9512.net
甜梦文库
当前位置:首页 >> 公务员考试 >>

The role of patenting activity for scientific research A study of academic inventors from China's na


Journal of Informetrics 4 (2010) 338–350

Contents lists available at ScienceDirect

Journal of Informetrics
journal homepage: www.elsevier.com/locate/joi

The role of patenting activity for scienti?c research: A study of academic inventors from China’s nanotechnology
Gangbo Wang a,1 , Jiancheng Guan b,?,1
a b

School of Management, Beijing University of Aeronautics and Astronautics, Beijing, PR China School of Management, Fudan University, Shanghai, PR China

a r t i c l e

i n f o

a b s t r a c t
Scientists from universities are becoming more proactive in their efforts to commercialize research results. Patenting, as an important channel of university knowledge transfer, has initiated a controversy on potential effects for the future of scienti?c research. This paper contributes to the growing study on the relationship between patenting and publishing among faculty members with China’s evidence in the ?eld of nanotechnology. Data from top 32 most proli?c universities in patenting are used to examine the relationship, consisting of 6321 con?rmed academic inventors who both publish and patent over the time period 1991–2008. By controlling for heterogeneity of patenting activities, patenting experience, institutional af?liation and collaboration with foreign researchers, the ?ndings in China’s nanotechnology generally support earlier investigations concluding that patenting activity does not adversely affect research output. Patenting, however, has negative impacts on both quantity and quality of university researchers’ publication output, when the assignee lists include corporations or scientists themselves. ? 2010 Elsevier Ltd. All rights reserved.

Article history: Received 28 October 2009 Received in revised form 1 February 2010 Accepted 8 February 2010 Keywords: Academic inventors Patent-publication tradeoff University–industry relations Nanotechnology

1. Introduction The interacting university–industry relationship has been recognized by not only scholars but also policymakers or practitioners as one of the most important characteristics in a knowledge-based economy. Scienti?c activities have increasingly played an important role in industrial innovation and more ?rms are relying on external sources of scienti?c knowledge generated mainly by universities. Besides other channels of university knowledge transfer like consulting, sponsored research, licensing and spin-offs, university patenting has long been a topic of keen interest in the literature and policy initiatives. The past few years has seen a surge in the number of patents which are generated by academic scientists and granted to universities. More and more scientists produce results which can be both published in academic journals and applied for ?ling patents. In recent years, the growing studies have focused on investigating the impacts of academic patenting for the future scienti?c research. Although the understanding of the effects of university patenting on scienti?c research remains open to debate theoretically, a large body of empirical studies on evaluating statistically the relationship between patenting and publishing have provided strong evidence that there is no negative effect of patenting activities on publication output of individual academic scientists, especially for star scientists (Agrawal & Henderson, 2002; Azoulay, Ding, & Stuart, 2006;

? Corresponding author at: School of Management, Fudan University, Shanghai, PR China. Fax: +86 21 65642412. E-mail addresses: guanjianch@fudan.edu.cn, guanjianch@buaa.edu.cn, guanjianch@sina.com (J. Guan). 1 These authors contributed equally to this paper. 1751-1577/$ – see front matter ? 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.joi.2010.02.002

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350

339

Azoulay, Ding, & Stuart, 2007; Breschi, Lissoni, & Montobbio, 2007; Breschi, Lissoni, & Montobbio, 2008; Buenstorf, 2009; Calderini, Franzoni, & Vezzulli, 2007; Carayol & Matt, 2004; Fabrizio & Di Minin, 2008; Meyer, 2006a, 2006b; Murray & Stern, 2007; Van Looy, Callaert, & Debackere, 2006). The general ?nding is that patenting activity does not affect publishing activities. Meyer (2006a, 2006b) showed that patenting faculty members apparently outperform their non-patenting peers in terms of both quantity and quality of publication in the ?eld of nanotechnology. As Klitkou and Gulbrandsen (in press) point, the impacts of patenting activity and other types of commercialization for the scienti?c output are highly context-dependent in the national, university or disciplinary level. A large body of investigations or case studies have concentrated on developed economies, e.g. Meyer (2006a) for three European countries, Murray and Stern (2007) for US, Czarnitzki, Gl?nzel, and Hussinger (2009) for Germany, Breschi et al. (2008) for Italy, Klitkou and Gulbrandsen (in press) for Norway and Chang and Yang (2008) for Taiwan. Little is known, however, about China’s status. To address this, we examined the effects of faculty patenting behavior in a panel dataset of nanotechnology scientists employed at Chinese 32 universities. These universities are representative because they are most proli?c in patenting with the patent number larger than 50 during the period of 1991–2008. This ?eld was chosen for three reasons. First, it is widely acknowledged that nanotechnology, as an emerging and rapidly evolving ?eld with the multidisciplinary nature, is perceived as proximate ?elds of science and technology (Meyer, 2006b). Scientists engaging in this ?eld may have the disciplinary advantage of both publishing and patenting their discoveries. Second, scienti?c research and technological development in China’s nanotechnology has attracted considerable attention from scholars and policymakers all over the world in the past few years. China has emerged as one of the key global players in this ?eld, producing the second largest number of nanotechnology papers following only the United States (Guan & Ma, 2007) and ranking third behind only the United States and Japan in terms of the number of nanotechnology patents granted (Liu & Zhang, 2005). Third, nanotechnology has been identi?ed as a main component and a priority mission area in China’s strategic plans for future developments in science and technology and has been given a high level of investment and signi?cant support from central and local governments (Bai, 2005; Hassan, 2005). The reinforce effect or the con?ict effect of patenting on future scienti?c research is increasingly central and of great interest to policy makers and university leaders. Therefore investigating an integrated quantitative perspective on this issue will provide an answer for them. We focus on the following factors in the context of China. First, we attempt to explore a potential difference between different levels of supporting by governments, by considering whether a researcher comes from the key university and the State Key Laboratory. The key research universities have established themselves as an important source of knowledge for ?rms (Wu, in press). At the same time, the State Key Laboratories have played a vital role in China’s scienti?c research system (Xue, 2006; Jin et al., 2006). Researchers from there are in a specialized and well-equipped environment and may face with better institutional culture. Second, China’s patent laws are designed to grant Intellectual Property Rights (IPRs) on public inventions to the employers emulating the Bayh-Dole Act.2 The regulation on protecting IPRs of higher education institutions established by Ministry of Education also points that the IPRs of employment inventions produced by a researcher belongs to his af?liated university (Ministry of Education, 1999). Thus most patents invented by faculty members are granted to universities in China. However, there is still heterogeneity of patenting activity (Czarnitzki et al., 2009), such as patents assigned to corporations due to joint or contract research, and assigned to the scientist himself besides the university under certain agreements on sharing revenue. We explore the effects of heterogeneous patenting activities on the scientists’ publication output by distinguishing both instances. Third, international scienti?c collaborations should be controlled in our model since China has bene?ted greatly from international scienti?c collaborations in improving its research (Guan & Ma, 2007). Following several studies on matching the data of publications and patents (Breschi et al., 2008; Boyack & Klavans, 2008; Meyer, 2006a, 2006b), we established inventor–author links and con?rmed 6321 academic inventors who both published and patented in the ?eld of nanotechnology over the time period 1991–2008. To further explore the publishing–patenting relationship with respect to China’s context, we performed the ?xed effects Poisson model. The remainder of this paper is organized as follows. In the next section, we summarize the empirical evidence on the relationship between patenting and publishing to develop our research hypotheses and also describe the related policies on academic patenting in China; Section 3 introduces China’s nanotechnology. The dataset and the model used in the paper are described in Section 4. Section 5 presents the results and analysis. Section 6 discusses the ?ndings and their implications; some directions for further study are suggested. 2. Backgrounds and hypotheses 2.1. The relationship between patenting and publishing Universities had long been seen as open science organizations, providing direct contributions to the creation and public dissemination of knowledge. Over the last 30 years, these tradition missions have been challenged by some emerging

2 We mentioned the Bayh-Dole act to show the fact that China has implemented the Bayh-Dole-like laws allowing universities to appropriate publicly funded research results, although the consequences of the Bayh-Dole act is still open to debate (Mowery & Ziedonis, 2002). Evaluating the effects of China’s patent system reform on patenting and licensing of university research will be discussed in our later paper.

