Pisces did not have increased heart failure: data-driven comparisons of binary proportions between levels of a categorical variable can result in incorrect statistical significance levels

Journal of Clinical Epidemiology

Volumn 61, Issue 3, 2008, Pages 295-300

  Austin, Peter C ^a,b   Goldwasser, Meredith A ^a  

^a INSTITUTE FOR CLINICAL EVALUATIVE SCIENCES   (Canada)

^b UNIVERSITY OF TORONTO   (Canada)

Author keywords

Astrology; Chi squared test; Contingency table; Maximal proportion; Significance testing; Type I error rate

Indexed keywords

ARTICLE; CANADA; COMPARATIVE STUDY; HEART FAILURE; HOSPITALIZATION; HUMAN; MONTE CARLO METHOD; MYSTICISM; PRIORITY JOURNAL; STATISTICAL ANALYSIS;

ASTROLOGY; CONFOUNDING FACTORS (EPIDEMIOLOGY); EPIDEMIOLOGIC METHODS; HEART FAILURE; HOSPITALIZATION; HUMANS; ONTARIO;

EID: 38349085947     PISSN: 08954356     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.jclinepi.2007.05.007     Document Type: Article

References (13)

1. Agresti A. An introduction to categorical data analysis (1996), John Wiley & Sons, New York, NY
2. Miller R., and Siegmund D. Maximally selected chi square statistics. Biometrics 38 (1982) 1011-1016
3. Halpern J. Maximally selected chi square statistics for small samples. Biometrics 38 (1982) 1017-1023
4. Koziol J.A. On maximally selected chi-square statistics. Biometrics 47 (1991) 1557-1561
5. 2statistics for k×2 tables. Biometrics 55 (1999) 317-320
6. Boulesteix A.-L. Maximally selected chi-square statistics for ordinal variables. Biom J 48 (2006) 451-462
7. Boulesteix A.-L. Maximally selected chi-square statistics and binary splits of nominal variables. Biom J 48 (2006) 838-848
8. Austin P.C., Mamdani M.M., Juurlink D.N., and Hux J.E. Testing multiple statistical hypotheses resulted in spurious associations: a study of astrological signs and health. J Clin Epidemiol 59 (2006) 964-969
9. Hurvich C.M., and Tsai C.-L. The impact of model selection on inference in linear regression. Am Stat 44 (1990) 214-217
10. Copas J.B., and Long T. Estimating the residual variance in orthogonal regression with variable selection. Statistician 40 (1991) 51-59
11. Austin P.C., and Tu J.V. Automated variable selection methods for logistic regression produced unstable models for predicting acute myocardial infarction mortality. J Clin Epidemiol 57 (2004) 1138-1146
12. Austin P.C., and Tu J.V. Bootstrap methods for developing predictive models in cardiovascular research. Am Stat 58 (2004) 131-137
13. Austin P.C., and Brunner L.J. Inflation of the type I error rate when a continuous confounding variable is categorized in logistic regression analyses. Stat Med 23 (2004) 1159-1178