Table of Contents
Foreword xv
Preface xvii
About the authors xxix
1 Getting Started with Data in R 1
1.1 What are R and R Studio? 1
1.1.1 Installing R and R Studio 2
1.1.2 Using R via R Studio 3
1.2 How do I code in R? 4
1.2.1 Basic programming concepts and terminology 4
1.2.2 Errors, warnings, and messages 6
1.2.3 Tips on learning to code 7
1.3 What are R packages? 8
1.3.1 Package installation 9
1.3.2 Package loading 11
1.3.3 Package use 11
1.4 Explore your first datasets 12
1.4.1 Nycflights13 package 12
1.4.2 Flights data frame 13
1.4.3 Exploring data frames 14
1.4.4 Identification and measurement variables 17
1.4.5 Help files 18
1.5 Conclusion 19
1.5.1 Additional resources 19
1.5.2 What's to come? 20
I Data Science with tidyverse 21
2 Data Visualization 23
2.1 The grammar of graphics 24
2.1.1 Components of the grammar 24
2.1.2 Gapminder data 25
2.1.3 Other components 26
2.1.4 ggplot2 package 27
2.2 Five named graphs - the 5NG 27
2.3 5NG#1: Scatterplots 28
2.3.1 Scatterplots via geom_point 29
2.3.2 Overplotting 31
2.3.3 Summary 35
2.4 5NG#2: Linegraphs 35
2.4.1 Linegraphs via geom_line 36
2.4.2 Summary 38
2.5 5NG#3: Histograms 38
2.5.1 Histograms via geom_histogram 40
2.5.2 Adjusting the bins 41
2.5.3 Summary 43
2.6 Facets 43
2.7 5NG#4: Boxplots 45
2.7.1 Boxplots via geom-boxplot 47
2.7.2 Summary 50
2.8 5NG#5: Barplots 50
2.8.1 Barplots via geom_bar or geom_col 51
2.8.2 Must avoid pie charts! 54
2.8.3 Two categorical variables 55
2.8.4 Summary 60
2.9 Conclusion 60
2.9.1 Summary table 60
2.9.2 Function argument specification 61
2.9.3 Additional resources 62
2.9.4 What's to come 62
3 Data Wrangling 65
3.1 The pipe operator: %>% 67
3.2 Filter rows 69
3.3 Summarize variables 72
3.4 Group_by rows 75
3.4.1 Grouping by more than one variable 78
3.5 Mutate existing variables 80
3.6 Arrange and sort rows 84
3.7 Join data frames 86
3.7.1 Matching "key" variable names 87
3.7.2 Different "key" variable names 88
3.7.3 Multiple "key" variables 89
3.7.4 Normal forms 90
3.8 Other verbs 91
3.8.1 Select variables 91
3.8.2 Rename variables 93
3.8.3 Top_n values of a variable 93
3.9 Conclusion 94
3.9.1 Summary table 94
3.9.2 Additional resources 96
3.9.3 What's to come? 96
4 Data Importing and "Tidy" Data 99
4.1 Importing data 100
4.1.1 Using the console 101
4.1.2 Using RStudio's interface 102
4.2 "Tidy" data 103
4.2.1 Definition of "tidy" data 106
4.2.2 Converting to "tidy" data 108
4.2.3 Nycflights13 package 112
4.3 Case study: Democracy in Guatemala 113
4.4 Tidyverse package 116
4.5 Conclusion 117
4.5.1 Additional resources 117
4.5.2 What's to come? 117
II Data Modeling with moderndive 119
5 Basic Regression 121
5.1 One numerical explanatory variable 123
5.1.1 Exploratory data analysis 124
5.1.2 Simple linear regression 133
5.1.3 Observed/fitted values and residuals 137
5.2 One categorical explanatory variable 139
5.2.1 Exploratory data analysis 140
5.2.2 Linear regression 147
5.2.3 Observed/fitted values and residuals 151
5.3 Related topics 152
5.3.1 Correlation is not necessarily causation 152
5.3.2 Best-fitting line 154
5.3.3 Get_regression_x() functions 157
5.4 Conclusion 160
5.4.1 Additional resources 160
5.4.2 What's to come? 160
6 Multiple Regression 161
6.1 One numerical and one categorical explanatory variable 162
6.1.1 Exploratory data analysis 162
6.1.2 Interaction model 166
6.1.3 Parallel slopes model 169
6.1.4 Observed/fitted values and residuals 173
6.2 Two numerical explanatory variables 175
6.2.1 Exploratory data analysis 176
6.2.2 Regression plane 181
6.2.3 Observed/fitted values and residuals 183
6.3 Related topics 184
6.3.1 Model selection 184
6.3.2 Correlation coefficient 188
6.3.3 Simpson's Paradox 188
6.4 Conclusion 191
6.4.1 Additional resources 191
6.4.2 What's to come? 191
III Statistical Inference with infer 193
7 Sampling 195
7.1 Sampling bowl activity 195
7.1.1 What proportion of this bowl's balls are red? 196
7.1.2 Using the shovel once 196
7.1.3 Using the shovel 33 times 198
7.1.4 What did we just do? 201
7.2 Virtual sampling 202
7.2.1 Using the virtual shovel once 203
7.2.2 Using the virtual shovel 33 times 206
