The CVT$x functions map character strings into numeric values.  CVT$% maps
  the  first  2  characters  of  a  string into a 16-bit integer.  The CVT$F
  function maps a 4- or 8-character string into a  floating-point  variable,
  depending on whether the program in compiled with /SINGLE or /DOUBLE.

  Examples

  A% = CVT$%(EMP_NAME$)
  A = CVT$F(EMP_NAME$)

1  –  Syntax

      int-var = CVT$%(str-exp)
      real-var = CVT$F(str-exp)