Restructuring the Social Sciences

Very interesting paper (pdf) from Gary King with practical advice on reorganizing centers and departments for the information age:

The social sciences are undergoing a dramatic transformation from studying problems to solving them; from making due with a small number of sparse data sets to analyzing increasing quantities of diverse, highly informative data; from isolated scholars toiling away on their own to larger scale, collaborative, interdisciplinary, lab-style research teams; and from a purely academic pursuit to having a major impact on the world. To facilitate these important developments, universities, funding agencies, and governments need to shore up and adapt the infrastructure that supports social science research. We discuss some of these developments here, as well as a new type of organization we created at Harvard to help encourage them — the Institute for Quantitative Social Science.  An increasing number of universities are beginning efforts to respond with similar institutions. This paper provides some suggestions for how individual universities might respond and how we might work together to advance social science more generally.

Comments

"from making *due* with a small number of sparse data sets"

Apparently making do with spell-check in lieu of an editor ...

While I too latched onto that, I double checked and found:

http://grammarist.com/usage/make-do-make-due/

"The idiom meaning to manage to get along with the means available is make do, not make due. Make do is short for make [something] do well enough, where do carries the rare sense to serve a specified purpose. So this do is similar to the one used in sentences such as, “I could use a cup of coffee, but tea will do.”

"While it’s tempting to call make due a misspelling and leave it at that, make due appears often enough (about once for every ten instances of make do in a current Google News search) to have gained some acceptance, and some people (including commenters on this post) find it at least as logical as make do. Perhaps due, which is mainly an adjective, could here bear the sense appropriate (as in, we have done due diligence), or perhaps it could mean sufficient (as in, we have due cause to be thankful). And because the phrase is an idiom, its logic can be loose.

"Still, that we can almost justify the use of make due doesn’t change the fact that make do is the standard form in edited writing from throughout the English-speaking world. If you are writing for an audience that might view make due as a misspelling, it’s probably best to go with the safer, more conventional spelling."

A standard University recipe: Make a case for a new center / interdisciplinary program. Make it seem like a large program by getting associate faculty co-opted from other programs to fill your roster.

Win win in terms of getting secretaries allocated, office space, support staff, budgets etc. and a great way to further careers by expanding the number of parallel-track opportunities to be Director, Head, etc.

Why call any of this "science" ?

Aspiration.

Enlightenment.

The alternative is to call it social religion, with all sorts of churches with their own dogma, their own bibles, their own revealed wisdom, prophets, cult leaders, perhaps leading to religious wars, inquisitions, forced conversions, dunkings and burnings at the stake.

Its called a science because it (society) shares many parallels with the natural world, and many social phenomenon can be understood using standard physics. Its metaphysical nature, that is 'the understanding of social interactions and relationships' tend us toward a more wishywashy / spiritual understanding of the knowledge it attempts to convey, but its science regardless.

The national accounts/NIPA have been exactly this for decades - there is a reason why they aren't estimated at universities but at specialized statistical institutes. Which, alas, also means that too many academic economists do not fathom them, see i.e. the troubled discussions about the S=I accounting identity ('Investing' in bonds or existing houses or other financial or already existing stuff does not count as an investment in the national accounts but as a more or less sterile transfer of ownership, 'savings' which are used to buy bonds of stocks are not considered to be 'productive' savings but as asset swaps, QE in fact being a kind of example of this, which does not, at least not in a direct sense, add to production, income or employment). There should be more internships for macro students at these institutes and more people moving from these institutes to a professorship at the universities. Hey guys, this is the way we estimate the macro-economy (streams of income, spending, production, debts, equity, hours, whatever). Surely, scientists have to know how their subject is estimated.

I find way too many economists who say or write "we can not run lab experiments to prove our theories like physicists do", obviously totally missing the point of the intro to physics or general science course they took to satisfy a prerequisite for their degree. The little lab experiments are simply exercises to learn how difficult it is to collect data and then find the "meaning". General science classes usually involve outdoors data collection, but without the illusion of precision and accuracy found in physics lab experiments - lacking much data and precision, the general science student likely forgets the lesson as being about data collection and analysis because the results are so vague and inconclusive. But if economists do any work to "prove a theory", they do so using the three hour general science lab as their standard of quality.

