Thanks! I tried the array option and got the following error message: The subroutine SYMPUTS is unknown, or cannot be accessed-- Any suggestion? ... View more

Oi. I see. I just took another look and I don't have any special characters including (), +, $%^*( etc. But you are right--the list contains space and dash which would cause issues with searching. Do you have any other suggestions? Thank you! ... View more

So SAS would treat 'smelling' and 'smell' differently because they aren't identical? Hmm.. if that's the case I should probably add a whole bunch of keywords based on proximity/lemmitization-- but that's for another day. Thanks for pointing that out. In the past, I had a slew of keywords ending with a particular string 'XYZ' and coded SAS to pick up all the words/phrases that end with 'XYZ' so thought it would work in a similar way.    If the same search term matches multiple times within a record I would count it as 1. However, if multiple search terms were found in one record, I'd like SAS to be able to identify them and list them as such; in turn, we can end up having multiple rows per ID in the final dataset. I am trying to identify how many of those records were asking about a certain topic using the list of keywords that are associated with the topic. These are really great questions and I appreciate your feedback. Thank you!   ... View more

I would imagine the final dataset would something like this:    ID      Keyword             Title                                                                       Description                                                                                     191      Smell                Questions about symptoms                                  I am having trouble smelling...                                                  200      brain                 inpatient reimbursement brain surgery                Where can I find the reimbursement form?......                              140     Paralysis           FW: lab sample                                                    Can you provide me guidance on paralysis resources    In addition, it would be nice to know how I can get SAS to grab additional variables (e.g. email or phone number) from the SAS dataset and include them in the final dataset if needed. Incidentally, would the length of the description matter? The description was exported from an email and I think some could go over 32k.  ... View more

I meant extra space between the text and the quote at the end ... View more

Thanks for your suggestion! Can you please explain why I would want keywords in quotes? Also, why would some keywords end up looking like this: 'Phenotype '. Thanks! ... View more

Sorry, I should have explained it better. ID, title, and description are variables from the SAS dataset. Matching doesn't have to be case sensitive. No special characters or quotes. I don't know how to upload a sample dataset here but I hope this helps.    The list in the excel spreadsheet has one column named 'keyword'. Examples: spinal tap, nerve, CNS, Guillain-Barre syndrome, Stroke, etc.  The SAS dataset has multiple variables including ID, title, description etc.: Apart from ID which was auto-populated, everything else is free text  I would like the final dataset to have a list of records from the SAS dataset that contains any of the keywords from the list along with its ID, title, and description.    I hope this clarifies things a bit.            ... View more

Thanks for the link! I guess the polychoric option didn't work because I am using SAS 9.3? Not sure. Anyway, thank you!  ... View more

Hi Rick, Thank you for your reply- Quick question in the Step 1 process: proc corr data=ordinal polychoric outplc=OUT; var x1-x5; run; I am not sure what 'outplc" does, did you mean outp? Also having 'polychoric' in PROC CORR gave me an error message- any suggestion? Thank you! ... View more

Hello.Thank you for your reply! My question is how do I specify the tetrachoric option in the PROC CORR (i.e. GPLOT)? I've used the tetrachoric in PROC FREQ in lieu of PROC CORR so I wasn't sure how to generate a plot/graph in this case. Hope this makes sense!! ... View more

Oh ok. I said that not because you suggested a different method, but I read somewhere that standard methods of performing FA (using a tetra/polychoric correlations matrix) assume that factors are continuous. I also noticed that having a three-level nominal var in the matrix generate some funky numbers (compared to only having binary vars). Thanks again for the suggestion! ... View more