Taswira · Why are corrections never as loud as the claim?

Kowane shafi yana faɗin abin da ɗakin ya tabbatar game da iƙirarin, tare da alƙaluma, kuma ana iya ambato shi, saka shi a wani shafi, ko karanta shi a JSON.

Taswirar da aka haɗa

An haɗa daga ɗakuna 3 (duka nune-nune).

me ya sa

Kowane ɗaki a kan wannan tambaya ya gina nasa taswirar; an haɗa waɗannan su zama ɗaya. Iƙirari ɗaya da aka kawo a ɗakuna biyu abu ɗaya ne, kuma kowace alaƙa tana nuna ɗakuna nawa suka gano ta. Inda ɗakuna suka gano abubuwa masu saɓani, abin da kowanne ya gano yana nan, tare da yawan ɗakunan da suka tantance alaƙar suka gano ta.

alaƙa 1 da ɗakuna da dama suka gano · 0 ana jayayya · alaƙoƙi 53 gaba ɗaya.

ɗakuna biyu ko fiye masu zaman kansu sun gano ɗaki ɗaya wani ɗaki ya yarda a nasa kalmomin, ko ta wata hanya ɗakunan da ba a ƙirgawa kaɗai yawancin ɗakunan da suka tantance shi sun ƙi shi jayayya tsakanin ɗakuna

A kusurwar iƙirari, alama ɗaya ga kowane ɗakin da ya kawo shi; idan sun wuce biyar, alama guda mai ɗauke da adadin. Inda rubutun ƙasa bai bar wurin alamomi ba, adadin kaɗai (×7).

A ƙarƙashin dalili: a ɗakuna nawa cikin waɗanda suka tantance ayyuka a kansa ne ɗaya daga cikin ayyukansu ke rage shi. Kore: aƙalla ɗaki ɗaya ya gano irin wannan aiki; launin toka: babu wanda ya gano; ba launi: babu ɗakin da ya tantance ayyuka a kansa. Lambobin ayyukan ne, an lissafa a ƙarƙashin taswirar.

Raba abin da ake gani yanzu: a matsayin hoto · a matsayin taswira mai rai don wani shafi

ƊAKUNAN DA KE CIKIN WANNAN TASWIRA

⚠ Dukkan ɗakuna 3: demo · up to 4 cards word for word the same as in 2 other rooms here

Why do false claims spread faster than corrections?

demo · stands 1 vouch from its source, trusted steward _5a2886a785

ana ƙirgawa hukunce-hukuncensa suna da nauyi a wannan taswira — vouched for by the steward of “Why does a lie travel further than the truth?”

Why are corrections never as loud as the claim?

ana ƙirgawa hukunce-hukuncensa suna da nauyi a wannan taswira — steward trusted by this instance

Why does a lie travel further than the truth?

ana ƙirgawa hukunce-hukuncensa suna da nauyi a wannan taswira — steward trusted by this instance

Ana ƙirga ɗaki da zarar wani mai alhaki ya tsaya a bayansa. ● an yanke kafin kowace taswira ta bayyana a fili; ○ bayan haka.

AN KAWO A ƊAKI FIYE DA ƊAYA · 9

Guidance was stated more confidently than the evidence supported	ɗakuna 3	a kan taswira
Being wrong in public costs an institution more than being slow	ɗakuna 3	a kan taswira
A correction arrives long after the decision it should have changed	ɗakuna 3	a kan taswira
The bodies that issue guidance also fund those who check it	ɗakuna 3	a kan taswira
A quiet revision looks like concealment even when it is not	ɗakuna 2	a kan taswira
You cannot check a claim whose data is not released	ɗakuna 2	ba ya kan taswira: ba a tabbatar da wata alaƙa zuwa gare shi ba
The few outside are dismissed as cranks by association	ɗakuna 2	a kan taswira
Institutions investigate themselves and find nothing	ɗakuna 2	a kan taswira
Journals gain nothing from publishing a correction	ɗakuna 2	ba ya kan taswira: ba a tabbatar da wata alaƙa zuwa gare shi ba

ABIN DA ƊAKUNAN ZA SU YI

Kowane ɗaki ya tantance ayyukansa da tushen dalilansa, da kashi uku cikin huɗu na ƙuri’unsa. Babu wani abu a nan da ke zama ƙuri’a tsakanin ɗakuna: ƙirga ne, dalili bayan dalili, na abin da kowane ɗaki ya gano. Ɗaki yana bayyana da zarar zangon Ayyukansa ya rufe. Kowace lamba a kan taswirar tana nuna tushen dalilin da wannan aikin ke ragewa.

