I'm analysing some ID50 data for 2 different groups and had already
calculated this by hand using Reed-Muench formula, when I came across the
dose.p function in R.
I have 2 queries:
1) dose.p gives me a different answer to Reed-Muench, and actually I suspect
wrong answer, given that the dose.p result dosage stated to infect 50% is
actually stronger than the dose used in my experiments caused above 50%
infection!
>SetA<-data.frame(c("Hi","Med","Lo"),c(8.44,7.46,6.22),c(31,27,35),c(21,14,6))
>names(SetA)<-c("Group","dose","total","infected")
>attach(SetA)
>y<-cbind(infected,total-infected)
>model<-glm(y~dose,binomial)
>library(MASS)
> dose.p(model,p=0.5)
# Dose SE
# > p = 0.5: 7.6056 0.2263418
BUT! The 50% result is Higher than the Med Dose (Log 7.46) which in test
produced 51.9%
(and by R-M formula the ID50 would be be Log 7.40)
2) How can I compare the ID50's of two sets to see if the difference between
their ID50 is signficant?
I also have one further Set B and will calculate ID50 for this (by R-M its
working out at Log 6.84), and am interested to see if Set A and Set B differ
in their ID50.
Many thanks in advance
--
View this message in context:
http://r.789695.n4.nabble.com/ID50-i-comparisons-ii-dose-p-vs-Reed-Muench-tp3579311p3579311.html
Sent from the R help mailing list archive at Nabble.com.
