Advice on appropriate test on proportions with three factors with replication


New Member
I am planning a study with three factors (salinity, parasite type and species). Can anyone suggest an appropriate statistical test? I am wondering if some form of chi square would be adequate. I tried to convey the essential information concisely below but would be happy to provide more information. I also attached a hypothetical data set in long form. Column titled "neg_pos" in data set represents individual fish that are positive for the parasite (1) and negative for the parasite (0). "ICH_Type" column is the parasite type.

Ho: There is no affect of species, salinity or parasite type on prevalence of infection.

- species (two levels): herring and trout

- salinity (salinity of the water fish are kept at) (three levels on trout; two levels on herring): low, medium, high (for trout) and medium and high (for herring, which would otherwise die at low salinity).

- parasite type (three levels): marine type, freshwater type and saline (no parasite) negative controls

Positive control: trout x low salinity x freshwater parasite

Each species x salinity x parasite combination has three replicates each with 15 fish

The prevalence of parasite infection among fish in each replicate will be counted.