How to Test Research Hypothesis

Research hypothesis is a tentative answer to a research problem. It is also referred as a scientific guest work. meaning; it is a testable prediction. It helps a researcher determine the relationship between two variables e.g., the relationship between demand and supply; property users and choice; property location and type of development etc. Test of hypothesis is crucial in research papers.

The main purpose of a hypothesis is enabling your study answer the research questions. To formulate a good research hypothesis, do the following;

Take cognizance of your research questions

Make sure they are testable

Make sure they can be measured

It must contain two variables (dependent and independent variables).

The hypothesis is of two types;

Alternate hypothesis: This is represented as H₁. It tries to test and give positive answer to a research question by fostering a significant or possible relationship between two variables.

Null hypothesis: This is represented as H_o. It usually gives a reverse view of the Alternate hypothesis. Null means Zero i.e., zero significance between two variables or zero relationship between two variables.

Note that formulated hypothesis (alternate or null) is subject to verification through investigation. It is after the investigation that your conclusion will either reject or support the formulated hypothesis.

To better answer your research questions; hypothesis must relate to the questions.

SAMPLE:

Hypothesis Formulation

The following hypothesis was postulated and tested.

H₁:  There is a significant relationship between User-Demand preferences and residential accommodations in New Haven Layout, Enugu.

H_o:  There is no significant relationship between User-Demand preferences and residential accommodations in New Haven Layout, Enugu.

The Alternate hypothesis states that the two variables have significant relationship while Null hypothesis gives an opposite view. With these two sides the researcher will have to rely on the conclusion drawn from investigations involving test of above hypothesis using reliable statistical models to decide whether to adopt the Alternate or Null hypothesis. This clearly means that one of the hypotheses will be rejected (i.e., if H₁ is rejected, H_o will be accepted, on the other hand if H₁ is accepted, H_o will be rejected)

After collecting and analyzing data from questionnaires, you will have to ask yourself these questions;

Which is the most appropriate test to use on the data, should One-Way ANOVA, Chi-square, Pearson’s Correlation, t-test, Regression etc. be adopted?

Are you going to use the manual way to test hypothesis or software such as SPSS, EViews etc.?

What research data will be most appropriate to test?

Where should test of hypothesis be positioned (after each table with data to be tested or at the end of the entire analysis)?

Steps to Testing Your Hypothesis:

State your research hypothesis (you can find this in your chapter one).

State your significance or Alpha level (5% or 10%)

State the determinant (criterion) for decision on outcome of result in your test of hypothesis(es)

Calculate and obtain the test statistics (this can be done manually or by use of software e.g., SPSS etc.);

Decide whether the result is significant (this depends on your adopted level of significance).

If your calculated figure is less than your critical value, then the result of your research is not significant, meaning that you will accept the null hypothesis. But if the calculated figure is equal to or more than the critical value in the statistical table, then your results are significant; hence you should reject the null hypothesis and accept the alternate hypothesis.

If p-value<0.05, it means that the probability of a result is less than 5%. If p-value<0.01 this means that the probability of the result due to chance is less than 1%. The lower the probability figure, the more your level of confidence in concluding that there is significance in your data. If p<0.05, it means that p-value is less than the significant level, hence the null hypothesis is rejected.

Write your conclusion: This could be either in favor of the null or alternate hypothesis depending on the outcome of the test.