|
<< Click to Display Table of Contents >> Navigation: Reference Manual > Windows and dialog boxes > Raw data treatment > Raw data treatment |
Use [Treatment] [Change file treatment] from the Raw data view window to alter the way the raw data file is read.
When closed the data file is read again using new settings.
You can set the file type yourself, see Files, Data file. You should only need to do this if the file name has the wrong extension.
Most data files store the data as lines, so the line lengths are known.
IBM 360 column binary CSI files do not have lines. Every record is 80 columns (160 bytes) and each record follows immediately after the previous one.
If the record length is not 80 you need to change this setting to the correct record length.
You can also set a padding after each record, in case the producer has added one or more line end characters to each record.
Normally blank (completely empty) lines in data are ignored.
You can set this to include empty lines in the display, and in any saved data file.
This does not work with CBE data files.
This setting causes all lines to be padded out to the same specified length in the display, and in any saved data file.
Any lines that are too long will be truncated and any lines that are too short will be padded.
If [Split records to new length] is set then lines that are too long are split into separate lines.
If [Combine short lines] is set then any lines that are shorter than the maximum will be combined with the following lines until the record is filled.
TIP: Combine short lines can be useful if a character file contains answers to text questions that might include a return. For example data exported from SPSS.