COMP

Integer Data in Binary Format (same as BINARY)

Single precision floating point in hexadecimal format, 4 bytes long

Double precision floating point in hexadecimal format, 8 bytes long

Integer data in Packed decimal format/ 8421 code / (Int((n/2)+1)))

Integer Data in Binary format (same as COMP/BINARY)

Integer Data in Native/Extended Binary (No Truncation due to PICTURE clause size)

Unsigned Packed Decimal (No Sign) (Int(n/2)) (RM/COBOL Only)

EBCDIC Char string with single byte (8 BITS) per character, Default clause

EBCDIC Char string with two bytes per character (DBCS or DBCS-EDITED)

INDEX

32 Bit Binary Format to hold the displacement of a table, Assigned by SET statement

Unicode Char string (UTF-16) with one byte per character

NATIVE phrase is syntax checked, but has no effect on the execution of the program

Integer data in Packed decimal / 8421 code  (same as COMP-3)

POINTER

32 Bit Binary Format to hold the address of a variable

64 Bit Binary data to hold the address of a primary/alternative entry point of a COBOL program

32 Bit Binary data to hold the address of a primary/alternative entry point of a COBOL program

Handle for accessing a class in OO Cobol