John W Tukey - Collected Quotes

"[We] need men who can practice science - not a particular science - in a word, we need scientific generalists." (John W Tukey, "The Education of a Scientific Generalist", 1949)

"[...] the whole of modern statistics, philosophy and methods alike, is based on the principle of interpreting what did happen in terms of what might have happened." (John W Tukey, "Standard Methods of Analyzing Data, 1951)

"Just remember that not all statistics has been mathematized - and that we may not have to wait for its mathematization in order to use it." (John W Tukey, "The Growth of Experimental Design in a Research Laboratory, 1953)

"Difficulties in identifying problems have delayed statistics far more than difficulties in solving problems." (John W Tukey, Unsolved Problems of Experimental Statistics, 1954)

"Predictions, prophecies, and perhaps even guidance - those who suggested this title to me must have hoped for such-even though occasional indulgences in such actions by statisticians has undoubtedly contributed to the characterization of a statistician as a man who draws straight lines from insufficient data to foregone conclusions!" (John W Tukey, "Where do We Go From Here?", Journal of the American Statistical Association, Vol. 55 (289), 1960)

"Today one of statistics' great needs is a body of able investigators who make it clear to the intellectual world that they are scientific statisticians. and they are proud of that fact that to them mathematics is incidental, though perhaps indispensable." (John W Tukey, "Statistical and Quantitative Methodology, 1961)

"If data analysis is to be well done, much of it must be a matter of judgment, and ‘theory’ whether statistical or non-statistical, will have to guide, not command." (John W Tukey, "The Future of Data Analysis", Annals of Mathematical Statistics, Vol. 33 (1), 1962)

"The most important maxim for data analysis to heed, and one which many statisticians seem to have shunned is this: ‘Far better an approximate answer to the right question, which is often vague, than an exact answer to the wrong question, which can always be made precise.’ Data analysis must progress by approximate answers, at best, since its knowledge of what the problem really is will at best be approximate." (John W Tukey, "The Future of Data Analysis", Annals of Mathematical Statistics, Vol. 33, No. 1, 1962)

"The histogram, with its columns of area proportional to number, like the bar graph, is one of the most classical of statistical graphs. Its combination with a fitted bell-shaped curve has been common since the days when the Gaussian curve entered statistics. Yet as a graphical technique it really performs quite poorly. Who is there among us who can look at a histogram-fitted Gaussian combination and tell us, reliably, whether the fit is excellent, neutral, or poor? Who can tell us, when the fit is poor, of what the poorness consists? Yet these are just the sort of questions that a good graphical technique should answer at least approximately." (John W Tukey, "The Future of Processes of Data Analysis", 1965)

"The first step in data analysis is often an omnibus step. We dare not expect otherwise, but we equally dare not forget that this step, and that step, and other step, are all omnibus steps and that we owe the users of such techniques a deep and important obligation to develop ways, often varied and competitive, of replacing omnibus procedures by ones that are more sharply focused." (John W Tukey, "The Future of Processes of Data Analysis", 1965)

"The basic general intent of data analysis is simply stated: to seek through a body of data for interesting relationships and information and to exhibit the results in such a way as to make them recognizable to the data analyzer and recordable for posterity. Its creative task is to be productively descriptive, with as much attention as possible to previous knowledge, and thus to contribute to the mysterious process called insight." (John W Tukey & Martin B Wilk, "Data Analysis and Statistics: An Expository Overview", 1966)

"Comparable objectives in data analysis are (l) to achieve more specific description of what is loosely known or suspected; (2) to find unanticipated aspects in the data, and to suggest unthought-of-models for the data's summarization and exposure; (3) to employ the data to assess the (always incomplete) adequacy of a contemplated model; (4) to provide both incentives and guidance for further analysis of the data; and (5) to keep the investigator usefully stimulated while he absorbs the feeling of his data and considers what to do next." (John W Tukey & Martin B Wilk, "Data Analysis and Statistics: An Expository Overview", 1966)

"The science and art of data analysis concerns the process of learning from quantitative records of experience. By its very nature it exists in relation to people. Thus, the techniques and the technology of data analysis must be harnessed to suit human requirements and talents. Some implications for effective data analysis are: (1) that it is essential to have convenience of interaction of people and intermediate results and (2) that at all stages of data analysis the nature and detail of output, both actual and potential, need to be matched to the capabilities of the people who use it and want it." (John W Tukey & Martin B Wilk, "Data Analysis and Statistics: An Expository Overview", 1966)

"In many instances, a picture is indeed worth a thousand words. To make this true in more diverse circumstances, much more creative effort is needed to pictorialize the output from data analysis. Naive pictures are often extremely helpful, but more sophisticated pictures can be both simple and even more informative." (John W Tukey & Martin B Wilk, "Data Analysis and Statistics: An Expository Overview", 1966)

