91Ó°ÊÓ

The null hypothesis, often denoted as \(H_0\), is a fundamental concept in hypothesis testing. It is a statement that there is no effect or difference in a particular situation. In the context of the problem, the null hypothesis asserts that the drug does not significantly alter reaction times. Essentially, it assumes that any observed changes are due to random chance rather than the drug's effect.
For example, in our psychologist's study, the null hypothesis (\(H_0\)) is that the mean difference in reaction times before and after the injection, \(\mu_D\), is equal to 0. This means firstly we presume there is no real change caused by the drug. Always remember, if our test results indicate it's unlikely to happen under the null hypothesis, then we may consider evidence against it.

A t-test is a statistical method used to determine whether there is a significant difference between the means of two groups. In this exercise, it helps assess if the drug has an impact on reaction times. The t-test is applicable here because we work with a small sample size and need to compare the before-and-after situation.
The general formula for calculating the t-statistic in our study is:

• Calculate the mean of the differences between reaction times before and after the drug is administered, denoted as \(\bar{D}\).
• Calculate the standard deviation of these differences, denoted as \(s_D\).
• Use the formula \( t = \frac{\bar{D}}{s_D / \sqrt{n}} \) where \( n \) is the sample size.

This calculation helps determine if the mean difference is substantial enough to suggest a true effect, compared to what would typically be expected by random variation alone.

The level of significance, often represented as \(\alpha\), is a predetermined threshold that guides decision-making in hypothesis testing. In this example, it is set at \(5\%\) or \(0.05\). It indicates the probability of rejecting the null hypothesis if it is actually true (a Type I error).
Choosing a \(5\%\) level of significance means you are willing to accept a 5% chance of incorrectly concluding that the drug affects reaction times.

• If the p-value (calculated probability of observing the test results under \(H_0\)) is less than \(\alpha\), we reject the null hypothesis.
• Conversely, if the p-value is greater, we do not reject the null hypothesis.

Thus, the level of significance provides a standard criterion for making decisions based on the t-test results.

A critical value identifies the cutoff point beyond which we will reject the null hypothesis. It is dependent on the chosen level of significance and the degrees of freedom in the data. In practice:

• The critical value is determined using a t-distribution table corresponding to the \(5\%\) significance level and \(n-1\) degrees of freedom, where \(n\) is the number of subjects.
• For a one-tailed test, it designates the threshold above which the t-statistic results in rejecting \(H_0\).

When the t-statistic computed from the experiment exceeds the critical value, it demonstrates that the sample data are significantly different enough from what is expected under \(H_0\), providing reason to reject \(H_0\). Conversely, if the t-statistic does not exceed this critical threshold, we do not have sufficient evidence to reject \(H_0\). This role makes the critical value a crucial deciding factor in hypothesis testing.