1970 HUNGARIAN CENSUS OF POPULATION Information on the data collection and processing (1977)

I. THE COMPREHENSIVE REWIEV OF THE DATA COLLECTION AND PROCESSING - 7. The preparation, programme and organization of the data processing

tern, each third or fifth had to be coded by the revisor, namely in a separate coding sheet. In another phase of the work, both codings have been compared i.e. collated and the disrepancies have been recorded. in the following the person doing super-revision - in the majority of the cases, a member of the census-staff - had to decide whether the error was made by the person performing the coding or the revision. According the formula given, he had computed the percentage of the errors and evaluated the quality of the coding. If the number of errors committed amounted to 4 per cent or more, the material had been entirely recoded. Between 1-3 per cent only the serial errors (the systematically performed errors) came to supervision and correction. The coded material remaining under 1 per cent of error was not corrected by manual techniques, but the procedure was entrusted to the auto.matic mechanical correcting. The coding has been performed, by small lagging, in separate phases, simultaneously with the material of the dwellings and buildings as well as of the persons. That ensured that the individual inquiry lists had not be removed from the registration lists, consequently, the material remained full and controllable. Practically, the situation was the same with the punching where the recording on the two sorts of questionaries has been performed parallel with the full material. The punched cards have been treated, on account of the different character and solution of the further steps, quite separately, and although the steps were going on parallel with each other, they have not influenced each other. In the following, the different data files did not meet until the preparation of the tapes. In this phase the personal, family- and household data were connected with the data of the flats, and the letter to the data concerning the families and households, and so on. The final data files were processed quite separately, independently of each other, on the basis of independent programmes. d/ THE ORGANIZATION OF THE PROCESSING The two main phases of the processing were: the full-scale processing and followingly, the processing of the sample of 25 per cent representativity. As to the full-scale processing, the computer made copies of the data material of the 25 per cent representativity, needed to the full-scale processing, and combined them with the material of 75 per cent representativity. The processing was performed on the following topics: Code number of the topic on the tables Demography 1 Occupation 2 Household and family 3 Fertility 4 Work-place and residence 5 Dwelling 6 Building 7 Other 9 60