How Science Can Ensure Integrity and Quality: A Thesis

Society has high expectations for the quality and efficacy of science. Research is grounded in the faithful acquisition of new insights, a process in which results have to be discussed and reproduced before they can be accepted as new knowledge. This process should be under the purview of doubt, allowing self-reflection.

Because the acquisition of new insights in science is tied to our social lifestyles as well as the nature and technology that surround us, science cannot be decoupled from everyday life but should be recognized as “science for humanity.”

If science distances itself from these goals and no longer serves the quest for truth, the relevance of its results becomes questionable, because the whole system of science becomes vulnerable to dishonesty and scientific misconduct. Recent, more commonly publicized violations of the rules of good scientific practice have drawn attention to the possible underlying structural problems of science. These include excessive pressure to publish, the rapid entrepreneurialization of academic institutions, a heightened pressure to raise third-party funds, and the increasingly required staging and marketing of scientific results. A visually attractive presentation of research goals is nowadays occasionally considered more important than the competent, solid handling of the issues. Crowd-pleasing partial results trump the serious global view. If science is increasingly assigned to institutions or organizations, and not to the actual researchers, then there is a threat that the awareness of the responsibility for research results and their scientific relevance will be lost.

Suggestions for Safeguarding the Integrity and Quality of Science

Our system of science needs a strategy of “deceleration” that will again afford researchers the opportunity for critical reflection.

1. Contain the deluge of publications. The number of publications worldwide (relative to the increasing number of scientists) should be reduced, and with that— against the economic interests of publishers—the number of journals. Only in this way can we ensure that this important method of assessment of the quality of research will once again consist of results that have been reflected on and carefully evaluated. And only in this way will researchers be able sufficiently to take note of relevant results and insights in their areas of specialization.
2. Fundamental research requires a continuous funding stream. Science needs a continuous and reliable funding stream, because in its quest for the novel and for an understanding of nature, it follows radically different laws than business enterprises. Academic institutions clearly need to manage their budgets prudently. However, we have to be firm in confronting the expectation that science should make a profit or should be evaluated by highly economic criteria. Rather, we should strive together, even more than in the past, to demonstrate more clearly the high intrinsic value for the general public good of gaining scientific insights.
3. Put more weight on evaluation of the content of scientific achievements. In the process of awarding research grants, the substantive concepts and goals of grants must be evaluated on their own merits, without depending on promises of immediate success or translation of the research into practical applications. The qualitative evaluation of a researcher’s scientific work should count as much as the numbers obtained from a bibliometric analysis. The pure number of publications is not a reliable criterion.
4. Condemn strategic authorship. Authorship for a scientific publication demands substantial contributions to the content of the work to be published. Authorship has, in this day and age, become a currency of science, rewarded with money. In a system of achievement-based grant funding, the actual contributions of each author should therefore be investigated; merely strategic authorship without an ascertainable content contribution should be condemned.
5. Require researchers to write their own grant proposals. Raising third-party funds is an important competitive component of the system of science. However, because of the trend of requiring high funding contributions from third parties, the pressure to raise funds has increased to a degree that a system of professional grant writers has emerged, in which scientists do not write their grant proposals anymore, and, in extreme cases, outside consultants produce boilerplate proposals. The scientists themselves should write scientific conceptualizations. “Ghostwriters” should not be tolerated, not even in collaborative proposals in which the parts of a proposal written by scientists are “smoothed out” by outside consultants.
6. Make the data-collection process transparent. Science needs transparency, in spite of its increasing complexity. Fast technical progress in combination with an excess of competition leads to ever more complex experiments that are hard to verify. Without a transparent and careful description of the datacollection process and the scientific approach, the number of untraceable errors and dishonesties that endanger the substance of science increases.
7. Good research takes time. The development and execution of well-grounded projects are incompatible with short, time-limited contracts. The pressure engendered by the latter encourages researchers to perform minimal projects without substantial new insights and to publish them piecemeal. Only contract periods with reasonable time frames that provide the possibility for long-term project planning ([for the senior] as well as for the junior scientific staff) permit high-quality research that is indispensable in international competition.

The original document was funded by the Robert Bosch Foundation and published under the title “Wie die Wissenschaft Integrität und Qualität sichern kann.” Translated with permission. The German text and the list of signatories are available online at http://www.bosch-stiftung.de/content/language1/downloads/Thesenpapier_BWG.pdf.

James Geller is a professor at the New Jersey Institute of Technology whose research focuses on medical terminologies and the Semantic Web. He has published more than 140 research papers and was designated NJIT Master Teacher in 2005.