Production of Scientific Research

Paula StephanGeorgia State Universitypstephan@gsu.eduEuropean Summer School on Industrial DynamicsSeptember 2006

Overview

How science is produced Why science is produced Provide some empirical examples Draws on material from reading list Focus primarily on scientists working in the

“public” sector One goal: convince you that generalizing

from research practices in economics to those in science can lead to faulty conclusions

How is Science Done: Necessary Ingredients

Production of scientific research involves: Scientists: bring effort and cognitive

resources Equipment: some extremely expensive,

some paper and pencil. Colleagues: importance of teams in

science Serendipity: Can play a role

How Science Is Done

R=f(Cg,R, t, e) Cg= cognitive resources R=other resources, such as equipment, lab

assistants t=time of researchers e is some error term, encompassing among other

things serendipity and uncertainty. “Any new idea … cannot be fully mastered, developed into the stage of a tentatively acceptable hypothesis, and possibly exposed to some empirical tests without a large expenditure of time, intelligence, and research resources.” Stigler

Scientist

Effort Science takes time; common observation is

that scientists work exceptionally long hours. Also requires motivation. “Informed

observers have long described high-producing scientists as driving and indefatigable workers.” (Fox.)

Dimensions of Cognitive Resources

Ability: studies document that as a group scientists have above average IQs.

Knowledge base: How well educated is scientist? Does scientist keep up? Raises possibility of obsolesence and related vintage effects.

Possibility that “too much” knowledge can be a bad thing. Encumbers discovery. “Outsiders” as exceptionally productive.

Cognitive resources enhanced by a team and by networks

Equipment

Science heavily influenced by availability of technology and access to equipment

Exceptions exist (David Quéré story) But, increasingly science requires access to equipment

In genetics: DNA gene sequencer and synthesizer, protein synthesizer & sequencer comprise the technological foundation for contemporary molecular biology. DNA sequencer has revolutionized genomics by allowing the rapid automated sequencing of DNA

Super Computers tunneling microscopy—key in nanotechnology Accelerators Cell lines Mice

Non-linear Model

Importance of equipment is one reason to stress non-linearity of scientific discovery

Not just that science affects technology Technology very much affects science:

The history of science is the history of how important resources and equipment are to discovery. Theme in research of Nathan Rosenberg; Joel Mokyr.

Cost

Equipment is expensive Examples: Sequencer 35,000 euros;

tunneling microscope $1 million plus; Animals: off the shelf “inbred” mouse is $50

per mouse; common to have facility with 1,000 mice or more.

Mice “per diem” $.65 per day A transgenic mouse for a specific disease

can be around $2000. Example follows

Joslin Mouse from JAX used to study Insulin

Tg(Ins2-tTA)1Doi encodes the tetracycline regulatable transactivator (tTA) regulated by the rat insulin promoter (Ins2, commonly designated RIP), and is expressed in the pancreatic beta cells. Mice hemizygous for this transgenic insert are viable, fertile, normal in size, display normal NOD diabetes onset, but do not display any other gross physical or behavioral abnormalities. When NOD.Cg-Tg(Ins2-tTA)1Doi/DoiJ mice are mated to a second transgenic strain carrying the target gene regulated by the tetO responsive elements, expression of the target gene is turned off by the tetracycline analog, doxycycline (dox). Dox can be administered orally in the food or water. This strain provides a Tet-Off tool that permits the inducible expression of genes in the pancreatic beta cells during various stages of diabetes development.

Equipment continued

Access to equipment is an issue How scientists working in the public sector gain

access to equipment varies by country Necessary condition is generally having a

research position: leads to cohort effects Research positions extremely scarce in many

countries currently; cause for concern Indivisible: accelerator; telescope Role of patenting in gaining access—rights can

be “traded” among researchers. Often acknowledged as a reason for seeking a patent.

US: Public Research occurs in Universities

Universities invest heavily in scientists they recruit. Start-up package for Assistant Prof in

chemistry is $489,000; in biology it is $403,071. High end is $580,000 in chemistry; $437,000 in biology.

