Thinking clearly with data : a guide to quantitative reasoning and analysis / Ethan Bueno de Mesquita, Anthony Fowler.

Format:

Book

Author/Creator:

Bueno de Mesquita, Ethan, 1974- author.

Fowler, Anthony, author.

Language:

English

Subjects (All):

Sociology--Statistical methods.

Sociology.

Sociology--Methodology.

Physical Description:

xxi, 377 pages : illustrations (black and white) ; 26 cm

Place of Publication:

Princeton, New Jersey : Princeton University Press, [2021]

Summary:

"This is an intro-level text that teaches how to think clearly and conceptually about quantitative information, emphasizing ideas over technicality and assuming no prior exposure to data analysis, statistics, or quantitative methods. The books four parts present the foundation for quantiative reasoning: correlation and causation; statistical relationships; causal phenomena; and incorporating quantitative information into decision making. Within these parts it covers the array of tools used by social scientists, including regression, inference, experiments, research design, and more, all by explaining the rationale and logic behind such tools rather than focusing only on the technical calculations used for each. New concepts are presented simply, with the help of copious examples, and the books leans towards graphic rather than mathematical representation of data, with any technical material included in appendices"-- Provided by publisher.

Contents:

Chapter 1 Thinking Clearly in a Data-Driven Age p. 1

What You'll Learn p. 1

Cautionary Tales p. 2

Abe's hasty diagnosis p. 2

Civil resistance p. 3

Broken-windows policing p. 5

Thinking and Data Are Complements, Not Substitutes p. 7

Part I Establishing a Common Language p. 11

Chapter 2 Correlation: What Is It and What Is It Good For? p. 13

What You'll Learn p. 13

What Is a Correlation? p. 13

Fact or correlation? p. 18

What Is a Correlation Good For? p. 19

Forecasting p. 20

Causal inference p. 23

Measuring Correlations p. 24

Mean, variance, and standard deviation p. 24

Covariance p. 27

Correlation coefficient p. 28

Slope of the regression line p. 29

Populations and samples p. 29

Straight Talk about Linearity p. 30

Chapter 3 Causation: What Is It and What Is It Good For? p. 37

What You'll Learn p. 37

What Is Causation? p. 38

Potential Outcomes and Counterfactuals p. 39

What Is Causation Good For? p. 40

The Fundamental Problem of Causal Inference p. 41

What is the cause? p. 42

Causality and counterexamples p. 44

Causality and the law p. 47

Can causality run backward in time? p. 47

Does causality require a physical connection? p. 48

Causation need not imply correlation p. 49

Part II Does a Relationship Exist? p. 53

Chapter 4 Correlation Requires Variation p. 55

What You'll Learn p. 55

Selecting on the Dependent Variable p. 56

The 10,000-hour rule p. 57

Corrupting the youth p. 59

High school dropouts p. 62

Suicide attacks p. 63

The World Is Organized to Make Us Select on the Dependent Variable p. 64

Doctors mostly see sick people p. 65

Post-mortems p. 65

The Challenger disaster p. 67

The financial crisis of 2008 p. 69

Life advice p. 69

Chapter 5 Regression for Describing and Forecasting p. 74

What You'll Learn p. 74

Regression Basics p. 74

Linear Regression, Non-Linear Data p. 79

The Problem of Overfitting p. 87

Forecasting presidential elections p. 87

How Regression Is Presented p. 89

A Brief Intellectual History of Regression p. 89

Chapter 6 Samples, Uncertainty, and Statistical Inference p. 94

What You'll Learn p. 94

Why Do Estimates Differ from Estimands? p. 96

Bias p. 96

Noise p. 97

What Makes for a Good Estimator? p. 98

Quantifying Precision p. 99

Standard errors p. 99

Small samples and extreme observations p. 101

Confidence intervals p. 102

Statistical Inference and Hypothesis Testing p. 103

Hypothesis testing p. 103

Statistical significance p. 104

Statistical Inference about Relationships p. 105

What If We Have Data for the Whole Population? p. 106

Substantive versus Statistical Significance p. 107

Social media and voting p. 107

The Second Reform Act p. 108

Chapter 7 Over-Comparing, Under-Reporting p. 113

What You'll Learn p. 113

Can an octopus be a soccer expert? p. 113

Publication Bias p. 118

p-hacking p. 119

p-screening p. 120

Are Most Scientific "Facts" False? p. 122

ESP p. 122

Get out the vote p. 123

p-hacking forensics p. 124

Potential Solutions p. 126

Reduce the significance threshold p. 126

Adjust p-values for multiple testing p. 127

Don't obsess over statistical significance p. 127

Pre-registration p. 127

Requiring pre-registration in drug trials p. 128

Replication p. 128

Football and elections p. 129

Test important and plausible hypotheses p. 130

The power pose p. 131

Beyond Science p. 131

Superstars p. 132

Chapter 8 Reversion to the Mean p. 138

What You'll Learn p. 138

Does the truth wear off? p. 138

Francis Galton and Regression to Mediocrity p. 139

Reversion to the Mean Is Not a Gravitational Force p. 142

Seeking Help p. 145

Does knee surgery work? p. 146

