Question

An experiment testing the effects of a new diet on student weight loss assigns the new diet to a group of randomly-selected students and has them exercise. Another trial was conducted in which a different group of randomly-selected students dieted without any exercise. The results did not match the expected results.

What should be done to improve this experiment?

Select the correct answer below:

The groups should be randomly selected from students and non-students.

The second trial should exercise an equivalent amount.

Nothing. The experiment is well-designed.

Students who are most overweight should be assigned to the new diet.

Answer 1

Ans: The second trial should exercise an equivalent amount.

Why? Since we want to test the effect of “new diet”, we do not want any other factor’s effect to ruin the experiment. Here two groups are randomly chosen but there’s a biased effect of exercise as one group is doing exercise while other isn’t. So there will be effect of exercise in the results. We want to confound the effect of exercise so that only difference between two groups should be the diet. Then we can test the treatment = new diet vs control = old diet. So if the 2nd group also does exercise, the exercise effect gets cancelled out and remaining effect if for diet.

This is why the other factors between two groups should be as close to each other as possible.

Thanks

Please let me know if you have any doubts in the comments below. Thumbs up if you like!!