A study of the dietary habits of a group of people who had recently developed cancer and a group without cancer found that during the previous five years the diets of the two groups' members closely matched each other in the amount of yogurt they contained.
Conclusion
If you have more galactose in your body than what the body can process, it is carcinogenic (cancer-causing).
Evidence
In a study, the group of people with cancer had such a low level of a certain enzyme that they couldn't process the galactose in the yogurt they were consuming.
Evaluate
The author is taking one fact she knows that is distinct to the cancer group and then concluding that that factor caused the cancer!
What if we had said, ? That would be a silly conclusion to reach. Maybe the hair loss is unconnected to the cancer. Maybe it's actually an effect of the cancer (or the chemotherapy treatment received by people who have cancer).
When an author presents a correlation, f.e. "the people who were X were also Y", she is likely to conclude that X causes Y. But we always evaluate causal arguments with the same two-pronged mindset:
- Is there any other way to explain this correlation? (maybe Y causes X. Maybe some third factor, Z, is causing both of them) - How plausible is the author's causal storyline? (is there any explanation of how X would cause Y? are there any data points that show when someone stopped having X, they stopped having Y?)
We are trained to react to correlations primarily by thinking,
Goal
Find an answer simply describing the author's illegal move of thinking, , or find an answer bringing up one of the alternative explanations the author has failed to consider: