What Hypothesis test is
A hypothesis test is a statistical procedure used to evaluate two mutually exclusive hypotheses about a population parameter. The two hypotheses are referred to as the null hypothesis and the alternative hypothesis. The null hypothesis is typically a statement of “no effect” or “no difference”, while the alternative hypothesis is typically a statement of “some effect” or “some difference”. The goal of a hypothesis test is to determine which of the two hypotheses is best supported by the data.
The steps for conducting a hypothesis test include:
- State the null and alternative hypotheses.
- Select the appropriate test statistic.
- Calculate the value of the test statistic.
- Determine the critical value and make a decision.
- Interpret the results.
Examples
-
An automobile manufacturer wants to determine if the fuel economy of their new car model is better than the industry average. They take a sample of cars and measure the fuel economy of each car. They then conduct a hypothesis test to determine if the observed mean is significantly better than the industry average.
-
An online retailer wants to determine if their new website design is more user-friendly than the existing website. They survey a random selection of customers and ask them to rate their experience on a scale from 1-10. They then conduct a hypothesis test to determine if the observed mean rating is significantly higher than the rating of the existing website.