The American Statistician back issues from August 2003:
Editor's Report for The American Statistician, Volume 56 (2002).(summary of contents and operation of magazine)
Aug 01, 2003; ... 1. INTRODUCTION This article summarizes the contents and operation of The American Statistician (TAS) for 2002.This year was a transitional year for editors--Lynne Stokes handled new paper submissions and revisions for the first half of the year and I managed new submissions the ...
Small samples and ordered logistic regression: does it help to collapse categories of outcome?(Statistical Practice)
Aug 01, 2003; ... 1. INTRODUCTION Medical studies often involve outcomes that are classified into ordered categories--for example, as "mild," "moderate," or "severe" toxicity. In small samples the question "would it be better to collapse categories of outcome" often arises. The rationale for ...
Adjusted coefficients of determination for logistic regression.(Statistical Practice)
Aug 01, 2003; ... 1. INTRODUCTION In linear regression, the coefficient of determination, [R.sup.2], defined as the proportion of variation in the outcome variable explained by the model, is a popular measure of the strength of association between the outcome and the predictors. One serious ...
Sample size: more than calculations.(Statistical Practice)
Aug 01, 2003; ... 1. INTRODUCTION In any protocol involving human subjects, an investigator must specify the number of subjects he intends to study. Similar information is often required for studies involving animals or industrial processes. To meet this need, there are multiple books on sample ...
Confusion over measures of evidence (p's) versus errors ([alpha]'s) in classical statistical testing.(General)
Aug 01, 2003; ... 1. INTRODUCTION Modern textbooks on statistical analysis in the business, social, and biomedical sciences, whether at the undergraduate or graduate levels, typically present the subject matter as if it were gospel: a single, unified, uncontroversial means of statistical ...
Nearly nonparametric multivariate density estimates that incorporate marginal parametric density information.(General)
Aug 01, 2003; ... 1. INTRODUCTION Frequently modelers do not want to be constrained by assuming a fixed model structure without validation from data. Nonparametric density estimation allows modelers to estimate probability densities without buying into a parametric model. In univariate data ...
History corner: introduction.(Brief Article)
Aug 01, 2003; ... Introduction Francis Galton (1822-1911) is well known as a leading figure in nineteenth century statistics, making important developments in regression--introducing the term correlation, for example. Arguably more a scientist than a methodologist, Galton was a major contributor ...
Galton's bend: a previously undiscovered nonlinearity in Galton's family stature regression data.(History Corner)
Aug 01, 2003; ... 1. INTRODUCTION Francis Galton devised his regression model to develop an evolutionary theory of heredity. As Stigler (1986) showed, the mathematics of linear least squares fitting date back at least to the early nineteenth century. But it was Galton's idea of regression based ...
A diamond-shaped equiponderant graphical display of the effects of two categorical predictors on continuous outcomes.(Statistical Computing and Graphics)
Aug 01, 2003; ... 1. INTRODUCTION Three-dimensional (3-D) bar graphs are commonly used in biomedical studies to portray how two categorical variables (predictors, risk factors) jointly contribute to an outcome (Klag et al. 1996; Huang et al. 1997; Farahmand et al. 2000). However, most 3-D bar ...
Section editor's notes.(Statistical Computing Software Reviews)(Editorial)
Aug 01, 2003; ... The software reviews in this issue continue our discussion devoted to exact statistics. Robert Oster presents a Part II to his review dealing with categorical data analysis using exact statistics. Part I appeared in the August 2002 issue of The American Statistician (vol. 56, p. 235). In ...
An examination of statistical software packages for categorical data analysis using exact methods--Part II.(Statistical Computing and Graphics, SAS 8.2, Testimate 6)
Aug 01, 2003; ... REVIEWED SOFTWARE SAS 8.2 Available from SAS Institute Inc., 100 SAS Campus Drive, Cary, NC 27513; phone: (919) 677-8000; fax: (919) 677-4444; E-mail: refer to Web page for appropriate address; Web page: http://www.sas.com. Released 2001. Academic and commercial ...
Comparing statistical software packages: the case of the logrank test in StatXact.(Statistical Computing and Graphics)
Aug 01, 2003; ... 1. INTRODUCTION An examination of statistical software packages, followed by a discussion and a list of motivated recommendations, is extremely useful for the practicing statistician. A recent example of such an examination is the article by R. A. Oster (2002), in which the ...
