This distribution is essential in executing the t-test, as seen when calculating the range of the P-value using Table B, and later finding the exact P-value using a calculator.