Irene Tuveng, Statistics Norway, itu@ssb.no

Timo Koskimäki, Statistics Finland, timo.koskimaki@stat.fi

Nina von Lachmann-Steensen, Statistics Denmark, NLS@dst.dk

Therese Hedlund, Statistics Sweden, Therese.Hedlund@scb.se

The Nordic National Statistical Institutes (NSIs) have been active players in development of statistical capacity in partner countries in the Global south for many years. We each have separate strategies and ways of funding and organising our work, but we share a common approach with a focus on lasting results through local ownership, building of solid foundations for a National Statistical System and development of expertise that stays in the partner institutions.

Today, the Nordic countries actively share information and experiences with each other. We try to avoid duplication and to put unnecessary burden on our cooperation partners. Even if we do cooperate in some partner countries, we do not have a framework nor a strategy of actively promoting and offering a Nordic institutional cooperation programme to partner countries.

In this paper we consider the possible gains of explicitly joining forces in our international development cooperation. We explore how we could organise and fund a Nordic programme and what could be the possible benefits compared with the current situation. We also investigate possible challenges of a Nordic institutional cooperation programme.

Some of the potential benefits of a Nordic cooperation programme we investigate in this paper are if it will make the Nordic NSI’s a more visible partner on the international arena, help promote the Nordic register based statistical system, increase the pool of experts, strengthen the Nordic statistical network, facilitate knowledge sharing across the Nordic NSIs, and enable larger, broader, and more flexible country programs for better results in partner countries. The co-operation model presented in this paper, could also serve as a model to organize ordinary statistical co-operation, in any domain of the statistical system.

Keywords: Global statistical development,Statistical co-operation.

When: Tuesday, August 23 at 15:00 - 15:30

Hotspots are short presentations during coffee breaks. They carry no paper.

Magnus Nørtoft, Statistics Denmark, mnt@dst.dk

European Statistics Code of Practice (CoP) is a guideline we all follow in our production of official statistics. And the statistics on local government economy in the Scandinavian countries are quite similar. However, an examination of the production process (from “Collect” to “Disseminate” in GSBPM ) of the statistics on local government economy in Denmark, Norway and Sweden show different priorities of the principles in CoP. These differences lead to different production and publication procedures and potentially different quality. I.e. Norway and Sweden do the first initial quality controls for obvious mistakes automatically even before accepting the report from the government bodies. In Denmark the NSI control and adjust the reports for obvious mistakes themselves. The Norwegian and Swedish approach priorities principle 10 (Cost Efficiency) while the Danish approach priorities principle 9 (Non-excessive burden). In other parts of the production, the same principle is interpreted differently. To live up to principle 13 and “timely manner” Sweden publish the data for specific municipalities whenever they are ready, in Norway and Denmark preliminary data for the whole population are publish to fulfill principle 13.5. This paper discuss these different approaches to production of local government statistics in relation to CoP, quality and best practice. In general, the paper finds that CoP is open to different priorities, thus different decisions in statistical production. At NSM2022 we would like to discuss if CoP is really open to different priorities and if this compromise the quality assurance system of European national statistics.

Keywords: European Statistics Code of Practice, Quality, Local Government economy, International comparison, best practice.

When: Tuesday, August 23 at 15:00 - 15:30

Hotspots are short presentations during coffee breaks. They carry no paper.

Annette Grasmo Bergman, Statistics Norway, bag@ssb.no

In a large organization, it can be challenging to carry out language washing and proofreading of texts that go out to the respondents. Some may forget to take clear language into account in planning, and others, like new employees, do not know the routines. How do we get clear language to be a mandatory part of survey planning?

January 1st, 2022, the Language Act entered into force in Norway. Statistics Norway is required to comply with section 9 of the Language Act: “Public bodies shall communicate in a clear and correct language that is adapted to the target group”. In December 2019, due to the upcoming act, the division for user contact and the division for housing, property, spatial and agricultural statistics started a collaboration to implement clear language in all texts for Census of agriculture 2020, which had 41 200 respondents.

The divisions mainly had two goals for the collaboration: Ensure clear and correct language in letters, e-mails, instructions and survey, and minimize the number of inquiries after the survey was sent to the respondents. A clear language also contributes to the overall data quality and validity of the survey.

The divisions collaborated on the project from December 2019 to October 2020. The project resulted in a clear improvement of the amount of questions, and gave valuable insight for future processes. Subsequently, a routine description has been made for clear language in surveys, based on the collaboration on Census of agriculture 2020.

The paper aims to highlight the benefits of working together cross-diciplinary between communications staff and statisticians from an early stage, and how to implement clear language in future survey planning.

Keywords: clear language, collaboration, respondents, survey, planners

When: Tuesday, August 23 at 15:00 - 15:30

Hotspots are short presentations during coffee breaks. They carry no paper.

Jarle Kvile, Statistics Norway, jarle.kvile@ssb.no

User-needs for new data sources in, for example, the collaborative economy are a pressing matter for all National Statistical Institutions (NSIs). The European Code of Practice (CoP) is a well-established framework for quality assurance but has shortcomings in the production of new statistics using alternative data sources. There has therefore been a rise of statistics labeled “experimental,” which try to meet users need for data, even though those statistics do not necessarily meet the quality standards that the Code of Practice assures. This paper shows a use-case scenario to complement best practices in quality assurance, when new data sources with no historical reference.

