Simpson's Paradox, Lord's Paradox, and Suppression Effects are the same phenomenon - The reversal paradox

Emerging Themes in Epidemiology

Volumn 5, Issue , 2008, Pages

  Tu, Yu Kang ^a   Gunnell, David ^b   Gilthorpe, Mark S ^a  

^a UNIVERSITY OF LEEDS   (United Kingdom)

^b UNIVERSITY OF BRISTOL   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

ANALYSIS OF VARIANCE; CONCEPT ANALYSIS; DATA ANALYSIS; EXPERIMENTAL STUDY; HUMAN; LORD PARADOX; MULTIPLE REGRESSION; PHENOMENOLOGY; RESEARCH; REVERSAL PARADOX; REVIEW; SIMPSON PARADOX; STATISTICAL ANALYSIS; STATISTICAL PARAMETERS; SUPPRESSION EFFECT; THEORY;

EID: 40149084499     PISSN: None     EISSN: 17427622     Source Type: Journal    
DOI: 10.1186/1742-7622-5-2     Document Type: Review

References (45)

1. Fetal origins of coronary heart disease. DJ Barker, BMJ 1995 311 171 4 7613432
2. Fetal origins of adult disease: strength of effects and biological basis. DJ Barker JG Eriksson T Forsen C Osmond, Int J Epidemiol 2002 31 1235 9 10.1093/ije/31.6.1235 12540728
3. Early nutritional origins of hypertension: a hypothesis still lacking support. N Paneth F Ahmed AD Stein, Journal of Hypertensio 1996 14 5 S121 S129
4. Fetal origins of adult disease-the hypothesis revisited. A Lucas MS Fewtrell TJ Cole, BMJ 1999 319 245 9 10417093
5. Unravelling the fetal origins hypothesis: is there really an inverse association between birth weight and subsequent blood pressure? RR Huxley A Neil R Collins, Lancet 2002 360 659 65 10.1016/S0140-6736(02)09834-3 12241871
6. Why evidence for the fetal origins of adult disease might be a statistical artifact: the "reversal paradox" for the relation between birth weight and blood pressure in later life. YK Tu R West GTH Ellison MS Gilthorpe, Am J Epidemiol 2005 161 27 32 10.1093/aje/kwi002 15615910
7. Invited commentary: Barker meets Simpson. CR Weinberg, Am J Epidemiol 2005 161 33 5 10.1093/aje/kwi003 15615911
8. Statistical issues in life course epidemiology. BL De Stavola D Nitsch I dos Santos Silva V McCormack R Hardy V Mann TJ Cole S Morton DA Leon, Am J Epidemiol 2006 163 84 96 10.1093/aje/kwj003 16306313
9. Causality: Models, Reasoning, and Inference. J Pearl, Cambridge: Cambridge University Press 2000
10. Confounding and collapsibility in causal inference. S Greenland JM Robins J Pearl, Stat Sci 1999 14 29 46 10.1214/ss/1009211805
11. Statistics for Epidemiology. NP Jewell, London: Chapman & Hall 2004
12. A general condition for avoiding effect reversal after marginalisation. DR Cox N Wermuth, J R Statist Soc B 2003 65 937 941 10.1111/1467-9868.00424
13. The interpretation of interaction in contingency tables. EH Simpson, J R Stat Soc Ser B 1951 13 238 41
14. Notes on the theory of association of attributes in statistics. GU Yule, Biometrika 1903 2 121 34 10.1093/biomet/2.2.121
15. Mathematical contributions to the theory of evolution: VI - Genetic (reproductive) selection: Inheritance of fertility in man, and of fecundity in thoroughbred racehorses. K Pearson A Lee L Bramley-Moore, Philos Trans R Soc Lond A 1899 192 257 330 10.1098/rsta.1899.0006
16. Intergenerational influences affecting birth outcome. II. Preterm delivery and gestational age in the children of the 1958 British birth cohort. E Hennessy E Alberman, Paediatr Perinat Epidemiol 1998 12 1 61 75 10.1046/j.1365-3016.1998.0120s1061.x 9690274
17. Website of the Department of Health, United Kingdom. http://www.doh.gov. uk
18. A graphical representation of a three-way contingency table: Simpson's paradox and correlation. M Paik, Am Stat 1985 39 53 54 10.2307/2683907
19. The "birth weight" paradox uncovered? S Hernandez-Diaz EF Schisterman M Hernan, Am J Epidemiol 2006 164 1115 1120 10.1093/aje/kwj275 16931543
20. Invited Commentary: The perils of birth weight - a lesson from directed acyclic graphs. A Wilcox, Am J Epidemiol 2006 164 1121 1123 10.1093/aje/kwj276 16931545
