[R] problem with effects : 'subscript out of bounds'

Nicole Marie Ford nmford at uwm.edu
Sat Mar 10 06:58:36 CET 2012


hello.  help with effects plots.  here's the last bit of code before running the model and then the effects, then the error.

nor.dem <- norway$v162  ## nor.dem is my DV & it is continuous.
nor.dem <- as.numeric(nor.dem)-5
str(nor.dem)

(i had to do a great deal of coding here so i am snipping down to the end)

tmp[which(norway$v128 == "trust completely" & norway$v127 == "trust completely")] <- "generalized"
tmp <- factor(tmp, levels=c("particular", "generalized", "other"))
norway$trust <- tmp
nor.trust <- as.factor(norway$trust)  ####  nor.trust is my IV, a factor as you can see.

n.diversity <- norway$v221 ## control

n.net <- norway$v228  ##control  (there are others but they were coded the same)


n.mod1 <- lm(nor.dem ~ nor.trust + n.diversity + n.age + n.sex + n.educ + famecon + n.net)  ### linear model.  all of these variables already are specific to the dataset which i called 'norway' so there is no need to specify in the model.

summary(n.mod1)

> plot(effect("nor.trust", n.mod1), multiline=T)
Error in plot(effect("nor.trust", n.mod1), multiline = T) : 
  error in evaluating the argument 'x' in selecting a method for function 'plot': Error in apply(mod.matrix[, components], 1, prod) : 
  subscript out of bounds



~~this ran perfectly on my previous dataset so i am unsure of the issue.  thanks in advance.



More information about the R-help mailing list