The Dunning-Kruger effect. What is it actually?

[tvtech]

Hi Jim

For me its like (don't do this..I showed you it wont help..)

Leave it alone. Try something else rather.

Regards,
tvtech

 

[misterT]

Once couple of students came to me and asked "can we replace these wires with bigger ones".. I was stunned... I asked: "what are you doing"... the answer was: "we are fitting this toaster into this wooden box.. is it ok to use thicker wires than the original ones"...They did not go far with that project.. I made sure of that.

Then there was this one group that wanted to make a portable welding machine. I like the fact that we let students play and they learn, but looking after them is sometimes scary.. I mean.. they have been through classess etc. but when they do projects, then they really learn. That is why finland is number 1 in education.. we fail and it is ok.. just learn from the failure.

 

[jpanhalt]

You are too nice. I would have waited for the blaze of glory.

On the second issue, letting a student fail is one thing, and if done properly is usually a good thing. Letting a student hurt themselves, and particularly hurt someone else, is something entirely different. I have never failed a student on the spot. In one case in my entire career, I reduced a student's grade one letter grade (A to B+), because the student let his/her temper endanger others. I would have reduced the grade more, if that had been in my power to do.

John

 

[misterT]

I have never failed a student on the spot. In one case in my entire career, I reduced a student's grade one letter grade (A to B+), because the student let his/her temper endanger others. I would have reduced the grade more, if that had been in my power to do.

Oh yes.. you are right. Just to clarify my previous story: The students with the "toaster in a wooden box".. I was not in the position to "fail them on the spot".. I told them to stop doing whatever they were about to do... and come up with a completely new plan for the project. I think they eventually did ok. Can't remember what the project was.

 

[MrAl]

Hi,

What i take from this is that we should all be aware of the fact that someone else can at least sometimes teach us something new. As long as we are aware that there is information out there that we dont know yet, we should be good to go. We cant know everything, and even that which we think we know may still be improved with the addition of more recent information.

Just for a quick example, how many people here already knew that the electrons that we think of when we think of current flow in a wire are not real electrons but are a fictitious particle we made up to simplify the calculations behind Ohm's Law and other theories?
Im sure some people already knew this, but it's interesting that it is something that most of us here use every day in one way or another yet we pay little attention to the details behind this.

 

[ericgibbs]

Just for a quick example, how many people here already knew that the electrons that we think of when we think of current flow in a wire are not real electrons but are a fictitious particle we made up to simplify the calculations behind Ohm's Law and other theories?

Afternoon Al,

You are joking of course.

Have you ever done any CRT beam current measurements or CRT paddle turning experiments using the accelerated mass of this fictitious particle.

Eric

http://www.r-type.org/articles/art-004e.htm

**broken link removed**

 

[tvtech]

Afternoon Al,

You are joking of course.

Have you ever done any CRT beam current measurements or CRT paddle turning experiments using the accelerated mass of this fictitious particle.

Eric

http://www.r-type.org/articles/art-004e.htm

**broken link removed**

Wow Eric. Great reading with the first link.

Many thanks,
tvtech

 

[MrAl]

Afternoon Al,

You are joking of course.

Have you ever done any CRT beam current measurements or CRT paddle turning experiments using the accelerated mass of this fictitious particle.

Eric

http://www.r-type.org/articles/art-004e.htm

**broken link removed**

Hi there Eric,

First, thanks for the links, they look very interesting and i think i'll have to look into those experiments a little closer.

But if you reread my post, you'll notice that i specified "in a wire". If the electron leaves the wire, it's a whole different story. I was stating it in this specific way because the fictitious electron is only fictitious while it is inside a solid. Once it leaves (somehow) it is considered a 'real' electron.
I was saying it this way also because most of the time we are measuring currents inside wires and things like resistors. That's where we use the theories that are basically riding on that non real particle rather than the real particle. But it is easy to imagine what is happening here.

We start with a real electron that lets say somehow enters a wire (a solid). It starts to interact with atoms that make up the wire. Some of those atoms have holes, some have electrons that are free to move about. But because there are so many atoms and electrons and holes and the laws of physics say that the electron is influenced by (almost) every one of those other electrons and holes, the exact calculation of what happens to that one electron that just entered the wire is so complicated that it would be impossible to calculate it's exact path even with the best super computers we know of today, all working on the same problem. That means we dont really know the current flow because we can not track each and every electron that enters the wire. We'd have to keep track of billions and billions of particles and their exact movement in three dimensions plus time.

So because the problem is so complicated, we invent a particle which we still call an electron, but we assign to it the property that it can flow through the wire freely without ANY interactions, but at a reduced speed. That's the electron which we think of when we use Ohm's Law and other simple principles of circuit analysis. If i am not mistaken, i think another way of saying this is that electron would be like a 'drift' electron, but we dont even have to think about 'drift' when we use the theories like Ohm's Law so it's even a little more basic than that.

 

[ericgibbs]

Morning Al,

I would argue that if we can measure the CRT electron beam current by metering the Cathode wire thats conducting the electron beam current to the Cathode of the CRT, that the current in the conductor is electron flow/drift.

