FORMAT Specification (MATRIX-END MATRIX command)

FORMAT indicates how the internal (binary) values of matrix elements should be converted to character format for output.

• The format descriptor is any valid IBM® SPSS® Statistics data format, such as F6, E12.2, or A6, or it can be a format type code, such as F, E, or A. It specifies how the written data are encoded and, if a width is specified, how wide the fields containing the data are. (See FORMATS for valid formats.)
• If you omit the FORMAT specification, the default is F.
• The data field widths may be specified either here or after BY on the FIELD specification. You may specify the width in both places only if you give the same value.
• An additional way of specifying the width is to supply a repetition factor without a width (for example, 10F or 5COMMA). The field width is then calculated by dividing the width of the whole output area on the FIELD specification by the repetition factor. A format with a digit for the repetition factor must be enclosed in quotes.
• If the field width is not specified in any of these ways, then the freefield format is used--matrix values are written separated by one blank, and each value occupies as many positions as necessary to avoid the loss of precision. Each row of the matrix is written starting with a new output record.
• Only one format descriptor can be specified. Do not try to specify more than one format; for example, '5F2.0 3F3.0 F2.0' is invalid as a FORMAT specification on WRITE.