Open Thread: What To Chat About Next?

It is once again Travel Day—or was: I’m writing this in state of advanced jet lag. So, in lieu of thought I’ve written this little introduction to nothing, hoping you will fill in the blanks. Especially about what should we chat about next.

Suggestion #1: Dammit. Where are all the rocket cars we’ve been promised!

Suggestion #2: How to override jitters from plane nearly shaking itself to pieces half way over the Pacific Ocean, thousands of miles from nowhere.

Suggestion #3: Global warming is ever popular. But exceedingly dull. We’ve seen a large number of unanswered criticisms advanced, and what fun is it to repeat them endlessly? We still haven’t seen any climate produce strong skill for yearly temperatures. And we have no clear idea, but plenty of statistical noise, about what might happen if it indeed does warm. It’s all still speculation. Dimly viewed, too.

Suggestion #4: Discuss this quote by Jacques Barzun (thanks to Rafe Champion). “If science students leave college thinking, as they usually do, that science offers a full, accurate, and literal description of man and Nature; if they think theories spring from facts and that scientific authority at any time is infallible, and if they think that science steadily and automatically makes for a better world — then they have wasted their time in the science lecture room and they are a plain menace to the society they live in”.

Suggestion #5, etc.: Up to you.

Comments

Open Thread: What To Chat About Next? — 35 Comments

  1. Suggestions:

    In the statistical realm, how about commentary on propensity scoring?

    In the systems realm, Constructal Theory (http://www.constructal.org/)

    As for #4, undergraduates in science curricula ought to hang around some graduate students before they leave school (as I did) to get an inside view. That ought to disabuse any but the dullest of an illusions about the pristine nature of science.

  2. Dr. Briggs,

    I have always been interested in the ethics and responsibilities of engineers, especially those that design public structures. What is their responsibility to you and me, and society as a whole? What about esthetics, comfort, safety, and sympathy for social norms and religious beliefs (outdoor open urinals for example)? What about the engineer’s role in social change? What should an engineering education curricula look like – should it be appended with additional philosophy and humanities coursework?

    After all, engineers design our living and working spaces and, through their designs, control how secure we are (seismic safety as an example), how we move through spaces, how close we must be to others, how much or little light we have, what temperature we must cope with, and even where and how (you know there are differences as to the how, having traveled in the East) we use the toilet.

    Just a thought for a future commentary.

    Bruce

  3. Bruce,

    It seems to me like you are looking at the split between the archetecht and the engineer.

    The architect is supposed to adress the design, use of spase and asthetics. The engineer evaluates whether such a structure is stable and safe. The contractor has to figure out how to build it.

    There are way too many examples of architects who lose sight of the function of their buildings and go overboard with there artistic / evironmental statement.

  4. Another possibility is to ponder #4 slightly deeper in an attempt to quantify the “religious” aspects involved in scientific reliance on a course of study bringing adherents to a complete[?] understanding of the meaning of life.

    What do they use in place of ‘the Word”?; or who/what is the highest authority?; what do they substitute for worship?; are there other rites or rituals?; and so forth.

    Lots of meat there.

  5. Dr. Briggs,

    I enjoy your writing style – your topics are… well… the prerogative of your blog, but seeing you asked for an opinion, my suggestion is to continue to lead to your strong suit of statistics. Alternatively if you could pass on tangents relating to Roman Catholicism, why this country should sell arms to Taiwan or generally on items that at some point boil down to individual beliefs or matters of faith. They are interesting, but on a personal level and so hard to distill into universal truths.

    I agree climate debates can get wearisome, but it is an important topic. If you are looking for a “cross to bear” continue to make this blog a place to read and discuss the poor use of statistics in the politicization of science. It should remain a target rich environment for quite a while.

    And I also enjoy your observations on academia.

    So that would be votes for #3 and #4. And I am sure you could work in #1 too.

    (#2 is best addressed by “Air France”, “pitot tubes” and “Airbus”).

