Mildly fascinated since I'm mixed race and from Yorkshire (although family full of medics rather than dentists) just how many mixed race dentists there are in Yorkshire. My gut instinct is not many and also not easy to find out.
Tangential to the spirit of your post, "Type 3 Error" reminds me of Gelman & Carlin's "Type M Error" (conditioned on being "statistically significant, a result exaggerates the effect size) and even worse "Type S Error" (conditioned on being statistically significant, a result has the opposite sign of the true effect).
These *sound* like social errors, and ultimately they are (because it can be expensive to run a study with sufficient power, and because publication is biased toward sensational results.) But proximately, they are just the impersonal mathematical selection effect of the significance test on underpowered studies.
I have considerable experience with two Type 3 Errors.
1. The Rock Game: Bring me a rock. That's the wrong rock.
2. The Number Game: I'm thinking of a number. No, that's not it.
They are much the same. You have insufficient information to be successful. Either the boss knows what they want and they won't tell you, or they don't know what they want and can't.
Noting that in public administration at least, the Judge Over Your Shoulder ought to be a barrier to Accountability Sinks because of the principle that you mustn't fetter your discretion (see the Civil Service guide).
I need to do a full post replying to that, I think, because my view of that principle (and equivalent in financial regulation that I'm more familiar with) is that it's quite a bad and slippery piece of public law which is often counterproductive in practice; it puts the civil servant on a notice that they are the ones carrying the can if anyone does, and so incentivizes them to create systems and situations in which nobody carries the can
" it puts the civil servant on a notice that they are the ones carrying the can if anyone does, and so incentivizes them to create systems and situations in which nobody carries the can"
Actually applies to all middle managers, government bureaucracies just got there thousands of years earlier and so get the blame, a lot of current anti-commons inefficiencies in info-tech (some are addressed by ISO conventions but propriety systems will ignore them as they chase the golden goose into the greener fields of wall gardens) and data disasters are the result of middle-management making-do (satisficing to save their arses really). Within the Microsoft ecosystem there are a lot of bad design choices which later stuff has to be backwards compatible with, causing user friction, (its a feature not a bug) and of course it is even worse if one has to crosswalk stuff across the various OSs and platforms. (Very few of these are government built). At a granular level these decisions are paid for by the emotional labour at the data entry level, and if these workers point out an issue, let alone a solution they will be told, you are not paid to think. I fthis is outsourced there is no way the inefficiencies can be reported on. Microsoft, Apple, google are too big to even begin to find out about this stuff.
It will have to be addressed if productivity is to be increased. Middle-management decisions lead to anti-commons tragedies, there are efficiencies to be gained there.
"Moral crumple zone" is from this excellent paper by Madeleine Elish https://estsjournal.org/index.php/ests/article/view/260
Mildly fascinated since I'm mixed race and from Yorkshire (although family full of medics rather than dentists) just how many mixed race dentists there are in Yorkshire. My gut instinct is not many and also not easy to find out.
Tangential to the spirit of your post, "Type 3 Error" reminds me of Gelman & Carlin's "Type M Error" (conditioned on being "statistically significant, a result exaggerates the effect size) and even worse "Type S Error" (conditioned on being statistically significant, a result has the opposite sign of the true effect).
These *sound* like social errors, and ultimately they are (because it can be expensive to run a study with sufficient power, and because publication is biased toward sensational results.) But proximately, they are just the impersonal mathematical selection effect of the significance test on underpowered studies.
I have considerable experience with two Type 3 Errors.
1. The Rock Game: Bring me a rock. That's the wrong rock.
2. The Number Game: I'm thinking of a number. No, that's not it.
They are much the same. You have insufficient information to be successful. Either the boss knows what they want and they won't tell you, or they don't know what they want and can't.
Noting that in public administration at least, the Judge Over Your Shoulder ought to be a barrier to Accountability Sinks because of the principle that you mustn't fetter your discretion (see the Civil Service guide).
I need to do a full post replying to that, I think, because my view of that principle (and equivalent in financial regulation that I'm more familiar with) is that it's quite a bad and slippery piece of public law which is often counterproductive in practice; it puts the civil servant on a notice that they are the ones carrying the can if anyone does, and so incentivizes them to create systems and situations in which nobody carries the can
Interesting! I look forward to it.
" it puts the civil servant on a notice that they are the ones carrying the can if anyone does, and so incentivizes them to create systems and situations in which nobody carries the can"
Actually applies to all middle managers, government bureaucracies just got there thousands of years earlier and so get the blame, a lot of current anti-commons inefficiencies in info-tech (some are addressed by ISO conventions but propriety systems will ignore them as they chase the golden goose into the greener fields of wall gardens) and data disasters are the result of middle-management making-do (satisficing to save their arses really). Within the Microsoft ecosystem there are a lot of bad design choices which later stuff has to be backwards compatible with, causing user friction, (its a feature not a bug) and of course it is even worse if one has to crosswalk stuff across the various OSs and platforms. (Very few of these are government built). At a granular level these decisions are paid for by the emotional labour at the data entry level, and if these workers point out an issue, let alone a solution they will be told, you are not paid to think. I fthis is outsourced there is no way the inefficiencies can be reported on. Microsoft, Apple, google are too big to even begin to find out about this stuff.
It will have to be addressed if productivity is to be increased. Middle-management decisions lead to anti-commons tragedies, there are efficiencies to be gained there.
When I am feeling particularly jaded, I tend to think "Bayesian Thinking" is a synonym of "Quantamental"
I always mentally pronounce it "Bajan Reasoning" in a Barbados accent.
Reminded of "you'll have more data to ignore when you make your decisions based on company politics" https://x.com/StopBusinessBS/status/1245978714892120071