Abstract
In the Corona pandemic, it became clear with burning clarity how much good quality statistics are needed, and at the same time how unsuccessful we are at providing such statistics despite the existing technical and methodological possibilities and diverse data sources. It is therefore more than overdue to get to the bottom of the causes of these issues and to learn from the findings. This defines a high aspiration, namely that firstly a diagnosis is carried out in which the causes of the deficiencies with their interactions are identified as broadly as possible. Secondly, such a broad diagnosis should result in a therapy that includes a coherent strategy that can be generalised, i.e. that goes beyond the Corona pandemic.
Similar content being viewed by others
Explore related subjects
Find the latest articles, discoveries, and news in related topics.Avoid common mistakes on your manuscript.
1 How statistics can help
Given the urgency of the Corona pandemic, it seems natural to focus the question of the role of statistics on this. In the article “On the role of data, statistics and decisions in a pandemic” (Jahn et al. 2022), a cooperation of authors from very different fields pursues the very commendable approach of compiling the essential findings for statistics in a kind of inventory after about two years of the pandemic. In this short contribution, however, a broader perspective and a longer horizon are advocated (Radermacher 2020, 2021a). It is above all the need for transformation towards sustainable development in many areas of life, in production and consumption, in mobility and nutrition, etc., that will call for new statistics as a basis for political decisions and discourse in the years to come. We must adapt to this as quickly as possible so that the quality of the facts meets the demands (in time). In a strategic approach that aims to help statistics achieve its potential and adequate validity in its role for modern societies in coping with their problems, it is crucial that the different facets of this discipline are meaningfully linked. It is by no means sufficient to string together method development, applications, etc. as isolated elements. Rather, it is important to combine these puzzle elements into an overall picture of a coordinated and balanced strategy contained therein.Footnote 1 The “Data Manifesto” (Royal Statistical Society 2019) of the Royal Statistical Society shows that it is possible to bring together the various sectors of the statistical community with their specific perspectives in a coherent and policy-oriented strategy.
2 What is ‘statistics’?
“Factual truth … is political by nature. Facts and opinions, though they must be kept apart, are not antagonistic to each other; they belong to the same realm. … Seen from the viewpoint of politics, truth has a despotic character. It is therefore hated by tyrants, who rightly fear the competition of a coercive force they cannot monopolize, and it enjoys a rather precarious status in the eyes of governments that rest on consent and abhor coercion. Facts are beyond agreement and consent, and all talk about them – all exchanges of opinion based on correct information – will contribute nothing to their establishment. Unwelcome opinion can be argued with, rejected, or compromised upon, but unwelcome facts possess an infuriating stubbornness that nothing can move except plain lies.” (Arendt, 1967, p. 7).
Statistics are by nature political in a special way; they are the object of opinion, but without themselves being influenced by opinions and interests. How can statistics do justice to this difficult task of being factual information in the sense of Hannah Arendt? How can facts be politically relevant but not politically driven?
The answer to these questions is multi-layered:
-
It includes first and foremost good mathematical-statistical methodology
-
and
-
a modern technology based on reliable data sources and efficient processing procedures.
But that is not all:
-
Statistical processes require a sustainably maintained infrastructure, be it for research on methods and development of technologies, be it for the ‘fabrication’ of statistical information.
-
Quality of statistical information proves itself in the ‘fitness for purpose’ test: Are the facts provided relevant, are they correctly understood, do they meet (realistic) expectations? This dimension of quality is achieved as a goal when all parties involved, i.e. the producers and the users of facts, data, trends, speak a common language.
It is therefore necessary to approach the task with several objectives, which are partly in conflict to each other, and considered from different angles at the same time (see Fig. 1). If statistical results are not measurements conceived according to a naïve realism, but are crucially convention-dependent quantifications, how can they nonetheless be objective, neutral, and of trustworthy quality? How can the quality of statistical products defined in this way be achieved, guaranteed and communicated? What conditions must be met so that this theoretically possible quality is actually realised? Last but not least, which ethics, values and governance matter, is there a statistics-friendly or hostile ‘culture’?
