Mathematics Asked by dshunevich on December 19, 2020
A coin was tossed $N$ times. The number of heads is 10% more than the number of tails. At what $N$ can we say that the coin is "dishonest" (heads and tails fall out with different probability)?
Sample size for a dichotomous variable with $p$ proportion of success:
$$N = p(1-p)bigg({Z over E}bigg)^2$$
where,
$Z$ is the $z$-value corresponding to your confidence interval (eg. $z$-value $1.96$ for $95%$ confidence interval),
$E$ is the margin of error (eg: $0.05$ for $5%$ MoE)
For the given example, let $p$ be the proportion of tails.
$$therefore 1 - p = 1.1 times p$$ $$implies 1 = 2.1p implies p = {10 over 21}$$
Substituting, we get
$$N = bigg({10 over 21}bigg)bigg(1 - {10 over 21}bigg)bigg({1.96 over 0.05}bigg)^2 = 383.28 approx 384$$
So, 384 trials will tell you with 5% margin of error in 95% Confidence Interval. This will change if you change your CI and margin of error parameters.
Correct answer by vvg on December 19, 2020
Get help from others!
Recent Questions
Recent Answers
© 2024 TransWikia.com. All rights reserved. Sites we Love: PCI Database, UKBizDB, Menu Kuliner, Sharing RPP