How do I concatenate several text columns while ignoring missing values?

[amanda_cr]

Hi,

Here is what my data looks like:

Unique ID      Col1      Col2     Col3     Col4

   1                Text1                  Text3

   2                Text1     Text2                 Text4

   3

   4                              Text2     Text3

I'm trying to create a new column that lists all the text from column 1 through 4, with commas and a single between the different texts, and that shows nothing when there is no text. Here's what I want:

Unique ID     NewCol

   1                Text1, Text3

   2                Text1, Text2, Text4

   3

   4                Text2, Text3

I can't seem to figure out a way with out creating ", , ," for line 3, and extra commas in the text strings (this creates a problem since I actually have 29 columns of text I'm trying to put in a text sting in a single column).

Any suggestions?

Amanda

[ballardw · Accepted solution]

In a datastep

new_col = catx(',',col1, col2,col3,col4);

the first argument in the catx function is the character(s) to place between the values indicated. Blanks are suppressed

[amanda_cr]

Works perfect! Thank you for your help!

[amanda_cr]

Works perfect and so simple! Thank you.