July 16, 2019

Today we visited the library at the Technical University in Delft, which is a very cool building! We first got a tour of the space, and I have to say that truthfully the study space itself didn't really impress me. It looked about the same as any average university

The entrance to the library is underneath the cone, but once you're inside you can sit all the way up the cone!

library, which is definitely not a bad thing, but after seeing so many outrageous libraries on this trip I have started to expect being blown away. One thing which our tour guide mentioned that did blow me away, however, was that TU provides a data archiving service so that all research being done at the university can be preserved, regardless of its publication status. This mention stuck out to me as a psychology major, because our field has a major issue with replication studies currently (so much so that we refer to it as the replication crisis). Basically, the issue is that many of the formative experiments in the field of psychology are no longer showing significant findings as present-day researchers attempt to conduct the same studies again. While some of this may be due to the passage of time, it seems unlikely that this replication issue would happen on such a large scale purely due to chance. One major contributor to this issue is the pressure to publish new findings. This is not just pressure on the part of academics, but also from the journals themselves. It's exceedingly rare that a journal will publish a replication study, when in fact a finding shouldn't be taken as reliable until a replication has been done. As a result, researchers who conduct replications are not able to release their findings easily to other researchers. Having a data archive such as the one set up by TU Delft is beneficial for other researchers, because they are able to see findings showing reliability across data sets without relying on published data. This is important for future research, as well, because research builds on what has been previously done, but if prior research is not as reliable as it should be, then all research branching from the original findings can also not be considered reliable.

Another widespread issue in psychology journals is their refusal to publish null findings. In reality, null results are integral to good research, because they prevent future researchers from conducting the exact same study and wasting time and resources, but also because null results can still give significant information. For example, if you wanted to explore whether hair color was related to intelligence, and found null results, that gives crucial information that hair color is not related to intelligence. Since journals are currently still opposed to publishing these findings, however, data archives like the one at TU Delft help researchers use null findings to inform their future research by allowing users to contribute their unpublished results.

This data archive at TU reminded me of one of the readings for today, in chapter 4 of Leading Public Sector Innovation by Bosan. In the chapter, he describes strategic innovation as a term to describe efforts to address the "what" of an issue. By identifying concrete problems, you can then come up with specific solutions and strategic innovations that help amend these issues. Initially I thought maybe I was seeing this data archive as strategic innovation because I had identified the issue that was most relevant to me (the replication crisis), but when I later went back to do more research on the data archive itself, the TU Delft website describes the archive as designed to address replicability issues, which I was really excited to see. To me, it is the mark of a very successful innovation that I, as an outsider, was able to trace the product back to the original issue that it was targeted to address. In the future, I would love to see this archive on a larger scale, because science is meant to be shared so that researchers from all over the world can collaborate. On a smaller scale, though, I think archives on the same scale as TU could be implemented at all major research universities, because even just on the UW campus there is enough research data to have a major impact on researchers employed by the school.

After our visit to the TU library, we had a break for lunch and then reconvened at the hotel for a class session and debrief of the speakers from the library.