Online Statistics Education

Livro digital

Título:
Online Statistics Education

Autor:
David M. Lane et al.

Categoria:
Tecnologia > Dados

Doador:
Raffaello D. N.

Sinopse:
Before defining a single term, this book hands you three claims that are numerically correct and interpretively wrong, and asks you to find the flaw yourself. A new ice cream advertisement was followed by a 30% sales increase over the next three months, so the advertisement worked. Cities with more churches have more crime, so churches cause crime. Interracial marriages are up 75% over twenty-five years, so society has grown accepting. The answers involve summer, population size, and a base rate nobody mentioned — and they set the tone for everything that follows. Across twenty-one chapters the book moves from descriptive statistics and graphing distributions through probability, research design, normal distributions, sampling distributions, estimation, hypothesis testing, regression and analysis of variance. It also goes well past where introductory texts usually stop. There is a full chapter on advanced graphs covering quantile-quantile plots, contour plots and 3D plots. Another covers transformations, including Tukey's ladder of powers and Box-Cox. A chapter on distribution-free tests works through randomization methods, Mann-Whitney U, Kruskal-Wallis and Spearman's rho. Effect size gets a chapter of its own rather than a footnote. The organizing idea shows up in the structure: every single chapter ends with a section called Statistical Literacy, where a real published claim gets examined rather than a formula drilled. The material was assembled by David M. Lane with colleagues at Rice University and the University of Houston, with named authors on individual sections, and it closes with case studies and a glossary. The whole work is in the public domain, which the title page states plainly — it can be copied and reproduced without limitation.