## The earth is round (p < .05 (1994)

Venue: | American Psychologist |

Citations: | 173 - 0 self |

### BibTeX

@ARTICLE{Cohen94theearth,

author = {Jacob Cohen},

title = {The earth is round (p < .05},

journal = {American Psychologist},

year = {1994},

pages = {997--1003}

}

### Years of Citing Articles

### OpenURL

### Abstract

After 4 decades of severe criticism, the ritual of null hypothesis significance testing—mechanical dichotomous decisions around a sacred.05 criterion—still persists. This article reviews the problems with this practice, including its near-universal misinterpretation ofp as the probability that Ho is false, the misinterpretation that its complement is the probability of successful replication, and the mistaken assumption that if one rejects Ho one thereby affirms the theory that led to the test. Exploratory data analysis and the use of graphic methods, a steady improvement in and a movement toward standardization in measurement, an emphasis on estimating effect sizes using confidence intervals, and the informed use of available statistical methods is suggested. For generalization, psychologists must finally rely, as has been done in all the older sciences,

### Citations

3812 |
Statistical Power Analysis for the Behavioral Sciences. Lawrence Erlbaum Associates
- Cohen
- 1988
(Show Context)
Citation Context ...nt that causal modeling proceeds with regression and not correlation coefficients. In the same vein, I should note that standardized effect size measures, such as d and / developed in power analysis (=-=Cohen, 1988-=-) are, like correlations, also dependent on population variability of the dependent variable and are properly used only when that fact is kept in mind . To work constructively with "raw" regression co... |

1291 | The Visual Display of Quantitative Information - Tufte - 1983 |

1191 | The Logic of Scientific Discovery - Popper - 1959 |

1074 | Exploratory Data Analysis - Tukey - 1977 |

751 | Envisioning Information - Tufte - 1990 |

653 | Statistical Methods for Research Workers - Fisher - 1925 |

425 | Visualizing Data - Cleveland - 1993 |

190 | Belief in the law of small numbers - Tversky, Kahneman - 1971 |

169 | The "filedrawerproblem" and tolerance for null results. Psychol Bull - Rosenthal |

136 | Rational Choice in an Uncertain World - Dawes - 1988 |

100 |
Things I have learned (so far
- Cohen
- 1990
(Show Context)
Citation Context ...shington Place, 5th Floor, New York, NY 10003. 997others, and, more recently and emphatically, by Meehl (1978, 1986, 1990a, 1990b), Gigerenzer( 1993), Falk and Greenbaum (in press), and yours truly (=-=Cohen, 1990-=-). The Permanent Illusion One problem arises from a misapplication of deductive syllogistic reasoning. Falk and Greenbaum (in press) called this the "illusion of probabilistic proof by contradiction" ... |

85 |
Dynamic Graphics for Statistics
- Cleveland, McGill
- 1988
(Show Context)
Citation Context ... informal, and largely graphic techniques aims for understanding the set of data in hand. Important contributions to graphic data analysis have since been made by Tufte (1983, 1990), Cleveland (1993; =-=Cleveland & McGill, 1988-=-), and others. An excellent chapter-length treatment by Wainer and Thissen (1981), recently updated (Wainer & Thissen, 1993), provides many useful references, and statistical program packages provide ... |

68 | Statistical inference: A commentary for the social and behavioural sciences - Oakes - 1986 |

66 | The problem of multiple comparisons - Tukey - 1953 |

55 | Consequences of prejudice against the null hypothesis
- Greenwald
- 1975
(Show Context)
Citation Context ...heses and subjecting them to challenge, as Meehl (1967) argued many years ago. With more evolved psychological theories, we can also find use for likelihood ratios and Bayesian methods (Goodman, 1993;=-=Greenwald, 1975-=-). We quantitative behavioral scientists need not go out of business. Induction has long been a problem in the philosophy of science. Meehl (1990a) attributed to the distinguished philosopher Morris R... |

47 | The Fallacy of the Null Hypothesis Significance Test - Rozeboom - 1960 |

45 | The tests of significance in psychology research - Bakan - 1966 |

37 |
Why summaries of research on psychological theories are often uninterpretable
- Meehl
- 1990
(Show Context)
Citation Context ...e, leisure time activities, and high school organizations. AH of the 105 chi-squares that these 15 items produced by the crosstabulations were statistically significant, and 96% of them a\p< .000001 (=-=Meehl, 1990-=-b). One might say, "With 57,000 cases, relationships as small as a Cramer <j> of .02-.03 will be significant at p < .000001, so what's the big deal?" Well, the big deal is that many of the relationshi... |

