1

I have a biased coin; it's going to return either heads or tails at some percentage. If I run a test and get back say 28 heads out of 40 flips, how can i best add error bars to indicate i don't have a lot of samples, and the real results could be higher or lower?

Knox
  • 111
  • 3
  • look, i am a total nube at statistics, I know enough to know this is very basic and i apologize in advance if it's too basic for this site. – Knox Mar 21 '14 at 14:01
  • 3
    Are you after something like a (binomial) [confidence interval for a proportion](http://en.wikipedia.org/wiki/Binomial_proportion_confidence_interval)? Or do you want error bars that represent something else? – Glen_b Mar 21 '14 at 14:11
  • 1
    Although this question is asked and answered at http://stats.stackexchange.com/q/4756, the language is technical and the answer is incomplete (from the point of view of someone without a statistical background). More elementary expositions of the same ideas are found at http://stats.stackexchange.com/q/3402. The theory behind all this is explained (in very elementary terms but in great detail) at http://stats.stackexchange.com/a/18609. – whuber Mar 21 '14 at 14:34
  • @Glen_b yes, i think that's what i'm after. The elementary explanations at related questions were helpful. – Knox Mar 21 '14 at 15:16

0 Answers0