Abstract: This overview is given on the Generalized Survey Function Development Project (GSFD) developments at Statistics Canada, particularly the Generalized Editing and Imputation System (GEIS). The paper contains a more detailed exposition of the newly developed GEIS system than would normally appear in a paper of this kind, since this represents Statistics Canada's latest efforts at supporting an editing system with a DBMS product.
Abstract: The objective is to present macro-editing methods as a weapon against over-editing. In experimental studies in data-processing environments and in production they have proved superior to micro-editing methods for editing quantitative data. Savings of manual verifying work of 35 up to 80 per cent are reported. The paper is mainly devoted to an overview of studies of macro-editing methods. Emphasis is given to the rational aspects of macro-editing as compared to micro-editing by presenting the results of several studies and by discussing the problems connected with micro-editing. The methods are described in a brief and schematic form to make…them easy to understand. This review serves as a basis for considering macro-editing methods when designing an editing system for a survey with quantitative data. Detailed descriptions of the methods and studies are found in the references given in the text and in the reference list. Stress is laid on the specific features of every method in order to facilitate a choice. The paper concludes with a summing-up discussion of macro-editing versus micro-editing methods.
Abstract: A macro-editing application developed on a PC using the SAS-system for the programming is described. In the paper a comparison is made between the actual production editing and the macro-editing. It is shown that by using the macro-editing instead of the production editing, the number of flagged records is reduced by 80 percent.
Abstract: Some experiences are reported of Federal Statistical Offices using SAS for data-editing application development in survey processing. SAS reduces the necessary resources needed for development and maintenance of data-editing applications.
Abstract: The impact of data editing on data quality is a theoretical and practical matter, studied by the Joint Group on Data Editing of the Statistical Computing Project Phase 2. In this context this paper focuses on the difference between Imputation at the Record Level and at the Weight Adjustment at the Aggregated Level . After describing two experiments carryied out using a large and complex labour survey, it is stated that redundancy is necessary for record level imputation and that two types of variables, flow variables and semantic variables, affect differently the edit process.