  6. Doug M

    I am not all that sure there is a comfortable split, as engineers who design bridges clearly maintain responsibility for many of the elements you assign to architects. Although modern firms employ both engineers and architects, it is often the engineer who must accept responsibility for the final outcome of the design. In Fritz Leonhardt’s book, Brücken / Bridges Ästhetik und Gestaltung / Aesthetics and Design, he accepts the responsibilities of the engineer regarding the design and esthetics of the structures they design and includes an entire chapter devoted just to these elements.

    However, I do not believe the point will be lost if we include architects in the discussion and hold them accountable, as you suggest, for their many design indiscretions.

    Best wishes for this season that can’t be named.

  7. Taking #1 and broadening–the art and science of prediction. Climate, is, right now, saturated, but yeah, flying cars, the “Singularity” and Strong AI, gene mapping and disease…I have read broadly in the science fiction of the 1930-60 era and it’s remarkable how closely certain things in those stories came to pass and how some, we’re still waiting on.

    Of course these were mostly presented as fiction and not scientific prediction. But the paid “futurists” have had the same sort of record: Iffy. “Statistical methods and modeling in relation to predicting technological and societal developmemnt.” You know, simple posts you can dash off in a couple of minutes. :)

  8. During a trip to Paris in 1968 I encountered outdoor urinals. There was a modesty barrier that covered you from about the knees to mid chest. But you were in fact standing there with your junk in your hand on Champs elysees nodding at passerbys within a couple of feet from you as you emptied your bladder. Oh! Did I mention the urine drained into the gutter in plain sight of everyone on the sidewalk?

  9. #4 is good.

    The “Contingency, Causality, Determinism, And Free Will” topic is important, IMO, for the foundation of a statistician or scientist and there is more which needs to be said along those lines.

  10. I’d be worried if even Arts students came away from college with that view of science.

  11. A topic big enough for 100 posts and a host of ferocious comments – sixty years of systemic statistical malpractice involving what most people believe are “scientific facts” about tobacco.

    Forget the drivel about secondhand and thirdhand smoke. What about putting a devil’s toothpick to your lips and inhaling (as I do many times a day)?

    Smoking is said to cause over a dozen types of cancer, and emphysema and bronchitis, plus heart disease and stroke, and those are just some of the fatal diseases. Smoking is also said to cause blindness (macular degeneration), hearing loss, erectile dysfunction, multiple sclerosis, psoriasis, wrinkled skin, snoring, and – my own favorite – “apple-shaped weight gain.” (There are easily 100 diseases linked to smoking – cigarette smoking. One of the embarrassments of the “smoking causes” theory is that the statistical association between tobacco and disease generally doesn’t apply to cigars and pipes.)

    The case against smoking is almost entirely statistical. Despite decades of trying, other than mouse painting experiments (“…the wrong material, in the wrong form, in the wrong concentration to the wrong tissue of the wrong animal…” R.C. Hockett, 1968), attempts to create disease in animals by exposure to smoke have been a bust. From the start (1950, the first studies on lung cancer) the theory assumed that chemicals in tobacco caused disease; sixty years (and 100 “smoking related” diseases) later, no one has identified any chemical that causes even one of the smoking related diseases. (Reliably, as in viruses cause influenza – and by the way, influenza is a smoking related disease.)

    Even before the first Surgeon General’s report (there are 30 now, some 1,000 pages long), every point alleged to establish smoking as the cause of any disease was contested. The first SG report was brilliant public relations. Create a 400 page, dreadfully written, book. Of the tiny number of people who actually have the tome in hand, most will just skim the conclusions and accept them as “facts.” The miniscule few that check the data and question its rational will be overwhelmed by massive “public health” and “charity” propaganda campaigns.

    But let’s get to the math. The link below is to a critique of the SG’s statistics by P. R. J. Burch. Judge for yourself if the points are valid.

    Burch: http://legacy.library.ucsf.edu/tid/hms61f00 He’s talking about the 1982 report which uses the same “rational” as 1964. (SG reports can be downloaded at the Center for Disease Control http://www.surgeongeneral.gov/library/reports/index.html.) As is typical, Brownlee’s 1965 paper (cited by Burch) is still – 17 years later – ignored by the SG.

