Data tools transform_multiple_dummy
- 1. Transforming multiple responses into a series of dummy variables Survey Methods
- 2. In Google Form, a “checkboxes” question allows respondents to choose multiple response categories. The dataset will look like this.
- 3. To separate items (response categories), go to DATA menu in Excel and choose “Text to Columns.” Click on Delimited.
- 4. Add Semicolon as Delimiter and voila! The font color has been changed.
- 5. To more effectively analyze this data, we want the same fruit (response category) to be in the same column. In the table below, one could quickly and easily see that respondents #1 and #4 like apples, but not respondents #2 and #3, etc. One could manually manipulate the data (sort, copy, paste) to do this, but it’s tedious.
- 6. There may be a number of solutions to this problem. One is to use function COUNTIF. Then look for COUNTIF. Look in the cells B2, C2, D2, E2 and count the number of times the word “Apples” appear.
- 7. The answer is 1. This respondent likes apples. 0 if s/he doesn’t like apples.
- 8. Do the same for the other three fruits. Use absolute reference (“$”) if necessary. Make sure the range is constant (B, C, D, E) F, G, H, and I columns contain dummy variables (1 if the respondent likes the fruit, 0 if not). Then we can count the total number of 1s. #1 and #2 both like two kinds of fruit. #4 likes all four.
- 9. We don’t need the columns B, C, D, E any more, do we? Let’s delete them. No, they cannot be deleted because the COUNTIF (and SUM) data depend on those original columns. Options -Hide the unwanted columns -Copy and paste the values (using Paste Special). Then the COUNTIF data are no longer dependent on the original data.