Hypotheses and Hypothesis Testing

• The hypothesis is a statement of your research question.

• The hypothesis should state a causal relationship or simple proposition that has at least one independent and dependent variable.
  • This relationship or proposition must then be subject to empirical testing.

• Your hypothesis must:
  • Have two or more variables
  • State a causal relationship between those variables

• It also should be:
  • Logically related to your research question and theory.
  • Falsifiable, meaning that it can be empirically tested to be true or false.

• Your research will test this hypothesis and eventually accept or reject the statement.
  • Evidence that supports your hypothesis can lead to retention of the statement.
  • Evidence against your hypothesis mandates its rejection.
  • For quantitative research, you will most likely be using the null hypothesis to test your stated hypothesis.

• The null hypothesis states that there is no relationship between the two variables which you want to test.
  • What the null hypothesis does is allows you to collect evidence that there is a relationship between the two (or more) things you are studying.

• Rejecting the null hypothesis (indicating that there is a relationship) allows your hypothesis to remain a possibility.

• While it does not prove a particular hypothesis, rejecting the null hypothesis can add support for the truth of your hypothesis.

• Tests for acceptance and rejection of a null hypothesis will be discussed later in the statistics section.

  Examples:

  Hypothesis: Sport participation reduces the chances of obesity.

  Null hypothesis: There is no relationship between participation in sports and obesity.

Hypothesis: Baseball attendance is related to income.

Null hypothesis: There is no relationship between baseball attendance and income.

 

Hypothesis: More high scoring games in professional baseball cause higher attendance.

Null hypothesis: There is no relationship between high scoring games and attendance.

• Links to other reading on the topic:
  • http://trochim.human.cornell.edu/OJtrial/ojhome.htm