Start-up for senior faculty: Chemistry is $983,929 (high-end is $1,172,222); biology it is $957,143 (high end is $1,575,000). (Ehrenberg et al)

U.S. Funding continued

After startup, funding is responsibility of scientist Grants play major role in U.S. science Scientist(s) apply; peer reviewed. Grants are not a “sure” thing; necessary condition

for tenure at research universities—currently expectation to buy out 70% of PI’s time.

“Preliminary data” important; track record important in terms of publications and honors

Access to space and equipment important Who trained you and who you trained are often

noted Common to review center grants PI by PI.

NIGMS R01 Applications

Funding Curves for NIGMS Type 1&2 R01’s:FY2000-2005

Grants take time to write and administer

2006 survey of U.S. university researchers Report spending 42% of “research time” on

administrative tasks Split between pre-grant (22%) and postgrant (20%).

Most time consuming Filling out grant progress reports Hiring personnel Managing laboratory finances

Recent changes have increased time requirement Health privacy laws Institutional review boards Accounting for “select agents” after 9/11

Decker: Northwestern University

Time for Peer Review

NIH grants each have two principal peer reviews and a third “reader” review; principal reviews are generally three to four pages

Conclusion: scientists in U.S. spend considerable time reviewing proposals as well as writing proposals

Other funding sources in U.S. for faculty directed research

Industry State University itself

Other Models for Funding “Public” Research

Funding from government comes through Research Institute; researcher often also holds a faculty position Grants and contracts can come from other

sources but don’t play same role—in terms of tenure/buying out salary and hiring staff for lab.

Funding distributed according to productivity of individual, lab, department or university

Colleagues

Others will speak about importance of colleagues and networking

But important to note that most science is not done in isolation.

Scientists work in labs Lab has staff and scientists of various rank How these positions are hired varies considerably

across countries Whether these positions are “permanent” also varies. U.S. has staffed laboratories through what could be

thought of as a pyramid scheme. Scientists network outside their lab

Staffing Issues

Staffing labs with doctoral students and postdocs provides a ready flow of “new” ideas and “temporary” workers.

Produces more than university can absorb; movement of scientists from academe to industry is a major way in which knowledge spills over.

Staffing labs in this way produces problems of over-supply if the other sectors of the economy cannot readily absorb the production of new PhDs

Collaboration Increasingly Important

Systems biology good example: physics, biology, engineering, computer science. Proximity important here.

One indication of this is change in mean number of authors on scientific papers.

Evidence Concerning Teams

Figure 4--Mean Number of Authors per Paper, for PapersWith at Least One Author In the Top 110 U.S. Universities, 1981-1999:

Adams et al 2002

2.40

2.80

3.20

3.60

4.00

4.40

81 84 87 90 93 96 99

Year

Aut

hors

Per

Pap

er

Networks and Funding

Government agencies have bought heavily into the importance of networks

Network funding in EU Network funding at NIH

Serendipity—the Happy Accident

Unanticipated events occur Follow up is not accidental: “Chance

favors only the prepared mind.” Pasteur

Example of Serendipity—short bio of Robert Richardson, Nobel Laureate, Physics 1996

“He (Richardson) obtained his PhD degree from Duke in 1966. His thesis advisor was Professor Horst Meyer. In the Fall of 1966 he began work at Cornell University in the laboratory of David Lee. Their Research goal was to observe the nuclear magnetic phase transition in solid 3He that could be predicted from Richardson’s thesis work with Horst Meyer at Duke. In Collaboration with Douglas Osheroff, a student who joined the group in 1967, they worked on cooling techniques and NMR instrumentation for studying low temperature helium liquids and solids. In the fall of 1971, they made the accidental discovery that liquid 3He undergoes a pairing transition similar to that of superconductors. The three were awarded the Nobel Prize for that work in 1996.”

Why Science is Done

The Puzzle

The Ribbon

The Gold

Evidence Concerning the Puzzle

Richard Feynman—Nobel laureate in physics—”Once I get on a puzzle, I can’t get off.”

