Demonstrating Cause-and-Effect Relationships
Researchers may use experiments when they want to demonstrate that a cause-and-effect relationship exists between variables. In order to show that a change in one variable causes a change in another, these three conditions must be fulfilled:
1. You must show that a correlation exists between the two variables. CorreIati~n exists when two variables are associated more frequently than could be expected by chance (Hoover. 1992). For exampie. suppose that you wanted to test the hypothesis that the availability of a crisis intervention center with a twenty-four-hour counseling "hotline" on your campus causes a change in students' attitudes toward suicide (see" Figure 2.4). To demonstrate correlation. you would need to show that the students had different attitudes toward committing
suicide depending on whether they had any experience with the crisis intervention center.
2. You must ensure that the independent variable preceded the dependent variable. If differences in students' attitudes toward suicide were evident before the students were exposed to the intervention center. exposure to the center could not be the cause of these differences.
3. You must make sure that any change ill the dependent variable was not due to an extraneous variable- one outside the stated hypothesis.
If some of the students receive counseling from off-campus psychiatrists. any change in attitude that they experience could be due to this third variable and not to rhehotline. This is referred to as a spurious correlatioll-the association of two variables that is actually caused by a
third variable and does not demonstrate a cause-and-effect relationship.