    For the other side, Lilienfeld on Burch: http://legacy.library.ucsf.edu/tid/lyh05c00

    Burch’s response: http://legacy.library.ucsf.edu/tid/gms61f00

    If you find Burch persuasive, try Brownlee: http://legacy.library.ucsf.edu/tid/zgb2aa00 T. Sterling (oft paid by the tobacco industry) is also worth a read http://legacy.library.ucsf.edu/tid/jgo56b00 And, what the hell, despite those damn p > .05 values, R. Fisher http://legacy.library.ucsf.edu/tid/zrd91f00

    Mike F.

    P.S.: if you look at actual deaths attributed to smoking (rather than relative risk) lung cancer is dwarfed by heart disease. (The latter, of course, is caused by cheeseburgers and French fries.)

  12. In an earlier post discussing models you mentioned the problem of ‘probability leakage’ and stated that “my experience shows it’s usually large and not-ignorable.”

    I’m not sure if the topic would cover a full post but a discussion and some examples would be wonderful.

    A possible second topic:

    Many people use the words ‘chance’ and ‘random’ as if they are synonyms. There are, however, subtle differences and I believe I’m correct in saying that mathematicians mean something quite different than the ordinary language use of ‘chance’. I’m guessing but I think someone could make the claim that most people’s use of the word ‘chance’ is not at all as clear as the speaker thinks it is.

    So, a post clearing up confusion on this topic woud also be appreciated.

    In case anyone is counting votes I’d also read, with enthusiasm, a post on the issue of ‘tobacco statistics.’

  13. Bruce Foutch,
    Happily in retirement after a 35 year career attempting architecture with projects in 14 states, and one overseas, I would defy you to identify any of my “design indiscretions.” This is not to suggest that they aren’t there, but most of mine, at least the ones of which I’m aware, would not be apparent to someone outside the profession.

    A perusal of The Building Code of the City of New York, or similar volumes from other cities and states, along with the shelf full of the technical references cited would give you a working understanding of the standards to which today’s construction is designed. One can no longer do anything one wants, no matter what you might think from looking at some buildings.

    Recent history, particularly with the increased understanding of the effects of high-winds on buildings gained following Hurricane Andrew in 1992, suggests that not all transitory building impacts are well understood before they happen somewhere. I can assure you that the aftermath includes the development of design standards addressing prevention of failures already encountered.

    The drawings required for construction permitting, and for bidding, are sufficiently complete that little latitude is allowed for contractor invention. The idea that architects concoct artistic designs which the engineers “make work” and the contractors figure out how to build is a canard which might have had some validity at some point in time, but not now.

    In a good design-bid project, where the drawings will be completed and then bid to by a number of contractors, the work evolves through a collaboration between the architects and engineers. the team leader could be an architect or engineer depending on the purpose of the project. On a design-build project, a contractor will be added to the team, usually with an eye to benefiting by the contractor’s greater familiarity with where the costs come from in construction.

    Although an architect, I would like to praise highly the work of the engineers who design bridges, particularly some of the recent cable-stayed bridges. I understand, possibly incorrectly, that no architects are involved in these works and yet they have an elegance and simplicity which is seldom seen, or possible, in building design.

    I suppose there are some projects out there which stray too far into fantasy land, but for some reason, I’m not offended by them. Frank Gehry’s building at MIT comes to mind.

    I would like to add, not knowing whether this condition is unique to building design, that the most innovative things that ever happened in many of my projects were never obvious when the job was completed. For example, I was design manager on a rail-transit station to be fit into existing track between existing stations. The geometry of the track, nearness of bridges, presence of a LARGE water-main, made the construction of this station almost impossible without major unanticipated extra expense. It’s out there, and no-one not involved with the project would ever realize that fitting it to the site was a matter of fewer than 4 inches laterally and maybe 6 or 8 vertically.

    That the hard-part doesn’t show when the project is completed clouds my memories of my career.

    How about the rest of you? Does any of this make sense?

