Principles of Biostatistics (2^nd Ed.)

This edition is a reprint of the second edition published in 2000 by Brooks/Cole and then Cengage Learning.

Principles of Biostatistics is aimed at students in the biological and health sciences who wish to learn modern research methods. It is based on a required course offered at the Harvard School of Public Health. In addition to these graduate students, many health professionals from the Harvard medical area attend as well.

The book is divided into three parts. The first five chapters deal with collections of numbers and ways in which to summarize, explore, and explain them. The next two chapters focus on probability and introduce the tools needed for the subsequent investigation of uncertainty. It is only in the eighth chapter and thereafter that the authors distinguish between populations and samples and begin to investigate the inherent variability introduced by sampling, thus progressing to inference. Postponing the slightly more difficult concepts until a solid foundation has been established makes it easier for the reader to comprehend them.

The supplements include a manual for students with solutions for odd-numbered exercises, a manual for instructors with solutions to all exercises, and selected data sets.

Marcello Pagano is Professor of Statistical Computing in the Department of Biostatistics at the Harvard School of Public Health. His research in biostatistics is on computer intensive inference and surveillance methods that involve screening methodologies, with their associated laboratory tests, and in obtaining more accurate testing results that use existing technologies.

Kimberlee Gauvreau is Associate Professor in the Department of Biostatistics and Associate Professor of Pediatrics at Harvard Medical School. Dr. Gauvreau?s research focuses on biostatistical issues arising in the field of pediatric cardiology. She also works on the development and validation of methods of adjustment for case mix complexity.

1. Introduction

2. Data Presentation

Types of Numerical Data

Nominal Data

Ordinal Data

Ranked Data

Discrete Data

Continuous Data

Tables

Frequency Distributions

Relative Frequency

Graphs

Bar Charts

Histograms

Frequency Polygons

One-Way Scatter Plots

Box Plots

Two-Way Scatter Plots

Line Graphs

Further Applications

Review Exercises

Measures of Central Tendency

Mean

Median

Mode

Measures of Dispersion

Range

Interquartile Range

Variance and Standard Deviation

Coefficient of Variation

Grouped Data

Grouped Mean

Grouped Variance

Chebychev's Inequality

Further Applications

Review Exercises

Rates

Standardization of Rates

Direct Method of Standardization

Indirect Method of Standardization

Use of Standardized Rates

Further Applications

Direct Method of Standardization

Indirect Method of Standardization

Review Exercises

Computation of the Life Table

Applications of the Life Table

Years of Potential Life Lost

Further Applications

Review Exercises

Operations on Events and Probability

Conditional Probability

Bayes' Theorem

Diagnostic Tests

Sensitivity and Specificity

Applications of Bayes' Theorem

ROC Curves

Calculation of Prevalence

The Relative Risk and the Odds Ratio

Further Applications

Review Exercises

7. Theoretical Probability Distributions

Probability Distributions

The Binomial Distribution

The Poisson Distribution

The Normal Distribution

Further Applications

Review Exercises

B Sampling Distributions

The Central Limit Theorem

Applications of the Central Limit Theorem

Further Applications

Review Exercises

Two-Sided Confidence Intervals

One-Sided Confidence Intervals

Student's t Distribution

Further Applications

Review Exercises

General Concepts

Two-Sided Tests of Hypotheses

One-Sided Tests of Hypotheses

Types of Error

Power

Sample Size Estimation

Further Applications

Review Exercises

Paired Samples

Independent Samples

Equal Variances

Unequal Variances

Further Applications

Review Exercises

One-Way Analysis of Variance

The Problem

Sources of Variation

Multiple Comparisons Procedures

Further Applications

Review Exercises

The Sign Test

The Wilcoxon Signed-Rank Test

The Wilcoxon Rank Sum Test

Advantages and Disadvantages of

Nonparametric Methods

Further Applications

Review Exercises

Normal Approximation to the Binomial Distribution

Sampling Distribution of a Proportion

Confidence Intervals

Hypothesis Testing

Sample Size Estimation

Comparison of Two Proportions

Further Applications

Review Exercises

The Chi-Square Test

X Tables

r X c Tables

McNemar's Test

The Odds Ratio

Berkson's Fallacy

Further Applications

Review Exercises

16. Multiple x Tables

Simpsons Paradox

The Mantei-Haenszel Method

Test of Homogeneity

Summary Odds Ratio

Test of Association

Further Applications

Review Exercises

The Two-Way Scatter Plot

Pearson's Correlation Coefficient

Spearman's Rank Correlation Coefficient

Further Applications

Review Exercises

18. Simple Linear Regression

Regression Concepts

The Model

The Population Regression Line

The Method of Least Squares

Inference for Regression Coefficients

Inference for Predicted Values

Evaluation of the Model

The Coefficient of Determination

Residual Plots

Transformations

Further Applications

Review Exercises

The Model

The Least-Squares Regression Equation

Inference for Regression Coefficients

Evaluation of the Model

Indicator Variables

Interaction Terms

Model Selection

Further Applications

Review Exercises

The Model

The Logistic Function

The Fitted Equation

Multiple Logistic Regression

Indicator Variables

Further Applications

Review Exercises

The Life Table Method

The Product-Limit Method

The Log-Rank Test

Further Applications

Review Exercises

Sampling Schemes

Simple Random Sampling

Systematic Sampling

Stratified Sampling

Cluster Sampling

Nonprobability Sampling

Sources of Bias

Further Applications

Review Exercises

Marcello Pagano is Professor of Statistical Computing in the Department of Biostatistics at the Harvard School of Public Health. His research in biostatistics is on computer intensive inference and surveillance methods that involve screening methodologies, with their associated laboratory tests, and in obtaining more accurate testing results that use existing technologies.

Kimberlee Gauvreau is Associate Professor in the Department of Biostatistics and Associate Professor of Pediatrics at Harvard Medical School. Dr. Gauvreau’s research focuses on biostatistical issues arising in the field of pediatric cardiology. She also works on the development and validation of methods of adjustment for case mix complexity.