The metainformation system and the conceptual level of statistical data modelling

Article type: Research Article

Authors: Šoltés, Dušan^{1; 2}

Note: [1] This paper is a modified and updated version of a paper prepared for the Seminar on Integrated Statistical Information Systems and Related Matters (ISIS '82) convened by the Conference of European Statisticians and organized by the Computing Research Centre, Bratislava, Czechoslovakia, 10–14 May 1982. The modifications concern the most recent results achieved in the Metainformation System development within the ECE/UNDP Intercountry ‘Statistical Computing Project’.

Note: [2] Dušan Šoltés is currently a research project leader for the Metainformation System in the Computing Research Centre, Bratislava. Before joining the Computing Research Centre in 1971 he was a programmer-analyst and a team leader in business and enterprise computing centres. Dušan Šoltés graduated from the University of Economy, Bratislava, in 1965, received a PhD degree in Computer Sciences from the same university in 1976 and the degree ‘Senior Scientific Worker’ from the Slovak Academy of Science in 1981. He is co-author of three textbooks on computer sciences for post-graduate courses. He published numerous articles in the fields of database-oriented information systems and metainformation systems. In 1981, he was appointed to the international joint group within the ECE/UNDP Intercountry Project ‘Statistical Computing’.

Abstract: Several problems arise in the use of the metainformation system (METIS) and its system of catalogues and dictionaries of formalized descriptions as a tool for the conceptual design of statistical data modelling. The general problems of statistical data modelling can particularly be identified in relation to the construction of statistical data base systems. Statistical data modelling is to be perceived as modelling of a certain already existing statistical data system rather than as modelling of the objective reality and its subsequent representation in the form of an ‘ideal’ data system. The system of catalogues and dictionaries of formalized descriptions should be used directly within the conceptual framework for statistical data modelling which is based on the comparison and analysis of aims and functions. METIS and its formalized descriptions can be seen not only as an efficient tool for the sphere which appears to be neglected in applied methods as well as in conceptual approaches to data modelling, but also as an efficient tool serving the purpose of user communication with large statistical data base systems. In the second case METIS can be used as a simple locator system (SLS), a system informing in detail (DIS) and/or as a system for retrieval and use of data (SRU) by various categories of users. One of the main advantages of using METIS in these ways is that a considerable number of methods and procedures of the formalized descriptions, generation of catalogues and dictionaries and procedures for their use have already been implemented on the computer.

DOI: 10.3233/SJU-1984-2109

Journal: Statistical Journal of the United Nations Economic Commission for Europe, vol. 2, no. 1, pp. 97-108, 1984