91Ó°ÊÓ

When conducting hypothesis testing in statistics, determining the null and alternate hypotheses is a crucial first step. The null hypothesis, represented as \(H_0\), is a default assumption that there is no effect or no difference, which we aim to test against the evidence. In our coin example, it states that the Belgian Euro coin has a fair chance of landing heads up \(p = 0.5\).

On the other side, we have the alternate hypothesis \(H_1\), which suggests that the initial assumption of the null hypothesis is false. It is usually set up to demonstrate what we are trying to prove; in this case, that the coin is biased \(p eq 0.5\).

The objective of hypothesis testing is to determine whether we can reject the null hypothesis, thereby providing evidence that supports the alternate hypothesis.

The test statistic is a value used to determine whether to reject the null hypothesis. It compares the observed data to what is expected under the null hypothesis. To calculate the test statistic for sample proportions, we use the formula:
\[ Z = ( \hat{p} - p_0 ) / (\sqrt{p_0*(1-p_0)/n}) \]
where \(\hat{p}\) is the sample proportion, \(p_0\) is the hypothesized population proportion, and \(n\) is the sample size. For our coin-tossing example, \(\hat{p}\) is 0.56, \(p_0\) is 0.5, and \(n\) is 250. Inserting these values into the calculation provides us with our test statistic, which is then compared to the critical value associated with our chosen significance level.

In hypothesis testing, the significance level \(\alpha\) represents the probability of rejecting the null hypothesis when it is actually true, known as a Type I error. It is a threshold for determining the statistical significance of the test. Common choices for \(\alpha\) are 0.05, 0.01, and 0.10.

If the absolute value of the test statistic is greater than the critical value associated with \(\alpha\), we reject the null hypothesis, suggesting the sample provides sufficient evidence that the null hypothesis may not be true. With our coin example, we initially used \(\alpha=.01\) but also considered the commonly used level of \(\alpha=.05\). The choice of \(\alpha\) reflects how stringent we are in our decision-making process; a smaller \(\alpha\), such as 0.01, requires stronger evidence to reject the null hypothesis than a larger \(\alpha\) would.

Sample proportion analysis deals with summarizing the proportion of items in a sample that have a particular attribute. In our scenario, we want to know the proportion of times the coin lands heads up. We use the formula: \(\hat{p} = x/n\) where \(x\) is the number of heads and \(n\) is the total number of spins.

In our calculation, there were 140 heads in 250 spins, thus \(\hat{p} = 140/250 = 0.56\). We analyze this sample proportion in the context of the null hypothesis and using the test statistic to determine if the observed proportion \(\hat{p}\) is significantly different from the hypothesized proportion under \(H_0\).

This analysis informs us whether the deviation observed (from \(p_0 = 0.5\)) is simply due to the randomness of drawing a sample or if it's statistically significant enough to suggest a true bias in the coin.