# Which statistics to use- HELP!!! - (Apr/09/2009 )

HELP!!! Sadistics....sorry Statistics is nightmare for me and I need help in choosing which test to carry out.

I have done some phenotypic cells counts, 15 fields with each sample and 3 phenotypic characteristics, I have each represented as a percentage of the total cell number.
What I want to check is whether there is a statistically significant change between the same phenotype when the cells are undifferentiated and then differentiated.
Please let it be an easy one! I only have access to excel at the moment. Something like a T-test?

a speedy reply would be great as I can add it to my talk for tomorrow.

Lost

-LostintheLab-

Because you are doing non-parametric stats, you can use chi-square 2x3 table to calculate. Here is a excel temple which allows you to do that as found here http://www.runet.edu/~biol-web/stats/chi-sq2X3template.xls

There is also an online tool allows you to calculate Fisher exact probability http://faculty.vassar.edu/lowry/fisher2x3.html

-pcrman-

Thanks!! you're a life saver....
Chi squared- yes! I know this one and I understand it, which is more important.

Seriously, I don't really understand most statistics, I have missed out on this in my previous education somehow.

A much happier Lost

-LostintheLab-

You are welcome.

-pcrman-