Scholion.ai

Sampling Control of Literacy Data*

S. S. Zarkovic

Journal of the American Statistical Association · 1954 · Volume 49 · Issue 267

DOI: 10.1080/01621459.1954.10483518

Abstract

* The author wishes to express his indebtedness to Mr. S. Krasovec, formerly director of the Federal Statistical Office, and to Mr. M. Macura, director of the Serbian Statistical Office, who spent a lot of energy to make this research possible. The research described in this paper belongs to the new field in statistics that could be labeled “The problem of the value of statistical data.” The most important work in connection with this problem has been done in the USA and India. So far obtained results and experiences can be found in the following papers: M. H. Hansen, W. N. Hurwitz, E. S. Marks, W. E. Mauldin: Response errors in surveys, Journal of the American Statistical Association , 46 (1951), 147–90, P. C. Mahalanobis: Recent experiments in statistical sampling in the Indian Statistical Institute, Journal of the Royal Statistical Society , 109; P. V. Sukhatme, G. R. Seth: Measurement of non-sampling errors, Journal Indian Society Agriculture , Vol. 4; P. V. Sukhatme: Measurement of observational errors in surveys, Revue del'Institute Internationale Statistique , Vol. 20; M. H. Hansen, W. N. Hurwitz, L. Pritzker: The accuracy of census results, American Sociological Review , 1953; W. E. Deming: On errors in surveys, American Sociological Review , Vol. 9; E. S. Marks, W. P. Mauldin: Problems of response in enumerative surveys, American Sociological Review , Vol. 15; E. S. Marks, W. P. Mauldin, A. Nisselson: A case history in survey design; The post-enumeration survey of the 1950 census, Journal of the American Statistical Association , 48 (1953), 220–43; G. L. Palmer: Factors in the variability of response in enumerative studies, Journal of the American Statistical Association , 38 (1943), 143–52; S. S. Zarković: Completeness of enumeration (in Serbian), Federal Statistical Office, Belgrade, 1954; A. Gosh: Accuracy of family budget data with reference to period of recall, Calcutta Stat. Assoc. Bul. , Vol. 5; M. H. Hansen, W. N. Hurwitz, W. G. Madow: Sample Survey Methods and Theory , New York, Wiley, 1953; W. E. Deming: Some Theory of Sampling , New York, Wiley, 1950; S. S. Zarković: Population Census Errors (in Serbian), Federal Statistical Office, Belgrade, 1954. An attempt is described to control the value of literacy data by the use of sampling methods. The reason for this research is the widely known unreliability of this sort of statistics in countries with a high rate of illiteracy. This research has been conducted as a part of the post-enumeration survey, taken in connection with the Yugoslav census of population as of March 31, 1954. The aim of the survey was the control of accuracy and value of different census results. The value of literacy data was checked on the sample of individuals by means of reading and writing tests. The results show ( i ) that literacy is a continuous variable, and ( ii ) the unreliable character of literacy statistics is connected with the difficulty of defining the limit between the; different levels of literacy. Since these limits cannot be defined in the census of population, the best method to check the value of literacy data seems to be the use of sampling methods.

Enable JavaScript for citation tools and additional paper interactions.