Introduction Section and Hypotheses Assignment
Creating Hypotheses

A hypothesis is a statement usually derived from a theory that makes a prediction. There are several types of hypotheses.

Difference and Association Hypotheses

Some hypotheses predict that two groups are different or that the exposure to something causes change. These are called hypotheses of difference . Here is an example.

The threat of mandatory arrest deters potential domestic abusers from engaging in domestic violence.

This hypothesis predicts that potential domestic abusers will change their behavior (e.g., decide not to abuse their spouses) when they believe they will be arrested for doing so. Furthermore, this hypothesis proposes that domestic abusers will not be deterred without the threat of an arrest.

Some hypotheses predict that things are associated, that is, they happen together. These are called hypotheses of association . Here is an example.

The adoption of a mandatory arrest policy for domestic abusers is associated with a reduction of domestic violence incidents in a community.

The distinction between hypotheses of difference and hypotheses of association is most critical during the analysis phase of the research process. The type of hypothesis and the level at which the data are measured determine the statistical techniques that are available to the researcher.

Null and Alternative Hypotheses

Hypotheses come in pairs. Research is intended to gather and analyze information or data in order to support or falsify a hypothesis. These hypotheses can either predict a difference or establish an association. The hypothesis the researcher wants to prove is normally referred to as the alternative hypothesis , but may also be referred to as the research hypothesis .

The null hypothesis is a statement of no difference or association. The structure of research requires the researcher to develop a null hypothesis for each alternative hypothesis. Here is an example:

The Null Hypothesis (Ho):

The threat of mandatory arrest does not influence potential domestic abusers from engaging in domestic violence.

The Alternative Hypothesis (Ha):

The threat of mandatory arrest deters potential domestic abusers from engaging in domestic violence.

Notice that in the previous example the null hypothesis is not the opposite of the alternative hypothesis. For the null hypothesis to be the opposite it would have to state that the threat of mandatory arrest encourages domestic abusers to engage in domestic violence. Instead, the null hypothesis merely states that the threat of mandatory arrest does not influence potential domestic abusers in any way.

The Researcher’s Decision

The focus of every research project is initially on the null hypothesis. After gathering and analyzing the information or data the researcher must answer the following question.

Does the information or data I collected and analyzed support the null hypothesis?

If the answer to this question is yes, the researcher accepts the null hypothesis and rejects the alternative hypothesis. In other words, the researcher failed to support his original alternative hypothesis. This, however, should not be considered a failure. Instead, the researcher has contributed to the literature and now should be challenged to revise his alternative hypothesis and begin a new research project.

If the answer to this question is no, the researcher rejects the null hypothesis as false and accepts the alternative hypothesis. In other words, the researcher supported his original alternative hypothesis.

Hypotheses Formulation

Hypotheses have at least two variables, one dependent (the effect) and one independent (the cause).

In our example:

· The independent variable (or cause of the dependent variable) is: The threat of mandatory arrest.

· The dependent variable (or the result of the independent variable) is: Engaging in domestic violence.