Originally Posted by

**Shakarri** The margin of error E is the difference between the sample mean and the upper limit, or the sample mean and the lower limit.

For a 95% confidence interval the significance level $\displaystyle \alpha$ is 0.05 (the amount of the population not in the confidence interval)

$\displaystyle \alpha /2=0.025$. $\displaystyle Z_{0.025}$ is the point in the normal distribution that has 2.5% of the population below it, that point is -1.96

$\displaystyle 1- \alpha /2=0.975$. $\displaystyle Z_{0.975}$ is the point with 97.5% of the population below it, that point is 1.96.

So it doesn't really matter, all that changes is the sign