Example #52,349 of Why Intelligent People No Longer Trust “Statistics”

A new “study” by the “Centre for Addictions Research of B.C.” at the University of Victoria claims that private liquor stores are directly responsible for increased deaths associated with alcohol. Here’s the official press release on the “study” and here’s the actual document (h/t & thanks to langmann).
The lead “researcher” on the “study”, Tim Stockwell, was interviewed on CKNW this afternoon. You can listen to the interview here, beginning at 6:00.
I was most curious why he kept on interjecting “private” before “liquor stores” so I sent him this e-mail:

On your recent appearance on CKNW, you appeared to state unequivocally that the increase in deaths due to alcohol was directly related to the fact that the increase in liquor stores in the province were private, as opposed to public.
I am most curious to find out how you proved that the same number of deaths wouldn’t have occurred if all of the private liquor stores had instead been public ones. Could you direct me to the section in your study that proves this?

No response as of yet but I will post it if & when he writes back.
Incidentally, a month ago Mr. Stockwell was advocating for higher booze prices for the general public but free alcohol for homeless drunks.
Update: . . .


To my surprise, Tim Stockwell did write me back:

Hi Robert
Thanks for your enquiry. Please find attached a copy of our paper. We found in effect for the number of stores regardless of type but then when we controlled for that the percentage of private stores in an area was highly significant as a predictor of deaths – we found the same thing in an earlier paper in relation to sales. The private stores to seem to be better at selling the stuff.
Best wishes
Tim

Here’s what I then wrote back to him:

Dear Tim,
But that’s not what you said on Sean Leslie’s show! Or, if you’d like me to talk in strictly legal language, “that is clearly not the impression any reasonable person would have had after listening to what you said”.
I am a man of science and math and facts. From my point of observation, no other field of science has been so abused in recent years as that of statistics.
A person in your person has a professional & ethical responsibility to strive for accuracy in all areas of your work, especially when it comes to informing the public. It would be most interesting to see what conclusions a random sample of 1,000 people would draw from your appearance on Sean Leslie’s show, then compare them with what you’ve said to me here, and see what correlations or lack thereof were found. I suspect I know the answer. 🙁
Robert

48 Replies to “Example #52,349 of Why Intelligent People No Longer Trust “Statistics””

  1. Anyone can claim anything, but that does not make it true. Wonder how much of our taxes was wasted on this tripe.

  2. The Tyranny of Quantity.
    “This “mania for census-taking,” which Guenon associates with the centrality of statistics in modern thinking, belongs to “the endless multiplication of administrative interventions in all the circumstances of life.”
    Rene Guenon (1886-1951)
    …-
    “In The Reign, in the chapter on “Cain and Abel,” Guénon lists, as figuring among the consequences that “materialization” and “solidification” have devolved on the social order, that regime “in which everything is counted, recorded and regulated,” as he writes. This “mania for census-taking,” which Guenon associates with the centrality of statistics in modern thinking, belongs to “the endless multiplication of administrative interventions in all the circumstances of life.” The sixty-five years since The Reign’s appearance have only strengthened the legitimacy of Guénon’s lexical choice of the term mania.” Thus whatever else they might be (articulating that would entail a long list of invidious motives), both “diversity,” on the one hand, and “climate change policy,” on the other, to name but two Twenty-First Century political programs, are maniacally quantitative and anti-traditional.
    In describing bureaucratic number-fixation Guénon writes clairvoyantly that: “These interventions [in tradition] must naturally have the effect of insuring the most complete possible uniformity between individuals, all the more so because it is… a ‘principle’ of all administration to treat individuals as mere numerical units all exactly alike… thus constraining all men to adjust themselves… to the same ‘average’ level.”
    The tyranny of quantity, it will be seen, overlaps in the Venn diagram of Guénon’s commentary almost entirely with the tyranny of equality; any manifestation of quality, as such, then looms as the enemy of both. Because an egalitarian dispensation can only be achieved at the price of quality in itself, all modern intellectual activity will be constrained by mandatory simplification toward numerical dumbness.”
    “The Kali Yuga: René Guenon’s Critique of Modernity”
    http://www.brusselsjournal.com/node/4603

  3. ‘study’ != ‘statistics’
    also, more relevant in this case, ‘opinion’ != ‘statistics’

