Measuring science-based science linkage and non-science-based linkage of patents through non-patent references

https://doi.org/10.1016/j.joi.2015.04.004Get rights and content

Highlights

  • This study analysed firms’ science linkage performance based on their absolute and relative science linkage indicators.

  • Rapidly growing non-science-based linkage of patents was observed.

  • Field-specific differences in science linkage performance were identified.

  • Firm-specific differences in the science linkage performance were identified.

Abstract

This paper analysed the citations of patents to science- and non-science-based references as an agency of the linkage between technology and science. A review of the literature identified a variety of techniques of measuring science linkage (SL) with various results. Therefore, this study aimed to compare the differences between science-based SL and non-science-based linkage (NSL). Patent data were collected from the United States Patent and Trademark Office database for the past two decades. Results showed a phenomenon of rapidly growing NSL of patents at different levels of technological fields and firms. In addition, field- and firm-specific differences in the linkages between science and technology were identified. This study analysed various types of SL performances of the top 20 firms in the Computers and Communications field and found that science–technology linkages were stronger in Lucent, Mitsubishi and Microsoft. It is worth noting that Texas Instruments (TI) was ranked thirteenth in science-based SL but third in Relative SL Ratio. Based on the Relative SL Ratio, TI's science-based SL was much higher than its NSL.

Introduction

Research has shown that invention and innovation originate in basic and applied research, progressing into technological and economic growth (Rosenberg and Birdzel, 1990, Narin and Olivastro, 1992, Narin et al., 1995, Narin, 2000, Acosta and Coronado, 2003, Klitkou and Gulbrandsen, 2010). This positive relationship between important technological advances and economic outcomes has been statistically measured by high citations between science (using papers as a proxy) and technology (using patents as a proxy).

Science linkage (SL) was first proposed and defined by Narin in 1991 as ‘the average number of “other references cited” on the front pages of U.S. patents’. However, the SL formula and meaning were not clarified in detail until the publication of the Tech-Line® Background Paper in 2000. While its importance and popularity are evidenced in the literature, there is no common agreement as to what counts as a “science-based reference” for the purpose of SL calculations. In Narin's (2000) definition of the SL indicator, he excluded non-science-based references from non-patent references (NPRs). Similarly, Meyer, 2000a, Meyer, 2000b, who explored the different motivations for citing non-patent literature, found that only some are science-related. In other words, not all NPRs are science-based references.

It is argued that counting all NPRs as science-based references cannot reflect real linkages between science and technology. It is also observed that science-based references include more than Thomson Reuters Science Citation Index (SCI) papers. Only considering SCI papers for SL calculations arguably cannot reflect real science–technology linkages either. The past two decades have seen an increasing growth of citations to non-science-based references, with the number of such citations overtaking the number of citations to science-based references. To this end, this study aims to compare the differences between science-based SL and non-science-based linkage (NSL).

This paper begins with a review of the literature on characteristics of SL, comparisons between science-based NPRs and non-science-based NPRs, and SL as an indicator of technological innovation. It goes on to explain the methodology employed in the empirical study. The patent data analysed are presented in diagrams and tables with textual explanations. Finally, implications, based on field- and firm-specific differences in the linkages between science and technology, are discussed in Section 4.

While substantial research has studied the links between science and technology, a review of the literature observed various methods of statistically measuring such links. A common quantitative method used to measure the relationship between science and technology is citation analysis (Coward and Franklin, 1989, Hicks et al., 2000, Tijssen, 2001, Acosta and Coronado, 2003, Bhattacharya and Meyer, 2003, Chen and Hicks, 2004, Gupta, 2006, Lo, 2010). Other methods used include: inventor-author name matches (Coward and Franklin, 1989, Boyack and Klavans, 2008), author–inventor co-publications (Breschi and Catalini, 2009, Klitkou and Gulbrandsen, 2010) and peer opinions through interviews and surveys (Fagerberg, 1987, Tijssen, 2002).

