Wikipedia nails it

Wikipedia entries on economics seem to be written mostly by graduate students, eager to share their knowledge with others, and a disproportionate number of Austrian, feminist and other heterodox economists. 

Sometimes they get it pretty much exactly right, as with this description of the practice of economics:

Yup, economics uses mathematics, and that's the most important thing to know about it.  What makes a good theory? Precision, not accuracy, nor conformity with experience. Plus, of course, the possibility of generating additional research - and so additional publications, grants, and so on. 

Clicking on the footnote suggests that these words on theory are one person's interpretation Milton Friedman's Methodology of Positive Economics. It's not exactly how I understand Friedman, but it's a pretty good description of how some people approach the discipline.

So WCIers, should I leave the Wikipedia entry on the practice of economics alone? Or should I modify it? If so, how?

Comments

[Britonomist]

A bunch of us needs to get together and clean up wikipedia, a lot of the pages on economics are full of thoroughly un-sourced (or badly cited) and one sided nonsense.

[Frances Woolley]

Britonomist - agreed. Which page is #1 on your hit list?

The difficulty is that the big pages, e.g. labour economics are such a mess that it would take a good couple of days of concentrated effort to get just one entry into shape - and it's work that's unacknowledged, and could be undone by some Wikivandal at any point. I go in and do little edits every so often, but for something major it's as easy to write something on here and put a link to the WCI post on Wikipedia.

Marginal Revolution U and Khan Academy though suggest that perhaps the better approach is to create a Wiki-competitor that doesn't have the problems of Wikipedia.

[Thomas]

It's not vandals that are the big problem (the bots are quite good at handling them). It's people who think they can improve the article, but are wrong.

I think it's worth trying, but you could also look at Scholarpedia, which does expert articles. They are seriously short of boring mainstream economics. I know the people who did the Bayesian Inference page and they seem to be happy with the process.

[Chris J]

For better or for worse, the world uses wikipedia. Try to fix it as best you can. (Easy for me to say - the physics sections are clear and correct.)

[Min]

One nit: I think that signal strength is analogous to effect size rather than statistical significance.

[JoeMac]

Frances,

Didn't you have a post a long time ago where you mentioned that you were not entirely in agreement with Friedman's views in his essay?

[Frances Woolley]

Thomas " It's people who think they can improve the article, but are wrong."

Yes, absolutely. I wonder how many hits Scholarpedia gets and if it's in google scholar?

Chris J - actually, take a look at the Wikipedia section on velocity. It looks like an econ Wikipedia entry - a somewhat intelligible introduction, followed by a bunch of equations, with little or no explanation of why these equations are there or what they mean.

[Frances Woolley]

JoeMac - "you were not entirely in agreement with Friedman's views". I'm not, but I agree with him that theories should be judged on the accuracy of their predictions, as well as the precision.

[Frank Restly]

"What makes a good theory? Precision, not accuracy, nor conformity with experience."

The reason that economic theory cannot be accurate is because any economic system must incorporate choice. The reason that economics should reject conformity with experience is not because experience is unimportant, it is because economic experience is very subjective.

Economic theory must ultimately confront two variables that cannot be modeled mathematically.
The first is liquidity preference - the willingness to two people two exchange one good (usually money) for another.
The second is value determination - how are the relative values of two goods (one being money) arrived at.

Precision is all that is left. Precision allows you to stipulate what choices and value judgements will be made, and then identify the likely outcomes with some degree of confidence.

Which is not to say that economics should be a totally theoretical discipline. Obviously an economic system could be developed in theory that has no resemblance to any known reality. It could be precise as hell, but only the wonkiest of the wonks would give a hoot about it. Nor should economic theory consider the "markets" as the only method by which value is or should be determined. The markets are a consensus value judgement, but that does not mean that all parties agree with that consensus.

[Evan]

@Frank:
Value determination has to be one of, if not the, most mathematically modelled concepts in economics. Starting with Jevons, Menger and Walras (the marginal revolution in the late 19th century), moving through Arrow and Debreu and ending up constituting a major chunk of the microeconomists bible by Mas-Collel Whinston and Green.

I am less familiar with the work on modelling liquidity preferences, but I'm pretty sure it has been covered to death by both monetary macro types and finance types.

[Simon van Norden]

Good question, Frances.

re: Practice of Economics
I'm disturbed by the lack of any reference to the scientific method. I think economics revolves around empirically falsifiable predictions; understanding how different models make different predictions, testing those predictions, and modifying our models in light of that empirical testing. Instead, I see some elements of that listed under "Theory", which I think is confused.

re: Theory
I think this definition is much more in need of improvement than that for the Practice of Economics. One of the things about it that I found disturbing included the absence of anything relating to qualitative (as opposed to quantitative) models. But that last sentence about the aim of economic theory really sucks the bag. Also, there is no discussion about the link between economic policy and economic theory. Looking around, I see substantial energy devoted to theorizing about situations in which the type of phenomena we observe should/shouldn't be the subject of a policy intervention (usually by government.)

[Ritwik]

"What makes a good theory? Precision, not accuracy, nor conformity with experience"

You have to be extremely confident about the unimportance of the loss function to make a statement like that.

[Frances Woolley]

Simon: "I think economics revolves around empirically falsifiable predictions; understanding how different models make different predictions, testing those predictions, and modifying our models in light of that empirical testing."

Wikipedia is a mirror; it reflects what we teach our students. To a very large extent, models are taught in the various theory and field courses, and testing is taught in econometrics courses. Students are left on their own to make the connection between the two. Methodological ideas, such as falsifiability, deduction v. induction, the nature of knowledge, aren't taught (in my limited experience) much at all.

Ritwik - should I have used the html < sarcasm > markup or were you o.k. without?

[Ritwik]

Frances

I was ok without, I should have left a smiley at the end I suppose. :)

[Frances Woolley]

Ritwik, sorry for doubting you.

[Jacques René Giguère]

Britonomist: the pages that are wrong an dneed edit are easy to spot: they're the one I disagree with.

FrankRestly: "Obviously an economic system could be developed in theory that has no resemblance to any known reality. It could be precise as hell, but only the wonkiest of the wonks would give a hoot about it." It would not be economics but metaphysics.. It fails the correspondance test. As a french general said while watching from afar the charge of the LIght Brigade: It's magnificient but it's not war.

FrancesL you can be a trusted editor. There is a process. I'll leave the solution to the reader...

[?]

How would you deal with stuff like this?

http://en.wikipedia.org/wiki/Andrew_W._Mellon#Mellon_plan

[Jacques René Giguère]

?: by telling your students that on technical articles, they should proceed with caution. But on anything touching history and politics, stay away till they are mature enough.

[Sina Motamedi]

+1 for a problem with: "What makes a good theory? Precision, not accuracy, nor conformity with experience."

I'll outsource it to my main man Richard Feynman: "It doesn't matter how beautiful your theory is, it doesn't matter how smart you are. If it disagrees with experiment, it's wrong."

[Sina Motamedi]

My bad, didn't notice the sarcasm!

[Simon van Norden]

Frances:

"Wikipedia is a mirror; it reflects what we teach our students."

Of course....but when you asked how it should be modified, I thought we were trying to correct its weaknesses.

[Unsettled Worker]

This is just a complete blasphemy. I do not believe economics has to do anything with preciousness. It is a social science which tries to explain human behavior. There is nothing more to add. And Milton Friedman´s theories led to military dictatorships in South America, not to economic prosperity. If we continue to understand economics as a exact science, we will be doomed. Good theory serves to simplify reality, and should not contain any exemptions. It is not a mathematical equation nor a graph. These are only implications which serve to verify the theory.

What I want to say, economics is talking all the time about the same. We have a state. It can raise the taxes and spend more, or it can lower the taxes and leave more space for the private enterprises. This is the theory. It can not be achieved both. We are having a massive debt. Our government wants to lower our taxes and promises us no more debt (good explanation can be found in How to Fight Global Economic Crisis. This is not possible. But they tend to use those fancy terms and stuff no one can really understand, using economy for that. Sometimes, I think the goal of economic theories is to complicate the situation, which is contradictory to the definition of the scientific theory.

[david]

All of Wikipedia's topics that have dedicated ideological camps are poorly-written; the real action is in finding enough collaborators to dominate the Talk Page.

[genauer]

"What makes a good theory? Precision, not accuracy, nor conformity with experience. Plus, of course, the possibility of generating additional research - and so additional publications, grants, and so on"

priceless

It was actually the rant of some southern US Jesuit on the Cobb-Douglas wiki page, which was wrong in its own way, which sparked some very interesting discussions, readings, and discoveries for me.

Of course, you, and all paid economists should be prohibited to write wiki pages about economics, you are part of the problem : - )

[Edmund]

What makes a good theory? Precision, not accuracy, nor conformity with experience.

This is the single worst thing that has ever happened to economics. Imagine if meteorology or biology or oncology or, yes, physics comported itself in accordance with such ideas. Worse than useless.

[Frances Woolley]

Edmund - I'm not sure that I understand your point.

Is your view that of an informed observer, saying that, yes, it is true that economists value precision over accuracy, and are unconcerned about how their theories stand up to real world testing?

Or are you making the mistake of believing everything you read in Wikipedia (or on a blog), and figuring that if it's on the internet, it must be true?

There is a lot of garbage out there in the world of economic research. But there's a lot of good stuff as well. This Wikipedia entry just took my fancy because it was so outrageous as a description of scientific method.

[John]

The only truth about economics is that most economists "are unconcerned about how their theories stand up to real world testing."

Proof: Greenspan, Taylor, Cochrane, Mankiw al carry the title and no means or method exists to defrock the same