  14. Mike F,
    As I have pointed out before in comments, the National Cancer Institute plainly states the cause of cancer is unknown. Also the SG 1967 report on smoking and health clearly shows that moderate smokers have lower disease and mortality rates than nonsmokers.
    The 1967 report is not on the SG website because it’s politically incorrect. It’s been disappeared. You can find a copy here. http://members.iinet.com.au/~ray/
    Go here, next to last para. http://training.seer.cancer.gov/disease/war/

  15. I’d like to see more discussion around model testing. Specifically for “self feeding” models; that is to say a model that uses its output for time T+1 as parameters for predicting T+2…N.

    I’m interested in this because these sorts of models can be tested at various resolutions (or octaves, depending on your field) and I’d like to learn more about various testing methods.

  16. How about an analysis of head & neck injuries, and deaths, associated with the use/non-use of helmets in conjunction with motorcycle use?

    Undoubtedly, many deaths have been prevented…but…at the cost of many debilitating/crippling neck injuiries that render the value of the ‘quality of life’ vs. ‘lives saved’ trade-off/outcome debatable.

  17. j ferguson

    Glad I raised your ire enough to respond and give me a chance to apologize. I did not intend to denigrate your profession with my last comment. Perhaps it came off too flippant and broad reaching. I meant it to apply to aesthetic considerations and not much else. However, I think, because of my closing comment, you have perhaps missed my point. I was not so much speaking about the technical excellence of the projects, but to their esthetics and social consequences. I am most familiar with building codes and the many layers of other standards and codes that must be adhered to and I believe, in general, what you say is true about this part of your field. However, I have walked in many cities around the world and esthetic indiscretions and socially inappropriate structures seem easy enough to see, even to me.

    I am also quite sad to agree with what you say about how the real innovations in your field, especially some of the incredible construction tools and techniques, are not always obvious to the layperson; unlike many bridges and even some buildings where the innovations (or lack thereof) stand naked of ornament and stark before our eyes. In this regard, I am happy to see that some books and even documentaries on TV giving us a glimpse at some of these hidden design gems.

    A last comment regarding architects and bridges. The Golden Gate Bridge out here on the left coast would not be regarded as one of the world’s most beautiful bridges had it not been for the wonderful art deco design elements and international orange color furnished by the architect, Irving F. Morrow – a graduate of Berkeley and student of Ecole des Beaux-Arts. Working with the true engineering genius behind the bridge, Charles Alton Ellis (who was by the way and further supporting my original post, a Greek scholar and mathematician), together created a piece of steel sculpture that still holds viewers spellbound and shows just how important a gifted architect can be to enhancing the aesthetic appeal and social relevance of any project.

  18. Bruce, It’s good to see you do know your way around this stuff. I went to school just after the transition from FL Wright inspired (?) to Bauhaus, Walter Gropius, Institute of Design (Armour Institute-Chicago).

    This evolved during that time to doing a sensible plan and then figuring out how to put a skin on it.

    Naturally, we had nothing but contempt for anything done outside this narrow schematic armature.

    This contempt finally got to one of the older (he might have been 60) professors who offered to give us a tour of Saint Louis buildings designed in the 20s and 30s. We did it. He showed us nifty things that these guys had done, and alerted us to the problems they had solved which we were unaware of.

    This experience made me much slower to condemn buildings which might be considered undistinguished by most. At the same time, there are certainly plenty of bad buildings out there that don’t do what they were intended to do and are an affront to all humanity.

    One project I like out your way is the Bank of America Building Plaza – the red granite one. There is a crease in the plane of the plaza which is needed so that it can be tilted to meet the adjacent streets. The crease crosses the granite paving blocks, and I think one of the slickest things I’ve seen, except maybe for the column bases at the Lincoln Monument in DC.

    I can easily imagine that the plaza design was shown on the drawings and that after award of contract, the contractor asked the architects if they were serious about this since it would require this “crease” to be cut into every paving stone it passed through at likely incredible expense.

    I’m really glad they did it.