340

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350

factors such as new mode of knowledge production, new partnerships, and more varied funding. In a knowledge-based economy, universities are demanded to play more active roles in fostering technology transfer and economic growth through application and commercialization of academic research. The mission of universities has been expanding, no longer simply including education, training and research, but now embracing producing and applying technological innovation. Many universities have adjusted the reward systems or the incentive structures to encourage faculty members to patent their research results and have also established technology transfer of?ces to manage IPRs. From a theoretical view, the rise in academic patenting, for universities themselves, may encourage the faster commercialization and exploitation of university inventions from public research and development, generate both industrial funding and licensing income from patents, spur new start ups and protect academic intellectual property. For scientists, patenting activities may bring the following positive consequences: satisfying their curiosity, helping to receive more peer recognition within the community and advance their career, gaining government supporting and additional funding from industry to build more effective and better equipped scienti?c team, and increasing their personal income (Geuna & Nesta, 2006). However, university patenting has raised a number of concerns on potential negative effects for the future of scienti?c research, especially for fundamental research. These concerns range from the impact of patenting on the direction of research (basic or applied), the substitution for publishing (reduce quantity and quality of publications), the decline of the quality of both research and teaching, to the effects on the diffusion of and access to publicly funded research results (Baldini, 2008; Van Looy et al., 2006). They are common not only to developed economies, but also to developing countries. Some work shows tradeoffs or con?icts between patenting and publishing, called the anti-common effects or the crowd-out effects. The more involvement of academic researchers in patenting may be possessed of a part of time and energy and make them undertake signi?cantly less basic research, which ultimately result in fewer publications (Chang & Yang, 2008). The patenting process often involves a delay of publications and requires researchers to keep related information con?dential at some time, which is a deviation from the academic norm of openness and dissemination of scienti?c knowledge (Merton, 1968) and also reduce the incentive to publish (Owen-Smith & Powell, 2001). Patenting activity may be detrimental to other researchers’ future work, since there are to some extent restrictions on data sharing, open discussion and usage of related research tools. The citation rate for a scienti?c publication may decline after patents associated with that publication are granted (Murray & Stern, 2007). Despite such concerns, however, there is a well-documented positive correlation between patenting and publishing activities of academic scientists in the empirical studies. For example, Meyer (2006a, 2006b) explored the relationship between nanoscience publications and nanotechnology patents of three European countries (United Kingdom, Germany, and Belgium) based on inventor–author analysis. His ?ndings supported the above conclusions that patenting activity does not appear to have a negative impact on the publication and citation performance of researchers. More importantly, inventor–authors apparently outperform their non-inventing peers in terms of both publication and citation frequencies. Similarly, Breschi et al. (2008) reported academic inventors published more and better quality papers than their non-patenting colleagues. More specially, positive effect seemed to be stronger for star scientists because at least some of the more proli?c and highly cited authors were also presented in the list of patent inventors. Although a large body of studies clearly showed positive impacts of patenting or other commercialization on publication quantity of academic scientists, the effects of patenting on publication quality are mixed. Agrawal and Henderson (2002) carried out a case study on publishing and patenting activities of faculty members from the Departments of Mechanical and Electrical Engineering at MIT and found that increased patenting activity was positively related to increased rates of paper citations. On the contrary, the study investigated by Murray and Stern (2007) of patent-paper pairs covering the same research result presented that citations to a paper decreased by between 9% and 17% after the patent related to the same content grant. Fabrizio and Di Minin (2008) also reported a decrease in average citations to publications produced by repeat patenters. Thus, the direction of the effects is hard to reach agreement. In sum, we lean to this line of arguments which lead to the following testable hypothesis. Hypothesis 1. Quantity of publications generated by a scientist will be higher following the application year for a patent by the researcher. Hypothesis 2. Quality of publications generated by a scientist will decline following the application year for a patent by the researcher. As Klitkou and Gulbrandsen (in press) point, the impacts of patenting activity and other types of commercialization for the scienti?c output are highly context-dependent in the national, university or disciplinary level. Furthermore, patent heterogeneity, namely the ownership of patents may play an important role. Fabrizio and Di Minin (2008) provided a further exploration on this issue by characterizing each patent as university-assigned, industry-assigned, or unassigned. Their results presented that besides university-assigned patents, the industry-assigned by faculty were also highly associated with more publications. In consistent with above results, Breschi et al. (2007) also reported that Italian academic inventors were more productive than their non-inventor colleagues. The reasons they explained were the ‘individual productivity effect’ and the ‘resource effect’. Particularly, the ‘resource effect’ became more clearly visible when patents were owned by industrial partners rather than by universities or the scientists themselves. Even collaborations with co-authors in business were positively related to the publication output of university scientists, as Breschi et al. (2008) pointed. Based on a large sample of German professors active in patenting, Czarnitzki et al. (2009) also distinguished between patents assigned to corporations

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350

341

and patents assigned to nonpro?t organizations including individual ownership of the professors themselves. Their ?ndings showed that patents assigned to nonpro?t organizations reinforced publication quantity and quality, while effects of patents assigned to corporations were adverse. Finally, our two hypothesizes on heterogeneity of patenting activity are provided: Hypothesis 3. Both quantity and quality of publications generated by a researcher will present a decline following the application year for a patent by the researcher, if the patent is assigned to corporations. Hypothesis 4. Both quantity and quality of publications generated by a researcher will be higher following the application year for a patent by the researcher, if the patent is assigned to the researcher himself. 2.2. China’s context in scienti?c research and technological development Public R&D support plays a critical role in shaping and guiding national innovative capacity of Asian latecomer countries (Hu & Mathews, 2008). The university sector in China has made a remarkable contribution to the reform of China’s national innovation system and the growth of China’s high-tech industry over the last decade (Xue, 2006). Universities enlarge their missions from education, training and knowledge creation to the commercialization of their research results. However, different supporting levels from governments exist due to limited resource. We attempt to explore these potential differences by considering whether a researcher comes from the key university and the State Key Laboratory. Since 1998, China’s central government has begun to conduct Project 985, which is a constructive project for funding world-class universities in the 21st century. In the initial phase, nine universities, selected as the best universities in China, were given grants in excess of 1 billion Yuan each, over a period of 3 years. The second phase, launched in 2004, expanded the project in cooperation with local government until it has now reached 39 universities. Many participating universities receive tens of millions of Yuan each year. All these universities, regarded as the State Key University, represent almost all the leading universities in China and are expected to be outstanding world-widely, 24 of which fall into our sample. A large part of the funding goes to not only upgrading of university infrastructure, staff capacity building, construction of research platforms, but also innovations in university operation mechanism, including improving the quantitative evaluation system by faculty members’ ability, performance, and contribution. Universities participating in this project have advantages of attracting and bringing together elite scientists, providing professional development of young staff, joint research programs with international leading universities. Thus, researchers from these universities could show a higher academic level on average than those from other universities, especially in some key disciplines. Hence, the preceding discussion suggests the following hypothesis: Hypothesis 5. Quantity and quality of publications by a researcher from universities participating in Project 985 are higher than other. Similarly, the main objective of the State Key Laboratories is also to make China’s research more visible and outstanding to the global scienti?c community. The State Key Laboratories have a list of university laboratories currently receiving funding and administrative support directly by the central government. Their construction and management is one of the results of a major S&T system reform in the country (Jin et al., 2006). Chemistry, Physics and Materials science, which make up more than a quarter of all the laboratories and where nanotechnology is classi?ed, are three key disciplinary areas specialized by the State Key Laboratories. They have many important research results in nanotechnology. Researcher from there may face with well-equipped working conditions including library software, digital campus, experimental areas and research stations. Another situation we consider is that these laboratories are signi?cantly more active in utilizing market mechanisms of technology transfer such as patenting, licensing, and consulting. Thus, we expected that quantity and quality of publications by a researcher from the State Key Laboratories are higher than other. Thus, we develop our last hypothesis as follow: Hypothesis 6. Quantity and quality of publications by a researcher from the State Key Laboratories are higher than other.