"Data analysis must be iterative to be effective. [...] The iterative and interactive interplay of summarizing by fit and exposing by residuals is vital to effective data analysis. Summarizing and exposing are complementary and pervasive." (John W Tukey & Martin B Wilk, "Data Analysis and Statistics: An Expository Overview", 1966)

"Summarizing data is a process of constrained and partial a process that essentially and inevitably corresponds to description - some sort of fitting, though it need not necessarily involve formal criteria or well-defined computations." (John W Tukey & Martin B Wilk, "Data Analysis and Statistics: An Expository Overview", 1966)

"The typical statistician has learned from bitter experience that negative results are just as important as positive ones, sometimes more so." (John W Tukey, "A Statistician's Comment", 1967)

"It is fair to say that statistics has made its greatest progress by having to move away from certainty [...] If we really want to make progress, we need to identify our next step away from certainty." (John W Tukey, "What Have Statisticians Been Forgetting", 1967)

"'Every student of the art of data analysis repeatedly needs to build upon his previous statistical knowledge and to reform that foundation through fresh insights and emphasis." (John W Tukey, "Data Analysis, Including Statistics", 1968)

"Every graph is at least an indication, by contrast with some common instances of numbers." (John W Tukey, "Data Analysis, Including Statistics", 1968)

"Nothing can substitute for relatively direct assessment of variability." (John W Tukey, "Data Analysis, Including Statistics", 1968)

"No one knows how to appraise a procedure safely except by using different bodies of data from those that determined it."  (John W Tukey, "Data Analysis, Including Statistics", 1968)

"The problems of different fields are much more alike than their practitioners think, much more alike than different."  (John W Tukey, "Analyzing Data: Sanctification or Detective Work?", 1969)

"[...] bending the question to fit the analysis is to be shunned at all costs." (John W Tukey, "Analyzing Data: Sanctification or Detective Work?", 1969)

"Data analysis is in important ways an antithesis of pure mathematics." (John W Tukey, "Data Analysis, Computation and Mathematics", 1972)

"Undoubtedly, the swing to exploratory data analysis will go somewhat too far. However : It is better to ride a damped pendulum than to be stuck in the mud." (John W Tukey, "Exploratory Data Analysis as Part of a Larger Whole", 1973)

"The greatest value of a picture is when it forces us to notice what we never expected to see." (John W Tukey, "Exploratory Data Analysis", 1977)

"[...] exploratory data analysis is an attitude, a state of flexibility, a willingness to look for those things that we believe are not there, as well as for those we believe might be there. Except for its emphasis on graphs, its tools are secondary to its purpose." (John W Tukey, [comment] 1979)

"There is NO question of teaching confirmatory OR exploratory - we need to teach both." (John W Tukey, "We Need Both Exploratory and Confirmatory", 1980)

"Finding the question is often more important than finding the answer." (John W Tukey, "We Need Both Exploratory and Confirmatory", 1980)

"[...] any hope that we are smart enough to find even transiently optimum solutions to our data analysis problems is doomed to failure, and, indeed, if taken seriously, will mislead us in the allocation of effort, thus wasting both intellectual and computational effort." (John W Tukey, "Choosing Techniques for the Analysis of Data", 1981)

"Detailed study of the quality of data sources is an essential part of applied work. [...] Data analysts need to understand more about the measurement processes through which their data come. To know the name by which a column of figures is headed is far from being enough." (John W Tukey, "An Overview of Techniques of Data Analysis, Emphasizing Its Exploratory Aspects", 1982)

"Exploratory data analysis, EDA, calls for a relatively free hand in exploring the data, together with dual obligations: (•) to look for all plausible alternatives and oddities - and a few implausible ones, (graphic techniques can be most helpful here) and (•) to remove each appearance that seems large enough to be meaningful - ordinarily by some form of fitting, adjustment, or standardization [...] so that what remains, the residuals, can be examined for further appearances." (John W Tukey, "Introduction to Styles of Data Analysis Techniques", 1982)

"The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data." (John W Tukey, "Sunset Salvo", The American Statistician Vol. 40 (1), 1986)

"The worst, i.e., most dangerous, feature of 'accepting the null hypothesis' is the giving up of explicit uncertainty. […] Mathematics can sometimes be put in such black-and-white terms, but our knowledge or belief about the external world never can." (John W Tukey, "The Philosophy of Multiple Comparisons", Statistical Science Vol. 6 (1), 1991)

"Statistics is the science, the art, the philosophy, and the technique of making inferences from the particular to the general." (John W Tukey)