Testing Measures of Associations

  • Bruno S. Paolino
  • Raphael L. C. Araújo
  • David Bristol


Choosing the best method for comparing variables is a crucial step in any research project and the understanding of the advantages and disadvantages of each test of measures of association allows the investigator take an optimal decision. The choice of test depends on the number of groups in the analysis, the number of subjects in each group, the nature of the variable, and the type of dispersion. The type of distribution will be a key point in determining whether a parametric test or a nonparametric test would be more appropriate. In this chapter we will discuss the characteristics of the tests of measures of association commonly performed in clinical cancer research, when to use the tests, and how to interpret the accuracy of their results.


Statistical tests Null hypothesis Parametric tests Nonparametric tests 


  1. 1.
    Whitley E, Ball J. Statistics review 3: hypothesis testing and P values. Crit Care. 2002;6:222–5.CrossRefPubMedPubMedCentralGoogle Scholar
  2. 2.
    Baker M. Statistician issue warning on p-values. Nature. 2016;531:151.CrossRefPubMedGoogle Scholar
  3. 3.
    Razali N, Wah YB. Power comparisons of Shapiro–Wilk, Kolmogorov–Smirnov, Lilliefors, and Anderson–Darling tests. J Stat Model Anal. 2011;2(1):21–33.Google Scholar
  4. 4.
    Kao LS, Green CE. Analysis of variance: is there a difference in means and what does it mean? J Surg Res. 2008;144(1):158–70.CrossRefPubMedGoogle Scholar
  5. 5.
    Paolino B, et al. Myocardial inactivation of thyroid hormones in patients with aortic stenosis. Thyroid. 2017;27:738.CrossRefPubMedPubMedCentralGoogle Scholar
  6. 6.
    Bewick V, Cheek L, Ball J. Statistics review 9: one-way analysis of variance. Crit Care. 2004;8:130–6.CrossRefPubMedPubMedCentralGoogle Scholar
  7. 7.
    Winer BJ, Michels KM, Brown DR. Statistical principles in experimental design. 3rd ed. New York, NY: McGraw-Hill; 1991.Google Scholar
  8. 8.
    Friedman M. The use of ranks to avoid the assumption of normality implicit in the analysis of variance. J Am Stat Assoc. 1937;32(200):675–701.CrossRefGoogle Scholar
  9. 9.
    Siegel S, Castellan NJ. Nonparametric statistics. 2nd ed. Belmont, CA: Wadsworth Publishing; 1989.Google Scholar
  10. 10.
    Araujo RL, Karkar AM, Allen PJ. Timing of elective surgery as a perioperative outcome variable: analysis of pancreaticoduodenectomy. HPB (Oxford). 2014;16(3):250–62.CrossRefGoogle Scholar
  11. 11.
    Araujo RL, Gaujoux S, Huguet F. Does pre-operative chemoradiation for initially unresectable or borderline resectable pancreatic adenocarcinoma increase post-operative morbidity? A case-matched analysis. HPB (Oxford). 2013;15(8):574–80.CrossRefGoogle Scholar
  12. 12.
    Winters R, Winters A, Amedee RG. Statistics: a brief overview. Ochsner J. 2010;10:213–6.PubMedPubMedCentralGoogle Scholar
  13. 13.
    Cochran WG. Some methods for strengthening the common x2 tests. Biometrics. 1954;10:417–51.CrossRefGoogle Scholar
  14. 14.
    Portney LG, Watkins MP. Foundations of clinical research. 3rd ed. Philadelphia, PA: F.A. Davis; 2015.Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Bruno S. Paolino
    • 1
  • Raphael L. C. Araújo
    • 2
  • David Bristol
    • 3
  1. 1.Department of CardiologyState University of Rio de JaneiroRio de JaneiroBrazil
  2. 2.Department of Upper Gastrointestinal and Hepato-Pancreato-Biliary SurgeryBarretos Cancer HospitalBarretosBrazil
  3. 3.Independent ConsultantWiston-SalemUSA

Personalised recommendations