Sometimes when updating projects with new or revised data, additional information in the revised data file contradicts conclusions about the variable type when importing the original file. There are two relatively common scenarios where this can occur:
- Numeric variables that become text variables
- Text variables that become numeric
Numeric variables that become text variables
If a variable contains only numeric data when it is originally imported, it will be assumed to be a numeric variable from that point onward, and any non-numeric data will automatically be converted to missing data.
If it is important to change the type of a variable, the process is:
- Open the original CSV or Excel file in Excel.
- Create a new column, ideally immediately to the right of the original variable.
- Make the corrections in the new variable.
- Import the updated data file by selecting File > Data Sets > Update.
- Hide the original variable by ticket the "H" in the Tags column of the Variables and Questions tab, and use the new variable in its place.
Text variables that become numeric
Where a variable contains information that causes Q to interpret the variable as being text, such as commas, dollars signs or any other non-numeric data, Q will continue to interpret the data as being text, even if the underlying data is changed to text and re-imported.
You can change the variable to numeric by changing its Variable Type to Numeric in the Variables and Questions tab.