This is said to be one of the several reasons why experiments cannot be usefully conducted on an economy or on human beings. But what does it mean? Why do people change their behavior when they know they are being watched? Apparently, in quantum mechanics, the process of observing sub-atomic particles entails shining light on them, and the light interferes with the behavior of the particles being observed. But what is the meaning of this sentence when we are dealing with human beings? Are we talking about toy "lab experiments" where researchers attempt to simulate the market or find out the effect of incentives on people? I mean, even here, who cares who's observing you?Callahan writes: "if the person running the experiment is liked by the subjects, they will often try to figure out what result he wants, and act to bring it about." That seems like a bit of a stretch. An experimenter need not appear in person in front of the subjects; he can give them all the instructions with the help of a computer. In addition, the experimenter presumably wants to learn from the experiment; he has no preconceived ideas or desires of how the experiment "should" turn out. Why should he want any particular result?
Callahan goes on to state the opinion that people learn from one experiment to the next and adapt, nullifying the results of the previous experiment. But they also learn in the real world. This collapses the "different behavior when observed" reason into "people themselves change" reason: in particular, their desires, beliefs, and powers change, unlike elementary particles. In other words, for humans past behavior is no guarantee of future results. I fully agree, but why make it look like observation changes human actions?Thanks.
dchernik: First thing, a link to this mysterious Callahan? Callahan writes: "if the person running the experiment is liked by the subjects, they will often try to figure out what result he wants, and act to bring it about." That seems like a bit of a stretch. No it's not. Most people want to please the authority figure. An experimenter need not appear in person in front of the subjects; he can give them all the instructions with the help of a computer. But they might know whathe wants from other places, like his reputation, plenty of things. Remember, he is liked, so they probably know something about him. In addition, the experimenter presumably wants to learn from the experiment; he has no preconceived ideas or desires of how the experiment "should" turn out. Why should he want any particular result? You haven't been reading up on Climategate, have you. Callahan goes on to state the opinion that people learn from one experiment to the next and adapt, nullifying the results of the previous experiment. But they also learn in the real world. This collapses the "different behavior when observed" reason into "people themselves change" reason: in particular, their desires, beliefs, and powers change, unlike elementary particles. In other words, for humans past behavior is no guarantee of future results. I fully agree, but why make it look like observation changes human actions? Of course it does. Let's say the experiment is: how many times do people pick their noses in ten minutes. Do you think they will act different if they know they are being watched? Same for all behavior that isn't politically correct. It's a well known phenom when polling people. They lie when it's about themselves to look better. So clever q's have been thought up to get around that . Also, stage fright is a well known thing. I mean really, I would have thought Callahan is being Captain Obvious on this one.
First thing, a link to this mysterious Callahan?
Callahan writes: "if the person running the experiment is liked by the subjects, they will often try to figure out what result he wants, and act to bring it about." That seems like a bit of a stretch.
No it's not. Most people want to please the authority figure.
An experimenter need not appear in person in front of the subjects; he can give them all the instructions with the help of a computer.
But they might know whathe wants from other places, like his reputation, plenty of things. Remember, he is liked, so they probably know something about him.
In addition, the experimenter presumably wants to learn from the experiment; he has no preconceived ideas or desires of how the experiment "should" turn out. Why should he want any particular result?
You haven't been reading up on Climategate, have you.
Callahan goes on to state the opinion that people learn from one experiment to the next and adapt, nullifying the results of the previous experiment. But they also learn in the real world. This collapses the "different behavior when observed" reason into "people themselves change" reason: in particular, their desires, beliefs, and powers change, unlike elementary particles. In other words, for humans past behavior is no guarantee of future results. I fully agree, but why make it look like observation changes human actions?
Of course it does. Let's say the experiment is: how many times do people pick their noses in ten minutes. Do you think they will act different if they know they are being watched? Same for all behavior that isn't politically correct.
It's a well known phenom when polling people. They lie when it's about themselves to look better. So clever q's have been thought up to get around that .
Also, stage fright is a well known thing.
I mean really, I would have thought Callahan is being Captain Obvious on this one.
My humble blog
It's easy to refute an argument if you first misrepresent it. William Keizer
Gene Callahan's "Economics for Real People"
I don't see the need to be snide, Dave, but your answer is actually great. Thank you.
dchernik: Gene Callahan's "Economics for Real People" I don't see the need to be snide, Dave, but your answer is actually great. Thank you.
TY for the link.
Sigh. Yeah, the anonymity of the internet allows my dark side some escape. My apologies.
[BTW, kinda proves Callahan's point, doesn't it].
YW.
dchernik:he has no preconceived ideas or desires of how the experiment "should" turn out. Why should he want any particular result?
First, it would be very hard to find an impartial scientist. The reason being the scientific process. One step is formulating a hypothesis. This is what the researcher expects to find in nature, under given conditions. This is what he is trying to prove. He thinks this will happen, and is looking for it to happen.
dchernik:why make it look like observation changes human actions?
I think obseration changing human action is innate and can be best explained by religion. Even those who lived long ago knew that observation changes behavior. There are so many things that can be done that are unknown to others. I can covet my neighbor and nobody would know. Except God. God is the ultimate observer. He sees everything and because of this causes the person to act in a more morally upstanding way. The person is accountable to someone.
On a side note, the argument that observation changes human action is an easy attack for someone who doesn't want to believe or wants to refute the results of the study. I think that is why the phrase is used so frequently.