How to do graphs with mean value plus error bars for replicates? St deviation or - (Apr/07/2008 )

Dear All,

I have been doing a time course experiment (analysed by Real Time PCR) and I have repeated it 3 times (independently), obtaining 3 different and independent sets of results for the same time course.

I want to represent it in a graph with columns and error bars, but I am not too sure of how to do it and what statistical commands to use in Excel.

As far as I know, the average between the 3 values for a certain time point would be used for the column, right? But how about the error bars? Do we use standard deviation for that? Or standard error? SEM? Or something else?

If we are supposed to use standard deviation, then which type should it be? I noticed in Excel there are at least 4 different ways of finding standard deviation (STDEVP, STDEV, etc)

I would really appreciate if you could help me out here, I've checked some tutorials on the internet but I still couldn't figure it out.

Thank you very much!

Julianne.

Error bars - ahhhh, MS-Excel can be hard to understand for this but still doable.

Firstly, for standard deviation use STDEV.

There is no easy command for standard error but you can set up formula like <=STDEV(A1:F1)/SQRT(6)> where cells A1 to F1 contain the values you have averaged and 6 is the number of values (n) i.e. A1, B1, C1, D1, E1 + F1 = 6.

Hope this helps,

AussieUSA.

I think most error bars are SEM.
So first you calculate your stdev.
then SEM = stdev/sqrt(n)

I think that's right It IS early in the morning!

Hi! Thank you so much for all the tips! It helped so much, now I have my graphs all nicely...

I talked to my supervisor and he asked me to calculate the P value now for the different values, to know if they are significantly different from each other. Any idea on how to do it?

I checked my old statistics book, but I am still a bit clueless of how to proceed... do I have to do also any kind of ANOVA or other fancy analysis, to know if they are significantly different from each other?

Thank you a lot!

Julianne.

Stats ... not doable in Excel.

One-way ANOVA a little too much for your experiment. Students T-test should be sufficient. Whether to do paired or unpaired ... is it the same cell undergoing different treatments or different cell types undergoing treatments? If same = paired. Different = unpaired.

Use any basic stats program.

Hope this helps,

AussieUSA.

lots of papers use the mann-whitney test to see if they have a significant p-value (the closer it is to zero the higher probability of your samples being different). it has the advantage it is a non parametric test, that way you don't have to standarize your data (in case it has a high std deviation)

Thank you so much for all your replies, it has been very helpful!! However, I have to admit I don't have a good statistical background.

Do you have any suggestions of a book/paper/websites to look up? I have my statistical books, but I find it very hard to apply to lab and real life experiments.

Thank u!

All the Best,

Julianne.

look for "experimental design" online or books. if you have a specific problem it might be easier to help you. if your experiment is not set properly from the beginning it might be a bit complicated (not impossible tho) to find a test that suits your data best.