Bureaucrats have not only taken over much of the science world, but even the parts of the bureaucracy designed to hunt out corruption in science are incapacitated with bureaucracy-at-its-worst too. This is second order corruption — even the checks and balances on corruption are corrupted.
As James Delingpole points out: Science is rife with corruption, incompetence, dishonesty and fabrication–and now, thanks to a frank resignation letter by the US’s top scientific misconduct official we have a better idea why.
Government science desperately needs auditing– or the free market solution, competition
One in 50 scientists admitted to have fabricated, falsified or modified data or results at least once. It’s not just about fraud, it’s about bias, and statistical sloppiness. Up to 30% admitted other questionable research practices. When asked about their colleagues, 14% said other scientists falsified results, and 70% used other questionable research practices (Fanelli 2009). In the modern electronic science world, not only are many results not replicated, but the raw data itself is not even available for checking most of the time. Research shows that scientists who withhold data are more likely to have published errors (see below). Half of the papers in high-end journals contained some statistical error (Wicherts 2011).
What we’re seeing here is how even government funded checks on government funded science don’t work. Without free market competition and private funding, the layers of corruption and perverse incentives just build on the previous layers rather than neutralize them.
In this corrupt climate the need for independent checks is even more important
The director of the U.S. government office that monitors scientific misconduct in biomedical research has resigned after 2 years out of frustration with the “remarkably dysfunctional” federal bureaucracy. David Wright, director of the Office of Research Integrity (ORI), writes in a scathing resignation letter obtained by Science Insider that the huge amount of time he spent trying to get things done made much of his time at ORI “the very worst job I have ever had.”
Science Mag has the letter from David Wright:
“The rest of my role as ORI Director has been the very worst job I have ever had and it occupies up to 65% of my time. That part of the job is spent navigating the remarkably dysfunctional HHS bureaucracy to secure resources and, yes, get permission for ORI to serve the research community. I knew coming into this job about the bureaucratic limitations of the federal government, but I had no idea how stifling it would be. What I was able to do in a day or two as an academic administrator takes weeks or months in the federal government, our precinct of which is OASH.
On one occasion, I was invited to give a talk on research integrity and misconduct to a large group of AAAS fellows. I needed to spend $35 to convert some old cassette tapes to CDs for use in the presentation. The immediate office denied my request after a couple of days of noodling. A university did the conversion for me in twenty minutes, and refused payment when I told them it was for an educational purpose.
Wright describes OASH (Office of the Assistant Secretary for Health) as “secretive, autocratic and unaccountable”.
Third, there is the nature of the federal bureaucracy itself. The sociologist Max Weber observed in the early 20th century that while bureaucracy is in some instances an optimal organizational mode for a rationalized, industrial society, it has drawbacks. One is that public bureaucracies quit being about serving the public and focus instead on perpetuating themselves. This is exactly my experience with OASH. We spend exorbitant amounts of time in meetings and in generating repetitive and often meaningless data and reports to make our precinct of the bureaucracy look productive. None of this renders the slightest bit of assistance to ORI in handling allegations of misconduct or in promoting the responsible conduct of research. Instead, it sucks away time and resources that we might better use to meet our mission. Since I’ve been here I’ve been advised by my superiors that I had “to make my bosses look good.” I’ve been admonished: “Dave, you are a visionary leader but what we need here are team players.” Recently, I was advised that if I wanted to be happy in government service, I had to “lower my expectations.” The one thing no one in OASH leadership has said to me in two years is ‘how can we help ORI better serve the research community?’ Not once.
I’m offended as an American taxpayer that the federal bureaucracy—at least the part I’ve labored in—is so profoundly dysfunctional.
Data-sharing is a basic requirement of science, yet only 40% of the top journals even require the raw data to be shared as a condition of publication and of papers published in a journal with some kind of data sharing policy, less than half complied with the policy. [Scientific American] Only 10% of papers had their full primary data sets available upon publication. (Alsheikh–Ali, 2011) A different study claimed things are improving, but still only found 35% of gene expression articles with raw data available by 2009. (Piwowar, 2011)
In psychology, one analysis found the utterly predictable and obvious implication. Scientists who are reluctant to share data are more likely to be hiding evidence that contradicts their conclusions and also more likely to have published errors (Wicherts et al 2011)
Fanelli D (2009) How Many Scientists Fabricate and Falsify Research? A Systematic Review and Meta-Analysis of Survey Data. PLoS ONE 4(5): e5738. doi:10.1371/journal.pone.0005738 [abstract]
Ioannidis JPA (2005) Why Most Published Research Findings Are False. PLoS Med 2(8): e124. doi:10.1371/journal.pmed.0020124 [Abstract]
Piwowar, H. A. PLoS ONE 6, e18657(2011). [ PubMed ]
Wicherts JM1, Bakker M, Molenaar D. (2011)Willingness to share research data is related to the strength of the evidence and the quality of reporting of statistical results. PLoS One; 6(11):e26828. doi: 10.1371/journal.pone.0026828. Epub 2011 Nov 2. [Full article]