Protocol Online logo
Top : New Forum Archives (2009-): : Scientific Presentation

How to present Standard error in the bar graph - (Nov/17/2016 )

Hello folks,

 

If I have two individual experiments and each experiment I did triplicate, can I include standard error from pool 6 raw data for showing bar graph?

 

Because someone told me that I cannot have standard error if I did just two individual experiments.

 

any suggestion ?

-Tai-

What you have done is two experiments (biological replicates) with 3 technical replicates in each. You would not usually treat the technical replicates as independent experiments - you would take the means of the technical replicates for each experiment and compare them between biological replicates. 

 

You can get a standard error with 2 replicates, but the error will be large (you lose power in this case as it is derived from the square root of the sample size), and as such require that there be a big difference between the biological replicates before you could say anything about their difference statistically speaking.

-bob1-