3. Nanotechnology in China Nanotechnology is de?ned as “understanding and control of matter at dimensions of roughly 1–100 nm, where unique phenomena enable novel applications” (PCAST, 2005). It has been recognized by not only scientists and technology developers but also policymakers as one of the key and transformative technologies of this century. To address the great potential of the emerging technology and promote its development, China’s government has identi?ed nanotechnology as one of priority mission areas in its national agenda of science and technology development and escalates investment in its R&D. Such investments have begun to translate into world-class research results, in terms of scienti?c publications and patents (Bai, 2005; Hassan, 2005). While China has emerged as one of key global players in nanotechnology, Chinese research community in nanotechnology needs to improve its now-limited research in?uence (Guan & Ma, 2007; Kostoff, Barth, & Lau, 2008; Youtie, Shapira, & Porter, 2008; Zhou & Leydesdorff, 2006). Taking the total number of patents granted as another indicator of research activity in this ?eld, China is behind the developed economies (Hullmann & Meyer, 2003; Li, Lin, Chen, & Roco, 2007). Therefore, China’s science and technology authorities have adjusted the evaluation procedure and criteria in order to encourage researchers in universities and public research institutions to publish original articles in international journals with high impact or obtain patents (Jin, Rousseau, & Sun, 2006).

342

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350

Among many factors in driving the rapid growth in publications, joint research with international colleagues cannot be neglected. It is accepted that joint research could make a contribution to strengthening research capacity of developing countries. For researchers from the largest developing country in the world, establishing cooperative relationships with and keeping in close touch with scientists from US, the EU countries or other major contributors to world science can help them expand their perspectives, improve their research techniques and provide access to the international scienti?c network, thus highlight both quantity and quality of research output. Guan and Ma (2007) provided an integrated bibliometric analysis of China’s in?uence and position in global nanotechnology research and con?rmed a positive effect of international collaboration on citations of publication. Thus, it would expect that the quality of international collaborative papers, especially in cooperation with nanotechnology-advanced countries, is good with higher average impact and citations as compared to rest publications. Another factor we must consider is that where to publish their discoveries and in which language is an important concern for scientists in China. Recently, a large proportion of papers by China’s scientists in nanotechnology have appeared in international journals originated from the scienti?cally advanced occident countries. The procedure of peer review in international journals creates a desirable exchange between reviewers and authors. This could highlight the original points, improve their writing and raise the quality of papers through the submission, revision and publishing process. Although more and more scientists of China have been paying attention to improving the international visibility of publications, Chinese journals are still the main channels for communicating their research results. To notify, Chinese journals, even those published in English and contained in the Science Citation Index, usually have a smaller readership and a lower international visibility. Thus, in order to address this issue, we distinct from Chinese journals and international journals in the following text when evaluating the performance of individual scientists. 4. Data and models 4.1. Collecting data Since nanotechnology is an emerging and rapidly evolving ?eld with the multidisciplinary nature, it is dif?cult to delineate its boundaries and harvest the relevant publications and patents of the ?eld. Different bibliometric search strategies of querying keywords and prominent terms in titles, abstracts and patent claims, are found to collect publications and patents of nanotechnology, including simple term search for the pre?x “nano”, complex and evolutionary lexical queries, citation analysis, bootstrapping techniques, the use of core journal sets based on Bradford’s Law and hybrid lexical-citation methods (Braun, Schubert, & Zsindely, 1997; Gl?nzel & Meyer, 2003; Leydesdorff & Zhou, 2007; Mogoutov & Kahane, 2007; Schummer, 2004; Zitt & Bassecoulard, 2006). There seems to be no agreements having been made on search approaches of nanotechnology in the above-mentioned studies. For conveniences of retrieving and identifying nanotechnology publications and patents, we employ the search strategy suggested by Porter, Youtie, Shapira, and Schoeneck (2008). Reviewing a variety of search efforts, they provided a two-staged modularized Boolean search strategy, with merits of the comprehensive search words combining with expert panel review and strong ability to search large-scale and multiple databases. The study exploits a database of nanotechnology patents from the Derwent Innovation Index (DII) and a database of nanotechnology publications from the SCI-Expanded (SCI-E). DII is the most comprehensive database covering the data of the main leading patent-issuing authorities including the United States Patent and Trademark Of?ce (USPTO), Japan Patent Of?ce (JPO), European Patent Of?ce (EPO), World Intellectual Property Organization (WIPO) and Sino Intellectual Patent Of?ce (SIPO). Further, it provides the descriptive titles and concise abstracts rewritten by subject experts linking to full-text primary patent records from a range of full patent sources, which can be retrieved easily and exactly. The latter is a widely accepted database covering most of the important in?uential journals in natural and medical science, which is used often to assessing the scienti?c performance of one country from the international perspective. It should be noted that each record in DII means a Basic Patent de?ned as a unique invention, enabling a global view of all Equivalent Patents referred to this particular invention in a patent family structure. A patent family is a group of published patent documents relating to the same invention and patented in different countries by way of the priority or priorities of a particular patent document. We treat the invention described in each record as unit of analysis. Applying the search strategy on DII, we harvest more than 180,000 records in the time frame of 1991–2008. The total number of patents obtained by world universities and Chinese universities was 20,522 and 7227 respectively. By incorporating change and variation of assignee names, 32 universities whose total number of patent granted are larger than 50 are found to be most active in patenting nanotechnology. The total number of patents acquired by these universities is 5274, showing a high percentage. These universities are identi?ed as our sample for further investigation. Although the types of patent of?ces targeted are different, we found that the distribution of patents acquired by these universities was highly skewed, with nearly 93% in SIPO and only 3.6% in USPTO. Researchers from these universities may have the language advantage and also have a lower cost in patenting with SIPO, but the quality and economic value may be lower than with others. Considering that the type of patent of?ce may play a role, we tried to capture it by adding the patent numbers applied with USPTO in our model, but only found the effects insigni?cantly. As above, we collect the publication data for them by conducting the search on SCI-E from 1991 to 2008. Based on the current practices in scientometrics, we limited the analysis of publications to research articles, in order to focus on the original research component of the SCI-E database. To include all articles that are relevant to 32 universities, we ?ltered out

