In quantitative research, we can’t make claims without having some statistics to support it. In other words, it can’t be determined if the null or the alternative hypothesis is supported until the hypothesis has been explored, using the data that was collected.
There are numerous tests that can be used to test a hypothesis. Which test you use will depend on a number of factors. The purpose of this section is to discuss the basics of hypothesis testing.
Steps of Hypothesis Testing:
- State the Null & Alternative Hypotheses. These should reflect the hypothesis test that you anticipate conducting. For example, an alternative hypothesis for an ANOVA will state that there are significant differences between one or more means.
- Determine α. This refers to the amount of Type I error you are willing to make or the chance of determining that the null hypothesis is false when it is in fact true.
- State the decision rule. The decision rule states the circumstances under which the null hypothesis will be rejected. For a research paper, this will be comparing the obtained p-value (level of significance) of the test statistic to the alpha set for the hypothesis. For example, “If p < .05, the null hypothesis will be rejected.”
- Calculate the test statistic. In other words, conduct your hypothesis test with the appropriate statistical analysis. In most cases, this will be done using a statistical analysis tool, like SPSS.
- State the decision regarding the null hypothesis. In statistics, it’s not possible to prove a hypothesis is true/false. There is either evidence to suggest the alternative is true, or there is not. The decision made is based on the decision rule stated in step 3.
- Reject the null hypothesis – In this case, the p-value for the test statistic is less than alpha. There is evidence to suggest the alternative is true.
- Fail to reject the null hypothesis – The p-value from the hypothesis test was greater than alpha. There is not sufficient evidence to suggest the null is not true.
- State the conclusion of the test in layman’s terms. This is where you connect the statistics back to the specific hypothesis being tested. For example, “The results of the hypothesis test indicate that the salary of factory workers varies based on gender.”