91Ó°ÊÓ

Hypothesis testing is a crucial statistical tool used to make decisions about data. It helps determine whether an observed effect is due to chance or significant enough to warrant further consideration. In this context, we are examining whether the mean income of those choosing Plan B is greater than those choosing Plan A.

To conduct a hypothesis test, you start with two statements:

• **Null Hypothesis (H_0):** This suggests there is no significant difference or that any observed difference is due to sampling variation. For our problem, it is stated as \( \mu_B \leq \mu_A \)\.
• **Alternative Hypothesis (H_1):** This indicates an actual effect or difference exists. In the example, it is \( \mu_B > \mu_A \)\.

After setting these hypotheses, the next steps involve calculating test statistics, which determine whether we should reject or fail to reject our null hypothesis.

Degrees of freedom (df) play a vital role in various statistical methods, including the t-test. They represent the number of independent elements that can vary within your sample data. In a two-sample t-test, they help adjust for sample size differences and are used to determine the appropriate distribution to refer to when calculating the p-value.The formula for calculating degrees of freedom in a two-sample t-test, especially when the variances of the samples are unequal, is:\[df = \frac{\left( \frac{s_A^2}{n_A} + \frac{s_B^2}{n_B} \right)^2}{\frac{\left( \frac{s_A^2}{n_A} \right)^2}{n_A - 1} + \frac{\left( \frac{s_B^2}{n_B} \right)^2}{n_B - 1}}\]

This expression accounts for the variability and size differences in the samples. In our classroom example, the degrees of freedom calculated approximately to 56, ensuring that we can accurately compare the statistics to the t-distribution for decision making.

The p-value is a statistical metric that helps determine the significance of your hypothesis test results. It tells us the probability of observing a test statistic as extreme as, or more extreme than, the observed value, assuming the null hypothesis is true.

If the p-value is low (typically below the significance level like 0.05), it suggests that the observed data is inconsistent with the null hypothesis, leading us to reject the null hypothesis in favor of the alternative. In our two-sample t-test example, the calculated p-value was approximately 0.05.

• If \( p \leq \alpha \), reject the null hypothesis.
• If \( p > \alpha \), fail to reject the null hypothesis.

Here, since the p-value matches the significance level of 0.05, it falls right on the boundary of significance, making the decision a bit more nuanced. It suggests that, under these conditions, we do not have strong evidence to proclaim Plan B's mean income higher.

The significance level, often denoted by \( \alpha \), is a threshold set by the researcher to determine when an effect is statistically significant. It reflects the probability of rejecting the null hypothesis when it is actually true (a type I error). Commonly used significance levels include 0.01, 0.05, and 0.10.

In our example with the cell phone plans, we used a significance level of 0.05. This means we are willing to accept a 5% risk of incorrectly concluding that Plan B's mean income is higher than Plan A's. When we set \( \alpha = 0.05 \), it serves as a guideline to assess our p-value.

• **\( p \leq \alpha \)**: Suggest significant results prompting rejection of \( H_0 \).
• **\( p > \alpha \)**: Indicates insufficient evidence against \( H_0 \), so we fail to reject it.

This threshold guides researchers in making more informed decisions about their hypotheses.