Robert Noyce described as having “an unbounded curiosity.”

Jack Kilby “fell in love with the creative process of discovery.” “I discovered the peer joy of inventing.” 

Feyman: p. 9, “Surely You’re Joking Mr. Feynman.”Mean from Technology Review, p. 78, September 2004

Priority

Merton argued that the goal of scientists is to establish priority of discovery by being first to communicate an advance in knowledge

Rewards to priority are the recognition awarded by scientific community for being first.

Interest in priority is not new—has been an overriding characteristic of science for over 3.5 centuries.

Recognition Depends on Being First

Being first requires sharing of knowledge No reward for being second or third;

science is a winner-take-all contest (congruent with zero social value of second and third place)

Not unknown to write and submit article in same day.

Argue with journal editors over priority Negotiate to be first on program

Recognition continued

Where the publication appears is important: top journal or lower tier journal.

Difficult to publish in top tier journals—some have acceptance rates of less than 10%.

Recognition Awarded Priority

Publication— necessary step. This reward is within the reach of most scientists.

Eponymy—practice of attaching name of scientist to a discovery.

Prizes—Nobel heads list, but 100’s of others exist; Lemelson-MIT Prize; Lasker Prize

Societies for luminaries—Academie des Sciences in France, NAS in U.S.

Recognition/reputation can be measured through citation analysis

Priority as ownership

Being first to publish bestows ownership on author Unusual kind of ownership in sense that what you own the

rights to (it’s yours because of priority) is not diminished by use of others;

Indeed, the more others use it, the more famous scientist becomes

Note: reward of priority only comes by disclosing what you have discovered. You cannot obtain priority without publishing. Darwin as an example

Priority provides incentives to produce something that others will use and you can’t exclude them from using the knowledge once it has been published—the “fee” they pay is the citation

Functionality of Priority

Hinges on two observations: Knowledge is a public good Scientists have a choice as to whether or

not to share knowledge

Scientific Research has Properties of Public Good

Knowledge, like lighthouse, not depleted with use—thus marginal cost of use approaches zero.

Once made public, others cannot be excluded from its use.

Economists (Arrow; Dasgupta & David, Johnson) have recognized this

Others long before, such as Jefferson

Thomas Jefferson

…an individual may exclusively possess (an idea) as long as he keeps it to himself; but the

moment it is divulged, it forces itself into the possession of everyone, and the receiver

cannot dispossess himself of it. Its peculiar character, too, is that no one possesses the

less, because every other possesses the whole of it. He who receives an idea from me,

receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening mine.

Implications for Science of Public Nature of Knowledge

Public nature of knowledge suggests that scientists cannot appropriate the benefits derived from discoveries.

Corollary is that knowledge will be under produced.

Merton and Dasgupta and David show that the reward structure of science, based on priority, functions to make the public good, scientific knowledge, private.

How Priority Solves the “Public Goods Problem”

Merton (1986) stated in Ghent: “I propose the seeming paradox that in science,

private property is established by having its substance freely given to others who might want to make use of it.”

“Only when scientists have published their work and made it generally accessible, preferably in the public print of articles, monographs, and books that enter the achieves, does it become legitimately established as more or less securely theirs.”

Solution

It is the public nature of knowledge that facilitates building of reputation in science.

Quickest way for scientist to establish a reputation among peers is to “share” knowledge by placing it in the public domain, preferably in print.

By sharing, idea is established as private property of the scientist.

Incentive to do so quickly, before others do so.

Citations Play a Special Role

Other scientists cite scientist’s work, thereby endorsing the work.

Process encouraged by a user fee that approaches zero.

Two part role of citations: bestowing property rights on one who made discovery; communication information.

Priority Solves Other Problems

Shirking is rarely a problem in science. Knowledge that multiple discoveries are commonplace makes scientists exert considerable effort; monitoring rarely a problem.

Priority requires peer evaluation, which discourages plagiarism and fraud and builds consensus.

