The Evidence-Based Paradigm

shakahislop

Well-known member
In the policy world in my experience you can't ignore how important 'evidence' and an appeal to it is as a discursive tool for swinging a conversation in your direction and ultimately making an institution do what you want it to do
 

padraig (u.s.)

a monkey that will go ape
In the policy world in my experience you can't ignore how important 'evidence' and an appeal to it is as a discursive tool for swinging a conversation in your direction and ultimately making an institution do what you want it to do
this is the opticratics mentioned above

or if you didn't want to use a neologism you could just call it an appeal to authority

I didn't read the whole thread but I assume that's what the paper Sufi linked originally is referring to by "evidence-based"

not against using evidence at all (of course) but against a fallacious "evidence-based" paradigm
 

shakahislop

Well-known member
because economics only works well retrospectively tho? or is that the case for all of this ??
It's the case for a lot of important stuff I think. Trying to get good data on whether x will lead to y is generally very hard, especially to do at the scale necessary to be confident about the y. This is where our beliefs, ideologies are important, coz that's how a lot of policy decisions are being made in reality I think
 

sufi

lala
don't even get me started on long range weather forecasts!

( edit: 5 days at best! )
lick yr finger and hold it in the wind
In the policy world in my experience you can't ignore how important 'evidence' and an appeal to it is as a discursive tool for swinging a conversation in your direction and ultimately making an institution do what you want it to do
to the extent that unevidenced activity might as well not exist, and in the end gets its funding slashed and ceases existence = quantitative destiny
 

shakahislop

Well-known member
Also a reason why the personal backgrounds, experiences of the people who get to be in the room (ie have influence) on the decisions which are made are important. Coz the evidence tends to be very inconclusive even before you get to the problems of research methodology that I mentioned before (ie whether that evidence really reflects reality or not)
 

padraig (u.s.)

a monkey that will go ape
obviously the further you get from pure numbers the greater the epistemological issues become

there is a great, apropos bit in Neuromancer that has always stuck with me wherein Case (a hacker) asks a ROM construct of his dead mentor about the difference between it and a true AI and the construct is basically like "I act like a person with self-awareness but I'm not (essentially a David Chalmers style philosophical zombie), I'm just the totality of my memories; the AI does have self-awareness but you can't discern its motives because it's not human"
 

shakahislop

Well-known member
lick yr finger and hold it in the windto the extent that unevidenced activity might as well not exist, and in the end gets its funding slashed and ceases existence = quantitative destiny
it depends where I guess. I've not got much direct experience of it but from what I've seen the French government for example is very far from the evidence paradigm, much closer to the 'well we just think it's a good idea' one. UK government definitely has the thing you described but it's a New Labour thing and onwards really, it's temporary and based on a certain set of beliefs, it's not forever
 

sufi

lala
obviously the further you get from pure numbers the greater the epistemological issues become

there is a great, apropos bit in Neuromancer that has always stuck with me wherein Case (a hacker) asks a ROM construct of his dead mentor about the difference between it and a true AI and the construct is basically like "I act like a person with self-awareness but I'm not (essentially a David Chalmers style philosophical zombie), I'm just the totality of my memories; the AI does have self-awareness but you can't discern its motives because it's not human"
just like a hipster
 

padraig (u.s.)

a monkey that will go ape
arguably the craziest modern Internet cult of them all to me is ultra-rationalists

sometimes you get effective altruism

sometimes you get eugenics and Nick Land type bullshit

sometimes from the same people

the idea that you can rationalize every decision if you just try to hard enough is fucking crazy
 

shakahislop

Well-known member
obviously the further you get from pure numbers the greater the epistemological issues become

there is a great, apropos bit in Neuromancer that has always stuck with me wherein Case (a hacker) asks a ROM construct of his dead mentor about the difference between it and a true AI and the construct is basically like "I act like a person with self-awareness but I'm not (essentially a David Chalmers style philosophical zombie), I'm just the totality of my memories; the AI does have self-awareness but you can't discern its motives because it's not human"

