Forum Discussion
Helper IMore efficient way to keep only numbers in multiple columns
the first thing that comes to mind is to merge all the columns through a character not present in the texts (#, for example) apply your rules to the union columns and then redo the division of the colonan through the added separator (# , in the example)
Helper IThanks for your reply Rocco,
I'm a complete beginner with power query and appreciate your help, but don't know how to upload data or save it in the binary fashion that you are showing at the beginning of your posts, or clean it quickly of non-commerically senstive data, so I will describe below:
The data is 210 columns, of which 140 contain survey response scores 1-7, in about 3,500 rows
Unfortunately, the 1s, 4s and 7s have text as well, like: "Completely agree7", or "7Fully understand", "Neutral4". I want to strip those out and save as numbers. But, I don't want to lose any text in other columns.
Those 140 columns are not next to eachother. I could manually re-arrange them of course.
The column names are also very long, because the reflect the survery questions, so that start like "Q_23..." but are up tp 200 characters long.
- If you don't post a meaningful example of your data, I can't give you a specific / complete answer.
- On the other hand, being, as you say yourself, you are inexperienced, you are unable to adapt my suggestion to your specific case.
- do you have any idea how to overcome the impasse?
- k1s1
Hello Rocco,
re 1 - I don't know what's not meanignfgul about the way I described the data
re 2 - I don't understand it
re 3 - No I don't, but thanks anyway for your efforts on this