1. Come up with a problem.

• For example, does drinking affect one's academic work?

2. Define your terms.

• How will you measure your independent variable, "drinking"?
• How will you measure your dependent variable, "academic work"?

Note:  your independent variable  causes the action.  Your dependent variable is the variable that is acted upon or influenced by the independent variable.

3. Make an hypothesis.

• This is a formal statement about exactly how you think your independent variable will affect your dependent variable. For example, "the higher one's weekly consumption of alcohol is, the lower one's GPA will be." Notice that an hypothesis specifies in what way the independent variable is related to the dependent variable: as drinking goes "up" grades will go "down." This hypothesis states that there is a "negative" or "inverse" relationship between the two variables.

4. Make a "model" of your hypothesis

• During this stage think of all possible "intervening variables" and include them in your research design. In this case, do you think that one's gender or age might influence how drinking affects grades?  If so, you should collect gender and age data so that you can "control" for the possible effects of gender and age.

 

 

Note:  your intervening variable influences how the independent variable affects the dependent variable.

5. Choose your research method.

• Will you run an experiment?
• Take a survey?
• Do participant observation?

6. Collect your data.

• Follow correct procedures. If taking a survey, try to draw a "random sample" of the population that you are curious about.  Ask unambiguous questions that are not biased.  If you are running an experiment, be sure to "randomly" distribute subjects into your experimental and control groups.  If you are engaged in participant observation, strive to remain objective and not identify totally with the group you are studying.

7. Interpret your data.

• Make sure to actually examine the relationship between the two variables in your hypothesis. The simplest way to do this is with a "cross tabulation table" of these two variable. A "three way" cross tabulation table will allow you to examine the possible influence of an intervening variable.


For example, you have surveyed 112 Fairfield University students about their drinking habits last Spring and their GPA for that semester. To test your hypothesis that more drinking will lead to a lower GPA you produced this graph that examines the relationship between drinking and GPA. You must now "interpret" it. Does it confirm your hypothesis?

8. Prove or disprove your hypothesis.

Make a conclusion about the validity of your hypothesis based on the data you collected. You can also mention what further research needs to be done.