Location of Tables for Tests of Significance | |

Preface | |

Biostatistics and Clinical Practice | p. 1 |

The Changing Medical Environment | p. 1 |

What Do Statistical Procedures Tell You? | p. 4 |

Why Not Depend on the Journals? | p. 6 |

Why Has the Problem Persisted? | p. 9 |

How to Summarize Data | p. 10 |

The Mean | p. 12 |

Measures of Variability | p. 13 |

The Normal Distribution | p. 14 |

Percentiles | p. 15 |

How to Make Estimates from a Limited Sample | p. 20 |

How Good Are These Estimates? | p. 21 |

How to Test for Difference between Groups | p. 31 |

The General Approach | p. 31 |

Two Different Estimates of the Population Variance | p. 36 |

What Is a "Big" F? | p. 38 |

Three Examples | p. 46 |

The Special Case of Two Groups: The t Test | p. 65 |

The General Approach | p. 67 |

The Standard Deviation of a Difference or a Sum | p. 69 |

Use of t to Test Hypotheses about Two Groups | p. 72 |

What If the Two Samples Are Not the Same Size? | p. 79 |

The Examples Revisited | p. 80 |

The t Test Is an Analysis of Variance | p. 84 |

Common Errors in the Use of the t Test and How to Compensate for Them | p. 86 |

How to Use t Tests to Isolate Differences between Groups in Analysis of Variance | p. 89 |

Other Approaches to Multiple Comparison Testing: The Student-Newman-Keuls Test | p. 95 |

Which Multiple Comparison Procedure Should You Use? | p. 101 |

Multiple Comparisons against a Single Control | p. 101 |

The Meaning of P | p. 107 |

How to Analyze Rates and Proportions | p. 113 |

Back to Mars | p. 114 |

Estimating Proportions from Samples | p. 119 |

Hypothesis Tests for Proportions | p. 123 |

Another Approach to Testing Nominal Data: Analysis of Contingency Tables | p. 132 |

Chi-Square Applications to Experiments with More Than Two Treatments or Outcomes | p. 139 |

The Fisher Exact Test | p. 144 |

Measures of Association Between Two Nominal Variables | p. 149 |

What Does "Not Significant" Really Mean? | p. 164 |

An Effective Diuretic | p. 165 |

Two Types of Errors | p. 169 |

What Determines a Test's Power? | p. 171 |

Power and Sample Size for Analysis of Variance | p. 184 |

Power and Sample Size for Comparing Two Proportions | p. 188 |

Power and Sample Size for Relative Risk and Odds Ratio | p. 192 |

Power and Sample Size for Contingency Tables | p. 193 |

Practical Problems in Using Power | p. 195 |

What Difference Does It Make? | p. 195 |

Confidence Intervals | p. 199 |

The Size of the Treatment Effect Measured as the Difference of Two Means | p. 200 |

The Effective Diuretic | p. 203 |

What Does "Confidence" Mean? | p. 207 |

Confidence Intervals Can Be Used to Test Hypotheses | p. 209 |

Confidence Interval for the Population Mean | p. 211 |

The Size of the Treatment Effect Measured as the Difference of Two Rates or Proportions | p. 212 |

Confidence Interval for Rates and Proportions | p. 217 |

Confidence Intervals for Relative Risk and Odds Ratio | p. 222 |

Confidence Interval for the Entire Population | p. 224 |

How to Test for Trends | p. 230 |

More about the Martians | p. 231 |

How to Estimate the Trend from a Sample | p. 238 |

How to Compare Two Regression Lines | p. 254 |

Correlation and Correlation Coefficients | p. 262 |

The Spearman Rank Correlation Coefficient | p. 273 |

Power and Sample Size in Regression and Correlation | p. 280 |

Comparing Two Different Measurements of the Same Thing: The Bland-Altman Method | p. 282 |

Experiments When Each Subject Receives More than One Treatment | p. 298 |

Experiments When Subjects Are Observed before and after a Single Treatment: The Paired t Test | p. 299 |

Another Approach to Analysis of Variance | p. 307 |

Experiments When Subjects Are Observed after Many Treatments: Repeated-Measures Analysis of Variance | p. 318 |

Experiments When Outcomes Are Measured on a Nominal Scale: McNemar's Test | p. 330 |

Alternatives to Analysis of Variance and the t Test Based on Ranks | p. 339 |

How to Choose between Parametric and Nonparametric Methods | p. 340 |

Two Different Samples: The Mann-Whitney Rank-Sum Test | p. 343 |

Each Subject Observed before and after One Treatment: The Wilcoxon Signed-Rank Test | p. 354 |

Experiments with Three or More Groups When Each Group Contains Different Individuals: The Kruskal-Wallis Statistic | p. 362 |

Experiments in Which Each Subject Receives More than One Treatment: The Friedman Test | p. 370 |

How to Analyze Survival Data | p. 387 |

Censoring on Pluto | p. 388 |

Estimating the Survival Curve | p. 391 |

Comparing Two Survival Curves | p. 400 |

Gehan's Test | p. 409 |

Power and Sample Size | p. 411 |

What Do the Data Really Show? | p. 416 |

When to Use Which Test | p. 417 |

Randomize and Control | p. 419 |

Does Randomization Ensure Correct Conclusions? | p. 427 |

Problems with the Population | p. 432 |

How You Can Improve Things | p. 434 |

App. A: Computational Forms | p. 438 |

App. B: Power Charts | p. 444 |

App. C: Answers to Exercises | p. 453 |

Index | p. 469 |

Table of Contents provided by Blackwell. All Rights Reserved. |