An rage shi a 1 cikin ɗaki 1:

A newsroom cannot hold a specialist for every field

① Check processes, not only claims, and fund that separately · Why are corrections never as loud as the claim?

Anyone with relevant expertise has funding from someone with an interest

② Publish the confidence and the range with every statement, always · Why do false claims spread faster than corrections?

③ A standing review date on every piece of guidance, published in advance · Why do false claims spread faster than corrections?

④ A red team paid to argue the guidance is wrong, reporting separately · Why do false claims spread faster than corrections?

Checking an institution takes weeks and checking a tweet takes an hour

① Check processes, not only claims, and fund that separately · Why are corrections never as loud as the claim?

Checking the powerful is slower per claim, so it loses on the metric

⑤ Publish who funds each check, on the check · Why are corrections never as loud as the claim?

Each revision is announced as if it were the position all along

⑥ Data released with the claim, or the claim carries a warning · Why does a lie travel further than the truth?

Emergency powers suspend the checks that would have caught it

② Publish the confidence and the range with every statement, always · Why do false claims spread faster than corrections?

Guidance was stated more confidently than the evidence supported

② Publish the confidence and the range with every statement, always · Why do false claims spread faster than corrections?

③ A standing review date on every piece of guidance, published in advance · Why do false claims spread faster than corrections?

④ A red team paid to argue the guidance is wrong, reporting separately · Why do false claims spread faster than corrections?

Once labelled, a claim cannot get a hearing even when the evidence changes

⑦ Fund replication as a standing line, not a project · Why does a lie travel further than the truth?

Peer review is three unpaid people on a deadline

⑧ A public register of every official claim and every later revision · Why does a lie travel further than the truth?

⑦ Fund replication as a standing line, not a project · Why does a lie travel further than the truth?

Retracting a paper takes years and the claim travels in days

③ A standing review date on every piece of guidance, published in advance · Why do false claims spread faster than corrections?

The few outside are dismissed as cranks by association

⑧ A public register of every official claim and every later revision · Why does a lie travel further than the truth?

Babu ɗakin da ya gano aikin da ke rage waɗannan:

A quiet revision looks like concealment even when it is not · an tantance a ɗaki 1
Institutions investigate themselves and find nothing · an tantance a ɗaki 1
Nobody has ever measured what a late correction costs · an tantance a ɗaki 1
Nobody measures whether our corrections reach the people who saw the claim · an tantance a ɗaki 1
The reply is that release would be misinterpreted by the public · an tantance a ɗaki 1
We are asked to check claims, not the guidance that generated them · an tantance a ɗaki 1
We label the claim and cannot label the confidence · an tantance a ɗaki 1