7.2.3 Using the virtual shovel 1000 times 209
7.2.4 Using different shovels 212
7.3 Sampling framework 216
7.3.1 Terminology and notation 216
7.3.2 Statistical definitions 219
7.3.3 The moral of the story 222
7.4 Case study: Polls 226
7.5 Conclusion 230
7.5.1 Sampling scenarios 230
7.5.2 Central Limit Theorem 231
7.5.3 Additional resources 232
7.5.4 What's to come? 232
8 Bootstrapping and Confidence Intervals 233
8.1 Pennies activity 235
8.1.1 What is the average year on US pennies in 2019? 235
8.1.2 Resampling once 239
8.1.3 Resampling 35 times 244
8.1.4 What did we just do? 246
8.2 Computer simulation of resampling 247
8.2.1 Virtually resampling once 247
8.2.2 Virtually resampling 35 times 249
8.2.3 Virtually resampling 1000 times 251
8.3 Understanding confidence intervals 254
8.3.1 Percentile method 255
8.3.2 Standard error method 256
8.4 Constructing confidence intervals 258
8.4.1 Original workflow 259
8.4.2 Infer package workflow 259
8.4.3 Percentile method with infer 267
8.4.4 Standard error method with infer 269
8.5 Interpreting confidence intervals 271
8.5.1 Did the net capture the fish? 272
8.5.2 Precise and shorthand interpretation 280
8.5.3 Width of confidence intervals 281
8.6 Case study: Is yawning contagious? 284
8.6.1 Mythbusters study data 284
8.6.2 Sampling scenario 286
8.6.3 Constructing the confidence interval 287
8.6.4 Interpreting the confidence interval 294
8.7 Conclusion 295
8.7.1 Comparing bootstrap and sampling distributions 295
8.7.2 Theory-based confidence intervals 300
8.7.3 Additional resources 305
8.7.4 What's to come? 305
9 Hypothesis Testing 307
9.1 Promotions activity 308
9.1.1 Does gender affect promotions at a bank? 308
9.1.2 Shuffling once 310
9.1.3 Shuffling 16 times 314
9.1.4 What did we just do? 316
9.2 Understanding hypothesis tests 317
9.3 Conducting hypothesis tests 320
9.3.1 Infer package workflow 322
9.3.2 Comparison with confidence intervals 328
9.3.3 "There is only one test" 332
9.4 Interpreting hypothesis tests 333
9.4.1 Two possible outcomes 333
9.4.2 Types of errors 335
9.4.3 How do we choose alpha? 336
9.5 Case study: Are action or romance movies rated higher? 337
9.5.1 IMDb ratings data 338
9.5.2 Sampling scenario 340
9.5.3 Conducting the hypothesis test 341
9.6 Conclusion 347
9.6.1 Theory-based hypothesis tests 347
9.6.2 When inference is not needed 356
9.6.3 Problems with p-values 358
9.6.4 Additional resources 359
9.6.5 What's to come 359
10 Inference for Regression 361
10.1 Regression refresher 361
10.1.1 Teaching evaluations analysis 362
10.1.2 Sampling scenario 364
10.2 Interpreting regression tables 365
10.2.1 Standard error 366
10.2.2 Test statistic 367
10.2.3 p-value 368
10.2.4 Confidence interval 369
10.2.5 How does R compute the table? 370
10.3 Conditions for inference for regression 371
10.3.1 Residuals refresher 371
10.3.2 Linearity of relationship 373
10.3.3 Independence of residuals 374
10.3.4 Normality of residuals 375
10.3.5 Equality of variance 376
10.3.6 What's the conclusion? 378
10.4 Simulation-based inference for regression 379
10.4.1 Confidence interval for slope 380
10.4.2 Hypothesis test for slope 384
10.5 Conclusion 386
10.5.1 Theory-based inference for regression 386
10.5.2 Summary of statistical inference 388
10.5.3 Additional resources 389
10.5.4 What's to come 389
IV Conclusion 391
11 Tell Your Story with Data 393
11.1 Review 393
11.2 Case study: Seattle house prices 396
11.2.1 Exploratory data analysis: Part I 397
11.2.2 Exploratory data analysis: Part II 404
11.2.3 Regression modeling 407
11.2.4 Making predictions 409
11.3 Case study: Effective data storytelling 410
11.3.1 Bechdel test for Hollywood gender representation 411
11.3.2 US Births in 1999 411
11.3.3 Scripts of R code 414
Appendix A Statistical Background 417
A.1 Basic statistical terms 417
A.1.1 Mean 417
A.1.2 Median 417
A.1.3 Standard deviation 417
A.1.4 Five-number summary 418
A.1.5 Distribution 418
A.1.6 Outliers 418
A.2 Normal distribution 418
A.3 Log10 transformations 421
Appendix B Versions of R Packages Used 423
Bibliography 425
Index 427
