El_Gringo_Grande

Robin Hanson is welcomed to his opinion. I somewhat disagree with that opinion and my experience is different.

What I do agree on is that many in the "hard" science to poo-poo the social sciences. They give no credit to what the social sciences have accomplished. (I include psychology when I say social sciences. Technically it isn't but for this discussion I think it is fine.) The general public also tends to take up this attitude. Probably for the reasons Hanson point out. They see psychologists taking the stand trying to get criminals off.

But I generally did not hear what experts said in testimony in public settings. I simply heard what professors said and wrote in profession seminars and peer reviewed papers. I also heard what professors taught to their students in the area of social sciences. They stressed often that causal relationships are very hard to come by. That is why statistics are so very, very important in the social sciences. And statistics only say there is a percentage that this is what is occurring within this population.

I studied computer science as well in college and currently write commercial software for a living. I was part of the group that started the HTIC (Human Technology Interaction Center) at OU. This basically tried to use the fuzzy, but pretty vast, knowledge of human behavior and cognition in the design of technology to create better design for human consumption.

You know what most designers that we talked to spoke about? The art of it all. And these where not silly art degree majors. These where people that had doctorates in psychology and related fields. They used their knowledge of psychology to help them. But often it comes down to the art of the design.

After working in the computer field for 10 years I have to agree. A large part of it is art. Engineering practices do help out loads. But usually the difference between a good programmer and a great one is the artistry.

Another example is memory. Tons was known about memory in humans. But an overall picture was lacking. There where many theories about what memory was and how it worked. All these theories tried to use the vast amount of knowledge that was at hand. But none of those theories could be actually proven.

Then along came neuroscience. This field is much more quantitative. It allows you to understand the actual nerves and the chemical reactions reactions actually affect memory. This field allowed the various theories of memory to become a bit more unified. When I left some exciting thing where taking place. This was due to the qualitative AND the quantitative studies going on. But it was the quantitative stuff that really made things so promising. And that was because when you do primarily qualitative work there is a lot of room for questioning.

An example I KNOW you are familiar with is the current climate debate. Why is there so much debate? I would guess there is less disagreement in the climatology field than there is in the economics field.

The reason is that most of the results are somewhat qualitative. They rely on models and random and incomplete measurements. They can't test their theories as well as some would like. Qualitative studies also allow for much more of the researchers biases to enter the picture. Because you are relying on the quality of the evidence. And that quality is subjective. Statistics are supposed to remove that bias but the very act of selecting a statistic to use is sometimes subjective.

If you ever thought that I meant to say that social science doesn't know anything you did not understand me. Of course they know a lot. The problem is, in part, this vast knowledge base without many bedrock theories that have been proven and repeatedly proven.

My whole point is that some (many?) social scientist, like all scientist, get so wrapped up in their work that they cannot see that they are simply wrong. And because you cannot create experiments that directly proves them wrong their line of thinking can become to influential among certain group.s

I believe this will change over time. But there is always going to be a big problem with them. The study is all about humans. And humans do evolve over time. I would bet that they evolve rapidly. Since it is very hard, or even impossible, to predict where the evolution will end up it becomes doubly hard to have good predictions of the institutions humans create.

For example what would happen if tomorrow somebody discovers a drug that would allow many to live for 400 years? Or if a plant was created that supplied all nutritional value but could grow absolutely any place? Or if cold fusion was a reality?

Those things would throw some economic models for a loop. I would bet there are far more subtle things going on every single day that works towards undermining the economic models out there.

Yes, I am sure there are things going on in the universe that have the same effect on physic theories. But the universe operates on a much larger and slower scale. That give the physicists some time in dealing with it all.

But, really, I want to know what your thought out opinion is. What experience do you have with social and physical science? And why do you think the way you do? Is it just a belief you have? Maybe you are satisfied with the fact that it just works and history? Or maybe you have some evidence that I lack?

__________________
I’m not racists, I have republican friends. Radio show host.
- "The essence of tyranny is the denial of complexity". -Jacob Burkhardt
- "A foolish consistency is the hobgoblin of little minds" - Emerson
- "People should not be afraid of it's government, government should be afraid of it's People." - Line from V for Vendetta
- If software were as unreliable as economic theory, there wouldn't be a plane made of anything other than paper that could get off the ground. Jim Fawcette

xTYBALTx

The experimental social sciences address this. Through experiments, false theories can be tested and shown to be false.