The A to Z of Mathematics: a Basic Guide.(Book Review)
Aug 01, 2003; ... Thomas H. SIDEBOTHAM. New York: Wiley, 2002, xi + 474 pp., $54.95 (P), ISBN: 0-471-15045-2. This dictionary of mathematical terms gives an overview of mathematics and a bit of statistics that will be helpful for those who suffer from mathematics anxiety or who have been away ...
Problems in Probability.(Book Review)
Aug 01, 2003; ... T. M. MILLS. Singapore: World Scientific Press, 2001, x + 181 pp., $28.00 (H), ISBN: 981-02-4598-X. Undergraduate students in a first course in probability are often technically adept at solving routine problems, yet may not be mathematically mature enough to tackle more complex ...
Duelling Idiots and Other Probability Puzzlers.(What are the Chances? Voodoo Deaths, Office Gossip, and Other Adventures in Probability)(Book Review)
Aug 01, 2003; ... Duelling Idiots and Other Probability Puzzlers. Paul J. NAHIN. Princeton, NJ: Princeton University Press, 2002, xxviii + 269 pp., $18.95 (P), ISBN: 0-691-10286-4. What are the Chances? Voodoo Deaths, Office Gossip, and Other Adventures in Probability. Bart K. HOLLAND. Baltimore, ...
(with InfoTrac and CD-ROM).(Basic. Statistics and Data Analysis 1st ed.Introduction to Probability and Statistics (11th ed.), Just the Essentials of Elementary Statistics (3rd ed.) )(Book Review)
Aug 01, 2003; ... Basic Statistics and Data Analysis (with InfoTrac and CD-ROM) (1st ed.). Larry J. KITCHENS. Pacific Grove, CA: Brooks/Cole Publishing Company, 2003, xiii + 637 pp., $91.95 (H and CD-ROM), ISBN: 0-534-38465-X. Introduction to Probability and Statistics (11th ed.). William ...
King, E. N., and Ryan, T. P. (2002), "A Preliminary Investigation of Maximum Likelihood Logistic Regression Versus Exact Logistic Regression," The American Statistician, 56, 163-170: comment by Greenland and reply. Replies also to Webber, W. F.; and Wang, J., and Brady, M. F. (2003), The American Statistician, 57, 147-148.(Letters to the Editor)
Aug 01, 2003; ... As an analyst frequently confronted with sparse tabular data, I read with interest the comparison of maximum-likelihood and exact logistic regression (MLR and ELR) by King and Ryan (2002), who noted the superiority of ELR to MLR in the contexts they examined. There are important ...
Vardeman. S. B., and Morris, M. D. (2003), "Statistics and Ethics: Some Advice for Young Statisticians," The American Statistician, 57, 21-26: comments by Ganju and Losinger.(Letters to the Editor)
Aug 01, 2003 ... GANJU The article by Vardeman and Morris, "Statistics and Ethics: Some Advice for Young Statisticians," (2003) that offered advice to "statistics graduate students and recent grads" is, despite its good intentions, objectionable on grounds of publication policy, its assessment ...
Callaert, H. (2003), "Comparing Statistical Software Packages: The Case of the Logrank Test in StatXact," The American Statistician, 57, 214-217: comment by Oster.(Letters to the Editor)
Aug 01, 2003; ... Callaert (2003, p. 214 of this issue) examined a problem that should be of interest to all users of statistical software, and particularly to those who use statistical procedures (such as the logrank test) that have several formulas associated with the corresponding test statistic. The ...
Austin, P. C., and Brunner, L. J. (2003), "Type I Error Inflation in the Presence of a Ceiling Effect," The American Statistician, 57, 97-104: comment by Kraemer.(Letters to the Editor)
Aug 01, 2003; ... Austin and Brunner examined an interesting issue important to biomedical research: the situation in which an independent variable, [X.sub.1] in a multiple linear regression is "subject to a ceiling effect." This term covers what has also been referred to as a variable being "truncated" or ...
Shoemaker, L. H. (2003), "Fixing the F Test for Equal Variances," The American Statistician, 57, 105-114: Comment By Shoemaker.(Correction)
Aug 01, 2003; ... It has been brought to my attention that there is an overlap of work that I presented in Section 4 (p. 107) of the article (the second multi-sample test proposed) and work done by Layard (1973). The test statistic that I proposed was [X.sub.2] = [summation][([Z.sub.i] - ...