2 ms·
It was a straw poll. A sampling bias is that none of the people who had died of cancer responded.
by odyssey7 2y ago
It was a straw poll. A sampling bias is that none of the people who had died of cancer responded.
- A_D_E_P_T 2y agoPeople don't die of cancer instantly -- in many cases, it's pretty indolent, and "alive with cancer"/"in remission" are fairly common statuses. This would surely be picked up in the respondent data.
- odyssey7 2y agoIf I were dying of cancer, odds are good that I would have ended up in the control group for not being especially proactive about visiting the hot springs. The authors of [2] explain this limitation themselves: "Some study limitations should be considered. ... healthier people may have been more likely to bathe in the radon hot springs."