Has scientific development slowed down today?

According to the report of the Natural Science Foundation of the United States, the world's scientific research output grew at an average annual rate of about 3.8 per cent from 2008 to 2018, based on the number of peer-reviewed scientific and engineering journal papers and conference papers. By 2018, the "output" had exceeded 2.5 million. In terms of the number of papers published, the scale of scientific research today is much larger than in the past. However, after analyzing a large number of published papers, a study published in the Proceedings of the National Academy of Sciences found that the large number of papers published does not reflect the development of science, on the contrary, they even hinder the development of science.

The researchers first investigated the citations of the paper. They obtained more than 90 million articles and more than 1.8 billion citation records from 1960 to 2014. The analysis results show that there is obvious inequality in the citation of papers, that is, a small number of papers have been cited a large number of papers.

The study also found that the more articles published in a field, the lower the exhaustion rate of highly cited papers. In other words, papers that were heavily cited in the previous year are more likely to be heavily cited in the next year, and their "popularity" will hardly decline with the passage of time. This shows that the papers with the highest citations have attracted the attention of most researchers, while the newly published papers are difficult to enter the field of vision of researchers. This is a "class solidification" phenomenon in the field of scientific research, and the more papers published in a field, the more serious this phenomenon will be.

After analyzing the concerns of the researchers by citation, this study continues to analyze the content of the paper. The study found that newly published papers tend to develop and improve existing ideas, rather than put forward innovative ideas that break with conventional ideas. And the more papers published in a field, the greater the proportion of papers with conservative tendencies. As a result, most researchers have repeatedly ploughed within the well-known theoretical framework, while few have tried to open up a new world.

The researchers concluded that the continuous emergence of a large number of new papers in a research field will make it impossible for reviewers and other researchers to study the articles carefully and identify truly valuable innovations, thus hindering the update and progress in this field. From this point of view, the number of papers published does not well reflect the level of scientific development.

So in order to evaluate the level of scientific development in today's world, what other criteria should we pay attention to? in addition to the number of papers published, there are many other forms of scientific research output. for example, the number of technology patents, the improvement of indicators of new products (such as the number of new drugs and the number of transistors on microchips), and so on, all need to be taken into consideration. Before getting the output, a lot of resources such as labor force and research funds need to be invested in the early stage of scientific research. Considering the input and output of scientific research comprehensively, the concept of "scientific research productivity" is obtained. This indicator can more truly reflect the development of science than just considering the number of papers published.

# productivescience

Whether scientific research productivity in today's world is higher or lower than in the past Merck commissioned the Oxford School of Economics (Oxford Economics) to conduct a survey of scientific research productivity around the world today. The survey interviewed a number of experts in the field of R & D around the world, as well as 3500 scientists from 7 countries, made an econometric analysis, reviewed the existing literature, and obtained a report on the current situation of scientific research productivity.