Search code examples
spss

Problem reading variables containing mix of numbers and strings

I am reading an Excel file (see syntax below) where some of the fields are text mixed with numbers. The problem is that SPSS reads some of these fields as numeric instead of string and then the text is deleted.
I assume this happens in cases where a large part of the first rows are empty or with a numeric value and then it defines the variable as numeric.
How can this be avoided?

  GET DATA
  /TYPE=XLSX
  /FILE='M:\MyData.xlsx'
  /SHEET=name 'Sheet1'
  /CELLRANGE=FULL
  /READNAMES=ON
  /DATATYPEMIN PERCENTAGE=95.0
  /HIDDEN IGNORE=YES.
Solution

  • When you use the get data command, the subcommand /DATATYPEMIN PERCENTAGE=95.0 tells SPSS that if up to 5% of the values in the field do not conform to the selected format it's still ok. So in order to avoid cases where only very few values are text and the field is read as number, you have to correct the subcommand to:

    /DATATYPEMIN PERCENTAGE=100