I am importing data from a .csv file in the form of 40,000 unique US zip codes, each one matched to one of 200 region codes. My goal is to produce a dictionary of each region code and its linked zip codes. I then want to pass to the dictionary several hundred thousand zip codes to get a list of the appropriate region codes. This seems like a good application of the new function Associations and key-value pairs. Specifically - I want to pass my list of values to get a list of keys. Here is where I am, using abridged and simulated data:

Import the two columns codes and skip the headers in the file

`codes =  Import["filepath"][[1, 3 ;;  ,  ;; 2]];`

This produces a list in the form of: {zip, region} such as this:

`{ {16505, 234}, {16504, 234}, {91521, 456}...}`

I then use the new GroupBy function to produce an Association of key-value pairs:

`groups = GroupBy[codes, Last -> First]`

which produces and Association in the form of:

<| 234 -> {16504, 16505}, 
    456 -> {91521} |>

My next step was to send my list of zip codes using the Position function. Testing with a single zip does return the key value:

In= Position[groups, "16505"]
Out= {{Key[234], 2}}

Which I can Part to get the Key as such:

  In= Position[groups, "16505"][[1, 1]]
    Out= Key[234]

Here is where I am running into trouble. When I submit a list of zip codes I get an unexpected result back:

In= Position[groups, {"16505", "91521"}]
Out= {}

I have tried using other approaches such as the PositionIndex function and Extract. Can anyone suggest a better way to approach this?