The impact of ISO standards on the diffusion of scientific knowledge

Technical standards have gained increased importance in supporting technological innovation. Within the standardization process, standard-setting organizations (SSOs) build upon scientific and technological knowledge. Consequently, scientific publications can be referenced in a standard. These standard-relevant publications (SR-pubs) remain under-researched thus far, as the literature mainly focuses on standard-essential patents (SEPs). Besides providing some basic patterns regarding SR-pubs, we focus our analyses on the citation patterns of these standard-relevant publications. We hypothesize that a) SR-pubs are more highly cited than the average publication and b) that this is, on the one hand, due to a selection effect, i.e., more highly cited publications are more likely to be referenced in the standard from the start, but on the other hand due to a “reputation effect,” i.e., being referenced in a standard further increases the citation rates of a paper due to a boost in visibility. We test our assumptions using a dynamic difference-in-difference (DID) estimator on a sample of 6754 unique, standard-relevant publications cited in standards published by the International Organization for Standardization (ISO) and a control group that is ten times the size. The results show that SR-pubs are more highly cited than the average publication from the control group. We further find evidence for a strong selection effect, particularly for publications that have already been highly cited before being referenced in a standard. We can also confirm the reputation effect, albeit on a smaller scale than the selection effect, especially for publications that have previously been lowly cited before being referenced in a standard. • Standard-relevant publications receive significantly more citations than average papers • The selection effect drives high citation rates of standard-referenced papers • We can also confirm the reputation effect, albeit on a smaller scale than the selection effect