There have been threads here before which posted links of the media attributing causality to correlational studies, and links to those studies have been posted. It seems as if we are always focusing on showing students how to identify correlations and not to confuse it with causality.

We are missing something important in the process though: Some students assume afterwards that causality can never be inferred, and absolutely everything is a correlation!

Does anyone have any links to studies in the media which actually DO show causal links?

Student understanding of the difference between correlation/causation in studies reported by journalists, can only be inferred when "causal relationship" sometimes is actually the correct answer.

  • 4
    $\begingroup$ Are you thinking of (1) experimental studies that demonstrate causation, or (2) observational studies that turned out on further investigation to have identified causal relationships, or (3) observational studies that in themselves provide strong evidence for causal relationships (having included all plausible confounders, &c.) - or something else? $\endgroup$ – Scortchi Mar 19 '14 at 16:23
  • $\begingroup$ Or why not an example that shows how the accumulation over time of various evidence from different studies (observational & experimental) leads to the acceptance of a theory born a speculation? The importance of any single scientific study is rarely apparent except in retrospect & the contemporary news reports can be relied on to distort its findings beyond recognition. How about the story of the link between smoking & lung cancer? You could bring in the Doll-Fisher debate & the Bradford Hill guidelines for inferring causal relationships. $\endgroup$ – Scortchi Mar 19 '14 at 17:12
  • $\begingroup$ Thank you very much for your response. It is an undergraduate class (Freshman, fresh out of high school) so I am referring to simply cause-and-effect studies in the media. I would like my students to be able to read a news article, and be able to tell if they can refer cause-and-effect from it. However, journalists seem to only report correlational studies, and then some students leave with the impression that cause-and-effect can never be inferred. $\endgroup$ – Shai Mar 21 '14 at 12:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.