Researchers selected a simple random sample of 4048 medical records of adults diagnosed with gum disease. In all, 2226 were current smokers, 891 were former smokers, and 931 never smoked regularly. Their research question is, Do these data indicate that gum disease is equally likely regardless of smoking status? Is the chi-square test valid in this situation?

Question

Quizplus · Accepted Answer

The chi-square test is valid in this situation because the sample is random and the expected counts are large enough (i.e., greater than or equal to 5 for all cells). Therefore, we can assume that the distribution of smoking status is independent of the distribution of gum disease status, and use the chi-square test to test for independence. Choice B is not the best choice because the validity of the chi-square test depends on both the sample size and the expected counts. Choice A is not the best choice because randomization is not necessary for the chi-square test, and the study does not need to be a randomized experiment. Choice C is not the best choice because the observed counts are not relevant to the validity of the chi-square test.

Researchers Selected a Simple Random Sample of 4048 Medical Records