Muutke küpsiste eelistusi

Intuitive Biostatistics: A Nonmathematical Guide to Statistical Thinking 4th Revised edition [Pehme köide]

(Founder and Chief Product Officer of GradPad Software, Inc., GraphPad Software, Inc.)
  • Formaat: Paperback / softback, 608 pages, kõrgus x laius x paksus: 155x231x31 mm, kaal: 757 g
  • Ilmumisaeg: 21-Dec-2017
  • Kirjastus: Oxford University Press Inc
  • ISBN-10: 0190643560
  • ISBN-13: 9780190643560
Teised raamatud teemal:
  • Formaat: Paperback / softback, 608 pages, kõrgus x laius x paksus: 155x231x31 mm, kaal: 757 g
  • Ilmumisaeg: 21-Dec-2017
  • Kirjastus: Oxford University Press Inc
  • ISBN-10: 0190643560
  • ISBN-13: 9780190643560
Teised raamatud teemal:
Intuitive Biostatistics takes a non-technical, non-quantitative approach to statistics and emphasizes interpretation of statistical results rather than the computational strategies for generating statistical data. This makes the text especially useful for those in health-science fields who have not taken a biostatistics course before. The text is also an excellent resource for professionals in labs, acting as a conceptually oriented and accessible biostatistics guide. With an engaging and conversational tone, Intuitive Biostatistics provides a clear introduction to statistics for undergraduate and graduate students and also serves as a statistics refresher for working scientists.

Arvustused

This splendid book meets a major need in public health, medicine, and biomedical research training. It is a user-friendly biostatistics text for non-mathematicians that clearly explains how to make sense of statistical results, avoid common mistakes in data analysis, avoid being confused by statistical nonsense, and how to make research more reproducible. Students may enjoy statistics for the first time! * Gilbert S. Omenn, University of Michigan * This book applies what I would call "scientific common sense" to the confusing world of statistical analysis and interpretation. If you want to really understand what a p-value is, read this book. * Louis G. Zachos, University of Mississippi * Intuitive Biostatistics places statistical concepts and practical issues of data analysis within an understandable light. The textbook helps the reader grasp the fundamentals and the pitfalls of data presentation and analysis. It should be on the "must-read" list of clinicians, journal reviewers, editors, and other consumers of scientific literature. * John D. Bonagura, The Ohio State University * I have already recommend the book to many colleagues. A concise, well written, and at times funny book that clearly explains the most important conceptual aspects about statistics, emphasizing proper interpretation of results and common mistakes to avoid. * Walter E. Schargel, The University of Texas at Arlington *

Preface   xxv  
  Who is this Book for?
 
  What makes the Book Unique?
 
  What's New?
 
  Which
Chapters are Essential?
 
  Who Helped?
 
  Who Am I?
 
Part A Introducing Statistics  
  1 Statistics and Probability are not Intuitive
  3 (11)
  We Tend to Jump to Conclusions
 
  We Tend to Be Overconfident
 
  We See Patterns in Random Data
 
  We Don't Realize that Coincidences are Common
 
  We Don't Expect Variability to Depend on Sample Size
 
  We Have Incorrect Intuitive Feelings about Probability
 
  We Find It Hard to Combine Probabilities
 
  We Don't Do Bayesian Calculations Intuitively
 
  We are Fooled by Multiple Comparisons
 
  We Tend to Ignore Alternative Explanations
 
  We are Fooled By Regression to the Mean
 
  We Let Our Biases Determine How We Interpret Data
 
  We Crave Certainty, but Statistics Offers Probabilities
 
  Chapter Summary
 
  Term Introduced in this
Chapter
 
  2 The Complexities of Probability
  14 (10)
  Basics of Probability
 
  Probability as Long-Term Frequency
 
  Probability as Strength of Belief
 
  Calculations With Probabilities Can be Easier If You Switch to Calculating with Whole Numbers
 
  Common Mistakes: Probability
 
  Lingo
 
  Probability in Statistics
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  3 From Sample to Population
  24 (7)
  Sampling from a Population
 
  Sampling Error and Bias
 
  Models and Parameters
 
  Multiple Levels of Sampling
 
  What if Your Sample is the Entire Population?
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
Part B Introducing Confidence Intervals  
  4 Confidence Interval of a Proportion
  31 (15)
  Data Expressed as Proportions
 
  The Binomial Distribution: From Population to Sample
 
  Example: Free Throws in Basketball
 
  Example: Polling Voters
 
  Assumptions: Confidence Interval of a Proportion
 
  What Does 95% Confidence Really Mean?
 
