91Ó°ÊÓ

The null hypothesis, denoted as \( H_0 \), plays a central role in hypothesis testing. It is typically a statement that implies there is no effect or no difference in the population being studied. In simpler terms, it's a statement of status quo or no change. This is why the null hypothesis often includes an equality sign. The null hypothesis is what we test to see if the data tells us something different about our assumptions.

Let's consider an example from our exercise: "In Canada, the proportion of adults who favor legalized gambling equals 0.50." This is a statement where we assume no change or effect in the population proportion regarding adults' opinions. Therefore, it is a classical example of a null hypothesis because it states that the proportion \( p \) equals 0.50, showing no variance from the past or what is currently perceived as true. By testing \( H_0: p = 0.50 \), we aim to see if there's any statistical evidence strong enough to reject this assumption.

The alternative hypothesis, denoted as \( H_a \), represents what we hope to prove or believe might be true about a population parameter. Unlike the null hypothesis, it suggests a difference or an effect, meaning a change from the current understanding or previous measurements.

The alternative hypothesis usually involves an inequality, indicating how the population parameter might differ from the null hypothesis. Let's break it down with statement b from our exercise: "The proportion of all Canadian college students who are regular smokers is less than 0.24, the value it was 10 years ago." Here, the statement suggests that the rate is now less than what was recorded previously, forming the alternative hypothesis \( H_a: p < 0.24 \). This hypothesis directs the testing towards proving or disproving the change or difference claimed since 10 years ago.

In hypothesis testing, a population parameter is a quantity that describes a numerical characteristic of an entire population, such as a mean (\( \mu \)), a proportion (\( p \)), or a standard deviation (\( \sigma \)). Identifying the relevant population parameter is crucial because it is directly linked to the hypotheses being tested.

For example, in both scenarios given in the exercise, the parameter is the population proportion, denoted by \( p \). In part a, we introduced \( p \) as the proportion of adults who favor legalized gambling in Canada. With part b, \( p \) represents the proportion of Canadian college students who are regular smokers. Each hypothesis in these examples is expressed in terms of \( p \), indicating what we perceive the population parameter to be or how we expect it might have changed. To test hypotheses about population parameters, we gather sample data and use statistical methods to decide whether to reject the null hypothesis based on the evidence presented by the sample.