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

I. THE COMPREHENSIVE REWIEV OF THE DATA COLLECTION AND PROCESSING - 8. The electronic data processing of the census of 1970

cording to two population concepts, i. e. to permanent and resident population. Each mistake made in this respect, or the erroneous definition of the head of the household brought about a chain of errors as to the criteria of the persons belonging to the household. The clearing up and the correction of these errors caused no greater difficulties, since the errors followed a pattern; consequently, by finding out a convenient algorithm, the computer was able to transform to the right code-numbers. In this phase, as compared with the respective notions, the 3,16 per cent frequency of the errors was equal to 15,16 per cent of the total corrections. In this phase of correction it became almost secured that from among the information correctly adjusted for contradictions, the computer should collect the records containing the notions of the households and the families in a later phase. The final correction of the demographic and occupational criteria The systems of the automatic correction of both controlling subsets were identical. While in the first the notions of the educational level, in the second those of the occupational criteria prevailed. Both topics were fully separable from each other, but as far as the testing of the accuracy of some criteria is concerned, they mutually relied upon each other. It was the spot where the most complex chains of linked combinations occurred in the greatest number, mainly in the examination of the occupational criteria (see Annex no. 1. ) When scrutinizing the educational level, the equalizing of the levels obtained in different types of school with their actual equivalent levels accounted for the majority of the corrections. Although the computer treated such cases as errors, they have been reckoned with in advance, since the equalizing procedure by means of computers seemed more convenient to be performed than by means of manual techniques on the part of the census-takers or of the coding staff. The 0, 33 per cent share of the demographic corrections (accounting for 17, 5 per cent of the total), including the expected errors as mentioned above, seemed to be rather acceptable. The 0, 51 per cent share of the corrections performed in the field of the occupational criteria amounted nearly to a third part of the total number of corrections made. This share had to be reckoned with anyway, since this phase of corrections was the most complex one, concerning both the registration and the coding, and required special attention. The largest number of connections are showing themselves here. Had we undertaken the description but of one of them in details, it would have been clear at once that it was impossible to require from the coding staff a fully correct display of the chain of the linked connections, within the short term available. h/ THE EVALUATION OF THE METHODS APPLIED On the basis of the experiences obtained, comparing the results with those of the electronic data processing of the census performed in 1960, the following statements are to be made. Notable differences as to the previously applied methods manifested themselves in the following: 83