  Are You Quantifying the Event You Care About?
 
  Lingo
 
  Calculating the CI of a Proportion
 
  Ambiguity if the Proportion is 0% or 100%
 
  An Alternative Approach: Bayesian Credible Intervals
 
  Common Mistakes: CI of a Proportion
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  5 Confidence Interval of Survival Data
  46 (9)
  Survival Data
 
  Censored Survival Data
 
  Calculating Percentage Survival at Various Times
 
  Graphing Survival Curves with Confidence Bands
 
  Summarizing Survival Curves
 
  Assumptions: Survival Analysis
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  6 Confidence Interval of Counted Data (Poisson Distribution)
  55 (8)
  The Poisson Distribution
 
  Assumptions: Poisson Distribution
 
  Confidence Intervals Based on Poisson Distributions
 
  How to Calculate the Poisson CI
 
  The Advantage of Counting for Longer Time Intervals (Or in Larger Volumes)
 
  Q&A
 
  Chapter Summary
 
  Term Introduced in this
Chapter
 
Part C Continuous Variables  
  7 Graphing Continuous Data
  63 (12)
  Continuous Data
 
  The Mean and Median
 
  Lingo: Terms Used to Explain Variability
 
  Percentiles
 
  Graphing Data to Show Variation
 
  Graphing Distributions
 
  Beware of Data Massage
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  8 Types of Variables
  75 (5)
  Continuous Variables
 
  Discrete Variables
 
  Why It Matters?
 
  Not Quite as Distinct as They Seem
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  9 Quantifying Scatter
  80 (9)
  Interpreting a Standard Deviation
 
  How it Works: Calculating SD
 
  Why n - 1?
 
  Situations in Which n Can Seem Ambiguous
 
  SD and Sample Size
 
  Other Ways to Quantify and Display Variability
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  10 The Gaussian Distribution
  89 (6)
  The Nature of The Gaussian Distribution
 
  SD and the Gaussian Distribution
 
  The Standard Normal Distribution
 
  The Normal Distribution does not Define Normal Limits
 
  Why The Gaussian Distribution is so Central to Statistical Theory
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  11 The Lognormal Distribution and Geometric Mean
  95 (6)
  The Origin of a Lognormal Distribution
 
  Logarithms?
 
  Geometric Mean
 
  Geometric SD
 
  Common Mistakes: Lognormal Distributions
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  12 Confidence Interval of a Mean
  101 (9)
  Interpreting A CI of a Mean
 
  What Values Determine the CI of a Mean?
 
  Assumptions: CI of a Mean
 
  How to Calculate the CI of a Mean
 
  More about Confidence Intervals
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  13 The Theory of Confidence Intervals
  110 (8)
  CI of a Mean Via the t Distribution
 
  CI of a Mean Via Resampling
 
  CI of a Proportion Via Resampling
 
  CI of a Proportion Via Binomial Distribution
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  14 Error Bars
  118 (11)
  SD Versus Sem
 
  Which Kind of Error Bar Should I Plot?
 
  The Appearance of Error Bars
 
  How are SD and Sem Related to Sample Size?
 
  Geometric SD Error Bars
 
  Common Mistakes: Error Bars
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
Part D P Values and Statistical Significance  
  15 Introducing P Values
  129 (16)
  Introducing P Values
 
  Example 1: Coin Flipping
 
  Example 2: Antibiotics on Surgical Wounds
 
  Example 3: Angioplasty and Myocardial Infarction
 
  Lingo: Null Hypothesis
 
  Why P Values are Confusing
 
  One- Or Two-Tailed P Value?
 
  P Values are Not Very Reproducible
 
  There is much more to Statistics than P Values
 
  Common Mistakes: P Values
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  16 Statistical Significance and Hypothesis Testing
  145 (12)
  Statistical Hypothesis Testing
 
  Analogy: Innocent Until Proven Guilty
 
  Extremely Significant? Borderline Significant?
 
  Lingo: Type I and Type II Errors
 
  Tradeoffs When Choosing a Significance Level
 
  What Significance Level Should You Choose?
 