3 Statistics: old school or trendy?
Statistics is called upon to position and assert itself as an independent discipline in association with other data sciences. This is about cooperation, new possibilities and synergies. From a purely methodological-technical perspective, the differences between data science, artificial intelligence and statistics appear less serious. Rather, it is other factors such as values, quality definition or professional self-image that make the difference.Footnote 2 Both on the side of the producers of quantitative information and on the side of the users, there are very different ‘data worlds’ in this respect, which becomes particularly clear in the comparison of the 'big data world' with official statistics (Radermacher 2021d).
4 What is on the political agenda?
When analysing the role of data, statistics and decisions in a crisis (such as the Covid-19 pandemic), it is necessary to include policy programmes. This can be done, for example, with the help of the following recent strategically oriented policy plans:
-
Coalition agreement of the German Federal GovernmentFootnote 3
-
European Data Strategy,Footnote 4 Europe’s Digital DecadeFootnote 5 of the European Commission
It is remarkable how absent ‘statistics’ is from these important plans, not only in terms of methods, but also in terms of infrastructure or language. Apparently, the dominant idea among those who develop and discuss such strategies is that nowadays data is ubiquitous (i.e. no longer needs to be collected), that intelligent algorithms automatically do the work of filtering the signals from the noise, and that it is a (meaningful) goal of digital education that every citizen has the skills of a data scientist and can produce his or her own statistics.
As a prerequisite for promoting improved data practices, there needs to be a deeper and broader analysis and understanding of what role data plays in contemporary society (Beaulieu and Leonelli 2021; Diaz-Bone and Horvath 2021; Henneguelle and Jatteau 2021). In this context, it is by no means sufficient to devote attention to the technical processes, the economics of data or similar. Likewise, it is not enough to focus the ethical and legal questions on the protection of individual privacy. With a proliferation of data, new dynamics are triggered that lead to the replacement of the previous well-ordered (infra)structures of statistical production based in the public sector by other forms of (decentralised and individual) use and production of data. The statistical logic of the past two centuries is being replaced by a data logic. Even if this statement comes across as too pointed, it nevertheless leads to an examination of the risks contained in this trend and measures to counter them effectively (Radermacher 2021b, 2021c).
5 Dare (more) progress
In order to achieve progress, which is made possible with and through statistics, different initiatives are needed at the same time. Only a short list of recommendation points and activities can be made here. Nevertheless, this list should give an overview of their breadth and diversity:
5.1 Methodology
-
Strengthening cooperation between method development and application
-
Promote links between data science and statistics
-
Use complementarity between survey statistics and analysis of existing sources.
5.2 Technology
-
Advance and connect efficient technologies in all possible forms (especially AI, machine learning)
-
Develop quality management for newer technologies
-
Apply consequently an Open Data Strategy (access for research while protecting privacy of data).
5.3 Infrastructure
Official statistics is a public infrastructure that (like rail transport) needs to be maintained, expanded and modernised. The following tasks and questions are up for solution:
-
Balance decentralisation and centralisation tendencies in the implementation of digitalisation offensives
-
Data Stewardship: make use of knowledge and capacities in existing institutions
-
From official to public statistics: coordinate the larger system by a chief statisticianFootnote 6
-
'Data Institute': make use of knowledge and capacities in existing institutions and enhance networks between research and application
-
Establish a forward looking programming of public statistics
-
Governance: modernise the German Federal Statistical Law adapting to modern standards, such as the EU Statistics Regulation 223 (European Union, 2015).
5.4 Language
-
Invest in Data Literacy, support activities such as International Standard DevelopmentFootnote 7 or Data Literacy ChartersFootnote 8
-
Include statistical expertise in political expert panels.