Among the various citation analysis techniques, SL has been widely adopted, to a varying extent, in the literature to measure the links between science and technology. The central idea of the indicator of SL is to use the “other references cited” (Narin & Olivastro, 1992) or “non-patent references” (Narin, 2000) cited on the front page of granted U.S. patents to directly measure the link. According to CHI Research, Inc. (Narin, 2000), SL was defined as “the average number of science papers referenced on the front page of the company's patents” to indicate “how leading edge [the] company's technology [is]”.

When Narin (2000) first proposed the SL indicator, he included only science-based references and excluded other non-patent literature references from the SL calculations. A complicated issue resides in what counts as science-based references, as not all NPRs cited on the front page of granted U.S. patents belong to science-based references. Although complicated, this distinction is considered to be essential, because linkage to science is the driving force behind many important areas of technology.

As mentioned in the previous section, the SL definition (Narin, 2000) clearly pointed out the necessity of distinguishing science-based references from non-science-based references among the NPRs. For example, science-based NPRs include scientific journal papers, meetings and books, whereas non-science-based NPRs include industrial standards, technical disclosures, engineering manuals and every other conceivable kind of published material (Narin, Hamilton, & Olivastro, 1997).

However, the application of SL in the following publications varied, with two distinctive approaches observed in the literature. When selecting science-based references for the SL calculations, one approach included all NPRs (e.g. Gupta, 2006, Lo, 2010) and the other used only papers listed in the SCI database (e.g. Tijssen et al., 2000, Tijssen, 2001, Verbeek et al., 2002, Hullmann and Meyer, 2003, Han, 2007). Somewhere between the two approaches are, for example, Park and Kang (2009), including academic journals and conference proceedings as science-based references.

It is clear that the indicator of SL has been widely applied in the literature after Narin (2000). Different from Narin's (2000) approach, many authors did not exclude non-science-based references from NPRs when calculating SL. However, it is observed that Narin (2000) included patent reports as science-based references. In order to reflect real linkages between science and technology, this study adopts Narin's definition of SL but argues that a more rigid criterion to distinguish science-based SL and NSL is needed.

Generally speaking, a lot of research employed the SL indicator and concluded that there is a positive relationship between technology and research (Narin et al., 1997, Sorenson and Fleming, 2004, Klitkou and Gulbrandsen, 2010). It was identified in the literature that the SL indicator was widely applied at different levels, e.g. at the firm level (Narin, 2000), at the country level (Narin, 1991, Narin, 1994, Narin and Olivastro, 1992, Narin et al., 1997, Narin et al., 2000, Gupta, 2006, Klitkou and Gulbrandsen, 2010, Park and Kang, 2009) and at the level of technological fields (Narin and Olivastro, 1992, Anderson et al., 1996, Hullmann and Meyer, 2003, Chen and Hicks, 2004, Lo, 2010). This study builds upon the work of Narin and other scholars and further explores the SL comparisons between firms, countries and technological fields.

Despite its popularity in the literature, some debates in the literature concern whether the science–technology relationship is causal. For example, Sirilli (1998) stated that “we have no explicit model capable of determining causal relations between science, technology, economy and society in a single synthesis.” In line with this, Meyer (1998) also argued that “it is necessary to corroborate these measures more carefully first and investigate the question whether a citation linkage really equals a causal link.” Such authors saw a need for a balanced analysis between quantitative and qualitative measurement of science–technology linkages.

Furthermore, SL was developed as a quantitative indicator to measure science–technology linkages. However, to what extent does it really reflect the linkages between science and technology? As discussed in the previous section, a key issue resides in the distinction between science-based SL and NSL.

Section snippets

Patent data and technological fields

In this study, patent data were collected solely from the United States Patent and Trademark Office (USPTO) database, which is generally accepted and is accessible to the researchers. While there exist different categories of patents (e.g. plant patents, design patents, reissues and continuations), this study, based on the recommendation by Narin (2000), collected the number of regular U.S. utility patents to keep the focus of the database on the key category of patents which contributes to

Results

This section starts with the citation trend of various types of NPRs, and proceeds to compare differences between science-based SL and NSL. It then examines the SL changes and growth rates of the top 20 patent assignees in the C&C field.