  Interpreting A CI, A P Value, and A Hypothesis Test
 
  Statistical Significance vs. Scientific Significance
 
  Common Mistakes: Statistical Hypothesis Testing
 
  Q&A
 
  Chapter Summary
 
  Terms Defined in this
Chapter
 
  17 Comparing Groups with Confidence Intervals and P Values
  157 (8)
  CIS and Statistical Hypothesis Testing are Closely Related
 
  Four Examples with CIS, P Values, and Conclusion about Statistical Significance
 
  Q&A
 
  Chapter Summary
 
  18 Interpreting a Result that is Statistically Significant
  165 (14)
  Seven Explanations for Results that are "Statistically Significant"
 
  How Frequently do Type I Errors (False Positives) Occur?
 
  The Prior Probability Influences the FPRP (A Bit of Bayes)
 
  Bayesian Analysis
 
  Accounting for Prior Probability Informally
 
  The Relationship Between Sample Size and P Values
 
  Common Mistakes
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  19 Interpreting a Result that is not Statistically Significant
  179 (7)
  Five Explanations For "Not Statistically Significant" Results
 
  "Not Significantly Different" does not Mean "No Difference"
 
  Example: a2-Adrenergic Receptors on Platelets
 
  Example: Fetal Ultrasounds
 
  How to Get Narrower CIS
 
  What if the P Value is Really High?
 
  Q&A
 
  Chapter Summary
 
  20 Statistical Power
  186 (7)
  What is Statistical Power?
 
  Distinguishing Power From Beta and the False Discovery Rate
 
  An Analogy to Understand Statistical Power
 
  Power of the Two Example Studies
 
  When does It Make Sense to Compute Power?
 
  Common Mistakes: Power
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  21 Testing For Equivalence or Noninferiority
  193 (10)
  Equivalence must be Defined Scientifically, not Statistically
 
  If the Mean is Within the Equivalence Zone
 
  If the Mean is Outside of the Equivalence Zone
 
  Applying the Usual Approach of Statistical Hypothesis Testing to Testing for Equivalence
 
  Noninferiority Tests
 
  Common Mistakes: Testing for Equivalence
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
Part E Challenges in Statistics  
  22 Multiple Comparisons Concepts
  203 (11)
  The Problem of Multiple Comparisons
 
  Correcting for Multiple Comparisons is not Always Needed
 
  The Traditional Approach to Correcting for Multiple Comparisons
 
  Correcting for Multiple Comparisons with the False Discovery Rate
 
  Comparing the Two Methods of Correcting for Multiple Comparisons
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  23 The Ubiquity of Multiple Comparisons
  214 (10)
  Overview
 
  Multiple Comparisons in Many Contexts
 
  When are Multiple Comparisons Data Torture or P-Hacking?
 
  How to Cope with Multiple Comparisons
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  24 Normality Tests
  224 (8)
  The Gaussian Distribution is an Unreachable Ideal
 
  What A Gaussian Distribution Really Looks Like
 
  QQ Plots
 
  Testing for Normality
 
  Alternatives to Assuming a Gaussian Distribution
 
  Common Mistakes: Normality Tests
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  25 Outliers
  232 (7)
  How do Outliers Arise?
 
  The Need for Outlier Tests
 
  Five Questions to Ask before Testing for Outliers
 
  Outlier Tests
 
  Is It Legitimate to Remove Outliers?
 
  An Alternative: Robust Statistical Tests
 
  Lingo: Outlier
 
  Common Mistakes: Outlier Tests
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  26 Choosing a Sample Size
  239 (24)
  Sample Size Principles
 
  An Alternative Way to think about Sample Size Calculations
 
  Interpreting a Sample Size Statement
 
  Lingo: Power
 
  Calculating the Predicted FPRP as Part of Interpreting a Sample Size Statement
 
  Complexities when Computing Sample Size
 
  Examples
 
  Other Approaches to Choosing Sample Size
 
  Common Mistakes: Sample Size
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
Part F Statistical Tests  
  27 Comparing Proportions
  263 (10)
  Example: Apixaban for Treatment of Thromboembolism
 
  Assumptions
 
  Comparing Observed and Expected Proportions
 
  Common Mistakes: Comparing Proportions
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  28 Case-Control Studies
  273 (11)
  Example: Does a Cholera Vaccine Work?
 