Money

Scientists also are interested in economic rewards. Stephen Jay Gould says scientists want “status,

wealth and power, like everyone else.” Henry Rosovsky asked one of Harvard’s most

eminent scientists the source of his scientific inspiration: reply (“which came without the slightest hesitation”) “money and flattery.”

Financial Rewards

Salary Consulting fees Prize money—some prizes have very

large purses:—Lemelson-MIT Prize is $500,000; Nobel over 1 million euros.

Royalties from patents Stock options

Compensation in Science

Winner-take-all nature of science places much of the risk on shoulders of scientist.

Compensation is generally composed in two parts: one portion paid regardless of individual’s success in races; other is priority-based and reflects value of winner’s contribution to science.

Find salary in academe in U.S. related to number of articles published and citations

Other Financial Rewards

Consulting income Bonuses related to publication in

prestigious journals—currently practiced in China

Rewards associated with start-up companies

Royalties from patents Bonuses for receiving grants

Beneficial Shares University Scientists Classified as “Insiders”52 Biotech Firms (Stephan & Everhart)

Value in US-Dollars Numbers of Scientists

<$100,000-500,000 5

500,001-1,000,000 6

1,000,001-2,000,000 2

2,000,001-3,000,000 4

3,000,001-4,000,000 5

4,000,001-5,000,000 6

5,000,001-6,000,000 2

6,000,001-9,000,000 6

>9,000,000 4

Emory Example

July 2005 Gilead Sciences, Inc. and Royalty Pharma bought Emory’s royalty interest in emtricitabine, also known as Emtriva® used in treatment of HIV

Emory received $525 million in cash Prior to deal, Emory had been receiving royalty

income since licensing the drug in 1996 Three Emory scientists involved: Dr. Dennis C.

Liotta, Dr. Raymond Schinazi and Dr. Woo-Baeg Choi

Emory’s intellectual property policy in effect at the time awards something like 40% of the amount to the three inventors.

Changing Role of Financial Incentives

Patenting is becoming increasingly important at universities and research institutions.

Monetary and resource rewards attached to productivity – especially outside U.S.--are changing incentives

Also present in US: “bonus” for research grants or loss of job without grant

Changes in Incentives May Explain Relative Decline of US Science

Number of articles published by US scientists has declined in recent years.

Big news: New York Times Why?

One argument is that incentives have changed in rest of world—both individual incentives and institutional incentives for funding.

Result is that competition from outside U.S. has increased.

Number of U.S. Papers 1981-1999

Adams et. al

Summary of how and why:

Science is rarely done in isolation; generally requires equipment and colleagues.

Scientist contributes effort and cognitive resources. In university setting in U.S. also responsible for getting the resources.

Scientists are motivated to “do” science by a desire to solve the puzzle, be recognized & earn rewards.

Inequality in Science

Defining characteristic of winner-take-all contests is extreme inequality in allocation of rewards. (Frank and Cook)

Science has extreme inequality Highly skewed nature of publications observed by

Lotka more than 75 years ago. Lotka found that approximately 6% of publishing

scientists account for 50% of articles written. Recent work shows distribution of patents to be

even more skewed. Come to group 4 for examples!

What Leads to Inequality?

Differences in ability and motivation—what we could call differences in possession of the “right stuff” or what some call “sacred spark.”

Cumulative advantage—what Merton called the Matthew Effect: “the accruing of greater increments of recognition for particular scientific contributions to scientists of considerable repute and the withholding of such recognition from scientists who have not yet made their mark.”

Right Stuff or Matthew Effect?

Need longitudinal data for studying scientific productivity at individual level.

Matthew Effect is consistent with work in winner-take-all contests. Frank and Cook observe that “in all their manifestations, winner-take-all effects translate small differences in the underlying distribution of human capital into much larger differences in the distribution of economic reward.”

Estimating Productivity

Issues involved: Cohort effects Vintage effects Fixed effects Disciplines differ considerably Highly skewed distribution; many zeros