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350 Table 1 Publication and patent numbers of 32 universities most active in patenting. University Tsinghua Univ Zhejiang Univ Shanghai Jiaotong Univ Fudan Univ Nanjing Univ Tianjin Univ Tongji Univ Shanghai Univ Beijing Univ Chem Technol Beijing Univ Sci & Technol Wuhan Univ Sichuan Univ Donghua Univ Jilin Univ Dongnan Univ E China Univ Sci & Technol Patent numbers 596 492 491 380 224 195 186 166 153 152 147 146 144 142 138 126 Publication numbers 4198 2917 2172 2618 3732 1113 667 526 744 105 1527 1039 309 2924 347 795 University S China Univ Technol Peking Univ Zhongshan Univ Xiamen Univ Nankai Univ Wuhan Univ Technol Dalian Univ Technol Shandong Univ Xian Jiaotong Univ E China Normal Univ Univ Sci & Technol China Harbin Inst Technol Beijing Univ Technol Hunan Univ Zhongyuan Univ Technol Shandong Normal Univ Patent numbers 123 120 116 109 107 91 90 86 83 80 75 74 62 58 54 51

343

Publication numbers 710 3305 1362 830 1432 691 913 1825 772 557 3854 1128 276 858 32 259

all papers where at least one author af?liation is located in one of 32 universities and identi?ed more than 28,000 records. Table 1 presents publication and patent data for these selected universities. It should be noted that there are three outliers in ratio of patents/publications: Zhongyuan Univ Technol, Donghua Univ and Beijing Univ Sci & Technol. All of them were developed from the institutes and nowadays concentrate in some engineering ?elds. Zhongyuan Univ Technol, formerly Zhengzhou Textile Institute, and Donghua Univ, formerly East China Textile Institute of Science and Technology, focus on the study of textile engineering and technology. Beijing Univ Sci & Technol, formerly Beijing Institute of Iron and Steel Technology, is renowned for its study of metallurgy and materials science. Thus these universities pay more attentions to applied research and patenting, showing large extremes in ratio of patents/publications. This interesting information may re?ect the different institutional policies concerned with Intellectual Property Right and deserves the further study.

4.2. Approaches of tracing patenting and publishing links Several approaches based on the informetric structure of publications and patents can be found in bibliometric or technometric studies to trace patenting and publishing links, such as citations, co-activity, classi?cation relations, shared topics (Bassecoulard & Zitt, 2004). The science-based citation based on the patent data is the classic and common way to establish the links, suggested by the pioneering works of CHI-Research (Narin & Noma, 1985) and some extensive works (Gl?nzel & Meyer, 2003; Schmoch, 1997; Verbeek et al., 2002). But links established through the reference ?eld of patents or publications are suspicious, mainly due to different citation behaviors and different citation motivations among authors, inventors and patent examiners. For example, these links are hardly direct and noisy (Meyer, 2006a) and their strength is somewhat limited (Czarnitzki et al., 2009). Besides this methodology, Murray and Stern (2007) provided another choice by making use of patent-paper pairs to establish linkages. However, patent-paper pairs may limit to some disciplinary in which research results can be served as a simultaneous foundation for future scienti?c research and commercialization. The approach of establishing links through collaborative knowledge production expressed by inventor–author relations suggested by Noyons, Van Raan, Grupp, and Schmoch (1994) and Meyer (2006a, 2006b), is not novel but much stronger and more meaningful. This approach was also used in Boyack and Klavans’s informetric study (2008) to measure interaction between science and technology at the level of individual researchers. Following above studies, our method used to match two datasets is to link inventor and author name from the same institution. Since DII’s name rules are different from SCI-E’s, it is dif?cult to perform a matching procedure based on the abbreviations of inventor names, especially for Chinese name, for example, ‘Qian, Y.T.’ in SCI-E and ‘Qian Yitai’ in DII point to the same scientist. Probably, this situation leads to be empty of empirical studies on China’s relationship between patents and publications. To solve this, more information about researcher’s full name is needed. We extract inventor full names of 5274 patents from the online databases of all responding patent of?ces and update their abbreviations in our database according to the name rules of SCIE. After matching the data in the same name from the same university only to prevent false connections, we get 6321 inventor–authors ultimately. Although our method is time-consuming in extracting information on full name, identifying name rules and coding for rewriting name abbreviations, it is the only way due to our limited knowledge. It is noted that we may miss some links, but we can ensure every link is identi?ed with precision. In the next step, we group all these university patents into two types in order to distinguish the different ownership of university patents. The one is assignment to a company, while the other is assignment to the inventor himself. It is worthy to note that the percentages of two types in total patents are nearly 15% and 5% respectively. The distribution of two types by patent application year can be seen in Fig. 1, which seems that numbers of patents assigned to company has been signi?cantly larger than those assigned to individual during recent years.

344

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350

Fig. 1. The distribution of faculties’ patents by type of assignee.

4.3. Indicators This study chooses both patenting and publishing activities by the academic researcher per year as the unit of analysis. Five different bibliometric indicators are developed to evaluate both the scienti?c productivity of researchers and the impact of their publications, which are illustrated as follows. The ?rst two measures are publication numbers per researcher per year on the basis of full counting and fractional counting, which are also used in Meyer’s (2006b) study. Our third publication outcome measure is the publication count per year in international journals, in order to distinct researchers’ choice where to publish their discoveries. The last two measures account for the quality of academic publications. The one is the publication count per year weighted by the journal impact factor (JIF) to account for the quality of academic publications. Impact factor is de?ned as the mean citations of the articles in the journal in a given period (Gar?eld & Sher, 1963; Gar?eld, 1972). The Impact factor published annually in the JCR has become a prominent international evaluation tool for assessing the quality, prestige and international visibility of a journal. It is also used to evaluate journal articles (Bordons, Fernandez, & Gomez, 2002; Czarnitzki et al., 2009), research activities (Moed, 2002), and researchers’ performance (Kostoff, 1997). In this study, we use it as a proxy to delineate the quality of production of Chinese nanotechnology researchers. The other is that the total number of citations received by publications. Year dummies are used in our econometric models to consider the fact that the impact of a given piece of research varies considerably with the time elapsed since initial publication. We use two variables to measure patent heterogeneity mentioned above. The ?rst is the number of patents by every academic inventor per year with the co-ownership by his university and a ?rm, while the second one is the number of patents by every academic inventor per year with inventors’ appearance in applicant list. With regard to supporting level from governments, we utilize the information on publications’ af?liated institution to identify whether the author comes from the State Key Laboratories based on the institutions’ title “State Key Lab” or “Natl Key Lab” and judge whether and when the university participates in Project 985 according to the related information from Ministry of Education’s website (as shown in Table 2). Another variable we pay attention to is the number of publications generated by a researcher per year in his (or her) main science ?eld. Our method of identifying one’s main science ?eld is to calculate which ?eld was published most frequently during the period of 1991–2008, since it could re?ect where the researcher has devoted his time. It is well known that journals in SCIE are categorized into 22 broad ?elds, among which there are different publication patterns. Nanotechnology is an emerging and complex ?eld with the multidisciplinary nature and the nano-related journals in SCIE are concentrated in Chemistry, Physics and Materials science. We add this variable to account for variation in the publication pattern across different science ?elds (Czarnitzki et al., 2009). We assume that more publications a researcher had in his (or her) main
Table 2 University sample participating in project 985. University Peking Univ Tsinghua Univ Nanjing Univ Fudan Univ Shanghai Jiaotong Univ. Xian Jiaotong Univ Zhejiang Univ Univ Sci & Technol China Harbin Inst Technol Nankai Univ Tianjin Univ Beijing Univ Technol Participating year 1999 1999 1999 1999 1999 1999 1999 1999 1999 2000 2000 2000 University Wuhan Univ Sichuan Univ Jilin Univ Dongnan Univ S China Univ Technol Zhongshan Univ Xiamen Univ Dalian Univ Technol Shandong Univ Hunan Univ Tongji Univ E China Normal Univ Participating year 2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 2002 2004