    Good for you for remembering the design of the Golden Gate Bridge. For some reason, there is a very large mock up of one of the towers at the Chicago Museum of Science and Industry.

    john

  19. j ferguson

    Sounds like you should be leading architectural tours for students of architecture. I think you hit upon a topic that is all too important – that of proper mentoring in the historical perspective of why you design the way you do. It is all too easy to buck tradition just because one thinks it is cool, rather than buck it with a full understanding of the whys and wherefores, and a sound understanding of the possible consequences. Also, as you eluded to, many things we think are new, have been done/tried before. Have you ever considered leading architectural walking tours? I would certainly sign up!

    Perhaps the reason a mock-up of the Golden Gate Bridge tower is in the Chicago Museum is because it illustrates perfectly how incredibly artistic architectural design elements can be part and parcel of the load bearing structure and not just some added-on whimsy. On the Golden Gate, even the rivets make an artistic statement…

    http://photoeverywhere.co.uk/west/usa/san_francisco/slides/golden_gate_bridge_details6334.htm

  20. Is heart disease “caused by cheeseburgers and French fries”? If it were then everyone in the West would be dead by age 20. Heart disease is caused by:
    1.Old age. If you live in a 3rd world country you probably won’t live to age 60, 70, 80 and die of “natural causes” (heart attack, stroke and cancer). If you live in the West and have good health care and good clean food and water you will avoid most of the illnesses that kill you at a young age and live long enough for your body to fail you, i.e. heart disease, cancer and strokes.

    2.Genetics. You need to pick your parents better. Genetics is the biggest factor in getting heart disease, (most) cancers and strokes at a young age (under 60). If you have good health and no genetic history of heart attacks at a young age eating cheese burgers and french fries won’t cause a heart attack. This is simply old wives tales and food biases. If your father and grandfather died of a massive heart attack in their 30′s and their cholesterol level was 300+ and yours is too then you have a problem that no amount of avoiding cheeseburgers and french fries will cure. Luckily modern medicine does have great options for you but the food faddists and other quacks are worse then useless to you. Following their advice will get you dead!

  21. Sorry, something didn’t come across right.
    Pick a disease. It is caused by the cures of all the other diseases.

  22. On #2

    N1000 If “ZZZZZZZZ” = true then End Program
    N1100 Ponder Probability Plane Plunges
    N1200 Order Irish Wiskey
    N1300 Drink Irish Wiskey
    N1400 Go to N1000

  23. Here is an example for a discussion regarding ethics of social responsibility for public design in engineering/architecture. I suppose this seemed like a good idea – on paper at least.

    The MVRDV Cloud high-rise apartments design for S. Korea (may have been more apt for N. Korea):

    http://www.property-magazine.eu/mvrdv-designs-two-residential-towers-in-korea-19004.html

    However, another interpretation:

    http://www.weeklystandard.com/blogs/proposed-s-korean-towers-resemble-exploding-world-trade-center_611802.html

    And, the apology:

    http://www.mvrdv.nl/#/news

    “…nor did we see the resemblance during the design process.” really???

    OK. I will stop here. Have ‘Occupied’ Dr. Briggs’ post enough today. He really shouldn’t offer the free license of an ‘Open Thread’. He really should know better… ;-)

  24. A bit late but…

    Suggestion 1: a discussion of the meaning and usefulness of the term “null hypothesis”.

    Reason: many people invoke the term in support of their metaphysical belief that if we are to be rational we must disbelieve that a phenomenon exists unless it has been proven to exist.

    Suggestion 2: a discussion of whether the simplistic, one-dimensional and abstract left-right view of politics harms political discourse; and whether a worthwhile improvement could be brought about by promoting a slightly more sophisticated view, with two or three dimensions. If so then what might/should those dimensions be?

    Reason: Although I’m generally of the broad, hand-wavy (and possibly utterly silly) view that the main distinction between intelligent and unintelligent people is that the former use words to express concepts, whereas the latter’s concepts are shaped by words I think we are all stuck with the left-right picture because it is ubiquitous. Anyone can produce some sort of multidimensional picture of political views – I think Eyesenck [sp?] did it decades ago – but it will be of little or no use in the pursuit of “better” political discourse.

