The importance of sample size and statistical power in experimental research. A comparative study

• Autorzy: Cristofolini C. , Testoni M.
• Języki publikacji: EN

Abstrakty

EN

The aim of this paper is to stress the importance of a proper statistical determination of the sample size in experimental research, and to underline the possible effect of an experiment with an inadequate number of cases on the results. The first part of the paper introduces the statistical concepts needed for the sample size calculation. Type I and II errors are defined and the associated probabilities are presented. Statistical power of a test is explained and its correlation with the sample size and experimental variability is defined. In a second section, the criteria for the calculation of the sample size are described. In the third part of the article, a statistical comparison between a real experiment and a numerical simulation is shown to highlight the consequences of the selection of different sample sizes. The risk of drawing mistaken conclusions caused by an inadequate sample size calculation is thus calculated.

Słowa kluczowe

EN

sample size; statistical power

PL

medycyna; statystyka

• Wydawca: Oficyna Wydawnicza Politechniki Wrocławskiej
• Czasopismo: Acta of Bioengineering and Biomechanics
• Rocznik: 2000
• Tom: Vol. 2, no. 1
• Strony: 3--16
• Opis fizyczny: Bibliogr. 21 poz., tab., wykr.

Twórcy

autor

Cristofolini C.

• University of Bologna, Engineering Faculty, Viale Risorgimento 2, 40136 Bologna, Italy

autor

Testoni M.

• University of Bologna, Engineering Faculty, Viale Risorgimento 2, 40136 Bologna, Italy

Bibliografia

• [1] ARMITAGE P., BERRY G., Sampling, [In:] Statistical methods in medical research, 3rd ed., Oxford, Blackwell Scientific Publications, 1994, pp. 78-92.
• [2] BETHED R.M., DURAN B.S., BOULLION T.L., Statistical inference: hypothesis testing, [In:] Statistical methods for engineers and scientists, New York, Marcel Dekker, Inc., 1995.
• [3] COLTON T., Statistics in medicine, Little, Brown & Co. , 1974.
• [4] FISHER L.D., VAN BELLE G., Statistical inference: populations and samples, [In:] Biostatistics: a methodology for the health sciences, New York, John Wiley & Sons, Inc., 1993, pp. 75-137.
• [5] FLEISS J.L., Determining sample size needed to detect a difference between two proportions, [In:] Statistical methods for rates and proportions, New York, John Wiley & Sons, Inc., 1981, pp. 33-49.
• [6] HABER M., Sample sizes for the exact test of " no interaction" in 2 x2x2 tables , Biometrics, 1983, 39:493-498.
• [7] KASTENBAUM M.A., HOEL D.G., Sample size requirements: one- way analysis of variance, Biometrika, 1970,57:421-430.
• [8] KASTENBAUM M.A., HOEL D.G., Sample size requirements: randomized block designs, Biometrika, 1970, 57:573-577.
• [9] LACHIN J.M., Introduction to sample size determination and power analysis for clinical trials, Biometrics, 1 981,2:93-113.
• [10] LACHIN J.M., FOULKES M.A., Evaluation of sample size and power for analyses of survival with allowance for nonuniform patient entry, losses to follow-up, noncompliance and stratification, Biometrics, 1 986, 42:507-519.
• [11] LIEBER R., Statistical significance and statistical power in hypothesis testing, Journal of Orthorpaedic Research , 1990, 8:304-309.
• [12] MANLY B.F.J., Synthesis: carrying out a research study, [In:] The design and analysis of research studies, Cambridge, Cambridge University Press, 1992, pp. 321 -325.
• [13] MAHHEWS D.E., FAREWELL V .T., The question of sample size, [In:] Using and understanding medical statistics, 2nd ed., London, Karger, 1988, pp. 197-205.
• [14] MONTGOMERY D.C., More about si11gle-factor experiments , [In:] Design and analysis of experiments , 3rd ed., New York, John Wiley & Sons, 1 991, pp. 95-133.
• [15] MONTGOMERY D.C., Design and analysis of experiments , New York, John Wiley & Sons, 1991 .
• [16] OBUCHOWSKJ N.A., Computing sample size for receiver operating characteristic studies , Investigative Radiology, 1994, 29:238-243.
• [17] OBUCHOWSKJ N.A., MCCLISH D.K., Sample size determination for diagnostic accuracy studies involving binormal ROC curve indices, Statistic in medicine, 1 997, 16: 1529-1542.
• [18] ODEH R .E., FAX M., Sample size choice, New York, Marcel Dekker, 1991.
• [19] PEARSON E.S., HARTLEY H.D., Biometrika: tables for statisticians, Cambridge, Biometrika Trustees at the University Press, 1972.
• [20] SAHAI H., KHURSHID A., Formulae and tables for the determination of sample sizes and power in clinical trials for testing differences in proportions for the two-sample design: are view, Statistics in medicine, 1996, 15:1-21.
• [21] VESCHI M., CRISTOFOLINI L., BALEANI M., TONI A., Progetto dell'esperimento per la caratterizzazione a flessione a 4 punti di cementa acrilico per uso chirurgico , E. Leone Publ. Catania, 1997; 203 p., XXVI Meeting of the Associazione Italian a Analisi Sollecitazioni.