string variable management

[Chiara Tasselli]

Good afternoon,
I would like to create a categorical variable from a string variable composed by several and heterogeneous words.

Is there a Stata command that works in this way (see below)?

example:

if X contains "a word that could be in the string " gen = y

Many thanks in advance for your time.

[Nick Cox]

* Example generated by -dataex-. For more info, type help dataex
clear
input str9 var1
"frog"     
"toad"     
"toad frog"
"frog"     
"newt"     
"dragon"   
end

gen wanted = strpos(var1, "frog") > 0

. list, sep(0)

     +--------------------+
     |      var1   wanted |
     |--------------------|
  1. |      frog        1 |
  2. |      toad        0 |
  3. | toad frog        1 |
  4. |      frog        1 |
  5. |      newt        0 |
  6. |    dragon        0 |
     +--------------------+

Watch out for spelling differences, including differences of upper and lower case.

[William Lisowski]

To Nick's advice about spelling differences, let me also add to watch out for "substring" matches - if you are looking for "ant" and your variable contains "phesant", for example, you will get a match you perhaps do not want. The strpos() function matches strings of characters rather than "words", but that is often sufficient.