9.7
Decision Making: Traditional Method
The process of hypothesis testing based on the traditional method includes calculating the critical value, testing the value of the test statistic usi…
The traditional or classical method involves using the critical value to conclude the hypothesis testing.
As a first step, a hypothesis is stated and expressed symbolically as follows.
For the proportion, mean, or standard deviation of a population, the null and alternative hypotheses are expressed as follows.
Further, a critical value is obtained for the chosen parameter in the hypotheses at a specific predetermined significance level α. For proportion, mean, or standard deviation, these critical values at α are the z, t, or chi-square values, respectively, which are calculated using the z, t, or chi-square distributions.
The critical value is then plotted to demarcate the critical region in the probability distribution.
Further, the test statistic is calculated using the sample data and plotted on the probability distribution curve.
The null hypothesis is rejected when the test statistic value falls within the critical region. However, we fail to reject it when the test statistic falls outside the critical region.
View the full transcript and gain access to JoVE Core videos
Q1: What are the main steps in the traditional method of hypothesis testing?
The traditional method involves stating null and alternative hypotheses symbolically, obtaining a critical value at a predetermined significance level, calculating a test statistic from sample data, and plotting both on a probability distribution. The null hypothesis is rejected if the test statistic falls within the critical region; otherwise, you fail to reject it.
Q2: How do you express null and alternative hypotheses in the traditional method?
Null and alternative hypotheses are expressed symbolically based on the population parameter being tested. By convention, the null hypothesis contains an equality symbol, while the alternative hypothesis contains >, <, or ≠ symbols. The null hypothesis represents a neutral statement, while the alternative hypothesis may have a specific direction or represent the original research claim.
Q3: What test statistic should you use when testing a claim about a population mean with unknown population standard deviation?
When the population standard deviation is unknown, use the t statistic to test a claim about a population mean. The t statistic is calculated from sample data and compared against the critical value obtained from the t distribution at your chosen significance level.
Q4: What significance level should you choose for hypothesis testing?
Common significance levels follow a general convention of 95% confidence (α = 0.05) or 99% confidence (α = 0.01). The significance level determines the critical value and defines the critical region boundaries on the probability distribution where you reject the null hypothesis.
Q5: How do you decide whether to reject the null hypothesis using the traditional method?
Compare your calculated test statistic to the critical region on the probability distribution. If the test statistic falls within the critical region, reject the null hypothesis. If it falls outside the critical region, fail to reject the null hypothesis. This decision does not require calculating the p-value.
Q6: What is the difference between the z statistic, t statistic, and chi-square statistic in hypothesis testing?
The z statistic tests claims about proportions or means when population standard deviation is known. The t statistic tests claims about means when population standard deviation is unknown. The chi-square statistic tests claims about population variance or standard deviation. Each uses its respective distribution to determine critical values.
Q7: Why is the critical value important in the traditional method of hypothesis testing?
The critical value demarcates the critical region on the probability distribution at your chosen significance level. It serves as the boundary for decision-making: if your test statistic exceeds this boundary, you reject the null hypothesis. Critical values are obtained from z, t, or chi-square tables depending on your test statistic type.