Even pure numbers, if you are talking about survey research on things like attitudes, perceptions, etc, have all kinds of issues. Polling is the most high profile example. And that is one quite simple question. Not saying that all polling is shit. Just saying that even numbers are hard and not always accurate. Not all of them though. Tracking online behaviour, stuff like RFID inventories or more automated things seem to be pretty good. But that's a narrow field all in all
 

shakahislop

Well-known member
arguably the craziest modern Internet cult of them all to me is ultra-rationalists

sometimes you get effective altruism

sometimes you get eugenics and Nick Land type bullshit

sometimes from the same people

the idea that you can rationalize every decision if you just try to hard enough is fucking crazy
The rational alturist guys are the worst. Haven't heard from them in a while, I guess it was a fad
 

shakahislop

Well-known member
Well they're not the worst. They were just young people from fancy universities who had a ridiculous amount of self-belief and faith, and the skills and connections to promote thier thing at the time
 

sufi

lala
And I think this may figure integrally into the future evolution of democracy, namely the transition from the rule of law to the rule of code, perhaps in the form of smart contracts.

If we think we're already data-driven, we may have a rude awakening awaiting us.
we need to pay more attention to how AI is already manifesting in network effects and access to knowledge at a gigantic scale, and how those effects are already acute in some spaces and also quite absent in others
 

padraig (u.s.)

a monkey that will go ape
here's a random thing I found - first search result - about evidence-based paradigm in HR

first paragraph
Evidence-based decision making is a process for making the best decisions possible using the evidence available. It avoids decision making that is based on gut feeling, intuition, or instinct and instead relies on data and facts. When you need to make a decision in any setting—but especially in a business one—it's essential to do so based on facts and not feelings if you want the best possible outcomes.
I mean obviously, use what you believe to be the best possible data

but 1) don't pretend that facts exist in some separate, sterile environment closed off from feelings. it's always both.

2) you have to constantly test the quality of your data-gathering tools just as you should question your own assumptions - they're both epistemological question and both unresolvable. you'll know never for a 100% certainty that what you're doing is the best or correct decision - or even if such a thing exists.

I assume if you actually talked to an HR person or whatever they'd be like "well yeah, of course, obviously" but it's again the abstract of a "paradigm"

I mean there's an entire industry devoted to selling companies (and presumably govts as well) these products - ATS, HRIS, HCM, etc - and it's obviously in their interests to convince the people making decision to buy their tools to help them make decision

really "evidence-based" doesn't even come off like opticratics or an appeal to authority so much as marketing
 

padraig (u.s.)

a monkey that will go ape
Even pure numbers, if you are talking about survey research on things like attitudes, perceptions, etc, have all kinds of issues. Polling is the most high profile example. And that is one quite simple question. Not saying that all polling is shit. Just saying that even numbers are hard and not always accurate.
absolutely

I meant more like, if you're trying to design a bridge or something. there pure numbers are the way to go.

maybe I should have said: the further you get into human input to generate your numbers the less reliable those numbers become
 

sufi

lala
here's a random thing I found - first search result - about evidence-based paradigm in HR

first paragraph

I mean obviously, use what you believe to be the best possible data

but 1) don't pretend that facts exist in some separate, sterile environment closed off from feelings. it's always both.

2) you have to constantly test the quality of your data-gathering tools just as you should question your own assumptions - they're both epistemological question and both unresolvable. you'll know never for a 100% certainty that what you're doing is the best or correct decision - or even if such a thing exists.

I assume if you actually talked to an HR person or whatever they'd be like "well yeah, of course, obviously" but it's again the abstract of a "paradigm"

I mean there's an entire industry devoted to selling companies (and presumably govts as well) these products - ATS, HRIS, HCM, etc - and it's obviously in their interests to convince the people making decision to buy their tools to help them make decision

really "evidence-based" doesn't even come off like opticratics or an appeal to authority so much as marketing
a lot of hr is about reducing the human factors, and that seems easily to slip over into the inhumane
 

william_kent

Well-known member
HR is a bit passe

our "Human Resources" dept has been taken over, and, well, needs a rethink

and guess what, we no longer have "HR"

we're beyond that
 

sufi

lala
we need to pay more attention to how AI is already manifesting in network effects and access to knowledge at a gigantic scale, and how those effects are already acute in some spaces and also quite absent in others
also how data gathering becomes an end and a determining factor, the unlimited overheads and rampant unintended consequences of datafication

at a relatively local level mandatory admin costs making stuff unviable for example,
or refugee management bio-data getting hacked and leaked
 
Top