You define delta twice, that's a no-no, for example, instead of for(k in 1:K){ delta[k]~dnorm(d.ethnic,prec.ethnic)}...
Ah, I can't install it if I'm using the R 3.1.0 version built for Mavericks, but...
I can build it from source, but just install.packages("rjags") works. Hmmm, it must...
I'm using 3.1.0 on OSX 10.9 and can install it right out of the box using install.packages("rjags")...
I can use the coda diagnostics after fitting the model using jags.model(), then update()...
I can use the coda diagnostics after fitting the model using jags.model(), then update()...