36 | Statistical significance in psychological research - Lykken - 1968 |

35 | The significance test controversy - Morrison, Henkel - 1970 |

33 | Appraising and amending theories: The strategy of Lakatosian defense and two principles that warrant using it
- Meehl
- 1990
(Show Context)
Citation Context ...e, leisure time activities, and high school organizations. AH of the 105 chi-squares that these 15 items produced by the crosstabulations were statistically significant, and 96% of them a\p< .000001 (=-=Meehl, 1990-=-b). One might say, "With 57,000 cases, relationships as small as a Cramer <j> of .02-.03 will be significant at p < .000001, so what's the big deal?" Well, the big deal is that many of the relationshi... |

28 | What do the data really mean? Research findings, meta-analysis and cumulative knowledge in psychology - Schmidt - 1992 |

21 | Theory testing in psychology and physics: A methodological paradox - Meehl - 1967 |

16 |
Graphical data analysis
- Wainer, Thissen
- 1981
(Show Context)
Citation Context ...data analysis have since been made by Tufte (1983, 1990), Cleveland (1993; Cleveland & McGill, 1988), and others. An excellent chapter-length treatment by Wainer and Thissen (1981), recently updated (=-=Wainer & Thissen, 1993-=-), provides many useful references, and statistical program packages provide the necessary software (see, for an example, Lee Wilkinson's [1990] SYGRAPH, which is presently being updated). Forty-two y... |

15 | Some difficulties of interpretation encountered in the application of the chi-square test - Berkson - 1938 |

13 |
Analyzing data: Sanctification or detective work
- Tukey
- 1969
(Show Context)
Citation Context ...a research is that A is larger than B (p < .01), we have not learned very much. And this is typically all we learn. Confidence intervals are rarely to be seen in our publications. In another article (=-=Tukey, 1969-=-), he chided psychologists and other life and behavior scientists with the admonition "Amount, as well as direction is vital" and went on to say the following: The physical scientists have learned muc... |

12 | Rational appraisal of psychological research: The good-enough principle - Serlin, Lapsley - 1993 |

11 |
p values, hypothesis tests and likelihood: implications lor epidemiology of a neglected historical debate
- Goodman
- 1993
(Show Context)
Citation Context ... as null hypotheses and subjecting them to challenge, as Meehl (1967) argued many years ago. With more evolved psychological theories, we can also find use for likelihood ratios and Bayesian methods (=-=Goodman, 1993-=-;Greenwald, 1975). We quantitative behavioral scientists need not go out of business. Induction has long been a problem in the philosophy of science. Meehl (1990a) attributed to the distinguished phil... |

11 | What Social Scientists Don't Understand - Meehl - 1986 |

8 | Confidence Curves: An Omnibus Technique for Estimation and Testing Statistical Hypotheses - Birnbaum - 1961 |

7 | On the probability of making type I errors - Pollard, Richardson - 1987 |

7 | The influence of Statistical methods for research workers on the development of the science of statistics - Yates - 1951 |

6 | Two and one-half decades of leadership in measurement and evaluation - Thompson - 1992 |

4 | Cumulating evidence - Rosenthal - 1993 |

3 | Statistical theory - Hogben - 1957 |

2 | The statistical power of abnormal-social psychological research: A review - unknown authors - 1962 |

2 | Theoretical risks and tabular asterisks: Sir Karl, Sir Ronald, and the slow progress of soft psychology - E - 1978 |

1 |
Do studies of statistical power 1002
- Sedlmeier, Gigerenzer
- 1989
(Show Context)
Citation Context ...archers that, given the fact that the nil hypothesis is always false, the rate of Type I errors is 0%, not 5%, and that only Type II errors can be made, which run typically at about 50% (Cohen, 1962; =-=Sedlmeier & Gigerenzer, 1989-=-). He showed that typically, the sample effect size necessary for significance is notably larger than the actual population effect size and that the average of the statistically significant effect siz... |

1 | SYGRAPH: The system for graphics - unknown authors - 1990 |