Standard deviation in Excel

Even if it's late, it will surely help someone else later... since I was looking for the same thing today.

→ When you have created the graph (for example, a histogram), double-click on the data series for which you want to make the standard deviation visible. The data formatting window will open. Choose the tab: Y Error Bars. You can select the standard deviation, the %, or custom data that you calculated earlier.

That's it ;-)
--
You have to pick the cherries by their stems...
I was already having trouble with my hands!

Oups a small mistake:

It's half the standard deviation that you need to add and subtract as the margin of error.

So when calculating the standard deviation, type instead: =0.5*AVERAGE.STDEV(series values)

Otherwise, it was displaying the standard deviation twice

Now it's correct

Have a nice day

Hello, a bit late but it can still be useful
To add error bars with the standard deviation values we calculated in Excel 2007:
After inserting the histogram, click on it, then on the layout tab (chart tools), then on error bars, in the list that appears choose more error bar options, then select custom: specify a value, select your series of values and click enter.

Hello,
First of all, thank you for the valuable information I've gathered from your conversations; it has been very helpful. However, I still have 2 questions that worry me (yes, I'm just really bad at stats ^^),
1) Why do you choose to calculate the mean deviation instead of the standard deviation? Is it like the standard error calculated automatically by certain software like Statistica? Or is it something else entirely?
2) I followed your logic of dividing the standard deviation by 2 to represent it as error bars, as visually it makes more sense (otherwise, I always ended up with standard deviations as large as my graph bars o_O). However, since I can't justify it by saying "it looked nicer," I started to question it and it seems strange to me. If I have, for example, a mean of 100 with a standard deviation of 15, that means subjects are considered within the norm when their score varies between 100 +/- 15, so between 85 and 115... Knowing this, I have doubts about dividing my standard deviations by 2 since, if I understand correctly, I end up with the relation 100 +/- 7.5, right?
Thank you in advance ;-)

Hello,
I'm using Excel 2007
and the proposed solution doesn't work, I can't find "data format" or "Y error bar" ...
I really need to put standard deviations in my average histograms (otherwise, no scientific value!!)
Thank you for any help.

Good evening,

I indeed have the same standard deviation displayed for all my histogram bars. When I right-click on a bar, I don't have "format data series," so I can't modify my error bars.
What should I do?

Hello,

Can someone explain to me why Nainternational halved the average gap?

- Is it when it depends on the type of orientation of the error bars in the Excel graph (orientation "plus" or "minus" or "both") in Dispo->Analyse->error bar->"other error bar options"

Thank you.

Hello,
I was browsing this topic and I would like to share a link to a free book: Statistics for Statistaphobes by Denis Poinsot, which has greatly helped me in my thesis.

http://perso.univ-rennes1.fr/denis.poinsot/Statistiques%20pour%20statophobes/STATISTIQUES%20POUR%20STATOPHOBES.pdf

On page 16 of the pdf, it says that it is the standard error that should appear on the histograms and not the standard deviation. I highly recommend his very interesting approach to statistics and the representation of results. It’s very easy to read and well-crafted for people like me who don’t really like it!!!!

As for the solution to put different error bars on the histograms, I insert multiple data sets. This way, each bar of the histogram is independent. It’s time-consuming, but it works. I will test the solution mentioned above!!!!

bye

Hello,
as "truc22" said, to match each standard error to each mean value (and this also applies to Office 10!), you should select not just one specific value, but the corresponding COLUMN (or row). If your column is directly aligned with your mean, it works perfectly.
For your calculations:
Calculate the standard deviation in a column using the formula STDEV.P (which takes into account the population and not a sample of values with the formula STDEV.S).
In the next column, you divide your left cell by the square root of the number of values, e.g.: AZ4/SQRT(COUNT(B4:AX4)). And that's it.

As for the graphical representation of the errors, I'm still indecisive (should I divide by 2 or not?!)

Well, it helped me! Thank you very much!!!

Your article has been a great help to me! Thank you very much :-D

I want to create a histogram representing the percentage of billing delays for 4 people, and I would like to add a curve to show the average of the 4 per month. When I do it for 3 people, it works, but when I add 4, it creates a curve instead of a bar for the 4th one.
How can I do that? Thank you for your answers, hoping I was clear enough!

Hello, I have the same issue displaying the significance of the test when my histograms are made like this and for the standard deviations. I have created the formula that calculates the significance. However, how can I display the "small stars" on my histogram?

Thank you in advance for your response

Hello,

Can someone explain to me why Nainternational halved the mean difference?

- Is it when it depends on the type of error bar orientation in the graph in Excel (orientation "plus" or "minus" or "both") in Dispo->Analyse->error bar->"other error bar options"

Thank you.

Question about the graphs, I know it's been a while but... what I want to see is the difference between the data... how do I do that?

For example: I have
num1 Num2
3 2
6 4
9 8

I would like to see 1(3-2), 2(6-4), 1(9-8)... and this for several columns if possible (num3, num4, etc..)
Thanks for the inputs

Please, I read some of your answers, I tried to add my error bars (Standard error to the mean) but it doesn't work, I have Office 2010. I'm stuck.
And for my calculation sheets, how do I calculate the standard error from means or standard deviation?
To represent them on the graphs + and - do I take half of the value or the whole thing?
Thank you for your answers.

The problem is that it's the standard error and not the standard deviation. In Excel 2997, can you create a histogram with a standard deviation? On another site, they say no, and honestly, I've been looking for a while, it doesn't seem possible unless you assign the same standard deviation to all the bars of the histogram.