Question

People who often attend cultural activities, such as movies, sports events and concerts, are more likely than their less cultured cousins to survive the next eight to nine years, even when education and income are taken into account, according to a survey by the University of Umea in Sweden.

Can this claim be tested by conducting a randomized experiment? Explain.

Answer 1

This claim can't be tested by conducting a randomized experiment.

A randomized experiment is useful to reduce the effect of unknown variables and their variance in the final result. But it requires that researchers don't interfere directly with the object of study.

To test the hypothesis of the survey, you require to make an experiment that follows several controlled groups of people for a long period. The controlled people could be separated into groups, people who attend cultural activities and people who don't do it. To test this claim you should ensure that the people of the first group attend to the cultural activities, while the people of the second group doesn't. This requires controlling the activities of people for all the duration of the experiment.

By controlling the activities and lives of the people in the experiment, you are affecting the randomizing factor of the study. To make a fully randomized experiment, you shouldn't control the activities of the subjects and let them act naturally. By doing this, you can't ensure that people of the original second group doesn't attend to cultural activities. Therefore the entire experiment won't prove or refute the main hypothesis.