Computing in the undergraduate curriculum•Teaching of Statistics Seminar•May 7, 2009Approaching Statistics 13•My focus today will be mainly on my experiences Winter Quarter 2008 with Statistics 13; it will be a report on my experiences and is, at best, a work in progress•I am new to teaching and really don’t have a clue about what works and doesn’t -- I am probably the least qualified member of the department to lead a seminar on teaching!•With those caveats...Approaching Statistics 13•In terms of background, it had been three years since I last taught this class, and in the meantime I had been rethinking the role that computing and data technologies should play at the undergraduate level1. In 2005 and 2006 we hosted a summer program for undergraduates that emphasized computing and visualization2. CENS introduced a core research focus in participatory urban sensing; this emphasized data collection and analysis by the public3. Together with Co-PIs from Geography, Computer Science and Information Studies, we mapped out a field “Data Science” for the NSF IGERT call4. In my graduate courses (seminars, Statistics 202a) I refined a story in which the fate of statistics was tied to that of information technologies...The human cannot reasonably be imagined as the center of this distributed networked system. Rather, data flows stream in from everywhere, and the vast majority of them remain invisible to human perception and direct control. Humans in this world become actors among many different kinds of agents, most of them non-human and non-biological. Human intelligence is prized not because it uniquely distinguishes us from other (biological) species, but because it occupies a niche within a complex ecology of interactions designed to optimize its capabilities and minimize its limitations. In such a world, technological expertise (which correlates only in complex ways with economic class) assumes increased importance. Those who can interrogate databases, and even more so those who control database design and the standards determining how they function within large network systems, set the rules by which the rest of us must play, whether we are conscious of it or not. In a way that Marx could not have imagined, data, capital, technological innovation, and information flows are setting new parameters for the exercise of power and control, and new landscapes defining what it means to be human. This is the real fear, and the real promise, of infinite data.N. Katherine HaylesReality Mining, RFIDs and Real Fears about Infinite DataThe more interesting and at the end maybe more important challenge is how to represent the personal subjective experience of a person living in a data society. If daily interaction with volumes of data and numerous messages is part of our new “data-subjectivity,” how can we represent this experience in new ways?Lev Manovich , The Anti-Sublime Ideal in Data ArtApproaching Statistics 13•In each case, I rediscovered my love for statistics and the important social-political-scientific-technical roles that our discipline plays as the acknowledged science of data•It’s almost a cliche at this point to say that the success of statistics as a discipline depends on our ability to compute; the corollary, however, being that our practice should make more explicit our dependence on information technologies and emphasize the need to continually track new programming languages and paradigms, new database technologies, new data formats•In terms of teaching, perhaps this sums things up...Today, software and hardware together provide far more powerful factories than most statisticians realize, factories that many of today's most able young people find exciting and worth learning about on their own. Their interest can help us greatly, if statistics starts to make much more nearly adequate use of the computer. However, if we fail to expand our uses, their interest in computers can cost us many of our best recruits, and set us back many years.Today, software and hardware together provide far more powerful factories than most statisticians realize, factories that many of today's most able young people find exciting and worth learning about on their own. Their interest can help us greatly, if statistics starts to make much more nearly adequate use of the computer. However, if we fail to expand our uses, their interest in computers can cost us many of our best recruits, and set us back many years. John W. Tukey, The Technical Tools of Statistics, 1964Approaching Statistics 13•And with all that as background, I cracked open my copy of Samuels and Witmer, the book that I had helped to select a few years back•Let’s have a look...A little retro •Fisher’s Statistical Methods for Research Workers was first published in 1925 and the table of contents is remarkably similar to the text I had been using for the previous two years of Statistics 13•To be fair, my scan of SMRW was from the most recent edition which appeared in 1973; although that’s still over 30 years old, perhaps there’s nothing very different about statistics since then•Let’s look back - first at the state of computing in 1973...1972 1973The first pocket calculator hits the market; Texas instruments and HP will follow suitThe ARPAnet connects 40 sites, the Internet consists of 25 computersXerox builds a personal computer (mouse, ethernet, GUI)but it’s too expensive for the general public1972-19731972 1973Dennis Richie at Bell Labs creates The Universal Product Code (UPC) is developedThe first working laser printer is produced1972-19731972 1973PONG is born!A little retro •Ah, PONG •Anyway, in the early 1970s we find the beginnings of many of the computing ideas we now take for granted; what about statistics and statistical computing in particular?•Let’s have a look...“Up to that time, the Annals had published not only papers in mathematical statistics,but also had been one of the main outlets for papers in probability theory. Now the editor, Ingram Olkin, felt that the theory of probability had developed into a subject that deserved its own journal. He persuaded the IMS to create a new journal, the Annals of Probability, and at the same time to broaden the
View Full Document
Unlocking...