Milton Friedman was "well known" to me from his Newsweek column, but when I found "A Monetary History..." in a small town public library and read it, his status zoomed - an economist who validated his hypothesis with data just like physicists and biologists. It took me a number of years before I realized he was not the scientist - she was Anna Schwartz.

I grew up when economists believed in a balance sheet with a zero sum:
labor plus capital = labor income plus capital income = labor consumption plus capital consumption.

Instead we have the idea that the way to increase GDP is to reduce labor cost to as close to zero while capital buys government bond, with consumption increasing because stock prices are higher because profits are higher.

No one ever says "we need to reduce labor costs to reduce consumption to reduce GDP growth." Somehow, consumers are neither labor nor capital. Government perhaps?

Sounds great. No doubt they will publish a list of the pproblems

Sounds great. No doubt they will publish a list of thr pproblems they solve... Otherwise, it would be a scam.

They may publish a list of problems they are working on.

It's the journey that's important not the destination. :)

Did you guys read the article? While I'm generally pretty sympathetic to the argument that interdisciplinary centers are just a means to increase budgets, prestige, etc., IQSS at Harvard is really different (note: I have never been affiliated with Harvard). Gary and his team identified several needs of social scientists that weren't getting addressed and then provided them through IQSS. These include:

Academic Grant Writing Support
Research website templates, hosting and support (for which they developed OpenScholar, a super-respected project in the Drupal world)
IT Support
Data Hosting for reproducible research

Not only does IQSS fulfill these roles, but it also provides consulting and server-hosting for folks in the Harvard Social Science community.

The point of the article is that IQSS definitely isn't like other interdisciplinary centers. It offers new software and services useful to scholars at other universities, while also making life easier for academics at Harvard. That seems like a pretty valuable part of the university to me...

Sure it may be valuable. Question is, does writing a software & providing servers + consulting need opening of a new institute?

Ok, even if it does, is that a good reason to extrapolate that to a broad ambitious necessity for restructuring the social sciences?

Academic Grant Writing Support ...

First you make the system so complicated that even smart people with a Ph.D. can't make head or tail of it. Then you provide massive numbers of Admin to help said smart people cope. Then, no doubt, you have to impose another layer of management to manage all those Admin.

Yes, this is good for the university's bottom line. Which is not, actually, based on Admin. And people wonder why the ticket price of higher education is running away.

The job of academia is teaching and research. Administration should be there to support both those goals. It is not as many in higher education management clearly think that Admin is the purpose of higher ed. That is why they have all the nice offices. If they want to do something to help research, move the Admin out of the nicest buildings on campus, put them in temporary offices somewhere near the gym. Put the philosophy department in instead.

Apart from that, the only useful thing I can see worth offering to the Social Sciences is support from the Statistics Department so that their figures are not such garbage.

Another interpretation is that institutes like this are realizing economies of scale...

Love love love the "not just studying but solving problems" assertion in the first sentence (surprised anyone read further).

Coupled with the "dramatic transformation" the social sciences are going through, they must be "solving" a dramatically increased number of "problems". Let's see the list! ...

More like dramatically increased beauracracy, advocacy, and capture of public funds. Solutions? Not so much.

Well, Gary King himself (along with 2 coauthors) just published a very exciting paper using text analysis to study Chinese censorship. Paper wouldn't have been possible to write 10 years ago.

http://gking.harvard.edu/publications/how-censorship-china-allows-government-criticism-silences-collective-expression
and a piece from NPR:
http://www.npr.org/blogs/alltechconsidered/2013/09/09/219721983/its-ok-to-protest-in-china-just-dont-march

I hope no one is saying that this paper solves a problem.

It's the usual academic message: "Give me da money".

With the modern coda "now that we have vast data sets we will find the truth".

Do the people complaining about this in the comments do so because they aren't the ones receiving the money?

Comments for this post are closed