excel should let you vary error bars for individual points

In scientific experiments it is often important to conduct an experiment in
triplicate. I find that I often need to plot these on a bar chart eg cell
number with two growth conditions at a variety of different time points. If
you count the cell number in three separate flasks for each growth condition
at each time point, then you can work out the mean cell number and the SEM or
standard deviation. I want to plot both the mean and the SD for each point by
using error bars. Obviously the SD is different for each point. Excel will
not let you do this and it makes me crazy because I have to plot cell counts
for each time point and each condition as a separate series. Why is there no
"vary error bar by point" command like there is for colour???????

Does anyone know of an easier way please?

----------------
This post is a suggestion for Microsoft, and Microsoft responds to the
suggestions with the most votes. To vote for this suggestion, click the "I
Agree" button in the message pane. If you do not see the button, follow this
link to open the suggestion in the Microsoft Web-based Newsreader and then
click "I Agree" in the message pane.

http://www.microsoft.com/office/comm...excel.charting

It does allow the specification of error bars by point.

Calculate the error bar values in some column. Then, double-click the
series in the chart and in the Format Data Series dialog box select the Y
Error Bars tab (or the X Error bars tab if appropriate). In there, select
the Custom option and specify the range containing the + and/or - error
values.


--
Regards,

Tushar Mehta
www.tushar-mehta.com
Excel, PowerPoint, and VBA add-ins, tutorials
Custom MS Office productivity solutions

In article <[email protected]>,
[email protected] says...
> In scientific experiments it is often important to conduct an experiment in
> triplicate. I find that I often need to plot these on a bar chart eg cell
> number with two growth conditions at a variety of different time points. If
> you count the cell number in three separate flasks for each growth condition
> at each time point, then you can work out the mean cell number and the SEM or
> standard deviation. I want to plot both the mean and the SD for each point by
> using error bars. Obviously the SD is different for each point. Excel will
> not let you do this and it makes me crazy because I have to plot cell counts
> for each time point and each condition as a separate series. Why is there no
> "vary error bar by point" command like there is for colour???????
>
> Does anyone know of an easier way please?
>
> ----------------
> This post is a suggestion for Microsoft, and Microsoft responds to the
> suggestions with the most votes. To vote for this suggestion, click the "I
> Agree" button in the message pane. If you do not see the button, follow this
> link to open the suggestion in the Microsoft Web-based Newsreader and then
> click "I Agree" in the message pane.
>
> http://www.microsoft.com/office/comm...excel.charting
>

I'm interested in the same query.

I've calculated the standard error margin (SEM) for each point in my bar graph - this varies as each point is the average of a varying number of samples.

This SEM value is represented in the same units as the data on the bar graph (second), and is in a column that corresponds to the values for the bars in the bar graph. (i.e. A1 = Bar 1 value, B1 = Bar 1 value SEM).

When I define the error bar using the 'custom error bars' option in the 'add error bars' tab I select the column (I drag from the top of the SEM data to the bottom) for the SEM data. However, Excel gives seemingly spurious error bars when I do this. I'm getting errors of several times the values according to the error bars, when the actual SEM is equal to zero, for example.

How do I define the error bars correctly? Is because I am working in seconds?

I have attached the sheet. The values of the bar chart are in column AU, and the values of each point's SEM is in column BB.

(You'll also notice the ridiculous formula I came up with for calculating the SEM! I'm not very good with Excel!)

Thanks for your help,

Adam

Here is the spreadsheet.

Still no answers?