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

Out comments on the comparisons are as follows: 1. Between the number of inhabitants calculated from the full-scale survey and by using the sample, a deviation amounting to 6 500 persons is manifesting itself. It results from the fact that the sample was adjusted according to the number of inhabitants as shown in the volume entitled "l. Preliminary data" which differs from the final result. (The difference between the preliminary and the final figures amounted to 6 499 persons on nation-wide level.

) 2. The last column of the table shows the confidence intervals according to the 95 p. c. significance level. That is to say that the deviations showing themselves, in 5 cases of 100 (i.e. once in 20 cases) may exceed the estimated limits. The computation was performed by means of simple random selection (see the table shown in the 2

n

C* volume, p. 230. ) 3. The deviation between the full-scale and the sample survey data according to sex and age groups remains in the case of each item, between the confidence intervals. When the data are disaggregated according to age groups of five years and family status, the deviation exceeds only once the intervals in question. Similarly, the deviation is also greater in the case of "other branches" of the national economy. c/ TESTING FOR THE RELIABILITY OF THE CENSUS RESULTS The reliability of the census results obtained in 1970 was controlled by means of a secondary registration using a 1/4 per cent sample. The post-enumeration control survey has been also preceded by an experimental survey on the basis of which the final drawing up of the inquiry forms (as far as their content and structure was concerned), as well as the plans of the organization and execution of the census were performed. The sample destined to controlling purposes, including 2, 5 per thousand of the dwellings, was centrally made. The sample was treated secretly till the launching of the operations, i. e. during the execution of the census. The regional sample was based on census districts. Each hundredth census district of the country has been selected by means of a mechanical procedure. In the second stage, the 0, 25 per cent sample was shaped by means of the regional quartering of the census districts. The division into quarters took place by using the map and the Directory as tools. The maps were also needful to the supervisor when exercising control on the spot. The post-enumeration control survey came into action on the 20^ January 1970, i. e. directly after the conclusion of the census itself. As far as the presence and absence are concerned, the status st prevailing on the 1 January 1970 as census moment had to be reconstructed. The operation covered 7 900 dwellings and 27 000 persons. As for registration purposes, a registration form of the dwelling and the personal data sheet of individuals have been used. The controlled groups of questions were as follows (in the respect of personal data): - age; - place of birth (the residence of the mother at the confinement); 54