2012, Number 1
PDF size: 297.46 Kb.
ABSTRACTClassical statistical inference is mainly based in p value calculation using the hypothesis contrast method, or estimating confidence intervals for punctual results. Hypothesis contrast permits the comparability between observed probability and predicted probability considering some grade of random variability. The null hypothesis can be accepted or rejected using the observed result with a certain grade of confidence. In this assay, an intuitive approximation of the hypothesis contrast method for statistical inference is presented, using a simple binomial problem for probability calculation, an analogy between the hypothesis contrast and a justice system, and a brief explanation about statistical significance.
Silverman ME, Murray TJ, Bryan CS, eds. The quotable osler. Philadelphia Pa, USA: American College of Physicians 2002.
Pita-Fernández S, Pértega-Díaz S. Pruebas diagnósticas: sensibilidad y especificidad. Cad Aten Primaria 2003; 10: 120-124.
Jaisingh L. Statistics for the utterly confused. 2 ed. New York, N.Y.: McGraw-Hill; 2006.
Altman DG. Practical statistics for medical research. London, UK: Chapman & Hall; 1991.
Pagano M, Gauvreau K. Fundamentos de bioestadística. México: Thomson Learning, Inc.; 2001.
Fisher RA. Design of experiments. London, UK: Macmillan Publishers Co; 1935.