easure the success of technology transfer from universities by using the total number of patents, licenses and spin-offs
Measuring the success (or efficiency) of technology transfer by using the revenue generated through the deals signed by universities may not be the best metric. Sometimes, there is a very big outlier that amounts to the majority of the money received by universities (imagine being at the other end of the Google licensing). It is possible to use the total number of Patents, Licenses and Spin-Offs as a metric for technology-transfer[@vinig2015Measuring the performance of university technology transfer using meta data approach: the case of Dutch universities]. However, this should be normalized, for example by the total number of published papers (on a given year). (see citation-based metrics incentivize competition over collaboration).
With this metric, MIT and Stanford are more efficient transferring technology than, for example, the university of Leiden1. The biggest problem with this approach is that patents and spin-offs do not have the same added value to the economy.
The paper has some mistakes (numbers wrongly added) and some methodological concerns: For example, they are obsessed calculating the 1%, 2%, 3% as the factor for the total number of published papers, which adds zero extra knowledge to their work. They also fail to disclose how they calculated the numbers of published papers for the MIT and Stanford. ↩
These are the other notes that link to this one.