Search

Technological change is categorized into two types » Comes to exist as entities that depart in some deep sense from what went before, and » Builds on and enhances its predecessors and therefore has the result of consolidating the status quo. No quantitative measure exist to examine the distinction. The article creates such a measure. Based on the network model of technological change, a measure is proposed called index, that quantifies the extent to which an invention consolidates or destabilizes the subsequent use of the components on which it builds. Then the was combined with an impact weight that characterizes the magnitude of an invention's consolidation or destabilization of the status quo. The indexes were applied in the context of university research commercialization. The results showed that when universities have greater commercial engagement, they tend to create technologies that consolidate the status quo but, when they receive more federal funding for academic research, they tend to produce more destabilizing inventions. Commercial engagement and federal support are both positively associated with forward citations of university patents. (90 refs.) Results: The article discusses some of the existing measures of technological change. Our measure has four core features. The index developed has four features: » It is structural in a network sense, » The measure is dynamic, » Continuous, and » Is valenced. The measures were developed using utility patents granted by the US Patent and Trademark Office. The article focuses only on utility patents to prevent complexities that arise from differences in citation practices across categories. The index characterizes how future inventions make use of the technological predecessors cited by a focal patent. The intuition was that citations of predecessors should decrease after a destabilizing invention is introduced because the technology entails a break with past ways of thinking. By contrast, consolidating inventions should be cited together with their predecessors and therefore increase citations of technologies on which they build. Figure 1 in the original article explains the idea. Based on the title tools of network science, the measure was developed by conceptualizing patents as nodes in tripartite networks. The illustrative calculations of the for three patents is presented in Figure 1 of the original article. The index captures the direction of an invention's effects on existing technologies, not the magnitude. The index introduces a weighting parameter (Equation 1), the magnitude of a focal invention's future use can be incorporated. The measure differs from the index by distinguishing among inventions according to their overall effect on a network of interlinked technologies. Whereas the index captures the direction of an invention's effects, whereas the index mixes both direction and magnitude. Based on a data on the 2.9 million U.S. utility patents granted between 1977 and 2005, the ability of the indexes to discriminate among inventions was made. The basic covariates were obtained from the patent network Dataverse and supplemented with data taken from the US Patent and Trademark Office. The analysis evaluated impact and the and indexes using forward citations made during the first five years after the focal patent's issue. These are labeled as mid , respectively. Table 1 in the original article reports descriptive statistics and correlations. The results showed that firms tend to produce more consolidating inventions, whereas universities and government laboratories. Companies tend to specialize in shorter-term, application-oriented development whereas public sector organizations more often focus on longer-term fundamental research. Larger teams may produce more destabilizing technologies In terms of within-year variability of the index, results showed that across technology categories, the average patent of today was less destabilizing than those of the late 1970s and early 1980s. The article examines how the measure was able to identify and classify particular breakthroughs through case studies. Table 2 in the original article reports the impact and and indexes measured five years after issue and as of 2010 for select inventions that have had a significant influence on their fields. The measure also identifies technologies that are known to have destabilized the use of their predecessors. A detailed consideration of three inventions - glyphosate-resistant soybeans, a method of ranking of online search results, and a eukaryotic co-transformation technique - that occupy different locations on the consolidating-destabilizing spectrum are chosen to explain. The results are presented in Figure 3 in the original article on the network diagrams on Monsanto. The case studies showed that the index is essentially uncorrelated with impact, and the technologies the approach identifies as consolidating or destabilizing are sensible and echo published evaluations of high-profile technologies. The paper then explores the value of the and indexes for substantive research. This was done by using sets of regressions that explore arguments from current literature about the determinants of important new technologies. Hence, the models at the patent level are presented followed by adaptations of the and indexes at the organization level in analyses of university patenting. The sample involved 2.9 million U.S. utility patents that » Were granted after 1976 but before 2006, » Were available in full-text form from the US Patent Office, and » Were assigned to one of the six NBER technology categories. The patent importance was measured using three indicators Mid mCD5 indexes and using citations received during the first five years after being granted. The regression involved several patent, team and US Patent Office covariates. Ordinary least squares regressions were the statistical approach used to model the mid indexes. Table 3 in the original article shows the estimates from four negative binomial models of patent impact . The results showed that patents awarded to a firm or university are more highly cited, whereas those with a government assignee are less acknowledged by future inventors. The relationship between technological importance and the use of existing knowledge is more complex than is apparent from models of impact. Both universities and firms tend to produce higher-impact inventions than other types of organizations, but that similarity masks important differences in how the technologies enter and change existing technological trajectories. Geographically distributed teams may have a greater tendency to create inventions that consolidate the status quo. Examiners require more time to evaluate patents that depart from the status quo and therefore their prior experience. The paper then discusses the ability of the and indexes on patenting by organizations. The analysis involved U.S. utility patents issued to the 110 most research-intensive American universities 55,322 high- and low-impact inventions utility patents awarded to these institutions between 1976 and 2010. A brief survey literature on the scale and impact of academic patenting was discussed. Six measures of portfolio importance were taken into consideration. The covariates were in terms of technology transfer experience, industry ties, government ties, scientific capacity. The statistical approach involved conditional fixed-effects negative binomial specification. The results showed that universities that have greater prior experience with patenting (patents) produce more and higher impact patents. A greater focus on Life sciences patenting, however, is associated with diminished patent impact. Increased industrial funding is modestly associated with more and higher-impact patenting, but campuses with more corporate alliances patent less, and this type of corporate connection has no significant relationship with patent impact. Increases in research funding from key public sources (NSF grants (log), NIH grants (log), and DoD grants (log)) demonstrate largely positive relationships with patent volume and impact. Universities that produce higher-impact portfolios of articles (Impact factor), however, also develop higher-impact patents that do more to destabilize the status quo. Both measures of engagement with corporate partners (Industry sponsored R&D) and Industry R&D ties) are strongly associated with patenting that consolidates existing technology trajectories. In contrast, increases in funding from all three sources of public research support (NSF grants, NIH grants, and DoD grants) are linked to more destabilizing patent portfolios. A higher volume of scientific papers is associated with higher-impact consolidating inventions. In contrast, a higher impact portfolio of scientific publications is associated with a more broadly used portfolio of patents that destabilize the use of the technologies on which they build. The paper then discusses some of the applications of the approach on social networks, industries and firms, science policy, geography and space, the evolution of management fads and fashions, the visibility of music artists and genres, the commercial and critical impact of films, and the larger legal implications of new court decisions. [ABSTRACT FROM AUTHOR]