Tell me more ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.
up vote 1 down vote favorite

Can anyone explain when should I add 0.5 to the z-score?

share|improve this question
2  
Could you give some context? It's going to be hard to answer your question without more information. – Mike Spivey May 6 '11 at 17:35
example Pr(z <= 1.35) = 0.5 + Pr(0 <= Z <= 1.35) = 0.5 + 0.4115 – optimus May 6 '11 at 17:37
and why is Pr(0 <= z <= 1.5) = 0.4332 without adding 0.5? – optimus May 6 '11 at 17:42
4  
Well, if the distribution is symmetric and continuous then $Pr(z\leq 0)=0.5$. – Rasmus May 6 '11 at 17:44

2 Answers

up vote 3 down vote accepted

Based on the example you give in your comment above, I think what's being used is that $P(z\lt 0)=0.5$ (and $P(z\gt 0)=0.5$, too), so that $P(z\le 1.35)=P((z\lt0)\text{ or }(0\le z\le 1.35))=0.5+P(0\le z\le 1.35)$.

share|improve this answer
well, can you further elaborate? – optimus May 6 '11 at 17:46
1  
$z$-scores are typically on datasets that are expected to be normally distributed and the $z$-score is the number of standard deviations from the mean. So, the probability of being below (or above) the mean should be 0.5. – Isaac May 6 '11 at 17:47
I think Isaac has it right. Some statistics tables give the $z$ score probability corresponding to $P(0 \leq Z \leq z)$, while some give it for $P(Z \leq z)$. I'm guessing OP is looking at one of the former. (See, for example, mathsisfun.com/data/standard-normal-distribution-table.html) – Mike Spivey May 7 '11 at 3:17
up vote 1 down vote

If your problem involves binomial probabilities, and you wish to use the normal approximation to the binomial, you would add .5 to the z formula (not the score) as a continuity correction factor.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.