Despite their countries' vast differences in culture and history, research ethics experts from opposite sides of the globe agreed at a recent AAAS meeting: teaching scientific integrity needs to start early — perhaps as early as elementary school.
A dozen U.S. and Chinese experts gathered 8-9 October at AAAS for the China-U.S. Scientific Morality/Integrity Development Seminar, the fourth such meeting convened since the landmark Scientists' Social and Ethical Responsibilities conference organized by the China Association for Science & Technology (CAST) and AAAS in 2007.
"Unquestionably, scientific integrity is — if not the most important — among the most important issues that are facing the worldwide scientific enterprise," said Alan I. Leshner, AAAS CEO and executive publisher of Science, during his welcome address.
"The only way the scientific enterprise can contribute to worldwide problems is for . . . scientists to be willing, able, and enthusiastic about collaborating on a worldwide scale. Central to any collaboration are issues like trust, shared values, norms, and standards that drive the practice," he said.
In China, massive efforts are underway to strengthen scientific integrity in the face of exponential growth in academic activities, said Prof. Guangxian Li, Executive Vice President of Sichuan University, speaking on behalf of Prof. Shen Yan, Vice Chairman of CAST and Deputy Director of the National Natural Science Foundation of China.
Total research funding for Chinese universities ballooned from RMB$47 billion (US$7.6 billion) in 2009 to RMB$78 billion (US$12.7 billion) in 2012, Li reported. Meanwhile, research papers published by Chinese universities saw a 10 percent increase — to more than 1.1 million — during the same period, buoyed by an academic system that values reputation and rewards publication.
The desire to achieve higher international standings has put enormous pressures on Chinese researchers to publish, said Li. The QS World University Rankings andTimes Higher Education World University Rankings, for example, base 60-70 percent of their scores on research and citations. The Academic Ranking of World Universities, produced by China's own Shanghai Jiaotong University, is entirely based on research output.
But the Second National Investigation of Scientific Researchers in China, a 2012 national survey of more than 30,000 Chinese researchers, found 51 percent of respondents admitted to frequently or occasionally improperly attributing sources, while 42 percent submitted manuscripts to multiple journals. Another survey of nearly 5,500 graduate students from 24 universities, conducted a year later, revealed a similar trend.
To that end, an ambitious national campaign launched 2011 by the Chinese Ministry of Education, Li said, is driving home the issue: Six thousand graduate students from 60 universities attended annual lectures on research ethics, held at the iconic Great Hall of the People. In less than four years, a staggering 20,000 lectures on the topic have been delivered to 4.4 million graduate students, 4 million undergrad students, and 37,000 young teachers and scientists.
Training in responsible research conduct is essential, but training alone is not enough, said Philip Langlais, professor of psychology and former Vice Provost of Graduate Studies & Research at Old Dominion University.
A 2006 survey he conducted with 222 faculty and 534 graduate students found that while up to 90 percent of faculty reported providing training in responsible research conduct, up to 35 percent of students claimed they received no training. Only 8 percent of students and faculty reported having discussed and reached consensus on topics such as authorship, ownership, and use of data.
The study also revealed that 70 percent of respondents never discussed misuse of research funds, 48 percent never spoke about "cooking" research data, and 56 percent never deliberated the implications of failing to present contradictory data.
"The lack of a clear understanding and a mutual agreement are frequently associated with disputes regarding authorship and use of data," Langlais said in an email.
"We've been targeting individuals, but not other layers of influence," he added. "Scientists who have received proper training in responsible conduct of research have violated best practices and principles of scientific integrity as a result of a 'toxic' research environment." Higher workloads and compliance burdens, shifting university priorities, increasing competition for funding, and changing societal expectations all require researchers to produce more while fighting temptations to cut corners.
Elizabeth Heitman, professor of medical ethics at the Vanderbilt University Medical Center, asked if the current focus on teaching responsible research conduct during graduate and postdoctoral years and the one-size-fits-all curriculum may be "too little, too late."
"We need to go back to elementary school, when they're learning about how things work," said Heitman, recalling a nine-month curriculum her daughter's fifth-grade science teacher designed: "The first two weeks introduced the concepts of research and discovery, the scientific method, the need for full and honest reporting and the threat of self-deception and bias."
Diange Yang, professor in Tsinghua University's Department of Automotive Engineering and Deputy Dean Responsible for Research Work and International Cooperation, concurred. Currently on an academic exchange at the University of Michigan, Yang recently attended his first American PTA meeting for his nine-year-old daughter.
"Written in big letters on the wall of the classroom was the definition of integrity: to act according to what is right and what is wrong," Yang recalled.
"We should encourage intelligent risk-taking and tolerate failure in research," he added. "Strengthening academic integrity education of graduate students . . . should be the responsibility of the whole society, and it should be done from childhood."
"It will be a long and difficult journey," he added.
Langlais agreed: "You can't change climate and culture overnight. You just don't flip a switch and make a change, but you can make small incremental changes."