error bars

[CEC]

Hi, I am trying to incorporate my actual error bars into a bar graph using excel 2008. In the 2004 version, you could highlight the spreadsheet cells containing the errors for each data point and add it into the Y error bars category. I am unable to do this. Is there any way that I can add the individual error bars for every point on my series?

 

[Corentin Cras-Méneur]

Hi, I am trying to incorporate my actual error bars into a bar graph using
excel 2008.

This feature is missing form Excel 2008. You can do it in 2004, open the
document in 2008 and see them, but you can't add the custom error bars
in 2008 since the user interface to do this is apparently missing.

Corentin

 

[Stephen]

This is absolutely idiotic. I'm sticking with 2007 running in a Parallels VM and Aabel for real scientific graphing. I think 2007 in Parallels is faster than 2008, anyway.

 

[Matthew Oskamp]

wow office 2008 is useless without error bars!

 

[Eric Waldbaum [MSFT]]

Hello,
There are error bars in charts in Office 2008 and we have been
investigating custom error bars as mentioned in the "We're working on it!"
(3/12/08) thread. Sorry for any inconvenience, but I still hope you can
enjoy some of the new features in Office 2008.

-Eric
MacOffice Testing
Microsoft

 

[gregman31]

In answer to Eric's answer. It is insane that you guys remove one of the only things that uses to work with your software... I would like to call this software Macrobogs Office instead of Microsoft... Any scientist who plots a graph needs to incorporate error bars...This is one of the base of science! A graph without error bars is not trustable. Oncce again you microsoft people give us a new reason to use other software than yours.
I remind to anybody who read that, that you can use Open Office that is free to share...Open office is a linux originated copy of microsoft office with the big difference that it is free of charge and also free of problems. There are no stupid default settings as in microsoft. You wont spent two hours to figure out how to put a legend on your graph for example....
Why paying for microsoft that is full of problem when you can have a well working version for free!
The only softwares that freeze my computer are from microsoft, any explanation?

 

[Carl Witthoft]

In answer to Eric's answer. It is insane that you guys remove one of the only
things that uses to work with your software... I would like to call this
software Macrobogs Office instead of Microsoft... Any scientist who plots a
graph needs to incorporate error bars...This is one of the base of science!

First off, that's bull. There's nothing wrong w/ graphs stating clearly
that the data plotted are mean values only, maybe with rms plotted
elsewhere.

But even more important ("Zero off" ? ) Any scientist who wants to
plot quality graphs won't use Excel. He'll take the time to learn R,
or GnuPlot, or Matlab, or any of dozens of good packages aimed at real
charting.

 

[angryscientist]

First off, that's bull. There's nothing wrong w/ graphs stating clearly
that the data plotted are mean values only, maybe with rms plotted
elsewhere.

But even more important ("Zero off" ? ) Any scientist who wants to
plot quality graphs won't use Excel. He'll take the time to learn R,
or GnuPlot, or Matlab, or any of dozens of good packages aimed at real
charting.

That's bull???????
I work in science as well as hundreds of people around me and we all use excel; where did you get you info, from Disney channel?
Customizing error bars is essential since we never get identical standard deviations between two or more different groups of data.
These guys at Microsoft screws up on a regular basis :sillygrin:

 

[Carl Witthoft]

First off, that's bull. There's nothing wrong w/ graphs stating clearly
that the data plotted are mean values only, maybe with rms plotted
elsewhere.

But even more important ("Zero off" ? ) Any scientist who wants to
plot quality graphs won't use Excel. He'll take the time to learn R,
or GnuPlot, or Matlab, or any of dozens of good packages aimed at real
charting.

That's bull???????
I work in science as well as hundreds of people around me and we all use
excel; where did you get you info, from Disney channel?
Customizing error bars is essential since we never get identical standard
deviations between two or more different groups of data.
These guys at Microsoft screws up on a regular basis :sillygrin:[/QUOTE]

This is an ancient post, so my response is probably lost, but:

A few points. Just because "... hundreds of people..." use Excel
doesn't mean they have a clue as to how much pain they're causing
themselves. Spend a couple hours with GnuPlot, or even the plotting
functions in R, and you'll see what real charting is about.
Error bars don't make or break a chart. It's all about how you want to
present the data, and what really matters. But if you love error bars,
and don't like the way Excel version[whichever] handles them, write
your own routines. For example, assuming you've got columns of + and -
error data, just plot them as symbols, do some gymnastics to plot lines
between pairs of points, and you're done.