The reform of classification evaluation is progressing steadily, and the tendency to rely too heavily on papers is being reversed. On February 23, the Ministry of Science and Technology issued a notice on “A number of measures (trial) on the bad orientation of “only the papers” in science and technology evaluation (hereinafter referred to as “measures”), referring directly to the scientific research evaluation system, put forward targeted and effective measures, to break the paper “SCI first” “only the thesis” dysentery, and establish a correct evaluation orientation. This series of initiatives related to the reform of the evaluation system has encouraged the confidence of the vast number of science and technology workers. At the same time, some voices of doubt have also emerged on the Internet.
Reporters found that the controversial for several specific measures, including the implementation of the paper evaluation of the representative system, the publication of not less than 1/3 of the domestic scientific and technological journal papers, strengthen the representative for peer review.
How can we more effectively remove the “paper-only” orientation and make the policy work? Scholars in the scientific and technological fields have come up with suggestions.
The representation system is not “representative”?
In an interview with the China Science Daily, Zhou Zhihua, a professor at Nanjing University, said that he is based at Nanjing University in the computer science, in recent years in recruitment, job title evaluation and other aspects have adopted the “representative work reporting” system.
“SCI itself is a document retrieval system, and SCI papers do not represent high-level papers. Top conference papers that are most important in computer science are not usually retrieved by SCI. He said.
In response to the “representative system for the evaluation of the thesis” pointed out in the Measures, Zhai Liping, a professor at Zhejiang University of Industry and Commerce, believes that the regulation is not universal.
“It’s okay to use representative work by super-class scientists, but for the other 95 percent or more of the average scientist, the representation does not reflect the problem, or needs to be quantified.” “And it’s closely related to the category of disciplines, ” He told The Chinese Science Daily.
He pointed out that for economics, management and other disciplines, representative works can not represent the level of scholars, for the development of slow-moving basic disciplines is not suitable.
In the view of Yan Da, a researcher at Yunnan University, the paleontological research he is engaged in belongs to the “slow development” of the subject. The representative composition system or can not reflect the original can not issue a high-level journal papers of many basic work value.
“For example, in some cold-calling bio-taxation studies, researchers need to do a lot of field collection, specimen finishing and production, morphological observation and statistics, drawings. Finally, the results of the publication are usually some ‘not high quality’ research papers – the results of the work of taxonomy are generally published in the journals of the three or four districts. Compared with those in the hot direction of the researchers, in the review is certainly at a disadvantage. He said.
“The representative system is very good, each take the director is generally acceptable, can avoid the repetition of low-quality results to the reviewer’s problems. Ma Jun, a professor at Lanzhou University of Technology, told The China Science Daily that how to rank representatives is a key issue.
“For example, A selects SCI papers as representative works, B selects invention patents as representative works, C selects monographs as representative works, D selects scientific and technological achievements as representative works, if ABCD also applies for competitive funding or talent projects, how to screen their position will be challenging. He said.
In addition, the system of representation should reflect periodicity, not rely on a representative for “all-eating”.
“If a scholar had published a paper in Nature, there would have been no basic scientific output for the next 10 years or more. In this case, how can researchers be monitored and motivated? Ma asked questions about this.
Peer review to see “human feelings”?
“Measures” put forward in the “domestic scientific and technological journal papers should be in principle not less than 1/3”, also aroused public concern and discussion.
For a long time, publishing papers in academic journals, especially core journals, has become the main or even the only condition for university, hospital and other units to evaluate titles, promotions, awards and graduate students.
Under the guidance of “reviewby”, Chinese academic circles pay more than 2 billion yuan a year to foreign countries for the layout of their papers.
In Mr Yu’s view, these are the “costs” of China’s international exchanges. It will take some time for local periodicals to develop into high-quality periodicals with international influence, especially in English.
“The policy is now in operation, and scholars may not be able to publish Chinese papers. Even if there are Chinese papers may not be of enough quality, because previous excellent papers are issued in English journals. “The policy should have a transition period of one to two years, ” Mr. Yu suggested.
At the same time, the “capacity” of domestic science and technology journals has become a concern for some scholars. Han Guangjie, a professor at Dalian University of Technology, told the China Science Daily that “domestic scientific and technological journal papers should in principle be no less than one-third” of the publication requirements or lead to the “monk more porridge less” situation.
He introduced that the core journals of computer science include the Journal of Computer Science and the Journal of Software. There are a limited number of periodicals and a limited number of articles published each year, but the number of people working on academic research is in the hundreds of thousands. “If the papers are turned to local journals, the hiring rate will be greatly reduced, resulting in fierce competition. “
In the interview, many experts pointed out that the “enhanced representative peer review” proposed in the Measures is their biggest concern. On the contrary, sci paper comes with the qualitative evaluation function, on behalf of the peer evaluation of the operation of the fair ness is not yet known.
How do you define and screen these peers? Do officially designated peers really have academic influence in this field? The division of labor in scientific research is getting more and more detailed, how to find interdisciplinary interdisciplinary scholars to act as peers? How is the academic conduct of peers supervised? In the interview, many experts have thrown out the above questions.
“Even those who have won the title of the national title of the Yangtze River scholars, Jie Qing to be peer judges can not be fully convinced.” Moreover, these national titles do not have much time to participate in a wide variety of academic evaluations. Ma Jun said.
He suggested that peer evaluation should not be limited to domestic experts within the scope of evaluation, relevant departments may consider combing the domestic database of internationally renowned experts in various fields, these well-known experts to consult, so that peer evaluation internationalization.
Strengthening the Construction of the Scientific Community
Reporters found that, although the specific implementation methods are still to be implemented in detail by universities and scientific research institutions, but the vast majority of scholars on the “measures” of the introduction of positive attitude.
Yu Dasaid that the introduction of the “not only SCI paper” proposal is particularly suitable for the application of science and technology workers and doctors, engineers and so on. “I’ve heard doctors say that their job title is also largely dependent on the SCI paper, which I think is unreasonable. “
Regarding the other programs involved in the Measures, Ma Jun believes that many reflect the progress of the scientific research system.
“For example, it would be nice to give a clear opinion on the amount of paid publishing for the reimbursement of scientific research expenses, which is a very big blow to the huge amount of foreign exchange that international ‘predatory journals’ get from low-level Chinese researchers.” This has a good enlightenment to further standardize and guide the domestic periodical charging standard. He commented.
“The introduction of such a document by the Ministry of Science and Technology is positive for reversing the orientation of the ‘paper-only’ system. Xue Lan, director of Tsinghua University’s China Science and Technology Policy Research Center and dean of the Su Shimin School at Tsinghua University, told the China Science Daily that the state has made great efforts to introduce measures that seem to be more reasonable, but different research institutions and different research fields may have different evaluation criteria because of their different evaluation functions, and it is difficult to use a set of evaluation standards.” The golden rule of jade “unity.”
He said that the establishment of a reasonable scientific and technological evaluation system, the core is to strengthen the construction of the scientific community, so that various disciplines and scientific research institutions in accordance with the characteristics of the discipline and the needs of the institution, gradually set up a more objective and fair evaluation system.
The second is to minimize the number of evaluations of all types.
“At present, too many evaluation projects, hat reviews, too many disciplines is too indiscriminate. In order to ensure fairness, a project to arrange seven or eight people to evaluate, so that the real peer review is completely diluted by the outsider’s review, and the layman review in the case of unfamiliar with specific areas of expertise, can only count papers. He said.