5.2 Data collection and analysis
5.2.1 Logistic regression
Since the dependent variables (opposed/not-opposed; litigated/not-litigated; settled/not-settled) are binary, logistic regression model with robust standard errors was used to estimate the likelihood of opposition, litigation, and settlement. A number of patent characteristics are used as controls.
The logistic regression coefficients by default do not measure the change in dependent variable by unit change in independent variable. Marginal effects are computed to study the exact amount of change in dependent variable caused by the independent variables.
Marginal effects pertaining to the logistic regressions are computed but not shown in the results. This is because I do not wish to study how much the dependent variable changes with changes in independent variable. The results are not meant to be interpreted in a causal way but to test our theory and hypotheses.
The unit of analysis for regressions studying opposition and litigation is a patent grant.
For studying settlements, regressions are carried out at case level. This is because one patent might be involved in several cases, each having a different settlement outcome. A single case may also involve multiple patents but their settlement outcomes can be coded as settled or non-settled (a binary variable). Since the independent variables are the procedural patent characteristics, the standard errors were clustered at the patent level involved in a case.
6 Results and discussion
This section summarizes the results from five selected publications, in light of the research questions introduced earlier. Some of the results stretch beyond the purview of the research questions but I found them necessary somehow for the discerning reader.
The publications are not necessarily summarized in a specific order. The following sub-sections are based on the concepts and constructs introduced earlier in the dissertation. I have added an extra section on my unpublished work related to differences between patents involved in infringement and invalidity lawsuits.
6.1
The innovation life cycleThe basic premise of this dissertation is that patents are not one-shot actions. They follow a life cycle of their own which is strongly linked to the life cycle of the corresponding innovation. Publication 1 is a compilation of findings from 11 semi-structured interviews.
The objective was to compare the views of various experts about the concept of patent value and the relevance of patent value indicators. Research question 1 seeks to answer
“how are patent value indicators related to the life cycle of innovations”. The list of interviewees presented in Table 3 represented different stages in the innovation life cycle that are divided among the following five categories.
1. Inventors (Professors owning patents; R&D managers applying for patents) 2. Technology transfer office personnel (University and small-medium firm level) 3. Patent attorneys (Individuals filing and maintaining patents on behalf of clients) 4. Patent consultants (Individuals involved in advice and/or transaction of patents) 5. Patent examiners (Individuals reviewing patentability at patent offices)
Table 3 Interviewee details
Serial Designation Category Years of Experience
1 Professor, Lappeenranta University of
Technology Inventor More than 20 years
2 Professor, Lappeenranta University of
Technology Inventor More than 30 years
3 R&D Manager, Industrial Firm Inventor More than 20 years
4 Project Coordinator, Lappeenranta University of Technology
Technology Transfer
professional More than 10 years 5 Development Personnel, Foundation for
Finnish Inventions
Technology Transfer
professional More than 10 years 6 Former Director, National Board of
Patents and Registration of Finland Patent Consultant More than 30 years 7 Private Technology Consultant Patent Consultant More than 20 years
8 Patent Attorney, Software Patent Attorney More than 10 years
9 IPR Specialist, Independent Patent Attorney Less than 10 years
10
Senior Patent Examiner, National Board of Patents and Registration of
Finland
Patent Examiner Less than 10 years
11
IP communication specialist, National Board of Patents and Registration of
Finland
Patent Examiner More than 10 years
Research question 1 is answered in Table 4 where views of different stakeholders regarding complementary market-based indicators are tabulated. Commercialization of innovation has a lot to do with market forces that influence decisions of inventors. In this regard, patents are not an end in themselves but complement various market forces at various points in time.
Table 4 Value indicators during different stages
Innovation Stage Value Indicators
Patent Based Market Based
Idea Market Potential
Invention announcement
Market reports
Theoretical estimate
Discussion about
patentability Preliminary novelty search
No. of competitors
Potential Market Size
No. of alternate solutions
No. of patents possible to infringe
No. of products patent can be applied
Patent application (priority filing)
Patenting route
Applicant references (if any)
No. of independent claims.
Text of claims
History of success of applicant
Experience of patent attorney
Past success of applicant
Working prototype Foreign filing (12
months) Family size Countries chosen
Search report
Citation categories (X’s, Y’s, A’s)
No. of IPC fields
No. of claims found novel or carrying inventive step
Likelihood of grant: Examiners comments
Publication (18 months)
No. of Claims
No. of backward citations
Text of claims
Citation categories
No. of IPC
No. of IPC of backward citations
Negotiation of investments and licenses
Opposition (if any) Opposition event raised
Opposition survival
Grant Forward citations
Readiness of invention for commercial purposes
Filing trademarks
The overall objective of the first research stage was to explore patent value indicators throughout this life cycle. This was achieved by looking at detailed procedural information of patents filed at the European Patent Office (EPO). While publication 1 sets the foundation for such an exercise. Detailed usage of relevant patent value indicators with an appreciation towards their availability in time is shown in Publications 4 and 5.
Publication 4 separates citations introduced in a patent document during different stages of the patent life cycle. Publication 5 uses various interactions of applicants and their rivals with the patent office.