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350

345

Fig. 2. Distribution of numbers of faculties by their ?rst patent application year.

science ?eld last year, more likely to highlight his (or her) international visibility. We imposed a logarithm form and 1year lagged value of this variable to control skewness and avoid endogeneity respectively. We also collect the number of internationally coauthored publications produced from the nanotechnology leading countries by every researcher per year to control the impact of international scienti?c collaborations on the research performance. Based on same considerations, we lag this variable by 1 year. The last variable is the experience in patenting activity of every researcher, also used in Czarnitzki and his colleagues’ study. The experience may help faculty be familiar with patenting procedures, know how to ?le a patent in an effective way and reduce future cost of patenting. To some extent, patenting experience may counteract the crowd-out effect of patenting activity on publishing activity of faculties. However, too much experience may mean that the faculty has shifted his research agenda towards applied research or commercialization which can produce more patents. Therefore, it is expected that publications follow a non-linear inverted U shape relationship with patenting experience. We use a proxy of the time elapsed since his ?rst patent application to measure patenting experience of every researcher and also include its squared term in the regressions to examine the non-linear relationship. Fig. 2 shows that the majority of faculties have participated in patenting activity since 2000, while only a small part started at the beginning of our investigating period. 4.4. Panel data regression model To further explore the publishing–patenting relationship in China’s context, we perform the ?xed effects Poisson model on the unbalanced panel dataset described above. This count panel data model was developed by Hausman, Hall, and Griliches (1984) on explaining patent applications by ?rms in terms of research and development expenditures. We adopt this econometric model based on three considerations as follows. Firstly, ?xed effects model vs random effects model. There is heterogeneity among researchers, but not all heterogeneity among researchers will be re?ected in the above observables. So we need to take unobserved effects of faculty members into account, such as individual motivations, skills, effort, and serendipity (Buenstorf, 2009; Czarnitzki et al., 2009). We expect a priori that these important individual unobserved effects are correlated with the right-hand variables. Based on this, the ?xed effects model is picked. Furthermore, all of Hausman tests for our data reject Random Effects speci?cations at the 0.1% level. Secondly, Poisson models vs negative binominal models. It has long been recognized that the basic assumption of Poisson model is equality of the conditional variance and mean, namely equi-dispersion. In the case of over-dispersion, some scholars appropriated the Negative Binominal Models (e.g. Chang & Yang, 2008). However, Wooldridge (1999) had shown some nice and strong robustness properties of the ?xedeffects Poisson model. He pointed that the Poisson model makes few restrictions, e.g. its estimator still consistent whenever the restrictive equi-dispersion assumption holds. Compared to the Negative Binominal Models, its functional form is quite ?exible when allowing for the possibility of correlation across observed variables and it also provides the calculation of fully robust standard errors to correct the biased standard errors. Thirdly, the ?xed-effects Poisson model is a type of count data model but it could also be applied to the situation where the dependent variable is a nonnegative continuous variable (Wooldridge, 2002, p. 676). While the publication full counts, the publication counts of international journals and citation times are actual count data, the publication fractional counts and the journal impact factor-weighted publication counts are not necessarily integers. Thus, the ?xed-effects Poisson model is more suitable for our data. The starting point of the ?xed-effects Poisson model is exponential mean function and multiplicative individual speci?c term as follows: yit ?P[ it = ?i it ] i = 1, . . . , n; t = 1, . . . , T, it = exp(xit ˇ), where ? refers to the individual speci?c effect. (1)

346 Table 3 Variable de?nition and descriptive statistics. Variable

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350

De?nition

obs. = 22,369 Mean S.D. 3.548 0.774 3.482 9.874 59.613 0.930 2.148 1.909 2.391 Min 1 0.002 0 0 0 0 0 1992 1992 Max 93 16.993 93 181.556 1513 26 56 2007 2007

PFUit PFAit PIJit PJIFit TCit InterCoAuthorPubit MainScienceFieldPubi,t?1 ApplicationYear FirstApplicationYeari

BPnumberi,t?1 CompanyPatentNumberi,t?1

IndividualPatentNumberi,t?1

isKeyUnivit isKeyLabit Experienceit

Publication full counts by faculty i in year t Publication fractional counts by faculty i in year t Publications in international journals by faculty i in year t JIF-Weighted publications by faculty i in year t Total cites by faculty i in year t Publications with foreign co-authors by faculty i in year t Publications in the researcher’s main science ?eld in year t ? 1 Application year of patent, for matching the publication year Application year of one’s ?rst patent, for calculating his or her experience on engaging in patenting activity Basic Patent numbers by faculty i in year (t ? 1) Basic Patent numbers with the company assignee by faculty i in year (t ? 1) Basic patent numbers with the individual assignee by faculty i in year (t ? 1) Whether the researchers from the 985 universities or not Whether publications are af?liated with the State Key Lab Experience of enagaging in patenting acitivity

2.929 0.603 2.736 6.732 24.046 0.384 1.158 2005.056 2003.921

0.457 0.025

1.138 0.587

0 0

52 48

0.006

0.147

0

12

0.836 0.362 0.584

0.370 0.481 1.360

0 0 0

1 1 15

Taken altogether, our speci?cation is given in Eq. (2): E(yit |xit , ?i ) = ?i exp(xit ˇ). (2)

Before performing our empirical analysis, the time window between publishing and patenting must be considered. Scientists engaging in nanotechnology may have the disciplinary advantage of both publishing and patenting their discoveries in parallel. It is assumed that the application date of a patent and the submission date of an article could be seen as the ?nished time of research work. The former can be easily extracted from DII, while the latter is unobservable in SCIE. To solve this issue, we ran correlation analysis to explore time lags between the dependent variables and patent numbers. The results show the correlation coef?cients when taking 1-year lag value are highest among others (see Appendix A). Thus, we time the patent variables by application year and include their 1-year lag value in the regressions. Finally, the de?nitions of all the variables in our models are shown in Table 3. 5. Empirical analysis Table 3 also presents some descriptive statistics in the last four columns, namely the means, standard deviations as well as minima and maxima of the variables used in the subsequent regression analysis. On average, each scientist in our sample has nearly 3 scienti?c publications per year but less than a half patent. Further, the mean of PIJ is approximate to that of PFU, which re?ects that publishing research results in international journals has become prevailing in China’s scienti?c community in the ?eld of nanoscience and technology. Table 4 presents the estimation results of conditional ?xed-effects Poisson regressions. The ?rst three columns present the results for the estimation for the publication quantity as measured by PFU, PFR and PIJ respectively. The subsequent columns show the estimated coef?cients for PJIF and TC, i.e. the journal-impact-factor weighted publication outcome of the academic inventors and the citations received. Both the standard errors and the fully robust standard errors of the ?xed effects Poisson model are reported in brackets, since in the case of over-dispersion, the former are biased and need to be adjusted while the fully robust standard errors is still consistent. Hypothesis 1 predicts the positive relationship between patenting activity and publication quantity. As shown in left three columns, whatever the indicators is used, the coef?cients are all statistically signi?cant and positive. The positive signs suggest that publication quantity and academic patenting are mutually reinforced, thereby supporting Hypothesis 1.

