Conferences

The author begins with a synthetic presentation of Neyman’s contribution to the theory and practice of sampling. First, Neyman’s initial contribution to the theory of sampling in the 1920s is emphasized, and second, Neyman’s famous 1934 paper (Neyman, 1934) is discussed in detail. It is stressed that despite many contributions of others to the technical development of sampling in the period from 1900 to 1926, Neyman’s 1934 paper played a crucial role in turning pure mathematics into real sampling plans for actual randomly selected large scale surveys. The Neyman’s 1934 paper played a catalytic role in the actual use of randomization in the utilization of methods for random sampling (Fienberg, S.E. and Tanur, 1996).

Neyman originally prepared the paper in 1932 in Polish (with an English summary) as a booklet growing out of his practical experience, and published in 1933 (Neyman, 1933). Summary of this paper is given and the contribution of J. Piekałkiewicz, another Polish statistician, to practical aspects of the research is stressed (Piekałkiewicz, 1934).

Special attention is paid to the co-operation of J. Neyman, after the Second World War, with Polish statisticians, and mainly with the Mathematical Commission of the Central Statistical Office of Poland (Fish, 1950; Zasępa, 1958). The author was one of the Polish statisticians who participated in that cooperation in 1958. Some personal reflections of this cooperation are mentioned.

Next the author considers the Deming’s (1950) approach to survey design, and discusses its impact on sample survey practice in Poland. He considers planning and preparatory work of survey design: defining major objectives, preparing first operation plans, method of data collection (questionnaire design), preliminary tabulation and analysis plans, frames and sample design, pre-test and pilot studies, preparing final survey documents; survey implementation (field work), data processing; analysis of survey data, publication and results dissemination. The author points out that the Mathematical Commission of GUS accepted the Deming’s approach that the statistician should keep in mind the root-mean-square error (MSE), which is the resultant of the two parts – sampling error, and the biases, and strive to reduce it. When the biases are large, it is uneconomical and ineffective to spend a lot of money to keep the sampling error small. As a matter of fact, often the most effective way to shorten the MSE is to decrease the bias while increasing the sampling error (Kordos, 1975,1985,). At the end some conclusions are drawn for future development of sample surveys in Poland and Deming’s 14 points are reminded (Deming, 1986).

REFERENCES

DEMING, W. E. (1950), Some Theory of Sampling, New York: Wiley.

DEMING, W. E. (1986), Out of the Crisis, Massachusetts Inst Technology.

FISZ, M. (1950), Consultation with Prof. Neyman and conclusions. „Studia i Prace Statystyczne”, nr 3-4. (In Polish).

FIENBERG, S.E. and TANUR, J.M. Reconsidering the Fundamental Contributions of Fisher and Neyman on Experimentation and Sampling, International Statistical Review, vol. 64, pp. 237-253.

KORDOS, J. (1975), 25 years of activities of the CSO Mathematical Commission, Przegląd Statystyczny, nr. 1. (in Polish).

KORDOS, J. (1985), Towards an Integrated System of Household Surveys in Poland, “Bulletin of the International Statistical Institute”, (invited paper), vol. 51, Amsterdam, Book 2, pp. 1.3.1 -18.

KRUSKAL, W.H. and MOSTELLER, F. (1980), Representative sampling, IV. The history of the concepts in statistics, 1895-1939, International Statistical Review, Vol. 48, pp.169-195.

NEYMAN, J. (1925), Contributions to the Theory of Small Samples Drawn from a Finite Population, “Biometrika”, nr 17, pp.472-479.

NEYMAN, J. (1933), An Outline of the Theory and Practice of Representative Method, Applied in Social Research, Institute of Social Problems, Warszawa (In Polish with an English summary),123 pp.

NEYMAN, J. (1934), On the Two Different Aspects of the Representative Method: The Method of Stratified Sampling and the Method of Purposive Selection, “Journal of the Royal Statistical Society”, nr 97, pp. 558-606.

NEYMAN, J. (1938), Contribution to the Theory of Sampling Human Population, “Journal of American Statistical Association”, nr 33, p.101-116.

PIEKAŁKIEWICZ, J. (1934), Sprawozdanie z badań składu ludności robotniczej w Polsce metodą reprezentacyjną (Report on Investigation of Composition of Workers’ Population in Poland by Sampling Method). Instytut Spraw Społecznych, Warszawa.

ZASEPA, R. (1958): Problems of Sampling Surveys of CSO in the light of consultations with Prof. J. Neyman), Wiadomości Statystyczne, nr 6, pp. 7-12 (in Polish).