Explaining Null Hypothesis in Simple Terms

At a first glance, null hypothesis appears to be one of those incomprehensible and seemingly unnecessary steps that a researcher must conduct in order to produce a legitimate work of academic research. In reality, of course, null hypothesis is of great use; if you can show its applicability in a research study you're planning, you will save yourself from spending loads of your time and effort exploring the phantom connection between your chosen independent and dependent variables - the connection that you have already proven is not there by showing that your null hypothesis is true.

Null hypothesis is a hypothesis of no effect - in other words, it directly contradicts the hypothesis of the proposed research by stating that the independent variable (cause) will not have any influence on the dependent variable (effect), and that any differences between the control and experimental groups in the chosen research are due to chance.

The main application of null hypothesis is to test the significance of the proposed hypothesis. The main goal of the researcher is to refute or nullify (hence the name) it so as to show that changes in the dependent variable indeed occur due to interaction with the independent variable, rather than due to some random causes.

As an example, lets take a study in which the researcher is hypothesizing the existence of a connection between specifics of a Russian cultural upbringing and instances of domestic violence among Russian-speaking immigrants in a major American city. In this case, null hypothesis will state that unique specifics of the Russian culture have no influence on the instances of domestic violence among the members of the group being studied. One of the ways to test it would be to compare life experiences of Russian-speaking immigrant women who are victims of domestic violence with those of native-born American women who also have experienced domestic violence - who, in this case, would be the control group. If no differences are found between the two groups in the frequency of domestic violence and reported reactions of female victims to it, then null hypothesis is correct, and cultural background does not influence domestic violence in the Russian-speaking community of the named city.

If this gives you an impression that you must conduct a whole separate research study before embarking on your own, you may be right. However, this would be true only if the subject of your research is so original and/or rare that no similar field studies exist that can provide you with empirical evidence. In other words, there is a very good chance that someone has already done something similar to what you are planning to do and has come up with solid numbers that you can use in your own study to see whether your null hypothesis actually applies.