Conclusion and discussion

This study has met the aims set at the outset of this paper. Specifically, this study divided NPRs into five categories and classified books, conference papers and journal papers into the group of science-based NPRs. It further considered not only science-based SL but also the ratio of science-based SL to NSL compared to the overall performance in the field. Interesting observations were also made regarding rapidly growing NSL of patents and differences of science-based SL and NSL at the levels

References (35)

  • J. Anderson et al.

    Human genetic technology: Exploring the links between science and innovation

    Technology Analysis & Strategic Management

    (1996)
  • S. Bhattacharya et al.

    Large firms and the science–technology interface: Patents, patent citations, and scientific output of multinational corporations in thin films

    Scientometrics

    (2003)
  • S. Breschi et al.

    Tracing the links between science and technology: An exploratory analysis of scientists’ and inventors’ networks

  • C. Chen et al.

    Tracing knowledge diffusion

    Scientometrics

    (2004)
  • H.R. Coward et al.

    Identifying the science–technology interface: Matching patent data to a bibliometric model

    Science, Technology, & Human Values

    (1989)
  • V.K. Gupta

    References to literature in patent documents: A case study of CSIR in India

    Scientometrics

    (2006)
  • B.H. Hall et al.

    The NBER patent citation data file: Lessons, insights and methodological tools

    (2001)
  • Cited by (21)

    • Probing into the interactions between papers and patents of new CRISPR/CAS9 technology: A citation comparison

      2021, Journal of Informetrics
      Citation Excerpt :

      Therefore, although there are still conflicting opinions about the role of scientific references in patents, this paper considers that they are still feasible means for analyzing the interactions between science and technology and should be taken seriously. Sung, Wang, Huang and Chen (2015) used science linkage (SL) indicator to measure the difference between science-based SL and non-science-based linkage (NSL). Their results indicate that NSL is growing and that dissimilarities are appearing across fields.

    • How pharmaceutical innovation evolves: The path from science to technological development to marketable drugs

      2021, Technological Forecasting and Social Change
      Citation Excerpt :

      In Glanzel and Meyer's (2003) study, scientific articles that cited patent documents revealed to the extent to which technological inventions had penetrated basic research. Sung et al. (2015) showed the reverse, using the citations from patents to science- and non-science-based references to show agency in the links between technology and science. They also identified field- and firm-specific differences in the linkages between science and technology.

    • The growing number of patent citations to scientific papers: Changes in the world, nations, and fields

      2020, Technology in Society
      Citation Excerpt :

      Also, this relationship has frequently been studied in technological fields with a more extensive number of citations to scientific papers, so the findings should not be generalized without caution [43]. Using NPRs, one group of research shows the degree of science-technology linkage, which includes non-science references in NPRs [44]. This paper aims to fill this gap by analyzing patent references to scientific literature over the 1998–2017 period in all disciplines.

    • Measuring the knowledge translation and convergence in pharmaceutical innovation by funding-science-technology-innovation linkages analysis

      2019, Journal of Informetrics
      Citation Excerpt :

      The non-patent literature cited by a patent document measures the translation from basic research to technology invention (Narin & Noma, 1985), while patent documents cited by scientific articles reveal the penetration of technology innovation into basic research (Glänzel & Meyer, 2003). Several studies have suggested that the biomedical patents cited more basic research publications and thus the biomedical industry relied on public science more than other industries (Jibu, 2014; McMillan, Narin, & Deeds, 2000; Sung, Wang, Huang, & Chen, 2015). Measuring citations to scholarly works in the global patent literature enables assessment of the influence of published research on invention, industry and enterprise, at the individual and institutional level.

    View all citing articles on Scopus
    View full text