tag:blogger.com,1999:blog-7937602330391830242.post5223283521835464359..comments2023-03-25T02:17:56.058-10:00Comments on Say Yes to the Honolulu Rail System: Scientific Opinion Polling Out of Favor in District 1Doug Carlsonhttp://www.blogger.com/profile/10191683240304122047noreply@blogger.comBlogger1125tag:blogger.com,1999:blog-7937602330391830242.post-33406383996616867492011-06-16T08:49:12.881-10:002011-06-16T08:49:12.881-10:00I haven't formally taken a statistics course, ...I haven't formally taken a statistics course, but I do know that polls need a RANDOM sample of the population. It is unfeasible to ask everyone, so the next best thing is to get a small group that is representative of the greater population. Basically, the number of the sample size and the margin of error are related. Bigger sample size, smaller margin of error.<br /><br />Pollsters just need enough of a sample size to get under a 5% margin of error. Beyond this 5% threshold, you need more people to get further accuracy; it gets exponential. 900 is a good number for Oahu's population.<br /><br />Anti-railers allege the poll was biased. It's ironic if they want in on a poll, because catering to one opinion is exactly what not to do in a scientific poll.sumwonyunohttps://www.blogger.com/profile/04694015523685165673noreply@blogger.com