Notes from contributors meetings at ASC/NZSA conference, July 5 2006

Tentative Timeline

  1. Nov. 1 2006: Authors provide detailed outline (subsections and a brief description) of their chapter.
  2. April 1 2007: Complete draft of chapters in.
  3. June 15 2007: Editors provide drafts to beta-test instructors for evaluation and comments.
  4. Fall semester 2007 (Northern Hemisphere), semester 1 2008 (Southern Hemisphere): beta testing of book (conditional on satisfactory evaluation at step 3).

Target

The target for this book is an introductory statistics class, appropriate for non-statistics majors in their first year at University. The emphasis is on learning how to use the techniques, and their motivation, but with a minimum of hand calculations or formal distribution theory. For instance, in the simple regression section, students would be expected to be able to carry out a simple regression using appropriate software, interpret the regression table, and use residual plots to diagnose problems with the regression assumptions. While they would have seen least squares formulae, they would not be expected to compute a regression line for a large data set 'by hand.' While they would have a general notion about how the F-statistic and it's p-value are related to the overall significance of the regression, they would not be expected to be able to prove that the sums of squares used to compute the numerator and denominator of the F-statistic have independent Chi-Squared distributions. We assume any course using this book will make extensive use of statistical software, but the choice of software is left to the instructor (see below).

Outline

See Draft Outline. As a result of the meeting we have:

Not all chapters in the outline are likely to be covered in a semester course. Chapters 1-2, 4-8, and 10 will be considered 'core' chapters, although perhaps with optional sections (eg joint distributions). Authors can for the most part rely on these chapters being covered previously when writing their chapter. Depending on the emphasis of a particular course, Chapter 3 may be covered superficially; chapters 8, and 11-14 may be covered superficially or not at all. Eg, an course for engineers may cover only the axioms of probability and ideas about independent events from chapter 3, and not use chapter 11 (Time Series) at all. A course for business majors exchange the time series chapter for the design of experiments material, and cover just the first few sections of the statistics for quality material.

Software

Initially plots and output shown in the text will be produced in R, with the assumption that similar plots or output can be produced in which ever software the instructor chooses to use. Instructions about how to use software will be restricted to an Appendix for each chapter. Production of appendicies for various software packages will be a second phase of the book production; at that time we will explore the possibility of automatic substitution the output/figures from the specific software packages into the text, so the book would effectively have an 'R version,' an 'Excell version' and so on.

Other notes

Authors requested guidance about the use of data/material from other sources. We will work on providing more information in the future, but for now look at the material on copyright at Wikipedia for guidance. Our current plan is to release the book under the same licence Wikipedia uses, GFDL. (Without invariant sections, for those concerned about that.)

Style notes: again, more later. For now: use imperial spellings. Don't start a sentence with a numeral. Short numbers that can be written as words are OK, eg "Two independent samples of size 50 were taken, one from each population."

Latex Style files coming soon.