Clinical research XVII. c2 test, from the expected to the observed
Main Article Content
Keywords
Chi-square Distribution, Statistics
Abstract
When you want to show if there is a statistical association or differences between categorical variables, it is recommended to use the c2 test. This nonparametric test is one of the most used in clinical research; it contrasts nominal or ordinal qualitative variables that are observed in clinical practice. This test calculates the p value that determines whether differences between groups are real or due to chance. The c2 test is the basis of other tests to analyze qualitative ordinal variables as c2 for linear trend, which compares three groups with two outcomes or McNemar test which contrasts two related samples (a before and afterward comparison) or Mantel-Haenszel c2, which controls for potential confounding variables. When using small samples, where the expected is less than 5, Fisher’s exact test should be used. These tests are the most widely used in the medical literature, however, do not give us the magnitude or the direction of the event and a proper interpretation that requires clinical judgment is needed.
References
Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst. 1959;22(4):719-48.
Henderson-Smart DJ, Davis PG. Prophylactic methylxanthines for endotracheal extubation in preterm infants. Cochrane Database Syst Rev. 2010;(12):CD000139.
Portney LG, Watkins MP. Foundations of clinical research: applications to practice. Third editon. New Jersey: Pearson/Prentice Hall; 2009.
Armitage P, Berry G, Matthews JNS. Statistical methods in medical research. Fourth edition. Massachusetts: Blackwell Science; 2002.
Peat JK, Barton B. Medical statistics: a guide to data analysis and critical appraisal. Massachusetts: Blackwell 2005.
Le Chap T. Introductory biostatistics. New Jersey: Wiley-Interscience; 2003.
Bland M. Introduction to medical statistics. Third edition. US: Oxford University Press; 2003.
Dawson B, Trapp R. Basic and clinical biostatistics. Fourth edition. US: Lange Medical Books-McGrawHill; 2004.
Talavera JO, Wacher-Rodarte NH, Rivas-Ruiz R. Clinical research III. The causality studies. Rev Med Inst Mex Seguro Soc. 2011;49(3):289-94. Texto libre en http://revistamedica.imss.gob.mx/index.php? option=com_multicategories&view=article&id=1410: investigacion-clinica-iii-estudios-de-causalidad&catid =501:temas-de-actualidad&Itemid=664
Talavera JO, Rivas-Ruiz R. Clinical research IV. Relevancy of the statistical test chosen. Rev Med Inst Mex Seguro Soc. 2011;49(4):401-5. Texto libre en http://revistamedica.imss.gob.mx/index.php?option=com_content&view=article&id=1432:investigacion-clinica-iv-pertinencia-de-la-prueba-estadistica&catid=139:2011&Itemid=511
Talavera JO, Rivas-Ruiz R. Clinical research V. Sample size. Rev Med Inst Mex Seguro Soc. 2011;49(5):517-22. Texto libre en http://revistamedica.imss.gob.mx/index.php?option=com_content&view=article&id=1452:investigacion-clinica-v-tamano-de-muestra&catid=139:2011&Itemid=511
Rivas-Ruiz R, Pérez-Rodriguez M, Talavera JO. Del juicio clínico al modelo estadístico. Diferencia de medias. Prueba t de Student. Rev Med Inst Mex Seguro Soc. 2013;51(3):300-3. Texto libre en http://revistamedica.imss.gob.mx/index.php?option=com_multicategories&view=article&id=2077: 51-3-11-investigacion&catid=785:temas-de-actualidad &Itemid=775
Rivas-Ruiz R, Moreno-Palacios J, Talavera JO Investigación clínica XVI. Diferencia de medianas con la U de Mann-Whitney. Rev Med Inst Mex Seguro Soc 2013;51(4):414-9.