Data editing perspectives

Abstract

Data editing, tracing and correcting errors in records, takes up a great deal of the avail­ able human resources for the production of statistics. As budget restraints become more severe, the pressure to obtain more efficient data editing procedures increases. There is a need for procedures that restrict the attention of human judgment to errors that are crucial for the quality of the statistical outcomes and that leave the remaining decisions to automatic procedures. The present contribution firstly deals with the types of errors that occur in survey research, then summarizes the capabilities of the state-of-the-art micro editing software systems and finally gives an overview of more efficient data editing methods. Next an overview is given of fully automatic editing, and of methods to edit records with influential errors only: selective editing, aggregate editing and graphical macro editing.