91Ó°ÊÓ

Understanding the null hypothesis is fundamental in hypothesis testing. It is denoted by the symbol \( H_0 \). The null hypothesis is like the baseline or default assumption about a population parameter that you want to test. It's the statement assumed to be true unless there's substantial evidence against it.
For example, if we're assessing a new drug, the null hypothesis might be that the drug has no effect, i.e., it performs the same as no treatment at all. From a statistical standpoint, the null hypothesis often posits no effect or no difference.
When we conduct a hypothesis test, we're often looking to gather evidence that challenges the null hypothesis. Remembering that the null hypothesis is the status quo can help you understand why it needs to be disproven rather than proven.

The alternative hypothesis is the complement of the null hypothesis. Denoted by \( H_1 \) or \( H_a \), this hypothesis introduces a notion contrary to the null hypothesis. It suggests that there is indeed an effect or a difference in the population parameter being studied.
A well-formulated alternative hypothesis provides a clear answer to the question being studied. It tells us what statement we should be ready to accept if the test statistic indicates significant results.
Consider a situation where you suspect a coin is not fair. The null hypothesis \( H_0 \) would say the coin is fair, while your alternative hypothesis \( H_1 \) would suggest it's biased. Using sample data, your aim is to find strong evidence against the null hypothesis to consider the alternative hypothesis true.

The test statistic plays a crucial role in hypothesis testing, serving as a decision-maker about the null hypothesis. It is a standardized value calculated from the sample data that allows us to determine whether there's sufficient evidence to reject the null hypothesis.
The nature of the test statistic depends on the type of data and the test being performed. Common examples include the z-statistic for large sample sizes from a normally distributed population and the t-statistic for smaller sample sizes.
To compute the test statistic, we use the sample data to measure how far the sample mean is from the hypothetical population mean postulated in the null hypothesis. The bigger the test statistic, the lower the likelihood the null hypothesis is true.

The critical value is a threshold that the test statistic is compared against in the hypothesis test. It is determined based on the selected significance level (\( \alpha \)) which represents the probability of rejecting a true null hypothesis.
Critical values mark the boundary or cutoff region where the null hypothesis is rejected. If the computed test statistic falls into this region, we reject the null hypothesis. The critical value can be found using statistical tables, software, or calculators, depending on the distribution of the test statistic.
For example, with a significance level of 0.05 in a two-tailed test, the critical value might cut off the outer 5% of the normal distribution, indicating that a test statistic beyond this point is unlikely under the null hypothesis.

The significance level, denoted by \( \alpha \), is a probabilistic threshold in hypothesis testing that defines the strength of evidence required to reject the null hypothesis. A common choice is \( \alpha = 0.05 \), meaning a 5% risk of incorrectly rejecting the null hypothesis if it's actually true.
This level reflects how "willing" we are to make a Type I error, which is mistakenly rejecting a true null hypothesis. Lower significance levels indicate a more stringent test; the evidence must be more substantial to reject the null hypothesis.
Choosing the significance level is a key decision in the hypothesis testing process, usually determined before conducting the test. It directly influences the critical value and, consequently, the decision about the null hypothesis.