91Ó°ÊÓ

The z-score is a crucial component in hypothesis testing as it measures how many standard deviations an element is from the mean. This tells us where our sample mean falls in the distribution. Knowing where your sample mean lies can help determine how far it is from the hypothesized mean and whether it supports your hypothesis. To calculate the z-score, use the formula: \[ Z = \frac{\bar{x} - \mu_{0}}{\frac{\sigma}{\sqrt{n}}} \] where:

• \(\bar{x}\) is the sample mean,
• \(\mu_{0}\) is the hypothesized population mean,
• \(\sigma\) is the population standard deviation,
• \(n\) is the sample size.

For a sample mean of 9.24, hypothesized mean 8.5, and a standard deviation of 5.40 with a sample size of 18, calculate: \[ Z = \frac{9.24 - 8.5}{\frac{5.40}{\sqrt{18}}} \] This results in a z-score of approximately 0.70. This value places the observed sample mean relative to the hypothesized mean in the distribution space. The z-score is then used to determine if the observed sample mean is significantly different from the population mean under the null hypothesis.

Critical values determine the boundary at which you decide to reject the null hypothesis. They separate the rejection region from the non-rejection region in a statistical test. To find the critical values, refer to a z-table or standard normal distribution table. The location of these critical values depends on the chosen significance level \( \alpha \), which in turn defines the probability of rejecting a true null hypothesis.For a two-tailed test with \( \alpha = 0.05 \), split the significance level into \( \alpha/2 = 0.025 \) for each tail. From the z-table, this gives critical values at \(-1.96\) and \(1.96\). These values define the cutoff points where if your z-score falls beyond them, you reject the null hypothesis. In a one-tailed test scenario with \( \alpha = 0.05 \), the critical value is determined directly from \( \alpha \), resulting in a single cutoff point at \(1.645\) or \(-1.645\) based on the test direction. The choice between a one-tailed and two-tailed test directly affects the selection and interpretation of critical values.

A two-tailed test is used when you are testing for the possibility of an effect in either direction. In other words, you are checking if there is a significant deviation from the hypothesized mean, regardless of the direction. Use a two-tailed test when the alternative hypothesis is of the form \( H_1: \mu eq \mu_0 \). Here, the critical regions are located in both tails of the distribution. This means you have two critical values: one on the left and one on the right.Consider a significance level of \( 0.05 \), split into two critical regions with probabilities of \( 0.025 \) in each tail. A z-score that lands outside the range of \(-1.96\) to \(1.96\) (critical values for \(\alpha/2 = 0.025\)) would lead you to reject the null hypothesis.This approach ensures you remain unbiased by testing for extremes in both directions. Two-tailed tests are commonly used when initial predictions about the direction of the effect or deviation are not appropriate or unknown.

One-tailed tests focus on finding an effect in a specific direction, either greater than or less than the hypothesized parameter. This focus can offer more power to detect an effect in that specified direction since the entire significance level is dedicated to that one side.If your alternative hypothesis is structured like \( H_1: \mu > \mu_0 \) or \( H_1: \mu < \mu_0 \), use a one-tailed test. This setup tests for the possibility of the value being either significantly greater or significantly less than the hypothesized mean.With a significance level of \( \alpha = 0.05 \), the critical value corresponds to \( 1.645 \) (or \(-1.645\), depending on the test's direction). If your calculated z-score surpasses this critical value (in the desired direction), reject the null hypothesis.One-tailed tests are beneficial when you have a clear prediction about the direction of the effect or simply want to test for an increase or decrease, allowing for a more focused statistical insight.