  25. Argh! Point 1: I accidentally used a non-existent email address for the above – don’t know if it matters. Point 2: I haven’t read your statistical posts, lately, [too brain-mangled] so maybe you’ve covered suggestion 1 already. Apologies if that is the case.

  26. Ray 12/9 at 1:38 PM

    Thank you for your comment on my 12/9 4:17 AM remarks.

    I think I know the 1967 report you reference, and no matter how much I dislike the Surgeon General and other health nuts, the report was not suppressed (but it’s not easy to find unless you know exactly what to look for). If I’ve got the right document, it was created by the National Center for Health Statistics, so it’s not part of the SG canon. The title is “Cigarette Smoking and Health Characteristics” (publication series 10 # 34, May 1967). It has your Table A on page 8. http://www.cdc.gov/search.do?q=site%3Awww.cdc.gov%2Fnchs%2Fdata+filetype%3Apdf+“Cigarette+Smoking+and+Health+Characteristics”&submit1=Search It’s the first item on the list.

    For anyone interested, there’s a critique by Sterling: http://legacy.library.ucsf.edu/tid/jdn02a00 and a response by the author plus Sterling’s reply http://legacy.library.ucsf.edu/tid/gaj32f00

    Finding light smokers healthier than non smokers certainly clouds a dose response relationship. (I think this has been found before [Framingham?]; I wouldn’t trust a single study.) But DR has another serious problem. The cohort studies (e.g., British Doctors) show non smokers with the least amount of lung cancer, pack a day smokers with more, and two pack a day smokers with more still. Looks like DR. But all three groups develop cancer at roughly the same age. This can’t (?) be DR.

    Links:

    Passey: http://legacy.library.ucsf.edu/tid/jar02i00

    Doll / Pike: http://legacy.library.ucsf.edu/tid/drn24a99 Sad to say, I’m a mathematical illiterate, not qualified to evaluate this paper. I would point out, however, that table IV – smoking 3,000 cigarettes (150 packs!) a day – is weird.

    Lees: http://legacy.library.ucsf.edu/tid/nhe44e00 Added for his comment on Doll. Lees wrote excellent critiques of the smoking / lung cancer studies. His wave theory of disease is either a) very interesting, or b) Numerology. (A topic to chat about?) In the 1960’s Lees made predictions of future cancer incidence. I don’t know if these proved true, but if they did….

    The above age anomaly is also a problem for the earliest papers that linked smoking to lung cancer. An increase in smoking early in the 20th century was followed 20 years later by an increase in lung cancer. (National per capita tobacco consumption / mortality statistics.) Elementary books warn against “smoking causes” statements based on this type of data. I’m not aware of pre 1950 studies, but the post 1950 cohort studies (e.g., British Doctors) are a (not very good) sample of national data. The studies asked when people started to smoke, and death certificates set the age of death. If the national data is true cause / effect, cohort studies should show roughly the same mortality pattern. They don’t. I’m not going to try for exact numbers, but most smokers start in their teens; those who die of LC die 40 or 50 years later. Either the cohort samples are seriously wrong (they are, but…) or the national 20 year result is coincidence rather than cause / effect. At any rate, something is rotten in Denmark. (And don’t expect the Surgeon General’s reports to discuss this; they have always cherry picked data.)

    Next topic; again Ray’s comment.

    Re “the National Cancer Institute plainly states the cause of cancer is unknown,” I’d agree, except maybe for viruses like HPV and cervical cancer. (By the way, cervical cancer is a “smoking related” disease – but see http://velvetgloveironfist.blogspot.com/2010/10/nuns-prostitutes-witches-and-toads.html )

    Ray, don’t take the following personally; I suspect you might concur.

    Nobody knows what causes cancer – except, of course, lung cancer. EVERYONE knows that smoking causes lung cancer. Statistics prove it! If the NCI disagrees, perhaps they should tell the Surgeon General, The American Medical Association, The American Cancer Society, The American Lung Association, all children in grades K to 12, and quite possibly 99% of the adult population of the United States, if not the entire world, including (!) the billion or so who smoke. (A rant, and I’ll not write the endless paragraphs I’m tempted to write.)

    Next topic, GoneWithTheWind 12/9 at 3:26 PM comment.

