# chi square problem.

#### jimao

##### New Member
hi all:

I tried google and the search function on this forum with no luck. I did the following experiment:

Differentiate a stem cell into four different types of cell, let's call them type A, B, C and D.

The control group usually has a 25% A, 25% B, 25% C and 25% D distribution.
The treatment group usually has 80% A, 5% B, 5% C and 10% D distribution.

However, the word "usually" is my problem. Each group (control and treatment) is done in triplicates, meaning I have 3 sets of distribution data for eatch group. The 3 sets of distribution ratio within each group are all similar to the ones shown above however they ARE different. I suppose if I carry out Chi-square analysis, assuming it is the right analysis, I need to take into account the mean and the standard deviation for both the control and the treatment group. But I just can't seem to find the right formulae. Every site I've been to gives me the following formulae: where Oi = an observed frequency (ie count) for the ith bin
Ei = an expected (theoretical) frequency for the ith bin

#### jimao

##### New Member
In other words, if I have something like:

The control group has a 25±3% A, 25±3% B, 25±3% C and 25±3% D distribution.
The treatment group has a 80±4% A, 10±4% B, 5±4% C and 5±4% D distribution.

How do I carry out chi-square analysis while taking the standard deviation into account?

#### jimao

##### New Member
one more question, if I have two different treatment in addition to the control, is there anything like a chi-square/ANOVA combo analysis for this?