3 ms·
When you say "Maybe you should talk to more poor people", unless you are advocating HN readers to conduct a randomized study, you are essentially telling HN rea
by BeetleB 8y ago
When you say "Maybe you should talk to more poor people", unless you are advocating HN readers to conduct a randomized study, you are essentially telling HN readers to engage in selection bias.
- fLflLflFlfl 8y ago“Seeking out perspectives that could combat your inherent selection bias” != selection bias. Are you arguing that poor people would be OVERrepresented in the minds of techies who go out and talk to them about their experiences?
- BeetleB 8y agoNo - I am arguing that the only way not to have selection bias is by doing a randomized survey - it is the very definition of selection bias. If andbberger wishes to make a nice, nuanced argument, I encourage him/her to do so. However, if instead (s)he wishes to provide two word answers, (s)he should expect push back. Furthermore, I highly doubt andbberger has done a randomized study, and thus the accusation of selection bias applies equally to him/her. Saying things like "perhaps you should talk to more poor people" is simply dismissive, and unjustifiably assumes that someone who is disagreeing with him/her hasn't talked to enough poor people or that the other person suffers from selection bias, and not him/herself. Look at the whole thread: andbberger's comments are short, lacking any citations, sarcastic, accusatory ("middlebrow dismissal", not talking to poor people, talking to the wrong poor people). Contrast with HN guidelines: >Be kind. Don't be snarky. Comments should get more thoughtful and substantive, not less, as a topic gets more divisive. (Two word responses don't help). >Assume good faith. (Implying people of being ignorant of the conditions of the poor doesn't help). >Please don't post shallow dismissals ("Selection bias") So no, I'm not commenting here to argue whether the conditions of the poor constitutes a sky falling. I'm commenting to counter low quality statements.