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350 Table 4 Conditional ?xed-effects Poisson regressions. Dependent variables Covariates BPnumberi,t?1 PFUit PFRit PIJit PJIFit TCit

347

Coef?cient (std. err.) (Robust std. err.) 0.048 (0.005)*** (0.007)*** ?0.048 (0.007)*** (0.008)*** ?0.044 (0.026) (0.013)** 0.027 (0.001)*** (0.003)*** 0.029 (0.009)** (0.013)* ?0.349 (0.097)*** (0.162)* 0.278 (0.028)*** (0.061)*** 0.279 (0.012)*** (0.017)*** 0.126 (0.004)*** (0.013)*** 20,725 4677 8435.12*** 2985.17*** 0.051 (0.010)*** (0.007)*** ?0.052 (0.015)** (0.008)*** ?0.017 (0.051) (0.022) 0.027 (0.003)*** (0.003)*** 0.025 (0.019) (0.015) ?0.318 (0.220) (0.178) 0.297 (0.061)*** (0.060)*** 0.245 (0.027)*** (0.020)*** 0.117 (0.008)*** (0.012)*** 20,725 4677 1554.87*** 2668.70*** 0.048 (0.005)*** (0.007)*** ?0.047 (0.007)*** (0.008)*** ?0.035 (0.026) (0.013)* 0.027 (0.001)*** (0.003)*** 0.025 (0.009)** (0.013) ?0.327 (0.100)** (0.169) 0.268 (0.029)*** (0.064)*** 0.270 (0.013)*** (0.018)*** 0.129 (0.004)*** (0.014)*** 20,677 4654 8671.01*** 3259.42*** 0.048 (0.003)*** (0.007)*** ?0.056 (0.004)*** (0.009)*** ?0.064 (0.014)*** (0.017)*** 0.023 (0.001)*** (0.003)*** 0.011 (0.006) (0.014) ?0.242 (0.064)*** (0.166) 0.291 (0.022)*** (0.067)*** 0.255 (0.008)*** (0.021)*** 0.144 (0.002)*** (0.018)*** 20,500 4630 30243.44*** 3371.22*** 0.064 (0.002)*** (0.016)*** ?0.068 (0.002)*** (0.017)*** ?0.038 (0.005)*** (0.032) 0.015 (0.001)*** (0.005)** 0.033 (0.004)*** (0.024) ?1.156 (0.051)*** (0.323)*** 0.360 (0.007)*** (0.121)** 0.215 (0.004)*** (0.036)*** 0.149 (0.001)*** (0.024)*** 20,480 4561 107912.26*** 9601.52***

CompanyPatentNumberi,t?1

IndividualPatentNumberi,t?1

MainScienceFieldPubi,t?1

Experiencei,t

ExperienceSquarei,t

isKeyUnivi,t

isKeyLabi,t

InterCoAuthorPubit

Number of observations Number of researchers Wald- 2 Wald- 2 for robust model

Note: (1) Although our sample is 6321 academic inventors, 1644 groups (1644 obs.) dropped because of only one obs. per group. (2) The last three columns are smaller because several groups of all zero outcomes dropped. (3) *Correspond to a 10% level of statistical signi?cance, **correspond to a 5% level of statistical signi?cance, ***correspond to a 1% level of statistical signi?cance.

Hypothesis 2 predicts that the relationship between patenting activity and publication quality would be negative. However, the coef?cients for publication quality measured by PJIF and TC respectively are still statistically signi?cant and positive. They show that patenting may have a positive impact on publication quality, thereby rejecting the argument for Hypothesis 2. To summarize, our ?ndings con?rm the positive relationship between patenting activity and both quantity and quality, in consistent with some previous studies (Breschi et al., 2008; Czarnitzki et al., 2009; Fabrizio and Di Minin, 2008). With respect to the heterogeneity of patent ownership, the results are not in line with a positive patent–publication relationship mentioned above. Hypothesis 3 supposes that both quantity and quality of publications generated by a researcher will present a decline following the application year for a patent by the researcher, if the patent assigned to corporations. We ?nd that scientists’ engagement in this type of activity has a negative and signi?cant effect on both publication quantity and quality. These results con?rm our concerns on publication delay, the anti-commons effect, and restriction on data sharing. This ?nding is partially consistent with an earlier study by Czarnitzki et al. (2009), who showed a negative but statistically only weakly signi?cant effect of company patents. Hypothesis 4 predicts a positive effect of individual patents on publication activity. But these negative signs still hold when the patent is assigned to the researcher himself. However, one must be careful not to rush to conclusions because the percentage of individual is very small and these negative signs appear signi?cant only if PJIF is applied. Hypothesis 5 and Hypothesis 6 predict that a researcher from universities participating in Project 985 or the State Key Lab would show a higher level on both quantity and quality of publications. In any cases, the coef?cients are positive and statistically signi?cant, thereby con?rming our supposition. We also control some factors which may in?uence faculty members’ research performance and are also presented in previous studies, such as publication counts in one’s major science ?eld last year, international co-authorship, patenting experience and its square term. The positive and highly signi?cant coef?cients of publication counts in major science ?eld hold in both cases in terms of quantity and quality. In line with Czarnitzki et al. (2009), it con?rms our viewpoints that more

348 Table 5 The results of testing hypotheses. Description Hypothesis 1 Hypothesis 2 Hypothesis 3

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350

Predicting sign Positive Negative Negative

Result Not rejected, positive signi?cantly Rejected, positive signi?cantly Not rejected, negative signi?cantly

Hypothesis 4

Hypothesis 5

Hypothesis 6

The effect of patent on publication quantity The effect of patent on publication quality The effects of company patent on both publication quantity and publication quality The effects of individual patent on both publication quantity and publication quality The effects of Project 985 on both publication quantity and publication quality The effects of the State Key Lab on both publication quantity and publication quality

