Hypothesis Testing

 What is hypothesis testing?

Hypothesis testing can be defined as an act in the field of statistics that allows a statistician to test an assumption concerning any parameter within a population. Deferent methodologies are used in testing a hypothesis. It is, therefore, the duty of the statistician to choose the methodology that is best suited the research they are conducting and the kind of data that would be used (Tao & Shi, 2008). The method chosen will then be applied to the data to help determine or test the hypothesis.

The general process and the significant steps in hypothesis testing

The general process involves testing a statistical sample, with the primary objective being rejecting or accepting a null hypothesis. This test will help the statistician know if their central hypothesis is true or null. If it is a false hypothesis, then a new hypothesis has to be formulated and tested until the data shows the true hypothesis (Tao & Shi, 2008). The four main steps used in hypothesis testing include:

1. The first step is to set up the two hypotheses with the aim of ensuring that only one turns out to be true
2. The second step will be to develop an analysis plan that will be used in evaluating the date 
3. The third step is to implement the plan and analyze the data
4. The final step would be to analyze the results, and after that, reject or accept the null hypothesis.

The difference between parametric and non-parametric hypothesis testing

There is one significant difference between parametric and non-parametric hypothesis testing. Parametric statistical hypothesis testing makes assumptions regarding the parameters of the population by defining different properties in a population distribution (Tao & Shi, 2008). These assumptions are not made by non-parametric hypothesis testing. 

References

Tao, J., & Shi, N. (2008). Statistical Hypothesis Testing: Theory and Methods. New York, NY:

World Scientific.