__________________
A policy of freedom for the individual is the only truly progressive policy. -F.A. Hayek

El_Gringo_Grande

Yes, yes. But testing some theories, as I pointed out, is very difficult. Only by implementing them does their validity become known.

Want to see some proof of this? Look to the educational system. There have been dozens of theories that, once adopted wide scale, ended up being wrong. Those theories when tested through experimentation seemed to be correct.

A classic example is lighting in the work place. Decades ago a psychologist had a theory that increased lighting in the workplace would increase productivity of workers. He did experiments and found that it in fact did increase productivity.

So workplaces around the world started adding lighting to the workplace and all was good.

Later some other psychologists came around and said hmmm. This guy didn't control for all variables. What would happen if we lowered the lighting levels. And what do you know? Productivity went UP!

So some theorized that change itself increased productivity. They found many things that, if changed, would increase productivity. It was also found that if some things changed productivity didn't change one iota. But some things if changed, things you wouldn't even imagine, would hurt productivity. So it isn't change in the workplace all by itself that affects productivity.

Another example is class size. It is hard to find a school where class size isn't kept under a certain number. This has cost schools many dollars.

But I am not convinced of this effect. Some other others are not either. If you notice I am using a economist to prove my point. Now also notice that these economists bring up the term "liberal" in their paper. Bias is definitely rearing it's ugly head.

The fact is you see a lot of bias when you read academic papers in the social sciences. I came across a number of economist papers when doing some research and I notice that many have a conservative bias. (I studied decision making and expertise. As you can imagine economists have much to say about this field.)

And many in the education field have a liberal bias.

Bias is not good when you are seeking to solve a problem scientifically. I doesn't invalidate the results automatically if the experiment is designed well. However, because social science has so many variables you have to pick and choose which ones are the important ones to control for. You get some of this information from those that have previously studied the area. And you hope that they are correct. But bias can absolutely kill you during this part of the design.

Experiment design is not as simple as one would think. And single experiments do not tell you much one way or another. If you would consider that many social science theories cover huge populations then you can see that the difficulties are vastly increased.

None of this even covers the cost of it all. And that cost has it's own traps and leads to some bad conclusions itself.

On the macro scale I think there is good information out there. But much of that information is not far from common sense.

Overall the social sciences need maturing and more effort. They are indeed very important to humanity and I don't want to sound like they should be ridiculed. But we need to understand their limitations.

__________________
I’m not racists, I have republican friends. Radio show host.
- "The essence of tyranny is the denial of complexity". -Jacob Burkhardt
- "A foolish consistency is the hobgoblin of little minds" - Emerson
- "People should not be afraid of it's government, government should be afraid of it's People." - Line from V for Vendetta
- If software were as unreliable as economic theory, there wouldn't be a plane made of anything other than paper that could get off the ground. Jim Fawcette

xTYBALTx

You just described the scientific process perfectly. Someone comes up with a theory, someone else comes up with an experiment to test the theory, someone else comes up with a theory as to what was wrong with the experiment, someone else comes up with an experiment to test THAT theory, and so on ad infinitum.

That is the way science is supposed to work - whether that science be "soft", "hard", or somewhere in between.

You make many good points, but what you fail to realize is that your critiques apply to both "hard" and "soft" science.

__________________
A policy of freedom for the individual is the only truly progressive policy. -F.A. Hayek

El_Gringo_Grande

Of course there are very strong similarities. They are both sciences, after all. So both must use the scientific method.

I would say that the "hard" sciences did resemble current "soft" science at one time. Physicists couldn't always take the measurements they can today. At one time scientists thought plate tectonics was a crazy idea. Partly because they couldn't take accurate measurements.

Most of the bad ideas of the past where caused by bad assumptions and enforced by lack of accuracy in observations.

Lack of accuracy in observations is one of the biggest problems in the social sciences. You have to rely on the subjects reporting correctly what is going on.

This will change over time. Just as we eventually developed lasers to measure the movement of the earth we will develop ways to measure what is actually going on in the human. Currently some use high resolution MRI's to help with understanding how the brain functions.

There is probably no way I can describe to you the difference. All I can say is the methods of the social sciences are much more crude than those of most physical sciences.

You where in the global warming debate in these forums, correct? If so do you remember the picture of the weather station next to building and it's air conditioner? There where other strange pictures of stations in the middle of a parking lot.

Kinda makes you question all the measurements, doesn't it?

Every single measurement you make in human based experiments are like that. Even seemingly straightforward ones like measuring reaction time to a stimuli are a huge pain in the butt. Are they reacting slower because of our manipulation of the variable or is it because people are walking by and they are being distracted by the shadows under the door? Or is it the fact the janitor replaced the bulbs with slightly lower wattage ones yesterday? Or maybe they are hungry today. Or maybe the cafeteria is serving fattening food on certain days making the subject more drowsy. (Statistics are used to weed these things out, hopefully. But you never know.)

And those are the simple measurements you dream about when designing an experiment. It is a joy to just track eye movement.

I know the "hard" sciences also have to worry about these things. But they also have tools that can make wonderful measurements. They can find out the chemical composition of materials before and after experiments. They can measure the movements of continents very accurately. They can very effectively seal off material from other variables without worried about it dying.

And they don't have to worry about the material being tested finding out it is being tested and changing it's behavior.

__________________
I’m not racists, I have republican friends. Radio show host.
- "The essence of tyranny is the denial of complexity". -Jacob Burkhardt
- "A foolish consistency is the hobgoblin of little minds" - Emerson
- "People should not be afraid of it's government, government should be afraid of it's People." - Line from V for Vendetta
- If software were as unreliable as economic theory, there wouldn't be a plane made of anything other than paper that could get off the ground. Jim Fawcette

leejo

It's interesting that you two have blended quantum mechanics and economics, because I have long thought they have a lot more in common than might be clear to the novice. The piece of QM that Einstein resisted so strongly is the Copenhagen Interpretation that dominates the theory today. Its main point is that we can know in great detail the behavior of particles in aggregate but not how an individual particle will behave. This interpretation has been supported by many experiments.

Human behavior seems similar to me. While an individual's behavior may be essentially random, professionals can predict human behavior with great accuracy in aggregate. Actuaries can tell you the odds that a 16 year old male driver will have an accident during the next 12 months, for example, but it's impossible to say whether or not little Stevie Smith will have an accident, or when.

Economics is, among other things, a study of human behavior, and so this dichotomy between individual and group behavior is an issue. I don't have anything productive to offer here, but it's something I've always thought interesting. In other words, once Physics discovers how quantum behavior manifests itself in this big world, the answer may be so obvious we'll be amazed that it took so long to figure out.

El_Gringo_Grande

Right. In some areas there is very good statistical data of what is going to happen. The thing is those actuary tables must be update regularly to reflect what has just happened. There is nothing predictive about them unless you mean predicting that tomorrow will be much like today and yesterday.

It is interesting and there are some similarities. But I think there are two different mechanisms at work. Of course I say that because I have only a layman's knowledge of particle physics and a "formerly intermediate" (at best) knowledge of a narrow aspect of cognitive psychology.

But I listen to podcasts!

Anyway with human behavior we are very certain that it all comes down to neurons and how those neurons interact with the rest of the body and the outside world. And we have a very good understanding of how those neurons work.

What is missing is the overall model of how the millions of neurons work together as a, seemingly, cohesive whole. Good models are emerging about how exactly various "parts" of the brain works. Memory, emotion, sight etc. are starting to be understood in detail. This all started only about about 30 years ago when they started to focus on the neurons and with the advent of measuring tools like CAT, MRI and PET scans.

Things at the quantum level have a different difficulty from what I understand. We just don't know what exactly is happening down there. But we do have a pretty darn good model about the effects caused by what is happening down there. We also have a decent idea of what the parts are down there. It is why we can design safe nuclear reactors and big nasty nuclear bombs.

__________________
I’m not racists, I have republican friends. Radio show host.
- "The essence of tyranny is the denial of complexity". -Jacob Burkhardt
- "A foolish consistency is the hobgoblin of little minds" - Emerson
- "People should not be afraid of it's government, government should be afraid of it's People." - Line from V for Vendetta
- If software were as unreliable as economic theory, there wouldn't be a plane made of anything other than paper that could get off the ground. Jim Fawcette

xTYBALTx

John Taylor discusses the Fed, inflation, monetary policy, and more.

http://www.econtalk.org/archives/200...taylor_on.html

__________________
A policy of freedom for the individual is the only truly progressive policy. -F.A. Hayek

leejo

Dollar Reaches 13-Month High as Credit Crisis Spreads to Europe

Like I was saying, you don't have to outrun the bear, you have to outrun your buddy.

xTYBALTx

Wow, nice nice nice prediction.

I get to eat some crow too.

__________________
A policy of freedom for the individual is the only truly progressive policy. -F.A. Hayek

leejo

OK. Now let's go make some money!