This is the conventional approach to making a prediction. It involves a statement that says there is no relationship between two groups that the researcher compares on a certain variable. The hypothesis also may state that there is no significant difference when different groups are compared with respect to a particular variable. For example, “There is no difference in the academic performance of high school students who participate in extracurricular activities and those who do not participate in such activities” is a null hypothesis. In many cases, the purpose of a null hypothesis is to allow the experimental results to contradict the hypothesis and prove the point that there is a definite relationship.
Certain hypothesis statements convey a relationship between the variables that the researcher compares, but do not specify the exact nature of this relationship. This form of hypothesis is used in studies where there is no sufficient past research on which to base a prediction. Continuing with the same example, a nondirectional hypothesis would read, “The academic performance of high school students is related to their participation in extracurricular activities.”
This type of hypothesis suggests the outcome the investigator expects at the end of the study. Scientific journal articles generally use this form of hypothesis. The investigator bases this hypothesis on the trends apparent from previous research on this topic. Considering the previous example, a researcher may state the hypothesis as, “High school students who participate in extracurricular activities have a lower GPA than those who do not participate in such activities.” Such hypotheses provide a definite direction to the prediction.
Some studies involve a measurement of the degree of influence of one variable on another. In such cases, the researcher states the hypothesis in terms of the effect of variations in a particular factor on another factor. This causal hypothesis is said to be bivariate because it specifies two aspects -- the cause and the effect. For the example mentioned, the causal hypothesis will state, “High school students who participate in extracurricular activities spend less time studying which leads to a lower GPA.” When verifying such hypotheses, the researcher needs to use statistical techniques to demonstrate the presence of a relationship between the cause and effect. Such hypotheses also need the researcher to rule out the possibility that the effect is a result of a cause other than what the study has examined.