Reversion to the Mean, the Placebo Effect, and Cosmic Habituation p. 147

The placebo effect p. 147

Cosmic habituation explained p. 148

Cosmic habituation and genetics p. 150

Beliefs Don't Revert to the Mean p. 150

Part III Is the Relationship Causal? p. 157

Chapter 9 Why Correlation Doesn't Imply Causation p. 159

What You'll Learn p. 159

Charter schools p. 160

Thinking Clearly about Potential Outcomes p. 163

Sources of Bias p. 168

Confounders p. 168

Reverse causality p. 169

The 10,000-hour rule, revisited p. 170

Diet soda p. 173

How Different Are Confounders and Reverse Causality? p. 174

Campaign spending p. 174

Signing the Bias p. 176

Contraception and HIV p. 179

Mechanisms versus Confounders p. 181

Thinking Clearly about Bias and Noise p. 183

Chapter 10 Controlling for Confounders p. 193

What You'll Learn p. 193

Party whipping in Congress p. 193

A note on heterogeneous treatment effects p. 197

The Anatomy of a Regression p. 198

How Does Regression Control? p. 201

Controlling and Causation p. 209

Is social media bad for you? p. 210

Reading a Regression Table p. 211

Controlling for Confounders versus Mechanisms p. 213

There Is No Magic p. 214

Chapter 11 Randomized Experiments p. 218

What You'll Learn p. 218

Breastfeeding p. 219

Randomization and Causal Inference p. 221

Estimation and Inference in Experiments p. 224

Standard errors p. 224

Hypothesis testing p. 225

Problems That Can Arise with Experiments p. 225

Noncompliance and instrumental variables p. 226

Chance imbalance p. 232

Lack of statistical power p. 234

Attrition p. 235

Interference p. 236

Natural Experiments p. 237

Military service and future earnings p. 238

Chapter 12 Regression Discontinuity Designs p. 243

What You'll Learn p. 243

How to Implement an RD Design p. 247

Are extremists or moderates more electable? p. 249

Continuity at the Threshold p. 251

Does continuity hold in election RD designs? p. 255

Noncompliance and the Fuzzy RD p. 256

Bombing in Vietnam p. 257

Motivation and Success p. 261

Chapter 13 Difference-in-Differences Designs p. 266

What You'll Learn p. 266

Parallel Trends p. 267

Two Units and Two Periods p. 269

Unemployment and the minimum wage p. 269

N Units and Two Periods p. 272

Is watching TV bad for kids? p. 273

N Units and N Periods p. 275

Contraception and the gender-wage gap p. 276

Useful Diagnostics p. 278

Do newspaper endorsements affect voting decisions? p. 278

Is obesity contagious? p. 279

Difference-in-Differences as Gut Check p. 282

The democratic peace p. 282

Chapter 14 Assessing Mechanisms p. 290

What You'll Learn p. 290

Causal Mediation Analysis p. 291

Intermediate Outcomes p. 292

Cognitive behavioral therapy and at-risk youths in Liberia p. 293

Independent Theoretical Predictions p. 294

Do voters discriminate against women? p. 294

Testing Mechanisms by Design p. 295

Social pressure and voting p. 295

Disentangling Mechanisms p. 296

Commodity price shocks and violent conflict p. 296

Part IV From Information to Decisions p. 303

Chapter 15 Turn Statistics into Substance p. 305

What You'll Learn p. 305

What's the Right Scale? p. 305

Miles-per-gallon versus gallons-per-mile p. 306

Percent versus percentage point p. 309

Visual Presentations of Data p. 309

Policy preferences and the Southern realignment p. 311

Some rules of thumb for data visualization p. 314

From Statistics to Beliefs: Bayes' Rule p. 314

Bayes' rule p. 317

Information, beliefs, priors, and posteriors p. 318

Abe's celiac revisited p. 319

Finding terrorists in an airport p. 322

Bayes' rule and quantitative analysis p. 325

Expected Costs and Benefits p. 328

Screening frequently or accurately p. 329

Chapter 16 Measure Your Mission p. 336

What You'll Learn p. 336

Measuring the Wrong Outcome or Treatment p. 337

Partial measures p. 337

Metal detectors in airports p. 337

Intermediate outcomes p. 339

Blood pressure and heart attacks p. 340

Ill-defined missions p. 341

Climate change and economic productivity p. 342

Do You Have the Right Sample? p. 343

External validity p. 343

Malnutrition in India and Bangladesh p. 343

Selected samples p. 344

College admissions p. 345

Why can't major league pitchers hit? p. 345

Strategic Adaptation and Changing Relationships p. 349

The duty on lights and windows p. 349

The shift in baseball p. 350

The war on drugs p. 351

Chapter 17 On the Limits of Quantification p. 357

What You'll Learn p. 357

Decisions When Evidence Is Limited p. 358

Cost-benefit analysis and environmental regulation p. 358

Floss your teeth and wear a mask p. 359

Floss your teeth p. 359

Wear a mask p. 360

Quantification and Values p. 361

How quantitative tools sneak in values p. 361

Algorithms and racial bias in health care p. 361

How quantification shapes our values p. 363

Think Clearly and Help Others Do So Too p. 367.

Notes:

Includes bibliographical references and index.

Other Format:

Online version: Bueno de Mesquita, Ethan, 1974- Thinking clearly with data

ISBN:

9780691214368

0691214360

9780691214351

0691214352

OCLC:

1246623843