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

I. THE COMPREHENSIVE REWIEV OF THE DATA COLLECTION AND PROCESSING - 11. The control of the reliability of the census

Floorspace of the dwelling The data figured only on the questionnaire of the 25 percent representative dwelling survey of the census. Thus also the survey of the post-enumeration control extended only to one fourth of the dwellings under survey: instead of the 7 750 dwellings in the case of other criteria here 1 800 dwellings figured. (Table 11). In view of the floorspace the gross error is very high owing to the uncertainty of declarations and estimates. (99 percent on the average). In connection with these percentage numbers it must be mentioned again that the gross error contains the deviations twice as a matter of fact. Then the majority of positive as well as negative errors is cancelled, thus the net error will be considerably lower: not more than 7, 5 percent. Hence it seems that the overwhelming majority of errors follows from lack of precision which is by all means of such degree that it exceeds the class intervals of processing - 10 sq. m. on the average. This is also shown by the fact that the net error exceeds double the standard error in two rubrics only. Notwithstanding also bias can be detected since the census contains more dwellings above the floorspace of 60 sq. m. than, the post-enumeration control which is more accurate or at least partly based on measurement. Also the misgivings of the data suppliers may have played a part beyond the inaccurate measurement because of the connection between the floorspace and the rent. For that matter the official survey of state-owned dwellings - as a secondary statistical source - sould surely supply more accurate data. However, in the course of censuses Hungarian statistics disregards original documents for reasons of principle. Table 11. INDEX-NUMBERS OF THE RELIABILITY OF THE DATA OF FLOORSPACE Total floor-space of dwellings sq. m. Percentage distr. Difference between percentages Identically grouped items Gross Net Index of inconsist ency Total floor-space of dwellings sq. m. Census P.C. Survey / Difference between percentages Identically grouped items errors Index of inconsist ency Total floor-space of dwellings sq. m. Census P.C. Survey / Difference between percentages in the percentage of the census returns Index of inconsist ency - 19 3,41 2,85 + 0,56 59,02 65, 57 16, 39 0, 679 20 - 29 10,28 10, 00 + 0,28 52,17 92,93 2,72 1, 040 30 - 39 19,16 18, 55 + 0, 61 54,23 88,34 3, 21 1, 093 40 - 49 18,83 18,32 + 0, 51 46, 59 104,15 2, 67 1, 283 50 - 59 17,93 16,15 + 1, 78 35,20 119,63 9,97 1,458 60 - 79 21, 73 23,35 - 1,62 59,13 89,20 7,46 1,140 80 - 99 6,20 7,99 - 1,79 50,45 127,93 28, 83 1, 364 100 2,46 2, 79 - 0,33 65,91 80,82 13, 64 0, 838 Total 100,00 100, 00 50,45 99,11 a/ 7,49 b/l,112 a/ The sum of the absolute numbers of net errors, compared with the total number of dwellings. - b/ Simple arithmetical mean. 111