The paper uses administrative data from the Norwegian tax authorities on rentals through online platforms. As this data source is brand new, and has few direct comparisons, we cannot use the standard data analysis methods to identify statistical “outliers.” What is an outlier, if we do not know what to compare the data to? This paper contributes to methodology by using commonly available distributions, households’ income, and revenue at hotels, as well as Lorenz curves, to create a framework for determining which observations in new data sources should be labeled “outliers.” By using this framework, in combination with webscraping of relevant sources, less than 1 per cent of the data was marked as outliers, in comparison to 5 per cent when using traditional methods. With over half a million observations, this is of huge benefit for NSIs and their users. This paper takes the reader through this methodology, with in-depth analysis of each step, in order to show others how to implement the methods.

Keywords: experimental, collaborative economy, bigdata, quality, webscraping

When: Tuesday, August 23 at 15:00 - 15:30

Hotspots are short presentations during coffee breaks. They carry no paper.

Anders Haglund, Statistics Norway, aeh@ssb.no

Anne Abelsæth, Statistics Norway, aae@ssb.no

Vebjørn Aalandslid, Statistics Norway, vaa@ssb.no

The Division of Development cooperation at Statistics Norway (SSB) runs several projects to give technical assistance to developing and transition countries. With a goal to establish a consistent platform for developing Business statistics, staff from the Business register (BR) are sent in as paratroopers to build the BR as a backbone for the rest of the statistics. In line with paratroopers in war the risk of failure is substantial.

In the first part of the paper we will highlight some of the dearly bought experiences SSB has achieved in the last decades of working in developing countries. We will assume that our Nordic colleagues will nod in recognition on some of Norway’s items for failure. Time optimism, lack of administrative sources and problems with convincing the statistical domains of using a central register are common obstacles. But unthreatened on the top of the list of reasons for not completing the mission is failure of implementing the technical solution for the BR.

In the second part of the paper we will embrace the problem of lack of IT skills in the statistics office of a typical developing country. SSB has developed a first version of a generic statistical BR (STATBUS) as a tool for better success in our projects abroad. We will briefly explain the functionality and possibilities with STATBUS. Handling administrative and statistical units, characteristics, relation between units and so on are key issues here.

At last we will promote a Nordic cooperation around STATBUS as our Nordic colleagues struggle with the same problems and we believe that together the Nordic countries can ensure a more sustainable solution in this area.

Keywords: Developing countries, Business Register, IT solutions, Nordic cooperation

When: Wednesday, August 24 at 10:00 - 10:30

Hotspots are short presentations during coffee breaks. They carry no paper.

Jukka Hoffrén, Statistics Finland, jukka.hoffren@stat.fi

The Agenda 20230 for Sustainable Development document was adopted United Nations in 2015. It includes 17 Sustainable Development Goals (SDGs) and 169 related targets to end poverty and protect the planet. The progress towards goals and targets are monitored by 247 SDG indicators. At national level statistical office in each country is responsible for their data collection.

SDG indicator compilation process challenges the principles of official statistics as the main goal is to increase indicator data coverage over data quality. The justification for this comes from the argument that for policy needs it is better that there is even a rough idea of the phenomena than no data at all. This assessment of usability of non-traditional data sources do require new data validation task. In Finland, the development of network cooperation, is used to strengthen our expertise base. The SDG data producer networks cover 80% of the available data. The remaining 20% of the data which comes from a wide variety of sources, remain as challenge.

Demand for SDG data in public debate is growing. Beside traditional statistics variety of ways, including international comparations, ad-hoc surveys, and expert insights, are used in current social and political debate as they seek deeper answers to common challenges and problems.

In Finland dissemination has been improved with the Environment and Sustainable Development Phenomena Group, which is planning to communication on the topic. In addition, a special phenomenon expert has been chosen to act as the speaker of the topic, participating in the debates.

As traditional statistical production has largely remained unchanged, have recent reforms changed the work content of experts in years. Together with lack of resources this change of practices have not been implemented without friction. It is our experience that raising the level of expertise and quicker reactions debates are needed.

Keywords: SDG, official statistics, non-traditional data, networks, phenomenon expert.

When: Wednesday, August 24 at 10:00 - 10:30

Hotspots are short presentations during coffee breaks. They carry no paper.

Lars Kirdan, SAS Institute A/S

The presentation takes you back to the Renaissance where science really accelerated and runs through some of the central figures of contemporary data- and fact-based platforms for the understanding of everything and the elaborated production model for new knowledge. Everything I depicted through historic paradigm shifts like these few examples:

After this quick run through 8-900 years of moving towards a secular and fact-based comprehension of our relation to everything as human beings the presentation briefly stops and look at the establishing of the Nordic Governmental Statistical Agencies within a short period from 1849 and 20 years and the notion of the new heroes.

The presentation ends by summing up 6 global challenges towards the very relevance of our belief system and mode of knowledge production. And a joint call for innovative ideas on how to move against these 6 challenges.

Keywords:

• How did we get to where we are as Governmental Statistical Agencies?

• Why were we so strong for 150 years?

• What challenges are we facing right now questioning the very relevance of Governmental Statistical Agencies?

• Governmental Statistical Agencies?

• Governmental Statistical Agencies?

• What are we going to do about it?

Hotspots are short presentations during coffee breaks. They carry no paper.

When: Wednesday, August 24 at 10:00 - 10:30

*This hotspot is sponsored by SAS.