I appreciate that electrons drift along a conductor is quite slow, IIRC a few cms/hr, but when we use Ohms law to calculate current we are really considering electron drift density.

The mass and charge of an electron are well known and by using a CRT, with the original paddle wheel experiments, the effect of the accelerated electron mass on the paddle and the fact that the electron beam within the CRT length can be deflected by using electric or magnetic fields convinces me that the current in the conductor is the electron flow/drift.

Eric

 

[MrAl]

Hi there Eric,

Oh yes i agree. What i am stating is not that the drift is not real, but that we think of the flow as a single electron and that translation makes this 'new' electron fictitious although still very very useful as you noted. This is a subtle point that we usually dont think about.
The current we 'measure' is the result of very unusual and complex interactions between a huge number of electrons, yet we only measure those which pass a certain 'point' in a given time. So we think of the flow as being a continuous flow of electrons that only move in one direction which is along the wire, and that's what we actually measure.

So let me see if i can come up with an analogy here

Say we pump water from one tank #1 to another tank #2, but the pump works in two directions so that if we reverse the DC voltage to the pump it can pump in reverse from tank #2 to tank #1. We also note the volume that passes a given point in the connecting pipe. Someone else is running the pump, and we are doing the measurement noting the volume that passes a given point in a given time.
Ok, so they turn the pump on, and we note that sometimes the flow goes to the right, and sometimes the flow goes to the left. When the flow goes to the right we call that positive, and when it goes to the left we call that negative. That makes sense because we want to know the flow from tank 1 to tank 2. So we might see 10 gallons pass to the right, and 1 gallon pass to the left, then 20 more to the right, then 2 more to the left. So the net flow from tank 1 to tank 2 is 9+18=27 gallons, and if that happened in 10 seconds that would mean we had a net flow of 2.7 gallons per second.

But now we look at what just happened. We had 'real' water passing from tank to tank, one way or the other, yet we dont consider every drop of water but only that net flow that passed that point. So the gallons we are accounting for is only part of the total flow because really 30 gallons had moved. So we call these new gallons 'fictitious' even though the 27 really did move from tank 1 to tank 2, and that would be possible if this was typical of what always happened when we pumped water (that is, always 27 out of 30 actually got moved from the first tank to the second tank). Thus, we make up the law that 2.7 gallons per second are always pumped, and in this way we are ignoring the 3 that moved the other way so we really are not counting 'real' gallons or monitoring 'real' flow.

This may not be the best example in the world but i hope it gets the point across. BTW this way of thinking is not something that i myself invented but is an accepted way of looking at particles.

So when we say we have 1 amp of current we are really counting the accumulation of only particles that have actually made it past a certain point and STAYED past that point, and they are also looked at as being not real when we make up a new formula using what we already know about them.
In other words, the current I=dq/dt only counts the electrons which actually made it past that given point (and stayed there), so it is acceptable to call them quasi particles rather than real particles because obviously dq/dt does not take into account ANY electrons moving diagonally or backwards. It's a statistical measurement and to simplify it we envision a new particle which only moves in one direction through the wire without any interference from other particles and we also call that an electron, but the proper term i think is quasi particle.

 

[ericgibbs]

hi Al,

As Sigmund Freud once said, 'sometimes a cigar is just a cigar'..

Paraphrasing: 'sometimes an electron is just an electron', it works for me.

I try to avoid using water analogies when explaining electronics, a wannabe can quickly get himself into deep water.

I think we both agree that all our scientific knowledge is based on assumptions, empirical results and approximations.

Eric

 

[jpanhalt]

@Mr Al

Can you provide a citation for your claim? This may not be it precisely, but is it reasonably close: https://www.electro-tech-online.com/custompdfs/2013/09/PB--nature-of-electricity.pdf

John

 

[killivolt]

Qed

I've been reading this?

I don't know if this would apply or not.

https://en.wikipedia.org/wiki/Quantum_electrodynamics

 

[MrAl]

Hi again,

John:
That looks like a different approach altogether where the flow is viewed as a single wave. I could be wrong because i read it rather quickly but i didnt expect to have to read several papers to get this point across because it is a much simpler concept than anything like that. But it could be related, because they appear to want to reduce billions and billions of waves into one wave that includes the effects of all the others. I could be wrong there though as i said.

Since Eric doesnt like water analogies and i dont blame him, i'll illustrate this very simple concept the way it really is. I will look for references too later on today.

First, imagine that we had a wire that was one electron in diameter. If we had 'current' flow, the electrons would move through that wire side by side say from left to right, and if we counted each electron as it passed a given point and tallied the result after one second and the charge on each electron was 1 unit each, we would see that we have:
I=dq/dt
units of charge passing from left to right per second. And as you can easily see, that's one of the definitions of current that we often refer to.

Now so far, we're in no trouble at all. If we have 10 real electrons passing a point x in 1 second then we have 10 units of this charge. As you can see, there are no problems yet. We count the electrons and we calculate the current. Simple.

Now lets say we have a wire with a diameter equivalent to 500 billion classical electron diameters, which is about the same diameter (roughly) of a number 10 gauge wire. Here we have electrons jumping up and down and left and right, with a certain number 'drifting' under the influence of a voltage to the right. The number that exit the end of the wire is again 10 per second. The question is then how do we apply:
I=dq/dt ?