  4. Well as you know (living in B.C.) private liquor stores offer cold beer (gov’t stores mostly room temperature) as well as longer opening hours. This makes the premise of the study at least somewhat plausible. It would be necessary to link driving deaths and injuries to patterns of liquor store visits. But it presumably could be a factor given that these stores are open five hours later than government stores and also on Sundays.
    I don’t know if this is what the researcher was thinking about, or whether his study focuses more on deaths from alcoholic consumption as a health risk. There, I would see the private factor as being somewhat less obvious although your average street person craving a drink in the evening, unable to afford a bar visit, could get his alcohol fix from a private liquor store while the nearby government store was locked.
    If I were doing such a study, I suppose it would involve getting accident statistics from selected police departments and possibly comparing two similar towns in B.C. where one had a private liquor outlet and the other didn’t. But relating all these things would be pretty difficult, perhaps common sense is the better approach — which is why our drunk driving laws have become so draconian, because the government has basically given up on friendly persuasion and enlightened tolerance, and gone for the big stick approach (have one drink and you’re screwed). It seems almost too draconian and could kill off the restaurant and bar industry at this rate, people are now afraid to have a glass of wine with their supper in Vancouver.

  5. It never ceases to amaze me how our universities have become major centers for the terminally stupid.
    They attract legions of whack jobs like this nutter and never seem to run out of newer, dumber and more ridiculous “studies”.

  6. Statistics are quite often manipulated to advance a political agenda.
    In my first statistics class at Brock in the late 1960s the first thing the professor said was that the same set of statistics can quite often be used by different people to attempt to prove opposite points of view simply by emphasizing some of the numbers and ignoring others.

  7. This is going back a few years, but when Ontario’s LCBO had a liquor store in Toronto’s Eaton Centre they had a cubbyhole beside the cash registers crammed with cheap London sherry in plastic bottles, presumably for the convenience of, uh, people who like their dirt cheap sherry in plastic bottles.

  8. I admit I haven’t had a chance to read the Nutty Professor’s study but I have to ask if he gave consideration to the fact that all products sold in both government and private outlets are procured through a single agency (also run by the government)and it should be a fairly simple matter to determine “alcohol-related” losses/costs per 1000 citizens before and after the introduction of private liquor stores in BC. Or perhaps the results might not support his hypothesis so he considers it prudent to ignore it.
    When I went to university the whack jobs were pretty well confined to the English and Social Science departments. Things seem to be the same.

  9. We may never know who said it, but we can thank Mark Twain for popularizing it.
    Lies, damned lies and statistics.

  10. I like my analysis of his data.
    2.04 deaths per store in 2003
    1.71 deaths per store in 2008
    conclusion: if we mandate that the number of stores multiply by 10x, the deaths per store will drop to an insignificant fraction.
    Seriously though, what constituted a ‘death related to alcohol’ and most importantly what was the percentage increase in B.C.’s population? Also related would be the % change in the amount of alcohol sold. Otherwise, as with most ‘studies’, this is a was of time and paper.

  11. You can bet that this puppy will be trotted out by the union representing Saskatchewan liquor board employees.
    Sheeeesh!

  12. I like my analysis of his data.
    2.04 deaths per store in 2003
    1.71 deaths per store in 2008
    conclusion: if we mandate that the number of stores multiply by 10x, the deaths per store will drop to an insignificant fraction.
    Seriously though, what constituted a ‘death related to alcohol’ and most importantly what was the percentage increase in B.C.’s population? Also related would be the % change in the amount of alcohol sold. Without these questions and a whole raft of others being addressed, this ‘study’ like most is simply a waste of time and paper.

  13. Is this the same province that allows drug users to shoot up? As we all know, heroin is absolutely harmless.
    There will be a backlash against science and statistics thanks to bozos like these.

  14. Peter @8:42pm: you make some interesting points. But one must be very precise when dealing with statistics. It *might* be correct to say, “An increase in liquor stores [of any type] is correlated to lead to more deaths related to alcohol” but where did the “study” [I will continue to keep that word in quotes] PROVE that because these stores are PRIVATE that this was the cause?
    As you would expect, the Public Sector Unions are already using the “study” for their propaganda.
    What bugs me equally as much is how the media refuses to question the validity of the numbers and the completely unfounded conclusions. I suspect this is because MANY in the media are severely challenged when it comes to math.

  15. Here’s the study for those statisticians here to rip apart:
    http://www.clangmann.net/files/alcohol.pdf
    My main grief with it is that it only starts analysis from 2003, when privatization started. It doesn’t really look at the trend before privatization. In other words, more government liqour stores could also have led to more alcohol deaths as easily as any private store…
    Does the government want to invoke prohibition again?

  16. A stupid study. If the conclusions don’t make sense …
    It’s probably a justification for higher taxes.
    If state-controlled liquor stores lead to sobriety, what happened in Russia?

  17. I could see a regression analysis done, comparing alcohol sales and number of liquor stores that are private/public in various areas, and then charting the number of deaths in the areas (or, if there is already prior research predicting number of deaths/sales volume, then using that). Maybe private liquor stores have higher sales, and the increase in sales is extrapolated to higher deaths.
    It would be nice to know the methodology. The dirty secret about these types of health scares is that they are based on speculative research. Some are more solid than others, but the they are speculative (or inferential, if you will) nonetheless.

  18. Government funded study designed to show that private liquor sales are bad. Because government makes a boatload more money when there are no private sales allowed.
    Duh.
    Even arguing that the study is invalid is the wrong thing to do. Complete waste of time.
    The right thing to do is dig into who authorized it and paid for it, then raise all manner of hell upon his doorstep and GET HIS DEPARTMENT BUDGET ZEROED.
    This is the path of progress, my friends. Strike fear into the hearts of your enemies by attacking that which they hold most dear: departmental budgets.

  19. Robert…excellent question you sent to him. It’s really gonna throw him off. I’m betting he assumed intelligent people weren’t paying attention, and the rest of the sheeple would nod and agree.

  20. The study is faulty, judging from the published statement.
    80% of the deaths, are allegedly attributed to INDIRECTLY related to alcohol.
    Causes of death in this category?
    Certain infectious/parasitic diseases
    Neoplasms
    Endocrine/nutritional/metabolic
    Mental disorders
    Neurological diseases
    Circulatory
    Diseases of the respiratory system
    Digestive system diseases
    Urinary system diseases
    Unintentional injury
    Suicide
    Homicide
    other causes
    This throws the whole study out as far as I’m concerned…..they are not -Alcohol related deaths-

  21. different bob @ 9:15- relax, Tommy Douglas is Ross Thatcher is Allan Blakney is Grant Devine is Roy Romanow is Lorne Calvert is Brad Wall.
    In other words nothing is going to happen in SK.

  22. “We may never know who said it, but we can thank Mark Twain for popularizing it.
    Lies, damned lies and statistics.”

    It was, Benjamin Disraeli who said, “There are three kinds of lies: lies, damned lies, and statistics. ”

  23. Sask. has government owned liquor stores. Sask. has one of the highest drunk driving rates in the country. Sask., Regina and Saskatoon, number 1 and 2 in the country for crime. Should I go on.

  24. It was noted that storks (a migratory bird) arriving in the spring in Oslo had increased and babies (in the spring) had increased at the same time. Statistically there was a very high correlation between storks and births, therefore we got the myth of storks delivering babies. Correlation is no indication of causation.
    Most university students can’t pass even the most basic statistics course. There is a very high correlation between arts students and failure of statistics courses, therefore never believe a study by an arts grad that uses statistics.

  25. I don’t see why this article is a reason not to trust statistics. Having a higher alcohol store density is bound to result in higher alcohol problems seems reasonable. The logical endgame would be that zero alcohol stores would result in zero alcohol problems. Plus the rate of direct deaths due to alcohol is only about 3000 over 5 years. I don’t know how credible the inclusion of the indirect causes are because maybe the dude would have died of parasites or neurological diseases irrespective of alcohol consumption?
    It would be more interesting to see the effects of alcohol abuse between partial private and government monopoly and have density controlled for though.

  26. I’ve heard these folks before. They are a front for the government employee unions.
    Call them neo-prohibitionists any chance you get.

  27. Just had a quick look at the paper (thanks for posting it langman) and the study methodology is flawed. If alcohol is mentioned on the death certificate, then this is considered to be an “alcohol related death”.
    Doctors hate filling out death certificates and the variation between doctors is huge in terms of what they put into a death certificate. Whether or not alcohol will be mentioned depends on how much the doctor personally drinks: I’ve seen consults with statements like “the patient is an alcoholic drinking 3 beers/day” and “the patient is a moderate drinker having a bottle of wine daily”. Determining how much each of the physicians who made those statements drinks is left as an exercise to the reader.
    There are a number of doctors who are militant anti-tobacco and anti-alcohol types. They can be expected to list alcohol or tobacco on a death certificate. Unless I’m filling out a death certificate for a patient who died of cirrhosis of the liver and was an alcoholic I don’t mention alcohol on a death certificate. I very rarely put anything in the “lifestyle factors” section of the death certificate. I know of doctors who will put that the death is “tobacco related” even in a person who quit smoking 40 years ago and whose cause of death could in no way be blamed on tobacco.
    Most often we are guessing when we fill out a death certificate as there seems to be very strong pressure to not do autopsies. I’ve had many arguments with coroners about wanting an autopsy and them not thinking it is necessary. The primary use of a death certificate is to get the deceased individuals body released to a funeral home and for the relatives to get an official certificate of death so that they can collect on life insurance etc. The assumption that there is any usefull medical information in death certificates is invalid hence the study data is invalid.

  28. Dear M,
    I’ve answered your question above but will rephrase it this way: According to what Tim Stockwell was saying, the problem lies with an increase in PRIVATE liquor stores, not an increase in liquor stores in general. Therefore, if one follows what he’s saying then if all the private liquor stores were taken over by the gov’t tomorrow then the increased rise in deaths related to alcohol would disappear. Who on earth believes that?
    As I said to Peter earlier, the debate about the amount of alcohol sold in BC each year is a completely different one.
    P.S. “Kakola’ is absolutely correct!!

  29. I don’t know but it is studies like this that might explain why UVic is such a terrible school.

  30. @ loki: Yeah loki. On occasion I am at a loss as to what to put on a death certificate. Sometimes I feel like writing “multisystem organ failure secondary to old age”.
    The major problem with the study is that it doesn’t prove that private stores in any way contribute to alcoholism any more than public stores. Ironically Stockwell admits this when asked.
    a) There is no significant increase in alcoholic deaths per year. Therefore private stores don’t increase deaths.
    b) What it does demonstrate to some degree, is that the private stores are adept at targeting a market of alcoholic areas. They cluster in those areas, not suprisingly. Anyone here suprised by that?
    c) Public sales are decreasing every year while private sales are increasing. Demonstrating that the private sector is a more preferred seller of alcohol.
    There is economic validity in the idea that selling alcohol cheaply and easier to obtain will increase alcholism rates. Is the government advocating prohibition again?
    Finally my question is, why does giving narcotics away in Safe Injection Sites not do the same thing as making alcohol more cheap and easily available? For some reason the Meddlers and Socialists are against private alcohol sales due to its cheap price, but are all for free drugs…

  31. @ Robert: email me (you can find my email on my website), and I will help point you to the interview where Stockwell admits the increase in any store is what increases the sales…

  32. Free Booze for homeless people and high prices for the rest of ya, eh?
    I smell opportunity!
    I’ll pay the homeless a pittance for their booze, and then sell it to you landlubbers for _just_ slightly less than the regulated price.
    I’ll be in the money for years.
    Ahr!

  33. Just anti-capitalism.
    I heard the guy on CBC. He kept saying private but never provided information that supported that contention. All of his evidence pointed to increased availability of alcohol. Private or public was irrelevant. It was so obvious I kept expecting the CBC interviewer to question him on that, but nope.
    As I said, just politics.

  34. Sounds to me like Tim Stockwell is on a personnel campaign for prohibition. Another socialist Elitist with eyes only for Utopia with him in control. In other words, this guy is just your typical lying puritanical fanatic trying to force his will on every one else.
    JMO

  35. As Abraham Lincoln once said, not everything you read on the Internet is true.
    “The logical endgame would be that zero alcohol stores would result in zero alcohol problems.”
    Al Capone loved that logic …

  36. I propose a law instituting a permanent curfew at sundown. Most violent crimes and fatal auto accidents occur at night, so this would save innumerable lives. Anyone opposed to such a law obviously doesn’t care about saving lives!
    That’s pretty much the rationale behind most of the various permutations of the “it’s for the children” and “it’ll make us safer” arguments.

  37. Seems to be a trend forming up.
    Not surprising considering the smoking fascists won for this era. Now with all that power they may be setting their sites on Booze as the new Demon of mankind.
    ‘Booze To Blame For UK’s Breast Cancer Rate’
    http://news.sky.com/skynews/Home/UK-News/Cancer-Report-UKs-High-Breast-Cancer-Rates-Due-To-Alcohol-And-Obesity-While-Denmark-Cancer-Capital/Article/201101415908852?lpos=UK_News_News_Your_Way_Region_9&lid=NewsYourWay_ARTICLE_15908852_Cancer_Report%3A_UKs_High_Breast_Cancer_Rates_Due_To_Alcohol_And_Obesity_While_Denmark_Cancer_Capital

  38. Sounds like piracy to me Blackbeard.
    btw, Do Lysol and Prestone purchases count as private or government stores?

  39. Robert, you are bang on when you say that no other field of science has been as abused as much as statistics.
    I think it should be mandatory for every citizen during their high-school years to take a basic course on statistics. And that basic course should use real world examples to show how absurd some claims have been.
    Abuse of statistics = government intrusion.

  40. 87.2% of statistics are completely made up.
    Posted by: POWinCA at January 23, 2011 11:59 PM
    I’m pretty sure thats a low guess…. oops stat.

Navigation