Bruce Levin

Preface.Preface to the Second Edition.Preface to the First Edition.1. An Introduction to Applied Probability.2. Statistical Inference for a Single Proportion.3. Assessing Significance in a Fourfold Table.4. Determining Sample Sizes Needed to Detect a Difference Between Two Proportions.5. How to Randomize.6. Comparative Studies: Cross-Sectional, Naturalistic, or Multinomial Sampling.7. Comparative Studies: Prospective and Retrospective Sampling.8. Randomized Controlled Trials.9. The Comparison of Proportions from Several Independent Samples.10. Combining Evidence from Fourfold Tables.11. Logistic Regression.12. Poisson Regression.13. Analysis of Data from Matched Samples.14. Regression Models for Matched Samples.15. Analysis of Correlated Binary Data.16. Missing Data.17. Misclassification Errors: Effects, Control, and Adjustment.18. The Measurement of Interrater Agreement.19. The Standardization of Rates.Appendix A. Numerical Tables.Appendix B. The Basic Theory of Maximum Likelihood Estimation.Appendix C. Answers to Selected Problems.Author Index.Subject Index.

There are two kinds of errors one must guard against in designing a comparative study. Even though these errors can occur in any statistical evaluation, their discussion here is restricted to the case where proportions from two independent samples are compared, that is, to sampling methods II and III. We present some means of specifying an important difference between proportions and provide mathematical results that will aide the investigator in finding the desired sample sizes. The later sections are devoted to the case where unequal sample sizes are planned for beforehand, and we present a discussion of some additional uses of the chapter's tables, including detectable effect sizes. A problem solving section appears at the end of the chapter.

Most studies compare two or more proportions, but occasionally we need to draw statistical inferences about a single problem. This chapter presents a brief survey of inferential methods, and lays a formal groundwork for statistical ideas use throughout this and other chapters. We begin with a discussion of methods of testing hypotheses about the parameter of a binomial distribution, using exact binomial calculations. We continue with a discussion of confidence intervals. We present methods that are approximate, based on large sample normal theory, which require only pencil-and-paper calculations, or at most a hand-held calculator. We then consider the important question of sample size planning for a single sample study. Next, we discuss how to calculate approximate standard errors by the delta method. We also discuss different ways of determining p-values and confidence intervals for discrete and asymmetrical distributions. A problem solving section appears at the end of the chapter.