4 ms·

This journalist appears to be unable to distinguish between correlation and causation.

• gpvos 11y ago
  According to the study, causation actually goes both ways. > What Canli finds really interesting about Cole's results is that people who felt lonely one year had increased gene activity around inflammation and norepinephrine later on. And people who had increased inflammation felt lonelier the next year. "It's a two-way street," he said. "Loneliness predicted biological changes, and biological changes predicted changes in loneliness."
  • rlpb 11y ago
    Again, this finds a correlation, not causation. There could be some external factor which influences both measured metrics. Canli as quoted clearly does understand the difference, because he states that one predicts the other, which is distinguishing it from causation. But the article as written implies causation without any justification for this additional leap.
    • gpvos 11y ago
      Yes, you're correct. Prediction is still not the same as causation.