Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

One thing to keep in mind with open ended polling questions like this is that there is that the questions can shape the response. One interesting display of this is that most Americans don't believe that we should take race into account for college admissions, but most Americans also believe that affirmative action for minorities is beneficial.

I don't know what the preferred narrative here has been on the subject, but I think it's been obvious the vast majority of black folks never wanted police protection to disappear, and the lack of assurance in our interactions with police (confirmed by polls mentioned in the article) demonstrates that reform is something that would be strongly desired.



A conclusion based on those polls for affirmative action in college is that people support the general principle of getting more minorities into college, but do not support the current methods applied under affirmative action. When specifics are outlined for respondents the support drops.

This has less to do with the style of shaping the response by what kind of questions are made, and more to do with what the question is. Support for a concept is not the same as support for a specific action. If you ask people to choose between multiple of bad choices in order to achieve a common good then the goal is likely to have a significant higher support than any of the bad choices.


Great explanation - and also why we should be wary of headlines touting polls that don't get into the specifics.


> the questions can shape the response.

An excellent illustration of this: https://www.youtube.com/watch?v=G0ZZJXw4MTA&t=28s

Although in this case, it doesn't look like they had bad intentions and they seem to have dug at least a little below the surface and drawn what seem like sensible conclusions. I wonder if more detailed results and the full list of questions are available somewhere.


I should say that I do assume there were no bad intentions and believe the data/poll is still useful, though I don't think it describes the sentiment on "Should we reform the police and how?" between demographics very well.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: