Dear Statalist,

I have been searching for ways to remove a sequence of characters from a string variable only if they appear in the beginning of the word, and I have not found a successful way to implement this.

My example data are as follows, with quotation marks added to make it clear where blanks are present in the string variable traintype:

obs traintype
1 " - Training on anesthesia machine - provider"
2 " - Training on anesthesia machine - trainer"
3 "Training on anesthesia machine - provider
4 "Training on anesthesia machine - trainer"

I would like obs 1&2 to appear as obs 3&4 however, I am unable to just replace the sequence " - " from the whole string because this would remove the sequence later in the variable that I want to maintain.

Thank you so much for considering this question.

Best,
Alison