The answer is that we can not apply that directly because there are billions and billions of electron interactions taking place, a truly unfathomable number. If however we could actually keep track of every single one, we would come out with the same result: 10 per second.

So instead of keeping track of every electron and/or constantly reminding ourselves that there are billions and billions moving that cause that 10 per second, we imagine a new particle that can travel through the wire without any bother from any other particles. Then we can apply directly:
I=dq/dt

and we get 10 per second again, but what happened was the ones that we do account for we imagine as flowing slower so that we can actually get that number again. So it's like we have many electrons flowing unimpeded through the wire, but they are moving so slow now that we still only see 10 per second pass a given point so we measure the same current.

So the point here is that if we used I=dq/dt on every single electron we'd have a lot of work to do. But if we imagine a similar particle that moves more slowly, we have just a very simple algebraic (or calculus) formula.

I believe this falls under the general category of the "many body problem". There are too many particles to consider so we imagine a quasi particle that can take the place of what we really want to know most of the time, the current level in this case.

I'll look for ref's.

 

[tvtech]

LOL....Dunning /Kruger to Electrons

Should have known this would happen

I don't mind. Keep discusssing away.

Regards,
tvtech

 

[MrAl]

Hi there tvtech,


He he, well that's the nice thing about a discussion like this, that we can discuss various things and bring out new information or maybe old information that we (or should i say 'i') forgot

I was mainly trying to get at what we can take from the DK study. To keep an open mind about new information and new ways of looking at things and stuff like that. We have to keep remembering that everything we see, do, feel, etc., is an abstraction in some way. We never get to actually 'touch' anything, it's just our electrons interacting with the objects electrons that we call a 'touch'.

 

[killivolt]

LOL....Dunning /Kruger to Electrons

Should have known this would happen

I don't mind. Keep discusssing away.

Regards,
tvtech

Well maybe, by you.

I explain exactly what I have done to the set so far. I tell him don't bother changing those spares again because IT WILL NOT SOLVE THE PROBLEM.

Three days later on that same set....and what is the result? Everything I told him not to bother changing he changed over and over. Hoping for a new result after being told it will not solve the problem. Wont listen. As per Dunning Kruger he has the following issues:

Dunning and Kruger proposed that, for a given skill, incompetent people will:

tend to overestimate their own level of skill;
fail to recognize genuine skill in others;
fail to recognize the extremity of their inadequacy

An Electron and Positron and or Photon will exhibit similar effects in relation to one another, you have just conjoined a phenomena. The example is similar.

* nature-of-electricity.pdf * As John showed, It seems to be that the oscillation charge represents the current of charges! But the speed of electricity would then be <<c! The oscillation charge causes polarization waves (AC) of the dielectric electron positron aether between the plates and these waves have phase velocity ~ c. By the way , the accelerated charge produce the polarization waves, these waves are accompanying waves of the electron. This phenomenon has nothing to do with the irrational duality of the electron and electron wave!

So, people will be people and an electron will be an electron, by phase or force they will neither bow or bend to reason. They will be one or the other.

So, in keeping to the thread. We attribute some of it, to the nature of physics, and General theory of relativity.

It all depends on your relative perspective.

kv

 

[MrAl]

Hi again,

Check this out:
https://en.wikipedia.org/wiki/Quasiparticle

killivolt:
Sorry i did not see your post #33 until now for some reason. All i can say is that this is a very simple concept.

In the wire, we have electrons 'bouncing' all around up and down and this way and that way. If we followed just ONE of those REAL electrons, we would find that we have to describe it in relation to time as well as to three dimensions which functionally would look something like:
p(t,x,y,z)

or at least as p(x,y,z).

So a particular electron could move from say point (1,2,3) to point (2,3,7) on a very small unit scale. But the only way we could have calculated that second point would be from a very big calculation involving perhaps a trillion other electrons and holes. And even then the 'current' would have to be described as being in a particular direction.
So rather than do this immense calculation which would be impossible anyway, we use a different particle with a different mass, that does not interact with any other particle so can travel through the medium without too much complexity. So we end up with a calculation that follows one dimension only so it is much simpler to think about ie I=dq/dt or something like that.

So this is a very very simple concept that does not involve thinking about waves and things like that. It's just replacing a huge complexity with a simpler solution that maps the same overall objective into one dimension. The simpler particle is easier to work with so it makes sense.

The reference should be able to explain this better.

 

[tvtech]

I know this little link of mine has nothing to do with Electrons and all...https://rationalwiki.org/wiki/Dunning-Kruger_effect

Quote" a little knowledge is a dangerous thing". Unquote.

At least it keeps the thread on topic

Regards,
tvtech

 

[tcmtech]

I think my neighbors dog is experiencing a variation the Dunning Kruger effect.

Big dummy got a shock collar the other day and is apparently too dumb to catch on to his actions being part of the cause and effect relationship between barking and shocking.

BARK yip yip yip yip. BARK. yip yip yip yip Bark. yip yip yip yip.