compositional analysis?

[ansepans]

Hi,

I have data on the bacterial compositions of skin swabs of 900 patients. I want to know if there is a difference in bacterial composition between males and females. I have stumbled upon compositional analysis but I don't know if it is the right way to do it and I don't know how to do it. Can any of you help me?

I have data that look like this:

female data set

sex sample count percent

1  CNS   511  63%

1  Gram-negative rods 63 8%

1  Gram-positive rods  63 8%

1  Micrococcus  127   16%

1  Other Gram-postive cocci  37 5%

1  Staphylococci   8  1%    

male data set

2 CNS   447  68%

2 Gram-negative rods 33  5%

2 Gram-positive rods 54 8%

2 Micrococcus  98  15%

2 Other Gram-positive cocci 18  3%

2 Staphylococci  5   1%

[nicobuettner · Accepted solution]

Hi,

just to maybe get things going: Can you specify what your desired outcome would be?

I have no idea of compositional analysis. My data-management-based approach would be to merge the datasets and just compare the male vs, female values with simple if statements or something.

Let me know if you need a code example for that should you need one.

Best,

Nico