Survey data - combined data types - value vs. text answers

Question

Dears,

would you be able to give me a hint for the below topic?

I have a practice survey where I have various kind of answers. Some of them are on scale 1-10, others are text answers (very mixed and individual as you can imagine). The thing is I need to divide answers to correspondent categories so I can work further with the data. In the end I need to have specific columns, e. g. Value (numerical answer), Others (text answers naming tools) and Good (things people are satisfied with).

But the text answers vary a lot - some including interpunction, special characters, spaces... But for some words I need to keep more words in one row, e. g. in answer "orders processing"... For some words I need to keep the number inside of the text, e. g. in answer go2.es etc.

Please see a simple example below. I will appreciate any advise on how to deal with text data processing for similar case and even for some more difficult inputs if you have some tips. Thank you a lot 🙂

dougperez · Answer

Usually I would try to clean this dataset with regex. One thing that you should have is a list of programs, and the "good" collumn I  would put the rest of the string.

TrevorS · Answer

Hello @KaMa

Thanks for posting on the Community!
In order for the Community to better assist you, can you please reply back with the sample data you are using and a copy of your workflow?

This will help our users know where they need to focus!
Thanks,

TrevorS