  Example: Isotretinoin and Bowel Disease
 
  Example: Genome-Wide Association Studies
 
  How are Controls Defined?
 
  How are Cases Defined?
 
  Epidemiology Lingo
 
  Common Mistakes: Case-Control Studies
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  29 Comparing Survival Curves
  284 (10)
  Example Survival Data
 
  Assumptions when Comparing Survival Curves
 
  Comparing Two Survival Curves
 
  Why Not Just Compare Mean or Median Survival Time: Five-Year Survival?
 
  Intention to Treat
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  30 Comparing Two Means: Unpaired t Test
  294 (12)
  Interpreting Results from an Unpaired t Test
 
  Assumptions: Unpaired t Test
 
  The Assumption of Equal Variances
 
  Overlapping Error Bars and the t Test
 
  How It Works: Unpaired t Test
 
  Common Mistakes: Unpaired t Test
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  31 Comparing Two Paired Groups
  306 (12)
  When to Use Special Tests for Paired Data
 
  Example of Paired t Test
 
  Interpreting Results from a Paired t Test
 
  The Ratio Paired t Test
 
  McNemar's Test for a Paired Case-Control Study
 
  Common Mistakes: Paired t Test
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  32 Correlation
  318 (13)
  Introducing the Correlation Coefficient
 
  Assumptions: Correlation
 
  Lingo: Correlation
 
  How It Works: Calculating the Correlation Coefficient
 
  Common Mistakes: Correlation
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
Part G Fitting Models to Data  
  33 Simple Linear Regression
  331 (19)
  The Goals of Linear Regression
 
  Linear Regression Results
 
  Assumptions: Linear Regression
 
  Comparison of Linear Regression and Correlation
 
  Lingo: Linear Regression
 
  Common Mistakes: Linear Regression
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  34 Introducing Models
  350 (7)
  Lingo: Models, Parameters, and Variables
 
  The Simplest Model
 
  The Linear Regression Model
 
  Why Least Squares?
 
  Other Models and other Kinds of Regression
 
  Common Mistakes: Models
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  35 Comparing Models
  357 (9)
  Comparing Models is a Major Part of Statistics
 
  Linear Regression as a Comparison of Models
 
  Unpaired t Test Recast as Comparing the Fit of Two Models
 
  Common Mistakes: Comparing Models
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  36 Nonlinear Regression
  366 (12)
  Introducing Nonlinear Regression
 
  An Example of Nonlinear Regression
 
  Nonlinear Regression Results
 
  How Nonlinear Regression Works
 
  Assumptions: Nonlinear Regression
 
  Comparing Two Models
 
  Tips for Understanding Models
 
  Learn More About Nonlinear Regression
 
  Common Mistakes: Nonlinear Regression
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  37 Multiple Regression
  378 (17)
  Goals of Multivariable Regression
 
  Lingo
 
  An Example of Multiple Linear Regression
 
  Assumptions
 
  Automatic Variable Selection
 
  Sample Size for Multiple Regression
 
  More Advanced Issues with Multiple Regression
 
  Common Mistakes: Multiple Regression
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  38 Logistic and Proportional Hazards Regression
  395 (12)
  Logistic Regression
 
  Proportional Hazards Regression
 
  Assumptions: Logistic Regression
 
  Common Mistakes: Logistic Regression
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
Part H The Rest of Statistics  
  39 Analysis of Variance
  407 (11)
  Comparing the Means of Three or More Groups
 
  Assumptions: One-Way Anova
 
  How It Works: One-Way Anova
 
  Repeated-Measures One Way Anova
 
  An Example of Two-Way Anova
 
  How Two-Way Anova Works
 
  Repeated Measures Two-way Anova
 
  Common Mistakes: Anova
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  40 Multiple Comparison Tests after Anova
  418 (13)
  Multiple Comparison Tests for the Example Data
 
  The Logic Of Multiple Comparisons Tests
 
  Other Multiple Comparisons Tests
 
  How It Works: Multiple Comparisons Tests
 
  When Are Multiple Comparisons Tests Not Needed?
 
  Common Mistakes: Multiple Comparisons
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  41 Nonparametric Methods
  431 (11)
  Nonparametric Tests Based on Ranks
 
  The Advantages and Disadvantages of Nonparametric Tests
 
  Choosing Between Parametric and Nonparametric Tests: Does It Matter?
 
