I did face some road blocks but was able to bypass them by including some libraries. But when i am in the step

all.results.list[[name]] <- get.constituency.results(constituency), it fails each time.
Error: could not find function "get.constituency.results". I am not very experienced in R. For verifying it, i took it out of the For Loop and hardcoded the index to [1] or [2] to generate Aberavon.rda or Aberconwy.rda. But for both steps i get the same result.
I understand that "constituency" is holding the JSON format from read_html. But i am clueless further.

1. Code can not find 'constituency_names.df'.
   I believe this is because the code checks to see whether "data/constituency_names.csv" exists but if it does (already exist) it doesn't actually load it.
2. Code can not find function credit() called on lines 303 and 356

Other than that code runs perfectly for me. I do get an error that I don't have a perl interpreter installed but since we never load an xls file I dont think that matters unless of course that is related to the credit() function call.

@kjhealy I'm curious, what was your thinking behind the '15 candidates standing' threshold for defining the main parties?

### parties by N candidates
main.parties <- data %>% group_by(Party) %>% tally() %>% arrange(desc(n)) %>%
filter(n>14) %>% data.frame(.)

I'm trying to decide if this level of aggregation makes sense for a graphic I'm working on. Thanks for the insight!

@kjhealy many thanks for sharing the dataset and the script. When I attempt to run the first two sections in RStudio I encounter this error:

Error in eval(expr, envir, enclos) : could not find function "read_html"

Hi,
I'm very new to R and to Github.
I installed the packages I was missing and ran the code only to find that I was still missing 'gdata' (called on line 154) and 'mapproj' so you may want to add these to the library calls at the top.
Also the code seemed to fail for me as it was trying to write to the 'figures' subdirectory but that directory didn't exist. Adding the directory manually fixed this.
Still working through some other issues.