Dear all,
I'm working with a dataset from a survey that has recently taken place. The original structure of certain questions in the survey were multiple choice. The ideal structure for these questions, in terms of variables, is that each possible option for the question (A,B,C,D... etc) have a separate value, with a 1 for yes (reported) and 0 for no (not reported). In other words, if Question1 has multiple possible answers (A,B,C...), then there should not be one variable created from this question, but rather, there should be a variable for each possible response: Question1_A, Question1_B, Question1_C, and so on. And within each of these variables, should be a 1 or 0 depending on if that option was reported or not.
The format I received the data in, though, was a single variable with a list of letters, (e.g. B,D,F) for each option reported.
Variable: Question1
Row 1 A, B, C
Row 2 B, E, F
Row 3 B, D, E
The data I have right now is housed in Excel prior to exporting to SPSS and/or STATA. I need to find a way to distribute these values into separate columns (variables). The tricky part, is that if I simply use "text to columns", it distributes the letters to new columns, but doesn't take into account that the first value reported might not be A. Is there a way to take the list of values in the single columns, and distribute them to assigned newly created columns?
Any help with this would be GREATLY appreciated!
Thank you in advance for any advice.
Best,
Jessie
Bookmarks