Main description:

This sophisticated package of statistical methods is for advanced master's (MPH) and PhD students in public health and epidemiology who are involved in the analysis of data. It makes the link from statistical theory to data analysis, focusing on the methods and data types most common in public health and related fields. Like most toolboxes, the statistical tools in this book are organized into sections with similar objectives. Unlike most toolboxes, however, these tools are accompanied by complete instructions, explanations, detailed examples, and advice on relevant issues and potential pitfalls - conveying skills, intuition, and experience. The only prerequisite is a first-year statistics course and familiarity with a computing package such as R, Stata, SPSS, or SAS. Though the book is not tied to a particular computing language, its figures and analyses were all created using R. Relevant R code, data sets, and links to public data sets are available from www.cambridge.org/9781107113084.

Contents:

Part I. Basics: 1. Statistical distribution; 2. Confidence intervals; 3. A weighted average; 4. Two discrete probability functions; 5. Correlation; Part II. Applications: 6. The 2 x 2 table; 7. Linear bivariate regression model; 8. The 2 x k table; 9. The log-linear Poisson regression model; 10. Two-way and three-way tables analysis; 11. Bootstrap analysis; 12. Graphical analysis; 13. The variance; 14. The log-normal distribution; 15. Nonparametric analysis; Part III. Survival: 16. Rates; 17. Nonparametric survival analysis; 18. The Weibull survival function; Part IV. Epidemiology: 19. Prediction, a natural measure of performance; 20. The attributable risk summary; 21. Time/space analysis; 22. ROC curve and analysis; Part V. Genetics: 23. Selection: a statistical description; 24. Mendelian segregation analysis; 25. Admixed populations; 26. Nonrandom mating; Part VI. Theory: 27. Statistical estimation; Part VII. R-Appendix.