21. A paradox in the interpretation of group comparisons. FM Lord, Psychol Bull 1967 68 304 5 10.1037/h0025105 6062585
22. Statistical adjustments when comparing preexisting groups. FM Lord, Psychol Bull 1969 72 337 8 10.1037/h0028108
23. When is baseline adjustment useful in analysis of change? An example with education and cognitive change. MM Glymour J Weuve LF Berkman I Kawachi JM Robins, Am J Epidemiol 2005 162 267 278 10.1093/aje/kwi187 15987729
24. Deconstructuring statistical questions. D Hand, J R Stat Soc Ser A Stat Soc 1994 157 317 56 10.2307/2983526
25. A primer on regression artefact. DT Campbell DA Kenny, Guildford: The Guilford Press 1999
26. Regression artifacts and other customs of dubious desert. LB Mohr, Eval Program Plann 2000 23 397 409 10.1016/S0149-7189(00)00029-X
27. Regression facts and artifacts. CS Reichardt, Eval Program Plann 2000 23 411 4 10.1016/S0149-7189(00)00030-6
28. Adjusting for differential base rates: Lord's paradox again. H Wainer, Psychol Bull 1991 109 147 51 10.1037/0033-2909.109.1.147 2006226
29. Statistics on the Table. SM Stigler, Cambridge, Massachusetts: Harvard University Press 1999
30. Analysing controlled trials with baseline and follow up measurements. AJ Vickers DG Altman, BMJ 2001 323 1123 4 11701584 10.1136/bmj.323.7321.1123
31. Statistical power for analyses of changes in randomized controlled trials. YK Tu A Blance V Clerehugh MS Gilthorpe, J Dent Res 2005 84 283 287 15723872
32. Suppression and enhancement in bivariate regression. JW Lewis LA Escobar, Statistician 1986 35 17 26 10.2307/2988294
33. A quirk in multiple regression: the whole regression can be greater than the sum of its parts. PV Bertrand RL Holder, Statistician 1988 37 371 4 10.2307/2348761
34. The relationship among sums of squares, correlation coefficients and suppression. NR Sharpe RA Roberts, Am Stat 1997 51 46 48 10.2307/2684693
35. Graphical views of suppression and multicollinearity in multiple linear regression. L Friedman M Wall, Am Stat 2005 127 136 10.1198/000313005X41337
36. Applied multiple regression/correlation analysis for the behavioural sciences. J Cohen P Cohen, London: LEA 1983
37. Multiple regression in behavioral research: Explanation and prediction. EJ Pedhazur, Fort Worth: Harcourt 1997
38. Blood pressure and birth weight in the first year university student aged 18-25. NP Stocks G Davey Smith, Public Health 1999 113 273 7 10.1016/S0033-3506(99)00179-1 10637518
39. Birth size, growth, and blood pressure between the ages of 7 and 26 years: failure to support the fetal origins hypothesis. S Williams R Poulton, Am J Epidemiol 2002 155 849 52 10.1093/aje/155.9.849 11978589
40. Blood pressure in relation to birth weight in twins and singleton controls matched for gestational age. G McNeill C Tuya DM Campbell P Haggarty WCS Smith LF Masson A Cumming I Broom N Haites, Am J Epidemiol 2003 158 150 5 10.1093/aje/kwg130 12851228
41. What is the effect of adjusting for more than one measure of current body size on the relation between birth weight and blood pressure? YK Tu MS Gilthorpe GTH TH Ellison, J Hum Hypertens 2006 20 646 657 10.1038/sj.jhh.1002044 16691183
42. The role of prediction variables which are independent of the criterion. P Horst, The Prediction of Personal Adjustment New York: Social Science Research Council, Horst P, 1941 431 6
43. Equivalence of the mediation, confounding and suppression effect. DP MacKinnon JL Krull CM Lockwood, Prev Sci 2000 1 173 81 10.1023/A:1026595011371 11523746
44. Growth, current size and the role of the 'reversal paradox' in the foetal origins of adult disease: an illustration using vector geometry. YK Tu GTH Ellison MS Gilthorpe, Epidemiol Perspect Innov 2006 3 9 16884533 10.1186/1742-5573-3-9
45. The scandal of poor epidemiological research. E Von Elm M Egger, BMJ 2004 329 868 9 15485939 10.1136/bmj.329.7471.868