I came across a post on Facebook which tried to celebrate how the "old ways" (when we were kids) are better, which went along the lines of:

Like this post if you used to cycle around on your bike as a kid without a helmet and survived.

(attempting to prove that kids these days are too "wrapped in cotton wool" and that the "need" for kids to wear helmet while cycling was just paranoia)

The obvious flaw is that anyone who cycled their bike without a helmet and didn't survive are not in a position to argue / do an angry face, so there will be an inherent bias in the results of the "survey".

(If we just ignore for a moment that people can argue on behalf of friends or relatives or people who were merely severely injured could raise their voices)

Is there a name given to this kind of bias / fallacy? I was thinking "selection bias" but the description of that on Wikipedia (https://en.wikipedia.org/wiki/Selection_bias) doesn't seem to quite fit.

  • > attempting to prove that kids these days are too "wrapped in cotton > wool" and that the "need" for kids to wear helmet while cycling was > just paranoia Are you inferring the above, or did the person who made the post spell it out? This doesn't sound like a fallacy at all to me. "Like this post" hardly constitutes a poll or survey. Moreover, there a various ways to interpret the post. It sounds to me like just another "Remember the good old days?" posts. I'm not sure what the laws are today, but, long ago, many people cycled without helmets. I was one of them. Whether it was safe or not is – David Blomstrom Oct 25 '19 at 10:59
  • Wrt the mentioned post: it sounds completely sarcastic to me. – Apollys supports Monica Oct 25 '19 at 21:13

The example with the bicycles would seem to fall into the category of survivorship bias (which can apply to more than just physical survival of people but also more general survival of data points past some selection filter that makes you more likely to come across some cases than others, see this xkcd comic for an illustration involving the issue of some people being more likely to get media spotlights than others). More generally, there is also the streetlight effect, a type of observational bias where you only look at samples that share some characteristic that makes them easier to discover, though it's also sometimes understood as a more qualitative problem with the type of evidence used to test a theory, as opposed to a purely statistical fallacy.


Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.