Formulation of the hypotheses- This involves identifying and clearing stating the null and the alternative hypotheses. It is important to ensure that the hypothesis are SMART. Usually, the hypothesis are developed to address the research question.
Choose the appropriate test –This is usually dependent on exactly what you want to achieve. Some of the most common statistical tests include ANOVA, t-test, regression test, correlation test, and discriminant analysis and decision trees amongst others. It is also at this point that the researcher can decide on which statistical software to use such as the SPSS, Excel, R, SAS, Minitab and STATA amongst others.
Determine your alpha level-The alpha level is simply the level of significance which you will use to either reject or “accept” the null hypothesis. The alpha level can be 90%, 95% or even 99%.It is important to note that the smaller the alpha level the less the risk of committing type I error.
State the decision rule-Often, the decision rule is associated with either the P value of the test statistic or the test statistic itself. When using the P value, it is usually compared to the level of significance. If the P value is less than the level of significance then the decision is to reject the null hypothesis and vice versa. If the test statistic is less than the critical value then the decision is to reject the null hypothesis and vice versa.
Collect relevant data collected perform the statistical test-It this point, the researcher has to make decision on the type of data to collect, type of variables and their levels of measurement, target population, sample size and the methods of data collection. Example of data collection methods include observation, experimentation, interviews and questionnaires. Data collected can either be primary or secondary/existing.
Decision making-This involves stating the test statistic and its P value then making the decision of whether to reject/fail to reject the null hypothesis.
Draw conclusion- What inferences about the target population can be made from the data analysis conducted? This is often referred to as the economic or market conclusion