Foundations of Social Inquiry
50 points possible
Experimental Design and Internal Validity
- A researcher finds that boys who join youth gangs are more likely to engage in delinquent behavior than are other boys. The researcher finds that this is true even after controlling for other, observable differences between gang members and other boys that might explain delinquency.
- If the above described an experiment, what would be the “treatment?” (3 points)
- Would an experimental design be appropriate for this study? Why or why not? (6 points)
- The researcher concludes from this research that gang membership causes delinquency. Discuss how the problem of selection bias might have, instead, produced the appearance that gang membership causes delinquency. (6 points)
2. A research team finds that compared to other 8th grade students, those attending private schools score better on a national aptitude test.
- What is the independent variable in this research? (3 points) Whether or not a
- Draw a path diagram (i.e., a path model) that depicts the team’s research finding. (8 points)
- In this study, the research team controls for differences among private-school and other students in family income. Even after taking income differences into account, the researchers still find that 8th graders in private schools score better on the aptitude test than do other students.
Is it safe for the researchers to infer that attending a private school causes 8th graders’ test scores to improve? Why or why not?
3. True Story:
Dr. Chavez was asked to evaluate and propose solutions for reducing student tensions and poor performance in local schools. Chavez believed that competition in the classroom generated these tensions, which in turn led to weak student performance in the classroom.
Dr. Chavez devised a new, cooperative method of learning called the “jigsaw” method. She decided to test this method by getting permission to run an experiment at a local school.
- Recall the Milgram experiments on obedience. What was the impact of the Milgram experiments on social research practice that, 40 years later, would require Dr. Chavez to seek special permission to run the experiment? (6 points)The Milgram experiments raised the question of
At the end of her first experimental trial, Chavez found big differences between children in classrooms with jigsaw-learning groups and kids in traditional classrooms. After 6 weeks, the kids in the jigsaw classrooms liked their peers more, showed more positive attitudes toward school, had more positive self-concepts, and showed an improvement in grades.
- At the start of the experiment, Chavez was unable to randomly assign students to the jigsaw classrooms vs. traditional classrooms. Instead, she randomly assigned some existing classrooms to be “jigsaw” and other classrooms (again randomly) remained “traditional.” What must you assume about the way students were enrolled in these preexisting classrooms for Chavez’s study to be a true experiment?
About two months later, Chavez ran a second 6-week experimental trial at the same school, but involving different classrooms of students (though in the same grades as the first experiment). This time, however, the expected differences between the “jigsaw” and the traditional classrooms failed to materialize. Instead, children in both the experimental and the control group showed the same increase in liking of peers, satisfaction with school, and so on.
- What might have happened? Assuming Chavez followed the same rules for conducting a classic or true experiment the second time around, what threat(s) to internal validity might explain why the results of the second trial were so different? (8 points)