@Mystik wrote:
Okay. Is there a way convert a variable to accept both character and numeric. This is because my variable set1, has '150', '569.8999', '.' , ' ' as well as '17/04/2011' as inputs in set1 column?
What is your variable set1 supposed to represent? Are you going to do arithmetic with the values? If so then your field should be numeric. Which brings up what numeric value you want for '17/04/2011'.
If you are not doing arithmetic then character should be fine and the earlier response using PUT is the way to convert numeric to text.
I would say that any variable that has a value that is almost certainly a date and other sort of random numeric values means that something was done poorly at a much earlier step as dates should be separate variables.
I would also be tempted to go back to the very source of the data and make sure that the columns actually align correctly with purpose and variable names. Perhaps Set1 is one data set means something else in the other. In which case they probably should not have the same variable name if you are going to combine the data.
Consider, I have two data sets. In the first set Name means the name of a person. In the second set Name is the name of city.
If I combine the two data sets with a set statement I could something like:
John Smith
Mary Jones
Boston, Ma
Chicago, Il
Which makes little sense. Perhaps something similar has happened with your data that is attempting to mix values that shouldn't even though they have the same variable name.
... View more
