9. New indicators and approaches in STI policy: beyond data and prediction to impacting the policy cycle

Convenors: Arho Suominen, Rainer Frietsch, Torben Schubert, Ad Notten, C.P. (Cees) van Beers, Scott Cunningham and Vilius Stančiauskas.

Session type: Full paper session.


The frontier of research in Science, Technology and Innovation has been moving rapidly within recent years. An important area in demand for immediate attention is the utilization and sense-making of increasing potentials to make use of big data. One of the aspects affecting both the depth and breadth of research has been novel methods, particularly quantitative approaches enabling the use of ever-expanding data sources but also analysis capabilities (Pirog 2014). Methodological advances have allowed several novel ideas to be tested on how to model economies, societies and policies affecting the direction of science and technology (e.g. Ranaei et al. 2019). However, new indicators and approaches are still lacking in their robustness, reproducibility and validity.

The ever-increasing amount of data offers new possibilities for discovering new relationships and inference on a multitude of problems. Most of the data that has gathered is unstructured and therefore in need of restructuring and cleaning prior to making use by existing machine learning methods (Kim, Trimi and Chung, 2014). New indicators and data come with new challenges involving reproducibility, complexity, security, and risks to privacy (Jarmin & O’Hara 2016), as well as a need for new technology and human skills. This is very much the case in public policy where we need to clearly identify where new indicators and data can add value in an ethical and trustworthy manner.

The proposed session will focus on the impact of new indicators and approaches to STI policy. The session will focus on how to address the challenges of producing new indicators and approaches that are policy-relevant. In specific, we will address the challenges highlighted by Giest (2017):

The increase in unstructured data calls for standardization of data and indicators construction. Creating robust indicators in a standardized manner from unstructured data is central to the reproducibility of the results. This challenge also extends to analysis methods, where novel machine learning methods have been critiqued on the lack of reproducibility (for discussion e.g. Glänzel et. al. 2017).

New indicators and approaches require that the policy developing institutions are able to utilize the tools provided (for discussion e.g Desouza and Jacob 2017). Policymakers need to interact easily with the data and analysis and the result needs to be attainable and robust (Zhang et al 2017).

Novel indicators often stem from the digitalization of public and private services. Would it be social media, websites, or open data, novel indicators stem from access to digitalized data (Jarmin & O’Hara 2016). Questions however remain, if these new data sources are sustained long term and will remain freely available also in the future. We also have questions on who and to what end curates different sources of data. There needs to also be better understanding when data collection, particularly when linked to secondary data, can infringe on the privacy of individuals (Stough and McBride 2014).

New indicators and approaches need to go “beyond predictions” *(Athey 2017) and demonstrate an impact on the policy cycle. How and to what extent new indicators and approaches enable new types of interactions to the policy cycle via e.g. continuously available tools or self-adapting systems to policy decisions.

In these, Giest (2017) highlights two issues, the substantive and the procedural role of new indicators and approaches in policy instruments. Procedural activities focus on regulatory activities, such as enabling open data, while the substantive actions relate collecting data to enhance for example evidence-based policymaking capacities, digitalization and the role of new indicators and approaches in the (substantive and procedural) policy cycle are core to the digital-era governance and evidence-based policymaking. The session hopes to address both of these issues.

To advance our understanding of the challenges mention, this session welcomes submission addressing in particular:

  • Novel indicators and analysis methods, where the submissions are particularly encouraged to focus on how to standardize data and ensure reproducibility of the analysis.
  • How to address issues arising from novel data sources such as using commercial data not designed for STI indicators, privacy issues or capturing benefits from digitalization of public services.
  • Case studies focusing on how big data has been used in public policymaking, where the submissions are particularly encouraged to focus on how the policy cycle was impacted.

The conference track will invite selected publications for consideration in a special issue with the journal of Technological Forecasting and Social Change. The special issue welcomes primarily empirical contributions adressing questions related to the effects of STI on productivity, competitiveness and growth. Submissions should have strong and important implications for STI policy. In terms of methodology, the issue expresses a preference for innovative approaches leveraging novel, big or unstructured data sources. Traditional approaches are also acceptable, if they demonstrate a potential to offer novel and highly relevant insights for STI policy. Qualitative approaches are welcome, if the work explicitly focuses on the integration of novel approached or measures into the decisionmaking process.

The session is proposed by the consortium members of the H2020 funded project “Addressing productivity paradox with big data: implications to policy making”.

 

References

Athey, S. (2017). Beyond prediction: Using big data for policy problems. Science, 355(6324), 483-485.

Giest, S. (2017). Big data for policy making: fad or fasttrack?. Policy Sciences, 50(3), 367-382.

Glänzel, W., Moed, H. F., Schmoch, U., & Thelwall, M. (2017). Springer Handbook of Science and Technology Indicators. Springer Nature.

Jarmin, R. S., & O’Hara, A. B. (2016). Big data and the transformation of public policy analysis. Journal of Policy Analysis and Management, 35(3), 715-721.

Kim, G. H., Trimi, S., & Chung, J. H. (2014). Big-data applications in the government sector. Communications of the ACM, 57(3), 78-85.

Pirog, M. A. (2014). Data will drive innovation in public policy and management research in the next decade. Journal of Policy Analysis and Management, 33(2), 537-543.

Ranaei, S., Suominen, A., Porter, A., & Kässi, T. (2019). Application of Text-Analytics in Quantitative Study of Science and Technology. In Springer Handbook of Science and Technology Indicators (pp. 957-982). Springer, Cham.

Stough, R., & McBride, D. (2014). Big data and US public policy. Review of Policy Research, 31(4), 339-342.

Zhang, Y., Porter, A. L., Cunningham, S., Chiavetta, D., & Newman, N. (2018, August). How is Data Science Involved in Policy Analysis?: A Bibliometric Perspective. In 2018 Portland International Conference on Management of Engineering and Technology (PICMET) (pp. 1-10). IEEE.