Positive

Not rejected, negative unsigni?cantly

Positive

Not rejected, positive signi?cantly

Positive

Not rejected, positive signi?cantly

publication output one has in his or her major science ?eld, higher level the publications show in terms of impact factors of journals and citations received. Similarly, international co-authorship with the leading countries helps China’s scientists increase their productivity and improve their research impact. In terms of experience measures, we predict a lifecycle effect of faculty members, which means that patenting experience generally leads to less distraction from publication activities, but more experience may re?ect the reverse effect. In our models, we con?rm a non-linear inverted U shape relationship between patenting experience and scienti?c research, though weakly signi?cant. Taken altogether, the results of testing hypotheses are illustrated in Table 5. Finally, we discuss several marginal effects of patent indicators. The marginal effects of basic patents mean that they will increase publication quantity by 5% and publication quality by 6%. However, if patents are assigned to companies, the positive impact of patents may be counteracted. Company patents may result in a reduction of publication quantity of 5% and a reduction of quality of 7%. At the average of the patent variable, the former corresponds to 0.5 less publication full counts, whereas the latter 1.7 less citations to publications. With regard to individual patents, we only discuss their effects on quality as measured by the JIF-weighted publications due to their signi?cant level. The marginal effects of individual patents amount to a reduction of quality of 6%, which corresponds to 0.4 the JIF-weighted publications. 6. Conclusion This paper contributes to the growing study on the relationship between patenting and publishing among faculty members with China’s evidence in the ?eld of nanotechnology. Following an interesting path on matching the data of publications and patents, we ?rstly establish China’s inventor–author links in the ?eld of nanotechnology. 6321 con?rmed academic inventors who are co-active in publishing and patenting over the time period 1991–2008 are used to construct panel dataset. By further exploring the publishing–patenting relationship in the ?xed-effects Poisson models, our ?ndings support that faculty members who patent their research do not generate fewer publications and lower quality after patenting. There are several reasons for a positive impact of patenting on the scienti?c productivity of individual scientists. First, patenting is most often regarded by scientists as a result of new scienti?c opportunities or by-products of scienti?c research and also provides the basis for the establishment of new scienti?c disciplines, especially for the emerging and rapidly evolving ?eld. Individual scientists who participate in patenting activity, have an advantage of being aware of research questions raised by technology. Knowledge of these questions may at the same time be the basis for the establishment of new scienti?c disciplines and of potentially economic value. This may cause research performed by the patenting scientists to be more ef?cient and more relevant to the community. Second, patenting process sometimes needs scientists to be closely interactive with industry, which may help them access additional resources. These resources include both technical and ?nancial supporting from industry, free access to expensive scienti?c instruments, and sharing related data. Much experience in patenting, also, may help them be familiar with patenting procedures, know how to ?le a patent in an effective way and reduce the time and cost of future patenting. Third, there may be the ‘Matthew’ effect operable. Since university leaders or laboratory directors have increased faculty members’ incentives to patent research results, e.g. including the patent indicators or increasing patent weights in their evaluation procedures in terms of research performance, patenting may bring more peer recognition within the scienti?c community to individual scientists and advance their career, which probably leads to the improvement of individual research performance. In one word, Scientists become more widely known after patenting, then are invited more to submit papers. It has been seen that more proli?c and higher cited scientists are also active in patenting, such as ‘star scientists’. Further, patenting scientists may gain more government supporting or additional funding from industry to build more effective and better equipped scienti?c team. Based on same reasons, not surprisingly, patenting could help to highlight publication quality of individual scientists, measured by citation times or the JIF-weighted publications.

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350 Table 6 Pairwise correlations. Bpnumber Current value 1-Year lag value 2-Year lag value 3-Year lag value
*

349

ln(PFU) 0.1693* 0.2231* 0.2149* 0.2133*

ln(PFR) 0.1720* 0.2203* 0.2068* 0.2100*

ln(PJIF) 0.1589* 0.2076* 0.2035* 0.2056*

ln(TC) 0.1546* 0.1290* 0.0654* 0.0215

ln(PIJ) 0.1638* 0.2267* 0.2135* 0.2066*

Correspond to a 1% level of statistical signi?cance.

Although a large number of patents generated by university scientists should be assigned to their af?liated university due to China’s patent laws and related regulations, we still ?nd the heterogeneity of patent ownership. By distinguishing between patents assigned to company and patents assigned to the scientist himself, our empirical analysis shows that the positive relationship between patenting and publishing indeed does not hold for two types of patents. Especially, patents assigned to company have a signi?cantly negative impact on publication quantity and quality. A possible explanation is that these patents come from joint or contract research, which requires scientists to delay research results and keep con?dential on related information. Thus, collaboration with industry could shift the orientation of scientists away from basic research, where they can produce more and higher quality publications. It is worthy noted that both the percentages of two types of patents are very small. Thus, one needs to be cautious about these conclusions. Additionally, we control two factors appeared in previous studies in driving the rapid growth in publications or improving research impact of China’s scientists, namely different supporting levels from government and international co-authorship with the leading countries. We con?rm that the key research universities and the State Key Laboratories have played a vital role in scienti?c research and technology development of China’s nanotechnology and international co-authorship does help China’s scientists improve their scienti?c research. It should be noted that the interaction between patenting and publishing is dual and complex, which deserves the further study. The effects of publishing on patenting cannot be ignored. Further, the issues on the effects of university patenting include not only quantity and quality of publications but also the contents of academic research. The main limitation of our analysis is that we only identi?ed the effects of patenting activity on quantity and quality of scienti?c research, which is a rather narrow perspective of the interaction. We acknowledge that it is overloading to put these issues into a single article. It is only our ?rst step in this study. To ?nd an ultimate and comprehensive answer, we need more information from investigation. Acknowledgements This research is funded by National Social Science Foundation of China (Project No. 08BJY031), National Natural Science Foundation of China (Project No. 70773006) and Shanghai Leading Academic Discipline Project (Project No. B210). The authors are very grateful for the valuable comments and suggestions of the anonymous reviewers, which signi?cantly improved the article. Appendix A. See Table 6. References
Agrawal, A., & Henderson, R. (2002). Putting patents in context: Exploring knowledge transfer from MIT. Management Science, 48(1), 44–60. Azoulay, P., Ding, W., Stuart, T. (2006). The impact of academic patenting on the rate, quality, and direction of (public) research. NBER Working Paper 11917. Azoulay, P., Ding, W., & Stuart, T. (2007). The determinants of faculty patenting behavior: Demographics or opportunities. Journal of Economic Behavior & Organization, 63(4), 599–623. Bai, C. L. (2005). Ascent of nanoscience in China. Science, 309(5731), 61–63. Baldini, N. (2008). Negative effects of university patenting: Myths and grounded evidence. Scientometrics, 75(2), 289–311. Bassecoulard, E., & Zitt, M. (2004). Patents and publications: The lexical connection. In H. F. Moed, W. Gl?nzel, & U. Schmoch (Eds.), Handbook of quantitative science and technology research: The use of publication and patent statistics in studies of S&T systems (pp. 665–694). Dordrecht: Kluwer Academic Publishers. Bordons, M., Fernandez, M. T., & Gomez, I. (2002). Advantages and limitations in the use of impact factor measures for the assessment of research performance in a peripheral country. Scientometrics, 53(2), 195–206. Boyack, K. W., & Klavans, R. (2008). Measuring science–technology interaction using rare inventor–author names. Journal of Informetrics, 2(3), 173–182. Braun, T., Schubert, A., & Zsindely, S. (1997). Nanoscience and nanotechnology on the balance. Scientometrics, 38(2), 321–325. Breschi, S., Lissoni, F., & Montobbio, F. (2007). The scienti?c productivity of academic inventors: New evidence from Italian data. Economics of Innovation and New Technology, 16(2), 101–118. Breschi, S., Lissoni, F., & Montobbio, F. (2008). University patenting and scienti?c productivity: A quantitative study of Italian academic inventors. European Management Review, 5(2), 91–109. Buenstorf, G. (2009). Is commercialization good or bad for science? Individual-level evidence from the Max Planck Society. Research Policy, 38(2), 281–292. Calderini, M., Franzoni, C., & Vezzulli, A. (2007). If star scientists do not patent: The effect of productivity, basicness and impact on the decision to patent in the academic world. Research Policy, 36(3), 303–319. Carayol, N., & Matt, M. (2004). Does research organization in?uence academic production? Laboratory level evidence from a large European university. Research Policy, 33(8), 1081–1102.