  Sample Size for Nonparametric Tests
 
  Nonparametric Tests that Analyze Values (Not Ranks)
 
  Common Mistakes: Nonparametric Tests
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  42 Sensitivity, Specificity, and Receiver Operating Characteristic Curves
  442 (10)
  Definitions of Sensitivity and Specificity
 
  The Predictive Value of a Test
 
  Receiver-Operating Characteristic (ROC) Curves
 
  Bayes Revisited
 
  Common Mistakes
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  43 Meta-Analysis
  452 (11)
  Introducing Meta-Analysis
 
  Publication Bias
 
  Results from a Meta-Analysis
 
  Meta-Analysis of Individual Participant Data
 
  Assumptions of Meta-Analysis
 
  Common Mistakes: Meta-Analysis
 
  Q&A
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
Part I Putting It All Together  
  44 The Key Concepts of Statistics
  463 (5)
  Term Introduced in this
Chapter
 
  45 Statistical Traps to Avoid
  468 (19)
  Trap #1: Focusing on P Values and Statistical Significance Rather than Effect Size
 
  Trap #2: Testing Hypotheses Suggested by the Data
 
  Trap #3: Analyzing Without a Plan-"P-Hacking"
 
  Trap #4: Making a Conclusion about Causation When the Data Only Show Correlation
 
  Trap #5: Overinterpreting Studies that Measure a Proxy or Surrogate Outcome
 
  Trap #6: Overinterpreting Data from an Observational Study
 
  Trap #7: Making Conclusions about Individuals when the Data Were only Collected for Groups
 
  Trap #8: Focusing Only on Means Without asking about Variability or Unusual Values
 
  Trap #9: Comparing Statistically Significant with Not Statistically Significant
 
  Trap #10: Missing Important Findings Because Data Combine Populations
 
  Trap #11: Invalid Multiple Regression Analyses as a Result of an Omitted Variable
 
  Trap #12: Overfitting
 
  Trap #13: Mixing Up the Significance Level with the FPRP
 
  Trap #14: Not Recognizing How Common False Positive Findings are
 
  Trap #15: Not Realizing How Likely it is that a "Significant" Conclusion From a Speculative Experiment is a False Positive
 
  Trap #16: Not Realizing That Many Published Studies have Little Statistical Power
 
  Trap #17: Trying to Detect Small Signals When there is Lots of Noise
 
  Trap #18: Unnecessary Dichotomizing
 
  Trap #19: Inflating Sample Size by Pseudoreplication
 
  Chapter Summary
 
  Terms Introduced in this
Chapter
 
  46 Capstone Example
  487 (15)
  The Case of the Eight Naked IC50S
 
  Look Behind the Data
 
  Statistical Significance by Cheating
 
  Using a t Test That Doesn't Assume Equal SDs
 
  Unpaired t Test as Linear or Nonlinear Regression
 
  Nonparametric Mann-Whitney Test
 
  Just Report the Last Confirmatory Experiment?
 
  Increase Sample Size?
 
  Comparing the Logarithms of IC50 Values
 
  Sample Size Calculations Revisited
 
  Is it Ok to Switch Analysis Methods?
 
  The Usefulness of Simulations
 
  Chapter Summary
 
  47 Statistics and Reproducibility
  502 (9)
  The Repoducibility Crisis
 
  Many Analyses are Biased to Inflate the Effect Size
 
  Even Perfectly Performed Experiments are Less Reproducible than Most Expect
 
  Summary
 
  48 Checklists for Reporting Statistical Methods and Results
  511 (6)
  Reporting Methods Used for Data Analysis
 
  Graphing Data
 
  Reporting Statistical Results
 
Part J Appendices   517 (16)
  Appendix A: Statistics with Graphpad
 
  Appendix B: Statistics with Excel
 
  Appendix C: Statistics with R
 
  Appendix D: Values of the t Distribution Needed to Compute CIs
 
  Appendix E: A Review of Logarithms
 
  Appendix F: Choosing a Statistical Test
 
  Appendix G: Problems and Answers
 
References   533 (15)
Index   548  
Harvey Motulsky is the CEO and Founder of GraphPad Software, Inc. He wrote the first edition of this text while on the faculty of the Department of Pharmacology at University of California, San Diego.