    My cheeseburger remark was partially a joke. Partially, however, it was serious since there are Eminent Scientists who have Statistics “proving” that “unhealthy” diets cause X% of cancer deaths (pick a big number) and cost society (pick a bigger number) Y dollars each year.

    Old age and genetics are very important. Of all “food faddists and other quacks,” how many are Eminent Scientists?

    Next topic, addressed to Mr. Biggs.

    Lest you think I protest too much about tobacco….

    You’ve done posts on fine particulate matter. That’s “tar” in cigarette smoke. Actually it’s smoke condensate; calling it “tar” is propaganda. Back in the ‘70’s, public health nuts ranted about reducing tar to produce “safer” cigarettes. Over the next few decades, most smokers switched to filter tips. Then (in a 2001 National Cancer Institute report) a new generation of health nuts decided that filtered cigarettes were no less deadly than non-filtered. (Filters were “…used by the tobacco industry as a tool to undermine prevention and cessation efforts.” 2010 Surgeon General’s report, Chapter 2, page 20.) If true, this implies that reduced inhalation of fine particulate matter has no effect on health. But FPM in the air sounds like something that will frighten people; this can be used to get research dollars and to enact specious regulations (and sell filter tip cigarettes).

    Re Fisher and p > .05. Speaking again as a mathematical illiterate, I’d call this genius. There may be different (or better) ways to analyze data, but statistical significance says that x and y could be related, so let’s do more (different) research (like biology) and see if they really are. Fisher provided an “educated guess” about the relationship between x and y. It’s not his fault that Eminent Scientists do Studies published in Peer Reviewed Journals proclaiming p > .05 values (without other evidence) as Truth That None May Dare Question. (Would Fisher data dredge for p > .05? I think not. Read his papers [linked in my 12/9 4:17 AM comment]; there’s more logic than math.) Bravo Fisher! Boo Eminent Scientists!

    P > .05? How about lowering the confidence interval from 95 to 90? After cherry picking 11 of maybe 100 studies, finding the 11 not significant at 95 CI, the Environmental Protection Agency lowered CI to 90 to “prove” that secondhand smoke “causes” 3, 000 lung cancer deaths a year. http://www.fraserinstitute.org/publicationdisplay.aspx?id=15937&terms=%22passive+smoke%22 (This is a 120 page book; you have to download a PDF at the bottom of the page.)

    There are scientists who use statistics with integrity. (Enstrom comes to mind.) All too many, however (in my opinion), begin with the truth (their theory) and then play with the numbers until they get confirming results. I’d call this statistical / scientific malpractice. It is rampant, and nowhere more so than in most studies about smoking done over the past 60 years. (Global warmers are newcomers and amateurs.)

    Besides the 100 + diseases attributed to primary smoking, secondhand smoke has its own laundry list of maladies. Statistics prove this.

    Then there’s thirdhand smoke. Eminent Scientists are now doing Studies that will link this to disease – more likely to multiple diseases. All will be “proven” with Statistics.

    If even half of this is true, The Most Lethal Substance in the Universe is cigarette smoke. But not cigar or pipe smoke; they’re only linked to a few diseases. And by the way, after 60 years of “research,” nobody can link any specific chemical(s) in smoke to any specific disease. The “proof” is all statistical.

    The template (shoddy science) developed against tobacco (dating back to the 1950’s) is now used against alcohol, “junk” food, salt, almost any type of fat (whole milk!), sugar, soda (children drink it!), and just about anything a normal person might enjoy without serious harm.

    Was the case against tobacco (primary smoking causes lung cancer) ever really proven? (If this is bogus, how much other medical wisdom might be equally false?) Have you read any of the Surgeon General’s reports, the British Doctors, American Cancer Society, or other studies? Maybe, and maybe you even agree with them. If you haven’t, however, I submit that there are 60 years of statistical studies to debunk, and I’d say that’s lots to chat about.

  27. Re 12/21 6:55 AM comment in moderation.

    Mr. Briggs, I spelled your name incorrectly.

    Dumb!

    Please fix if you accept the comment.