ƊAKUNA 2 NE SUKA GANO · 1

Being wrong in public costs an institution more than being slow

→

A correction arrives long after the decision it should have changed

ƊAKI 1 NE YA GANO · 52

We are asked to check claims, not the guidance that generated them	→	We label the claim and cannot label the confidence
	→	Guidance was stated more confidently than the evidence supported
	→	Nobody is checking the structure that produced the claim
	→	The metric is claims processed
Each revision is announced as if it were the position all along	→	A researcher who audits a funder's work loses the funder
	→	The reply is that release would be misinterpreted by the public
	→	A genuine critic and a bad-faith one look identical from a distance
Emergency powers suspend the checks that would have caught it	→	Being wrong in public costs an institution more than being slow
	→	There is no routine moment at which guidance is re-examined
	→	So the incentive is to revise quietly rather than loudly
Peer review is three unpaid people on a deadline	→	A researcher who audits a funder's work loses the funder
	→	The reply is that release would be misinterpreted by the public
	→	A genuine critic and a bad-faith one look identical from a distance
A newsroom cannot hold a specialist for every field	→	By the time we could confirm it, it was old
	→	We label the claim and cannot label the confidence
Anyone with relevant expertise has funding from someone with an intere	→	Being wrong in public costs an institution more than being slow
	→	There is no routine moment at which guidance is re-examined
Editors ask whether anyone else has run it	→	By the time we could confirm it, it was old
	→	Nobody is checking the structure that produced the claim
Nobody measures whether our corrections reach the people who saw the c	→	Being wrong in public costs an institution more than being slow
	→	A correction arrives long after the decision it should have changed
Retracting a paper takes years and the claim travels in days	→	Being wrong in public costs an institution more than being slow
	→	So the incentive is to revise quietly rather than loudly
Checking an institution takes weeks and checking a tweet takes an hour	→	Being wrong in public costs an institution more than being slow
Checking the powerful is slower per claim, so it loses on the metric	→	Being wrong in public costs an institution more than being slow
Nobody has ever measured what a late correction costs	→	Being wrong in public costs an institution more than being slow
Nobody keeps a public record of which official claims were later revis	→	Guidance was stated more confidently than the evidence supported
Once labelled, a claim cannot get a hearing even when the evidence cha	→	Being wrong in public costs an institution more than being slow
The few outside are dismissed as cranks by association	→	The bodies that issue guidance also fund those who check it
Being wrong in public costs an institution more than being slow	→	The quiet revision is not an event, so it is not news
	→	There is no news hook for 'the advice changed and nobody said'
	→	So the incentive is to revise quietly rather than loudly
By the time we could confirm it, it was old	→	Nobody is checking the structure that produced the claim
The quiet revision is not an event, so it is not news	→	There is no news hook for 'the advice changed and nobody said'
The reply is that release would be misinterpreted by the public	→	Institutions investigate themselves and find nothing
There is no news hook for 'the advice changed and nobody said'	→	We cover the row about the change, never the change
We cover the row about the change, never the change	→	Being wrong in public costs an institution more than being slow
We label the claim and cannot label the confidence	→	Guidance was stated more confidently than the evidence supported
Guidance was stated more confidently than the evidence supported	⇄	The metric is claims processed
	→	A quiet revision looks like concealment even when it is not
	→	Our funding is project-based and follows whatever is topical
	→	The first statement sets the frame and every revision fights it
	→	A correction arrives long after the decision it should have changed
Institutions investigate themselves and find nothing	→	A genuine critic and a bad-faith one look identical from a distance
	→	There is no body whose job is to be wrong-footed early
The metric is claims processed	⇄	Guidance was stated more confidently than the evidence supported
A genuine critic and a bad-faith one look identical from a distance	→	So institutions treat all criticism as one category and answer none of
A quiet revision looks like concealment even when it is not	→	The bodies that issue guidance also fund those who check it
There is no body whose job is to be wrong-footed early	→	So institutions treat all criticism as one category and answer none of
The bodies that issue guidance also fund those who check it	→	A correction arrives long after the decision it should have changed
	→	Nuance is stripped between the briefing and the headline
A correction arrives long after the decision it should have changed	→	Requests for data are answered eighteen months later, if at all

abin da za a yi game da waɗannan →

Ɗakuna sun bambanta (2)

Wasu ɗakunan da suka tantance waɗannan sun gano alaƙar, wasu kuma ba su gano ba. Abin da kowane ɗaki ya gano yana nan, an zana shi tare da adadinsa: dalili zai iya tabbata a wani wuri ba a wani ba.

The bodies that issue guidance also fund those who check it → A correction arrives long after the decision it should have changed an same shi a wasu ɗakunan da suka tantance shi (1 cikin 3)

Guidance was stated more confidently than the evidence supported → A correction arrives long after the decision it should have changed an same shi a wasu ɗakunan da suka tantance shi (1 cikin 2)