I have some intermittent data on age of host (Perca fluviatilis) against occurrences of Diplostomum and Tylodelphis. The occurrences appear to 'oppose' each other but I need something to say if this is merely an 'artefact' of the data or not.
Not sure to have fully understood your question: if I have, maybe you can try to build a generalized linear model (a negative binomial regression for abundance and a logistic regression for prevalence) for both Diplostomum and Tylodelphis using age and, respectively, Tylodelphis and Diplostomum as independent variables, in order to simultaneously evaluate the effect of both age and the other parasite abundance (or prevalence) on each parasite species
Thanks for the suggestion, I think you have understood the problem, but I wonder if you have any specific software in mind for that. I appreciate you responding. Tricky question!
STATA, for example. There is the function in "statistics->count outcome->negative binomial regression: for parasite abundance it's ok. In statistics->binary outcome->logistic regression: for parasite prevalence, coded 0-1 (o=negative; 1=positive). Also in R you can build generalised linear models http://127.0.0.1:15506/library/stats/html/glm.html
Good job!
look also here:
https://www.researchgate.net/post/STATA_GLM_and_negative_binomial_regression1
Hi Laura, thank you for the suggestions. Would you know if this can be done in SPSS / PASW?
Thanks!
Hi Laura,
actually I see that you can build a GLM in SPSS.
thanks for all the help.
Thomas
Dear Laura,
From the figures, can you tell if I have done the calculation properly? If you can offer advice I would be very grateful, and I would in that case invite you to be co-author if you wish. In any case I will acknowledge your advice.
best wishes,
Thomas
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size