# How do I determine which standard deviation to use after normalizing a set of data?

#### verges_prime

Ok, so I'm sure that I worded the topic of this thread poorly, but I'm a little lost as to exactly how to explain my problem. As such, I'll just lay out everyting in some detail and hope those of you with more stats expertise will understand )

The problem:

* I have data from 5 subjects

* Each subject gives 2 sets of data, 1 set is "pre-treatment" and the other "post-treatment"

* The "pre" and "post" data for all subjects cannot be lumped together in it's raw form, and it is not possible for me to take a mean and stdev over all subjects by just lumping all responses

* I have normalized the data for all subjects, and am now able to compare "post-treatment" responses across subjects. However, because that normalized data only takes into account the original mean value for each subject, and not the stdev of the raw data, I'm not sure how to compute the error bars for the normalized post-treatment responses. Any ideas?

* I would also (in a perfect world) like to be able to calculate an average response and stdev which lumps all subjects together. Can I calculate this from the normalized means I have now? How do I get the stdev for this value?

Thanks for taking the time to read through this. Any help would be appreciated.

Related Set Theory, Logic, Probability, Statistics News on Phys.org

#### EnumaElish

Homework Helper
How did you normalize the data without taking account of the raw stdev?

### Physics Forums Values

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving