Dana's Blog

Marlene Scardamalia and Carl Bereiter are pioneers in Knowledge building theory. Much of their work relates to understanding how people use information and how they construct knowledge. They also wrote a book entitled Education and Mind in the Knowledge Age. I haven’t had a chance to more than cursorily glance at it, but it seems to be an agenda for how to structure pedagogy so that people are prepared for creative, knowledge-intensive work work. It seems like a “how to remake education for the 21st century” book that actually has substance.

I came across their work while researching different theories of writing. Their particular theory emphasized the importance of knowledge transformation and how the process of writing encourages ideas to build upon themselves and evolve in the course of writing.

The below quotation is from some of their work on knowledge communities and discusses the sociological nature of what constitutes knowledge. It is also interesting from the perspective of being a researcher and trying to decide which topics to pursue.

Whether they are scientists working on an explanation of cell aging, engineers designing fuel-efficient vehicles, nurses planning improvements in patient care or first graders working on an explanation of leaves changing color in the fall, knowledge builders engage in similar processes with a similar goal. That goal is to advance the frontiers of knowledge as they perceive them. Of course the frontiers perceived by children will be different than those perceived by professionals, but professionals may also disagree among themselves about where the frontier is and what constitutes an advance. Dealing with such issues is part of the work of any knowledge building group, and so students must learn to deal with these issues as well. Identifying the fonrtier should be part of their research, not something preordained. The knowledge building trajectory involves taking increasing responsibility for these and other high-level, long term aspects of knowledge work.

Steven Johnson has been investigating where good ideas come from. His TED talk highlights several aspects of innovation and previews findings from his upcoming book.

Johnson begins by challenging the iconic image of “lone genius” as innovator and claims that innovation happens more when people come together. He highlights how England’s first coffeehouse was an example of a “liquid network” where people came together and shared ideas and which he. He also discusses research done by Kevin Dunbar  who found that most ideas happen at lab meetings as people discuss their work rather than by themselves. (Dunbar has more interesting work that is discussed in a Wired profile). In short, innovation happens with other people.

The next idea he challenges is the eureka moment. Instead, he argues for the “Long Hunch” where good ideas have long incubation periods. Moreover, after an innovation or discovery happens, people tend to compress the time-frame over which an innovation happens. For example, Darwin writes in his autobiography of a single day in October when the idea of evolution all came together. This, however, contradicts the finding of historians who, pouring over his scientific journals, found traces of the idea in notebooks from many months before.

Johnson links these ideas together by noting that many great ideas sit in the back of people’s heads as incompletely formed concepts. Often, they are missing small ingredients that reside in other people’s heads and that by being more connected, this long hunch period can be shortened.

One of the primary barriers for more “open” innovation is that scientists worry about protecting their ideas. Johnson mentions this as well and says that people spend too much time on “protecting” vs. “connecting”. People worry about people “scooping” them and cut off valuable opportunities for exchange in the meantime.

Links to articles in economics about these topics:

Of Mice and Academics: Examining the Effect of Openness on Innovation
The Increasing Dominance of Teams in Production of Knowledge

Today was my first day of real classes at MIT.

The day began especially well with Development Economics. It was really quite amazing to have Abhijit Banerjee and Esther Duflo as professors. As Prof. Banerjee laid out the grand themes for the course, I could not help but feel incredibly privileged to be at MIT and to have the opportunity to study such an important topic with such great thinkers.

About thirty seconds later, my mind returned to the subject at hand — development economics. Banerjee was discussing his and Duflo’s work on The Economic Lives of the Poor. The basic message was that even among the poorest of the poor, people who live on less than $1 a day, people have choices. The main evidence for this is that people do not spend 100% of their money on food (it’s more like 60-70%). Further, the money they do spend on food isn’t just going to the cheapest food and the kind with the most calories, people are buying tasty food (not to mention alcohol and sweets!)

The important point that this all illustrates is that even the poorest of the poor are not merely at the mercy of external forces  – the poor are not behaving mechanically and spending each additional dollar in a way determined by survival. They still have some degree of agency and are making  choices, albeit very limited ones. This study also reminded me of an old paper by George Stigler where he attempted to find the lowest cost diet that would still provide sufficient nutrients. His finding was that many dieticians recommend diets that are much more expensive and that most any diet necessarily has a physiologically necessary component, but that much is driven by palatability and cultural appropriateness and that people are spending more than “necessary” on food.

These low-cost diets of the professional dieticians thus cost about two or three times as much as a minimum cost diet. Why do these conventional diets cost so much? The answer is evident from their composition. The dieticians take account of the palatability of foods, variety of diet, prestige of various foods, and other cultural facets of consumption. Primarily on such grounds can one explain their emphasis on meats and the inclusion of sugar. ¹

Finally, I appreciated that the class seems to have a certain gravitas that I imagine cannot be matched by other classes.

Let’s see if having a nicer webpage will encourage me to update it more frequently. I’ve been writing quite a bit, but somehow I’ve found it difficult to turn these writings into a blog.

Hopefully this will change!

Lukas Biewald has kindly posted a recent blog post that I and my co-author Adam Kapelner wrote about our recent paper on motivation in crowdsourcing markets…

See the full post here.

It’s at Dana’s research page.

The title is “Breaking Monotony with Meaning: Motivation in Crowdsourcing Markets” and here’s the abstract:

Most of this speaks for itself… I particularly like the following suggestions.

• Get in the habit of always writing something every week for idea generation. It should be short, and done in a structured way — Akerlof gives suggestions below.
• Have 10 (or more!) ideas always on the backburner — “5 won’t work, 4 have probably been done…”
• Have an agenda. Have your own opinions, follow your interests, don’t let the Faculty set your research agenda.
• Have economic models in your mind as you observe, so that you can be a more careful observer. Use models and real-world observation to drive your research.

The following is excerpted from lecture notes that are part of a Macroeconomic Theory course taught by George Akerlof.

I think that economics is a much more eclectic and opportunistic field than what would be indicated by the standard Graduate Curriculum. What further, and especially, worries me is that we do not give you sufficient opportunity and encouragement to come up with your own ideas and opinions.

So there will be an informal section of my part of the course that is devoted to your own ideas and opinions.

First, I want each of you to keep a log of your opinions about economics and about economic issues that you find interesting. I want you to make at least one log entry per week. It should typically be about one page.

• Find a theoretical idea that you like, or that you dislike. Explain why you like it or dislike it. If you like it, you can explain how it could be used further. If you dislike it, you can explain how it is abused. You might further explain how you could do better.
• Look through the economics literature or elsewhere and try to find an interesting use of data. You can say what makes that use of data interesting. Don’t take something from another class unless you personally have something original to say.
• Look through some data and see whether anything interesting is happening in that data. If so, then you can explain why that data is interesting.
• You could engage in a very hard task. Try to think of an ideal data set that you would like to have, and explain how you would use it to answer some question.

… before you panic, that I have suggested that you re-write The General Theory in your first term of graduate school, let me give you four ways in which you can fulfill this assignment. I think that every one of you can do an excellent job at this.

What I want to encourage in all of you is that you should have your own opinions and your own subjects of interest, and that as students you should not let the Faculty be setting your agenda. I want you to have an agenda, or to develop, an agenda of your own.

I want you to write the logs for six separate reasons.

First, I think that this will make this a better course. I want to erase the opinion in your minds that a good program for economic research is one that attempts to re-write The General Theory …

Second, insofar as I can, I want to develop in your mind a critical aspect. I would be very happy if everyone here developed into a Paul Krugman or Robert Barro.

Third, pragmatically, the major problem in the PhD program is that people find the courses so daunting that they fail to apply themselves to the development of ideas. And they find it difficult later to find a thesis topic. I want to develop in you habits of mind that will make it easier to find a thesis topic. I want you to be thinking about what is and what is not a good idea for a research paper.

Fourth, I view it my duty to try to indicate to you how I personally think about economics and research in economics. This is in fact how I work. I always have a list of possible projects that I might work on. And my major intellectual activity is to actively work on generating this list. I want you to have your list of the ten ideas that you would work on. The reason that you need ten is that you will decide you do not like five of them and the other 4 have probably already been done, so you need to generate a lot of different ideas.

Fifth, I think that it is important that economists adopt a methodology that is much closer to the natural sciences. That is that we should derive our theories more from careful observation. I think that the best work is that which begins by observing the real world, and then constructing theories that explain how it works.

Good logs report such observations.

Sixth, the graduate economics program is in many ways an attempt to alter your identity. I do not think that you can do very good research in economics unless you also have your own identity. I want you to preserve and also to actively construct your own identity.

The other ingredient of this course is the models.

The typical economics paper makes a model of some phenomenon, which may also be empirically simulated or tested. By going over in some detail the key models of macroeconomics, and also by having you generate your own ideas, this course duplicates in toto what is needed to be a research economist.

So the modeling that is presented here can be of some use as a starter for being a research economist. As observers, you should have as large a tool-kit of models as possible. That way you can make your observations fit the facts.

I made a resolution for 2010 that I would create a proper blog…

Hoorah!