An AP story (July 22, 2003) described a study conducted over four years by Dr. Martha Morris and others from Chicago’s Rush-Presbyterian-St. Luke’s Medical Center involving 815 Chicago residents aged 65 and older ( Archives of Neurology , July 21, 2003). Those who reported eating fish at least once a week had a 60% lower risk of Alzheimer’s than those who never or rarely ate fish. However, the story also quoted Dr. Rachelle Doody of Baylor College of Medicine as warning, “Articles like this raise expectations and confuse people. Researchers can show an association, but they can’t show cause and effect.” She said it is not known whether those people who had a reduced risk had eaten fish most of their lives, and whether other dietary habits had an influence. Using this example, describe how you would explain to someone who has never taken statistics
(a) What a lurking variable is,
(b) How there can be multiple causes for any particular response variable, and
(c) Why they need to be skeptical when they read new research results.