Is it possible to find SD, having n, mean, median, 5% and 95% quantile?

Germano Reis

New member
Joined
Oct 27, 2017
Messages
4
Dear sirs, is it possible to find standard deviation and standard error having the following information? Thanks in advance. Germano :confused:

n = 87 mean = 31.1 median = 30.7 (Q50)Q5 = 7.7 Q95=48.1
 
Dear sirs, is it possible to find standard deviation and standard error having the following information? Thanks in advance. Germano :confused:

n = 87mean = 31.1median = 30.7 (Q50)Q5 = 7.7Q95=48.1
What are your thoughts?

Please share your work with us ...even if you know it is wrong.

If you are stuck at the beginning tell us and we'll start with the definitions.

You need to read the rules of this forum. Please read the post titled "Read before Posting" at the following URL:

http://www.freemathhelp.com/forum/announcement.php?f=33
 
What are your thoughts?

Please share your work with us ...even if you know it is wrong.

If you are stuck at the beginning tell us and we'll start with the definitions.

You need to read the rules of this forum. Please read the post titled "Read before Posting" at the following URL:

http://www.freemathhelp.com/forum/announcement.php?f=33

(corrected)
Hi thanks for your answer. Although the aritmetic mean is 31.1, the median shows that 50% of the cases are above 30.7 , while other 50% are bellow this value. Moreover, 5% are bellow 7.7 , while ohter 5% are higher than 48.1.

I think it would be much easier to calculate SD if distibution was normal (using z score: z = (x - μ)/σ). However - due to the difference between median and mean - it is clear to me that normality can not be assumed in this case. Thus, I suppose that some kind of adjustment/correction is needed. Unfortunately Im stuck here !:(
 
Although the aritmetic mean is 31.1, the median shows that 50% of the cases are above 30.7 , while other 50% are bellow this value. Moreover, 5% are bellow 7.7 , while ohter 5% are higher than 48.1.

I think it would be much easier to calculate SD if distibution was normal (using z score: z = (x - μ)/σ). However - due to the difference between median and mean - it is clear to me that normality can not be assumed in this case. Thus, I suppose that some kind of adjustment/correction is needed.
Your first post made it sound as though you were needing to answer the question, "Is the following possible?" Now it sounds as though you are needing to answer the question, "Do the following", but don't know how.

To help us understand what's going on, please reply with the full and exact text of the original exercise, the complete instructions, and recent topics of study in your statistics class. Thank you! ;-)
 
Your first post made it sound as though you were needing to answer the question, "Is the following possible?" Now it sounds as though you are needing to answer the question, "Do the following", but don't know how.

To help us understand what's going on, please reply with the full and exact text of the original exercise, the complete instructions, and recent topics of study in your statistics class. Thank you! ;-)


hello, many thanks for your answer. The whole context is as follows.
I´m doing a meta-analysis and intend to run a meta-regression. A meta-analysis synthesizes the results of previous research.

For this, I must use means and standard errors for all the analyzed variables. Although most of the studies disclose mean and SE, one of them does not do that. It provides just n, mean, median, Q5 and Q6 instead. Thus, a conversion (or some estimation) is needed here.

For instance:

Stocking density (kg/m2 ):
n = 87 cases
mean = 31.1
median (Q50) = 30.7
Q5 = 17.7
Q95 = 48.1
Where: Q = quantile (percent)

So the question is: What is the Standard Error? How do I calculate it with the provided information ?

I hope this is clearer now. thanks in advance. Germano
 
Last edited:
hello, many thanks for your answer. The whole context is as follows.
I´m doing a meta-analysis and intend to run a meta-regression. A meta-analysis synthesizes the results of previous research.

For this, I must use means and standard errors for all the analyzed variables. Although most of the studies disclose mean and SE, one of them does not do that. It provides just n, mean, median, Q5 and Q6 instead. Thus, a conversion (or some estimation) is needed here.

For instance:

Stocking density (kg/m2 ):
n = 87 cases
mean = 31.1
median (Q50) = 30.7
Q5 = 17.7
Q95 = 48.1
Where: Q = quantile (percent)

So the question is: What is the Standard Error? How do I calculate it with the provided information ?

I hope this is clearer now. thanks in advance. Germano

Hi guys any ideas on how to solve this challenge? thanks. Germano
 
Top