5.5 Ethics, values
-
Promote the existing principles and guidelines for professional ethicsFootnote 9 and good governanceFootnote 10
-
Review the guidelines and revise them as needed to cover newer methods in data science.
5.6 Data culture
Overall, it is about actively promoting a data culture. This goal goes far beyond conveying ‘digital skills’. In fact, in many cases, corresponding activities have already been initiated and have produced concrete results. Such activities should be built upon and further networked and strengthened (Radermacher 2022).
6 Improving the representation of scientific interests
All in all, the statistical community should become aware of and face the task of proactively offering its contribution to solving the upcoming critical transformation processes in the crises of the coming years. Statistics has a lot to offer in terms of potential solutions. Power brings with it responsibility.
Dealing with this will be the challenge of the statistical societies in the near future. Strategic guidelines such as a data manifestoFootnote 11 could provide us with orientation.
Notes
The book “Kunst Aufräumen” (Wehrli, 2002) may serve as an illustration.
See for example https://www.youtube.com/watch?v=uHGlCi9jOWY.
See W. Radermacher Technical Discussion in https://download.statistik-berlin-brandenburg.de/0e2a6bd7d6fd44ce/b43aa17fd047/Zeitschrift-20210304.pdf, p. 38.
References
Arendt, H.: Truth and Politics. The New Yorker (1967)
Beaulieu, A., Leonelli, S.: Data and Society: A Critical Introduction. SAGE Publications (2021)
Diaz-Bone, R., Horvath, K.: Official statistics, big data and civil society. Introducing the approach of “economics of convention” for understanding the rise of new data worlds and their implications. Stat. J. IAOS 37(1), 219–228 (2021). https://doi.org/10.3233/SJI-200733
Henneguelle, A., Jatteau, A.: Sociologie de la quantification. La Découverte (2021)
Jahn, B., Friedrich, S., Behnke, J., Engel, J., Garczarek, U., Münnich, R., Pauly, M., Wilhelm, A., Wolkenhauer, O., Zwick, M., Siebert, U., Friede, T.: On the role of data, statistics and decisions in a pandemic. Adv. Stat. Anal. (forthcoming) 36 (2022)
Radermacher, W.J.: Governing-by-the numbers—Résumé after one and a half years. Stat. J. IAOS 37(2), 701–711 (2021a). https://doi.org/10.3233/SJI-210819
Radermacher, W.J.: Standardisation and statistics. Surey Stat. 84, 24–31 (2021d)
Radermacher, W.J.: How statistics can help—going beyond COVID-19. Medium. https://medium.com/data-policy/how-statistics-can-help-going-beyond-covid-19-22bb2ce92440 (2020)
Radermacher, W.J.: Official statistics: language for public discourse (Part 1). Medium. https://medium.com/data-policy/official-statistics-language-for-public-discourse-part-1-294d31e56b6e (2021b)
Radermacher, W.J.: Official statistics: language for public discourse (Part 2). Medium. https://medium.com/data-policy/official-statistics-language-for-public-discourse-part-2-b9e912621523 (2021c)
Radermacher, W.J.: Statistical awareness promoting a data culture. Stat. J. IAOS (forthcoming) 13 (2022)
Regulation (EC) No 223/2009 of the European Parliament and of the Council of 11 March 2009 on European statistics, Regulation (EC) No 223/2009 C.F.R. (2015)
Royal Statistical Society.: The Data Manifesto. Royal Statistical Society RSS. Retrieved 26.02.2022 from https://rss.org.uk/policy-campaigns/policy/our-data-manifesto/ (2019)
Wehrli, U.: Kunst aufräumen. Kein & Aber (2002)
Funding
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Radermacher, W.J. Comment on: On the role of data, statistics and decisions in a pandemic statistics for climate protection and health—dare (more) progress!. AStA Adv Stat Anal 106, 391–397 (2022). https://doi.org/10.1007/s10182-022-00447-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10182-022-00447-7