Question

Rick’s class plans to study the effect of eating different breakfast foods on grades in morning classes. Students in the class will be the test subjects for the investigation. Which of the following would most likely make the results of the experiment invalid?
Half of the students ride the bus to school, and half ride in cars.
Some students eat breakfast two hours before other students do.
Some students miss school during the investigation due to illness.
All students in the investigation have morning classes in the same subjects.

Answer 1

Some students eat breakfast two hours before other students do.

Timing of eating is a confounding variable: how long before class someone eats affects blood sugar, alertness, and therefore grades. To test the effect of food itself you must control when meals are eaten. Missing students or different travel modes can be managed or accounted for, and having the same morning subjects is actually a good control.