Understanding Statistics and Experimental Design : : How to Not Lie with Statistics.

• Superior document: Learning Materials in Biosciences Series
• Author: Herzog, Michael H.
• TeilnehmendeR: Francis, Gregory.; Clarke, Aaron.
• Place / Publishing House: Cham : : Springer International Publishing AG,, 2019.; ©2019.
• Year of Publication: 2019
• Edition: 1st ed.
• Language: English
• Series: Learning Materials in Biosciences Series
• Physical Description: 1 online resource (146 pages)

Table of Contents

Intro -- Preface -- Science, Society, and Statistics -- About This Book -- Contents -- Part I The Essentials of Statistics -- 1 Basic Probability Theory -- Contents -- 1.1 Confusions About Basic Probabilities: Conditional Probabilities -- 1.1.1 The Basic Scenario -- 1.1.2 A Second Test -- 1.1.3 One More Example: Guillain-Barré Syndrome -- 1.2 Confusions About Basic Probabilities: The Odds Ratio -- 1.2.1 Basics About Odds Ratios (OR) -- 1.2.2 Partial Information and the World of Disease -- References -- 2 Experimental Design and the Basics of Statistics: Signal Detection Theory (SDT) -- Contents -- 2.1 The Classic Scenario of SDT -- 2.2 SDT and the Percentage of Correct Responses -- 2.3 The Empirical d -- 3 The Core Concept of Statistics -- Contents -- 3.1 Another Way to Estimate the Signal-to-Noise Ratio -- 3.2 Undersampling -- 3.2.1 Sampling Distribution of a Mean -- 3.2.2 Comparing Means -- 3.2.3 The Type I and II Error -- 3.2.4 Type I Error: The p-Value is Related to a Criterion -- 3.2.5 Type II Error: Hits, Misses -- 3.3 Summary -- 3.4 An Example -- 3.5 Implications, Comments and Paradoxes -- Reference -- 4 Variations on the t-Test -- Contents -- 4.1 A Bit of Terminology -- 4.2 The Standard Approach: Null Hypothesis Testing -- 4.3 Other t-Tests -- 4.3.1 One-Sample t-Test -- 4.3.2 Dependent Samples t-Test -- 4.3.3 One-Tailed and Two-Tailed Tests -- 4.4 Assumptions and Violations of the t-Test -- 4.4.1 The Data Need to be Independent and Identically Distributed -- 4.4.2 Population Distributions are Gaussian Distributed -- 4.4.3 Ratio Scale Dependent Variable -- 4.4.4 Equal Population Variances -- 4.4.5 Fixed Sample Size -- 4.5 The Non-parametric Approach -- 4.6 The Essentials of Statistical Tests -- 4.7 What Comes Next? -- Part II The Multiple Testing Problem -- 5 The Multiple Testing Problem -- Contents -- 5.1 Independent Tests.

5.2 Dependent Tests -- 5.3 How Many Scientific Results Are Wrong? -- 6 ANOVA -- Contents -- 6.1 One-Way Independent Measures ANOVA -- 6.2 Logic of the ANOVA -- 6.3 What the ANOVA Does and Does Not Tell You: Post-Hoc Tests -- 6.4 Assumptions -- 6.5 Example Calculations for a One-Way Independent Measures ANOVA -- 6.5.1 Computation of the ANOVA -- 6.5.2 Post-Hoc Tests -- 6.6 Effect Size -- 6.7 Two-Way Independent Measures ANOVA -- 6.8 Repeated Measures ANOVA -- 7 Experimental Design: Model Fits, Power, and Complex Designs -- Contents -- 7.1 Model Fits -- 7.2 Power and Sample Size -- 7.2.1 Optimizing the Design -- 7.2.2 Computing Power -- 7.3 Power Challenges for Complex Designs -- 8 Correlation -- Contents -- 8.1 Covariance and Correlations -- 8.2 Hypothesis Testing with Correlations -- 8.3 Interpreting Correlations -- 8.4 Effect Sizes -- 8.5 Comparison to Model Fitting, ANOVA and t-Test -- 8.6 Assumptions and Caveats -- 8.7 Regression -- Part III Meta-analysis and the Science Crisis -- 9 Meta-analysis -- Contents -- 9.1 Standardized Effect Sizes -- 9.2 Meta-analysis -- Appendix -- Standardized Effect Sizes Beyond the Simple Case -- Extended Example of the Meta-analysis -- 10 Understanding Replication -- Contents -- 10.1 The Replication Crisis -- 10.2 Test for Excess Success (TES) -- 10.3 Excess Success from Publication Bias -- 10.4 Excess Success from Optional Stopping -- 10.5 Excess Success and Theoretical Claims -- Reference -- 11 Magnitude of Excess Success -- Contents -- 11.1 You Probably Have Trouble Detecting Bias -- 11.2 How Extensive Are These Problems? -- 11.3 What Is Going On? -- 11.3.1 Misunderstanding Replication -- 11.3.2 Publication Bias -- 11.3.3 Optional Stopping -- 11.3.4 Hypothesizing After the Results Are Known (HARKing) -- 11.3.5 Flexibility in Analyses -- 11.3.6 Misunderstanding Prediction.

11.3.7 Sloppiness and Selective Double Checking -- 12 Suggested Improvements and Challenges -- Contents -- 12.1 Should Every Experiment Be Published? -- 12.2 Preregistration -- 12.3 Alternative Statistical Analyses -- 12.4 The Role of Replication -- 12.5 A Focus on Mechanisms.