If there is a better venue for this question, please advise me.
I am looking for methods to analyze categorical data similar to that
shown below. If the results were quantitative, I believe that an
analysis of covariance would be appropriate. However, with
categorical data and relatively small samples, I am at a loss. Any
help would be appreciated.
The purpose of the experiment was to discover whether or not two
groups of infectious organisms differ in there ability to infect a
host over time. The two genetically different groups of infectious
organisms (G1 and G2) are each subdivided into three subgroups based
on smaller genetic differences. They are G1-S1, G1-S2, G1-S3, G2-S4,
G2-S5, and G2-S6. The hosts must be sacrificed to discover which ones
are infected. This results in counts of infected and non-infected
hosts. (A critical biological point is that an infected host can
become uninfected with time.) For each subgroup an unequal number of
hosts are sampled at each of 4 time points such that the results look
something like this for one type of host organism.
Time point 1 Time point 2 Time point 3 Time point 4 Hosts
Inf Not-Inf Inf Not-Inf Inf Not-Inf Inf Not-Inf Tested
G1-S1 1 14 11 4 11 1 13 2 57
G1-S2 7 8 12 3 14 2 15 8 69
G1-S3 1 24 6 18 8 15 9 15 95
G2-S4 3 12 12 4 10 4 14 2 61
G2-S5 5 10 5 6 8 7 11 14 57
G2-S6 2 26 12 12 11 16 14 12 105
The questions are how can group 1 (G1) be compare to group 2 (G2) and
how can subgroups be compared. I maintain that the heterogeneity
within each group does not prevent pooling of the subgroup data
within each group, because the groupings were made a priori based on
genetic similarity.
--
R. Mark Sharp, Ph.D. [EMAIL PROTECTED]
Southwest Regional Primate Center Tel: 210-258-9476
Director of Biostatistics and Scientific Computing Fax: 210-258-9883
Southwest Foundation for Biomedical Research
P.O. Box 760549
7620 West Loop 410 at Military Drive
San Antonio, TX 78245-0549
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================