Graziano & Raulin
Research Methods (9th edition)
Judging Non-Equivalent
Control-Group Results
Listed below are several examples of non-equivalent control-group studies. Think about each study and decide how close the groups are to being equivalent and whether you can feel reasonably confident of the results that were found.
-
A college professor want to evaluate the effectiveness of two textbooks. She is scheduled to teach the same course in both the first and second semesters, so her plan is to use one textbook in the first semester and the second textbook in the second semester. Each class will take a final exam that covers material that is included in both of the texts to avoid biasing the results in favor of one text. (answer)
-
A state legislature wants to stiffen the penalties for drunk driving in the hopes that it will decrease drunk driving and decrease the number of deaths on the highway that are due to drinking. The legislature wants to know if the change in the law makes a difference, so the Division of Motor Vehicles is asked to monitor the results and to use a nearby state as a basis for comparison. (answer)
-
The government wants to measure the long-term emotional impact on New York City residents of the September 11th attack on the World Trade Center. A random sample of residents is drawn and evaluated for emotional distress three years after the event. As a control, another random sample is drawn, this one from Chicago. (answer)