本书是探索性数据分析（Exploratory Data Analysis，EDA）提出者、“数据科学之父”约翰·图基（John Tukey）的经典著作，书中尤其强调对数据分析采取更灵活态度以及仔细探索数据以了解其中可能包含哪些结构和信息的重要性。探索性数据分析是现代数据科学的鼻祖，建立了数据科学的关键基础。

1 SCRATCHING DOWN NUMBERS (stem-and-leaf)
Comments about the index page
1A Quantitative detective work
1B Practical arithmetic
1C Scratching down numbers
ID Doing better with stem-and-leaf
1E Using the right number of stems
lF How to count by tallying
IG What does it mean to "feel what the data are like"?
1H How far have we come?
1K How to use stem-and-leaf to pick up additional information (optional technique)
1P Additional problems
2 SCHEMATIC SUMMARIES (pictUres and numbers)
2A Extremes and median
2B Hinges and 5-number summaries
2C Box-and-whisker plots
2D Fences, and outside values
2E Schematic plots
2F Pros and cons; the Rayleigh example
2G Eighths, sixteenths, etc.
2H How far have we come?
3 EASY RE-EXPRESSION
3A Logarithms = logs
3B Quick logs
3C Comparisons of two batches
3D Quick roots and quick reciprocals
3E Looking quickly
3F Counted data
3G Relation among powers and logs (optional)
3H How far have we come?
3K How to think about logs (background)
3P Additional problems
4 EFFECTIVE COMPARISON (Including well-chosen expression)
4A Alternative forms of display of summaries
4B Comparing several batches (continued)
4C A more extensive example
4D The meaning of comparison
4E Adjustments, rough and exact
4F Residuals
4H How far have we come?
4P Additional problems
5 PLOTS OF RELATIONSHIP
5A How to plot y against x
5B Looking at subtraction
5C Subtracting straight lines
5D Plotting the population of the U.S.A.
5E Plotting the ratio of births to deaths
5F Untilting defines "tilt"
5H How far have we come?
5P Additional problems
6 STRAIGHTENING OUT PLOTS (using three points)
6A Looking at three points
6B Re-expressing y alone
6C Re-expressing x alone
6D A braking example
6E The vapor pressure of HzO
6F Re-expressing the second variable
6G Wise change of origin as a preliminary
6H How far have we come2
6P Additional problems
7 SMOOTHING SEQUENCES
7A Medians of 3
7B Eye resmoothing
7C Looking ahead
7D Copying-on--and more, usually.
7E Blurring the smooth--and setting the fences
7F Splitting peaks and valleys
7G Harming
7H How far have we come?
7+ OPTIONAL SECTIONS FOR CHAPTER 7
7I Breaking a smooth
7J Choice of expression
7K A two-section example
7M How much more may we have learned?
8 PARALLEL AND WANDERING SCHEMATIC PLOTS
8A Parallel schematic plots
8B Smoothing the cross-medians
8C Smoothing broken hinges
8D Dealing with the two questions
8E Wandering schematic plots
8F A more demanding example: Governor's salary and bank deposits
8G Further questions/analysis in the example
8H How far have we come?
8I The need to smooth both coordinates (optional)
9 DELINEATIONS OF BATCHES OF POINTS
9A E-traces and D-traces
9B Simple delineation--Twin Rivers again
9C Reduced and schematic delineations
9D What our schematic plots and delineations have missed
9E Three variables at once--or more
9H How far have we come?
10 USING TWO-WAY ANALYSES
10A Two-way residuals; row-PLUS-column analysis
10B The row-PLUS-column fit
10C Some points of technique
10D Row-TIMES-column analysis
10E Looking at row-PLUS-column fits and their residuals
10F Fitting one more constant
10G Converting PLUS to TIMES; re-expression
10H How far have we come?
11 MAKING TWO-WAY ANALYSES
11A Taking medians out
11B Alternative organizations of the arithmetic
11C Making the core of a two-way plot
l1D Going on with the residuals
11E Coding residuals; condensing fits and residuals
11F We can combine!
11G Guidance for expression
11H How far have we come?
11+ OPTIONAL SECTIONS FOR CHAPTERS 10 AND 11
11I Exploring beyond PLUS-one (extends Chapter 10)
11J Taking out any summary
11K An example of re-expression--city killings
IlL An unusual fit
11M How much more may we