Hacker News new | past | comments | ask | show | jobs | submit login

1900 is too low a number to sample. And there is no mention as to whether that 1900 is actually random enough. I doubt this statistic.



For your information, about all polls are done on about 1000 people (slightly less to slightly more). You need to poll a lot more people to reduce significantly the statistical error level.

See http://en.wikipedia.org/wiki/Margin_of_error




Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: