[FRIAM] anthropological observations
thompnickson2 at gmail.com
thompnickson2 at gmail.com
Fri Apr 17 17:58:58 EDT 2020
Hi, everybody,
This is Cranky-Nick, talking.
This conversation is reminding me of 538's constant reiteration that they actually accurately predicted Hillary's election in 2016. It's just that the electoral college didn't play along. If expert X tells me that the expected value of variable A is K, then, when it's all over and the data are in, and A did not equal K, I expect that expert to admit that what he expected did not happen. Only after that confession has been made, should a conversation begin about whether the expert’s prediction process was faulted or not. It seems to me that the shaded area is part of that second conversation.
But all of that is moot because I think IHME’s optimistic predictions have largely panned out. New York sent ventilators of Maryland yesterday!!!! I certainly would never have expected that, but I think it follows from IHME’s optimistic predictions. The were right to predict “K”, and I was wrong to doubt them.
Can I apologize for being cranky today. I realize I am riding for a fall. Thompson’s First Law: He falls hardest who falls from his highest horse.
But while I am being cranky, let me make a huge confession. I must be a determinist at heart because I instinctively believe that to say an event is random is to confess one’s ignorance, one’s laziness, or both. >>Pause for moaning, eye-rolling, and temple-pressing by all quantum-enthusiasts on the list.<< I am pretty sure that what I just said contradicts my Peirce-worship, but there it is! So when 538 says, our model was wonderful, we just got unlucky, The Leetle Voice In My Head, that voice that my behaviorism denies me, says back, “No, Son; you just didn’t work hard enough.”
Nick
Nicholas Thompson
Emeritus Professor of Ethology and Psychology
Clark University
ThompNickSon2 at gmail.com
https://wordpress.clarku.edu/nthompson/
-----Original Message-----
From: Friam <friam-bounces at redfish.com> On Behalf Of u?l? ?
Sent: Thursday, April 16, 2020 11:37 AM
To: FriAM <friam at redfish.com>
Subject: Re: [FRIAM] anthropological observations
I've quote-included what I think is the most important part of Dave's rant below your comment on consolidation of local outlets by right-leaning organizations. Dave's comment about "hard data" showing the model *extrapolations* being 20-50% higher than the numbers shown in local media.
To my eye, this seems similar to Nick's comments about the data at the IHME's site < <https://covid19.healthdata.org/united-states-of-america> https://covid19.healthdata.org/united-states-of-america>. That *huge* shaded area on each of those graphs seems to be totally ignored.
And this inability to grok uncertainty seems similar to our inability to grok exponential growth. Those of us who deal with such things (statistical power vs significance, confidence intervals vs error, standard deviation vs coefficient of variation, even median vs mean -- not implying *I* do so, but some of us do) MUST have an intuition about such things that prevents them from ignoring the uncertainty bands around the center/trend.
I also suppose the underlying cause of such judgement problems drives climate deniers' conflation of meteorology with climate science (or meteorologists vs climate scientists).
My own failure to grok large-scale, systemic iteration/evolution slapped me in the face in the form of an assertion I once made to a biologist friend, something like 20 years ago(?). I see/saw the democratization of music production (including the emergence of things like Acid Loops or LMMS < <http://lmms.io/> http://lmms.io/>, dub/rap/glitch/noise/circuit-bending/algorithmic music, as well as the emergence of the celebrity-not-a-musician-at-all musician/pop-star) as a sign that our music landscape was diversifying. The biologist retorted something like "Maybe *your* friends are making their own music. But that doesn't imply that more people are making their own music." And, over the years since he made that retort, I now *see* a drastic dearth of diversity all around me (though, again, in *my* circles, there's an increase in diversity ... it's just swamped by the homogeneity surrounding it).
Had Dave (or Nick) *mentioned* that -50% was well within the cone of uncertainty, along with condemning *both* the over-simplifying nature of local and national media, his assertion of "hard data" would have been credible. Hard data is always accompanied by a statement of how soft/hard it actually is. A claim of "hard data" (or optimism) with no accompanying claim of relative hardness is not credible. But what can we do? It's the nature of specialization and social trust that expert knowledge has to be simplified in order to percolate out.
On 4/14/20 3:48 PM, Steven A Smith wrote:
> 3. I have some experience (working in local Radio in the early 70's and investigative journalism in the late 70's) and basis to believe that Local Media is no less biased nor more given to reporting facts than the National Media. At *best*, a local bias (aligned with local ownership and/or local advertisers, real or aspirational) replaces the national bias. I believe bias is always nearly invisible to those who share the bias in place. At *worst* the local bias is in lock-step with the national bias which is often not just handed down from the affiliated network/syndication but in fact through a media conglomerate consolidation which has gobbled up a huge portion of the local print and broadcast media. This often comes without the change of ownership being made strongly evident to the consumers of that media. My personal bias/opinion is that the Right has done a bang-up job of gathering up local media around the country in the last decade or three to the purpose of
> subtly influencing public opinion, in a similar way to the way
> they have tried to hijack social media.
> <https://www.gsb.stanford.edu/insights/media-consolidation-means-less-l> https://www.gsb.stanford.edu/insights/media-consolidation-means-less-l
> ocal-news-more-right-wing-slant
> On 4/12/20 11:51 AM, Prof David West wrote:
>> 2) Models, projections and actual. In Europe I encountered almost none of the "the models predict and hence we are doomed unless ..." kind of articles that seem to dominate US mainstream media. Instead, "spreadsheet models" with data were published in tables by date, country, and raw number. European readers were left to make their own conclusions about how Netherlands data compared to Italian (for example) and make projections or draw conclusions as appropriate.
>>
>> In local newspapers in Nevada, Wisconsin, Minnesota, and Nebraska I
>> saw articles that compared projected numbers from the models touted
>> by CDC, Fauci, et.al. with actual local numbers. Local numbers varied
>> from national model projections by as much as -50% and never less
>> than -20%. (That is actual was dramatically lower than projected.)
>>
>> I saw and heard numerous editorial commentaries with regard the discrepancy between what the 'experts" were saying and what was locally observed and questioning why the variance. This leads immediately to questions about "hidden agendas" on the part of the Federal government and the "experts."
>>
>> CONCLUSION: A population that already mistrusts the Federal government and the"intelligentsia" is given one more reason, backed by hard data, for that mistrust. Also very clear — the population is NOT anti-science but IS very mistrustful of "authoritarian scientists" — those prone to saying "you wouldn't understand, but I do and you should trust me."
--
☣ uǝlƃ
.-. .- -. -.. --- -- -..-. -.. --- - ... -..-. .- -. -.. -..-. -.. .- ... .... . ...
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam unsubscribe <http://redfish.com/mailman/listinfo/friam_redfish.com> http://redfish.com/mailman/listinfo/friam_redfish.com
archives: <http://friam.471366.n2.nabble.com/> http://friam.471366.n2.nabble.com/
FRIAM-COMIC <http://friam-comic.blogspot.com/> http://friam-comic.blogspot.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20200417/58a186b6/attachment.html>
More information about the Friam
mailing list