Standard deviation of the sampling distribution

This number says how far a statistic typically lands from the parameter across repeated samples, and it becomes the denominator of every z and t statistic you compute. For the sample mean it is $\sigma/\sqrt{n}$ (sigma, the population standard deviation, over the square root of n), and for the sample proportion it is $\sqrt{p(1-p)/n}$; swap in a sample estimate for the parameter and the same quantity is called the standard error. Because $n$ sits under a square root, shrinking it is slow work: with $\sigma = 12$, going from $n = 36$ to $n = 144$ takes it from $12/6 = 2$ down to $12/12 = 1$. Quadrupling the sample size only halves the variability.