Hypothesis testing in statistics refers to the process by which a testable belief or opinion is tested by statistically analyzing data collected from an observational study or controlled experiments. Hypothesis testing is used to verify beliefs about an attribute or some measure of a population based on analyzing data from a sample drawn from the population. Hypothesis testing is quicker and saves time needed to collect data and make business decisions. Secondly, hypothesis testing is a cheaper means of verifying certain opinions about a population since it involves collecting information from only a small proportion of the population. Lastly, in tests that are destructive, hypothesis testing may be the only feasible means of verifying presumed information concerning the entire population.
Null hypothesis, which is normally denoted by Ho, is often stated to imply that no variation exists between the expected population parameter and the parameter obtained from the sample. It is stated to comply with the opinion that is assumed before the hypothesis test is conducted. Alternate hypothesis, which is often denoted by H1, is stated as the opposite of the null hypothesis. In this case, the null hypothesis will be; there is no significant difference between the presumed average satisfaction score of the 4 million customers and the sample satisfaction score of 4.5. The alternate hypothesis will be; there is a significant difference between the presumed average satisfaction score of the 4 million customers and the sample average satisfaction score of the 1,100 customers.
It is essential to frame both the null hypothesis and alternative hypothesis properly because it is the foundation of statistical hypothesis testing. The null hypothesis should always reflect what we will conclude, that is the presumed population parameter, if there is no reason for the researcher to think differently. This enables the researcher to define the decision rule and hence a clear hypothesis criteria.
Healey, J. F. (2011). Statistics: A Tool for Social Research (9 ed.). London: Cengage Learning.
Johnson, R. A., & Bhattacharyya, G. K. (2009). Statistics: Principles and Methods (6 ed.). New York: John Wiley and Sons.