What is a Hypothesis? Steps in Testing Hypothesis

A hypothesis (hypotheses) is a statement, a proposition (suggestion) about how something might work or behave.

You can develop your own hypothesis on the grounds of informal observation or your own experience if you wish to do so. You may also develop it from an examination of the existing literature.

The intention of your research should be to prove or disprove your hypothesis.

Again, similar to a research question, you do not necessarily need to provide an exact answer, but you must ensure that you have covered the issue at length and provided critical analysis of the outcomes.

Good Hypotheses must be:

• Clear and easy to understand
• Specific, with a definite focus
• Answerable – it must be possible to collect the necessary data
• Substantively relevant to your area of study

It is likely that different subject areas and research approaches will have tradition of using either research questions or hypotheses, often related to quantitative or qualitative studies.

How to come up with a hypothesis

To come up with a testable hypothesis, you should consider your broad research question or aim, and make it more specific by adding details of method, target population, etc.

Here’s an Example:

You work for a medium-sized chain of electronic products. Marketing surveys suggest customer satisfaction levels are low, and you have been tasked with improving customer satisfaction.

Your research question is ‘How can I improve customer satisfaction?’ Discussion with colleagues leads you to believe that part of the problem lies with staff training.

Your initial hypothesis is: ‘training staff in customer service skills will improve customer satisfaction.’

So how can you test this hypothesis?

You can ‘train staff in customer service skills’ by training them to a higher level in customer service (a higher NVQ level). You will know if customer satisfaction has improved by compare the results of customer service questionnaires before and after the staff training programme (you need to use the same questionnaires to compare the results).

So your updated hypothesis will be: ‘training retail staff to NVQ level 2 in customer service skills will improve customer satisfaction, as measured by standardised questionnaires.’

Writing the hypothesis in this way will enable you to start thinking about how to implement your research, research design, and so on.

Steps in Testing Hypothesis

A hypothesis is a proposition about a relationship between two or more variables. A hypothesis is subject to testing since it consists of two or more variables that are measurable or potentially measurable.

The important steps in testing the hypothesis are as follows:

Step 1: State the Research Hypothesis. (H1)
Example: With regard to the reading habits, there is a significant difference between undergraduate and post-graduate students.

Step 2: Formulate the Null Hypothesis (H0)
With regard to the reading habits, there is no significant difference between undergraduate and post-graduate students.

Step 3: Choose a statistical test
Use Chi-square statistic (X2) to test the relationship between the variables considered in the research hypothesis.

Step 4: Specify a significance level. You can suppose to test the hypothesis at 0.05 level of significance.

Step 5: Compute the statistical test
The researcher cross-tabulate the data and compute Chi-square test. Let us assume that on computation of the test the yielded value of 6.78, df =1.

Step 6: Reject/accept the H0
The null hypothesis is rejected since the calculated value of Chi-square is more than the critical value.

Step 7: Draw the inference i.e., accept/reject H1

The research hypothesis is accepted as the null hypothesis is rejected. Thus, we infer that there is a significant difference between undergraduate and post-graduate students with regard to their reading habits.