350

G. Wang, J. Guan / Journal of Informetrics 4 (2010) 338–350

Chang, Y. C., & Yang, P. Y. (2008). The impacts of academic patenting and licensing on knowledge production and diffusion: A test of the anti-commons effect in Taiwan. R&D Management, 38(3), 321–334. Czarnitzki, D., Gl?nzel, W., & Hussinger, K. (2009). Heterogeneity of patenting activity and its implications for scienti?c research. Research Policy, 38(1), 26–34. Fabrizio, K. R., & Di Minin, A. (2008). Commercializing the laboratory: Faculty patenting and the open science environment. Research Policy, 37(5), 914–931. Gar?eld, E. (1972). Citation analysis as a tool in journal evaluation. Science, 178(4060), 471–479. Gar?eld, E., & Sher, I. H. (1963). New factors in the evaluation of scienti?c literature through citation indexing. American Documentation, 14(4), 195–201. Geuna, A., & Nesta, L. J. J. (2006). University patenting and its effects on academic research: The emerging European evidence. Research Policy, 35(6), 790–807. Gl?nzel, W., & Meyer, M. (2003). Patents cited in the scienti?c literature: An exploratory study of ‘reverse’ citation relations in the Triple Helix. Scientometrics, 58(2), 415–428. Guan, J. C., & Ma, N. (2007). China’s emerging presence in nanoscience and nanotechnology: A comparative bibliometric study of several nanoscience ‘giants’. Research Policy, 36(6), 880–886. Hassan, M. (2005). Small things and big changes in the developing world. Science, 309(5731), 65–66. Hu, M. C., & Mathews, J. A. (2008). China’s national innovative capacity. Research Policy, 37(9), 1465–1479. Hausman, J., Hall, B., & Griliches, Z. (1984). Econometric Models for Count Data with an Application to the Patents-R&D Relationship. Econometrica, 52, 909–938. Hullmann, A., & Meyer, M. (2003). Publications and patents in nanotechnology: An overview of previous studies and the state of the art. Scientometrics, 58(3), 507–527. Jin, B. H., Rousseau, R., & Sun, X. X. (2006). Key Labs and Open Labs in the Chinese scienti?c research system: Their role in the national and international scienti?c arena. Scientometrics, 67(1), 3–14. Klitkou, A., Gulbrandsen, M., in press. The relationship between academic patenting and scienti?c publishing in Norway. Scientometrics. Kostoff, R. N. (1997). Citation analysis cross-?eld normalization: A paradigm. Scientometrics, 39(3), 225–230. Kostoff, R. N., Barth, R. B., & Lau, C. G. Y. (2008). Quality vs. quantity of publications in nanotechnology ?eld from the People’s Republic of China. Chinese Science Bulletin, 53(8), 1272–1280. Leydesdorff, L., & Zhou, P. (2007). Nanotechnology as a ?eld of science: Its delineation in terms of journals and patents. Scientometrics, 70(3), 693–713. Li, X., Lin, Y., Chen, H., & Roco, M. C. (2007). Worldwide nanotechnology development: A comparative study of USPTO, EPO, and JPO patents (1976–2004). Journal of Nanoparticle Research, 9(6), 977–1002. Liu, L. W., & Zhang, L. D. (2005). Nanotechnology in China—Now and in the Future. Nanotechnology Law & Business, 2(4), 397–403. Merton, R. K. (1968). The normative structure of science. In R. K. Merton (Ed.), The sociology of science: Theoretical and empirical investigations. Chicago, IL: The University of Chicago Press. Meyer, M. (2006a). Knowledge integrators or weak links? An exploratory comparison of patenting researchers with their non-inventing peers in nanoscience and technology. Scientometrics, 68(3), 545–560. Meyer, M. (2006b). Are patenting scientists the better scholars? An exploratory comparison of inventor–authors with their non-inventing peers in nanoscience and technology. Research Policy, 35(10), 1646–1662. Ministry of Education, 1999. The Regulation Regarding the Protection and Management of Intellectual Properties in Higher Education Institutions. Act 3, No. 8120. Moed, H. F. (2002). Measuring China’s research performance using the Science Citation Index. Scientometrics, 53(3), 281–296. Mogoutov, A., & Kahane, B. (2007). Data search strategy for science and technology emergence: A scalable and evolutionary query for nanotechnology tracking. Research Policy, 36(6), 893–903. Mowery, D. C., & Ziedonis, A. A. (2002). Academic patent quality and quantity before and after the Bayh-Dole act in the United States. Research Policy, 31(3), 399–418. Murray, F., & Stern, S. (2007). Do formal intellectual property rights hinder the free ?ow of scienti?c knowledge? Journal of Economic Behavior & Organization, 63(4), 648–687. Narin, F., & Noma, E. (1985). Is technology becoming science? Scientometrics, 7(3–6), 369–381. Noyons, E. C. M., Van Raan, A. F. J., Grupp, H., & Schmoch, U. (1994). Exploring the science and technology interface—inventor author relations in laser medicine research. Research Policy, 23(4), 443–457. Owen-Smith, J., & Powell, W. W. (2001). To patent or not: faculty decisions and institutional success at technology transfer. Journal of Technology Transfer, 26(1–2), 99–114. PCAST. (2005). The national nanotechnology initiative at 5 years. Washington, DC: President’s Council of Advisors on Science and Technology. Executive Of?ce of the President Porter, A. L., Youtie, J., Shapira, P., & Schoeneck, D. J. (2008). Re?ning search terms for nanotechnology. Journal of Nanoparticle Research, 10(5), 715–728. Schmoch, U. (1997). Indicators and the relations between science and technology. Scientometrics, 38(1), 103–116. Schummer, J. (2004). Multidisciplinarity, interdisciplinarity, and patterns of research collaboration in nanoscience and nanotechnology. Scientometrics, 59(3), 425–465. Van Looy, B., Callaert, J., & Debackere, K. (2006). Publication and patent behaviour of academic researchers: Con?icting, reinforcing or merely co-existing? Research Policy, 35(4), 596–609. Verbeek, A., Debackere, K., Luwel, M., Andries, P., Zimmermann, E., & Deleus, F. (2002). Linking science to technology: Using bibliographic references in patents to build linkage schemes. Scientometrics, 54(3), 399–420. Wooldridge, J. M. (1999). Distribution-free estimation of some nonlinear panel data models. Journal of Econometrics, 90(1), 77–97. Wooldridge, J. M. (2002). Econometric analysis of cross section and panel data. Cambridge: MIT Press. Wu, W. P., in press. Managing and incentivizing research commercialization in Chinese Universities. Journal of Technology Transfer. Xue, L., 2006. Universities in China’s national innovation system. Presented at the Second International Colloquium on Research and Higher Education Policy, UNESCO Headquarters, Paris, 29 November–1st December 2006. Youtie, J., Shapira, P., & Porter, A. L. (2008). Nanotechnology publications and citations by leading countries and blocs. Journal of Nanoparticle Research, 10(6), 981–986. Zhou, P., & Leydesdorff, L. (2006). The emergence of China as a leading nation in science. Research Policy, 35(1), 83–104. Zitt, M., & Bassecoulard, E. (2006). Delineating complex scienti?c ?elds by an hybrid lexical-citation method: An application to nanosciences. Information Processing & Management, 42(6), 1513–1531.


赞助商链接

更多相关文章:
更多相关标签:

All rights reserved Powered by 甜梦文库 9512.net

copyright ©right 2010-2021。
甜梦文库内容来自网络,如有侵犯请联系客服。zhit325@126.com|网站地图