Accuracy and Timeliness in Large-Scale Data-Entry Subsystems

Abstract

Errors generated at the input to automated information systems threaten the utility of the outputs produced by such systems—i.e., the classic GIGO (Garbage In—Garbage Out) problem. Large amounts of effort and attendant costs are involved in dealing with these errors once they are in the system; however, more attention should be given to a coordinated solution of the problem at the source—the data entry point. A large-scale data-entry subsystem model is presented, together with statistics on error and timeliness. The concept of a Source Data Entry Module (SDEM) is presented along with considerations of design criteria and a design approach. An example is given, paralleling a testbed for which data collection is planned. It is suggested that a generalized SDEM design may be achieveable, providing a basis for the solution of a wide range of automated information system data entry problems.