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

I. THE COMPREHENSIVE REWIEV OF THE DATA COLLECTION AND PROCESSING - 6. Sampling techniques applied in the census

Confidence intervals of the sample data The confidence intervals may be easily defined in the case of an individual simple random sampling and, considering that in the sample given, they depend only on the order of magnitude of the sample data, they can also be tabulated with common validity. The 2

n

C' volume of the Census presents these figures in tables and diagrams. In our case, however, the selection took place according. to flats (and not to persons), the method applied being a geographically systematic (i. e. automatically stratified) sample survey. Therefore, in order to supply sufficient information, at first the tables showing the errors obtained from individual simple random sampling were presented, by means of which the approximative evaluation of the confidence intervals might be attempted. Thereafter, on the basis of the surveys carried out hitherto, the estimated figures reflecting the effect of the sampling techniques were presented for several criteria, in order to define the impacts of the stratified selection in the case of the flats, and also for similar criteria. According to the examinations performed in this respect, with the criteria displaying a high dispersion within the flat (sex, age, family status, number of earners and dependants), the dispersion of the group mean will be low, and the results obtained as to the latter will be by far more exact than in the case of the individual selection. In the case of other criteria however, the composition of the persons living in the same flat is more homogenous (as to occupational and cultural characteristics), thus the accuracy of the data of this kind will be reduced as compared with the individual selection. The evaluation of the data originating from the sample survey of 1 per cent representativity As to the evaluation of the accuracy of the preliminary data based on the sample of 1 per cent representativity, cf. the detailed data shown in the publication entitled "2. Detailed data" (pp. 234-238.).The projected results of the sample have been compared there with the data registered according to the sex, the age and the family status. Consequently, there is a possibility of the comparison with the full-scale census results. The respective data are shown on the following tables. 51