Abstract
National statistical offices (NSOs) and academia benefit from establishing partnerships and collaborating in different ways by bringing together their respective expertise. Collaborative alliances of this nature appear to offer numerous advantages for both the partners and the public and seem to be essential for unlocking opportunities within the evolving data ecosystem. Establishing good and fruitful collaboration between academia and NSOs requires a collaborative environment where each partner can see the benefits of the collaboration and how they could contribute. Different areas of collaboration are presented within four categories: education and learning, research, promotion of data use in society and providing services to each other. The article further discusses the benefits and conditions of a successful partnership. Examples from Brazil, Colombia, Ghana, and Norway showcase practical-level experiences and some lessons learned at the country level.
Introduction
The Data For Now Initiative [1] commenced in 2019 to strengthen countries’ capacity to use innovative data sources (such as satellite imagery and social media data), methods (such as machine learning and web scraping) technologies (such as data warehouses and geographical information systems), and partnerships (such as with the private sector, academia, etc.) for better, more timely and disaggregated data for sustainable development. The project thus aims to support countries in tapping into the new data ecosystem and moving away from the traditional survey-based official statistics production. The new data ecosystem represents a new data culture where collaboration between the NSOs, academia, and other partners will be the new way of working [2] bringing in the needed expertise from different partners.
The United Nations Statistics Division (UNSD) has as a core partner of Data For Now, engaged in strengthening collaboration between national statistical offices (NSOs) and academia including at the 64th ISI World Statistics Congress 2023 in Ottawa during the session: “Mobilizing Statistical Innovation Through Academic Partnerships for More Efficient and Targeted Policies” [3]. Collaborative alliances of this nature appear to offer numerous advantages for the participating partners and the general public. They seem essential for unlocking opportunities within the evolving data ecosystem.
NSOs around the world have different experiences with academic collaboration, some a lot, some less. This is, however, not well documented in a global context. This article seeks to gather some of the thoughts, experiences and lessons learned from this type of collaboration. It presents benefits, areas of collaboration, and conditions for a successful partnership. It includes concrete examples from Brazil, Colombia, Ghana, and Norway showcasing practical-level experiences and some lessons learned at the country level.
It is crucial to understand how academia and NSOs are different types of institutions when pursuing collaboration between the two. NSOs function as an “institution” and will therefore usually seek partnerships with other institutions. However, NSOs need to understand that academia besides being an “institution” also consists of individual researchers with their funding, research programmes and interests. Collaboration with academia might therefore require a new approach to partnerships as they can be made of individual researchers who see the benefits for their research and students. Individual researchers can eventually help drive a partnership towards an institutional approach if needed.
For the drafting of this article, the term “academia” is used below to cover both individual researchers and academic institutions.
Benefitting from collaboration
Collaboration and partnership between the NSO and academia could bring benefits for both institutions, bridging expertise and comparative advantages of each. A common starting point is that the institutions and their employees love data. While academia is often more focused on data use and innovation and being able to publish research; the primary objective of an NSO is to ensure a regular and efficient statistics production of high quality that responds to user needs. The emergence of many new opportunities within the data ecosystem (new types of data, methods, technologies, partners etc.) could strengthen the mutual benefits, with a common interest in making use of the new opportunities and learning what works.
It might be the case that the NSO has to initiate the idea of collaboration. In doing so, the NSO should focus on collaboration mechanisms or topics that fit into existing incentives of individual researchers and universities and thus where they have an interest and see the benefits of collaboration.
Collaboration between NSOs and academia could yield numerous benefits for the institutions and ultimately, the general public.
The benefits of academia could include:
Access to anonymized microdata that advance research. Opportunities for students including courses related to official statistics, internships at the NSO, research on real data problems and future work opportunities. Recruitment of skilled staff from the NSO. Learning from specialized expertise of the NSO in areas such as data management, internationally based statistics standards, as well as dissemination and communication techniques. Opportunities for conducting new research and case studies that can lead to publications. Possibility of increasing the impact of academic research on policy and decision-making. Contributing to official statistics.
The benefits of the NSO could include:
Building in-house capacity by learning from academia’s expertise as a technological and methodological frontrunner. Benefitting from academia’s research capacity to address areas of interest that the NSO does not have the capacity, knowledge or resources to address. Expanding the use of new methods, technologies and tools to improve the production of official statistics. Improved recognition among graduating students, potentially increasing the likelihood of them applying for jobs at the NSO. Recruitment of experienced academic staff for higher-level positions or as part-time experts at the NSO.
The general public also seems to benefit from the collaboration including increased access to research and statistics, improved official statistics, enhanced statistical literacy, and heightened interest in working with statistics and data science. Thus, this type of collaboration seems to increase the value for society [4, 5].
A partnership between the NSO and academia could take several forms and include different areas of collaboration. Some collaborations might require funding before academia is likely to be interested, while others might only require recognition of mutual benefits. The proposed areas of collaboration listed below are organised into four categories:
Education and learning Research Promote the use of data in society Providing services to each other
The areas within each category are ranked from easier to more extensive forms of collaboration.
Neither the categories nor the areas are exclusive, and others might exist that have not been identified during the writing of this article.
The country examples from Brazil, Colombia, Ghana, and Norway in section 4 provide details on several of the areas of collaboration presented below. A reference to the country name is made when an area is included in the country example. Examples from other countries are inserted as references but the list is by no means exhaustive.
This category includes collaboration which promotes educational goals in support of official statistics and contributes to the capacity building of students, researchers and employees of the national statistical system.
Teaching students
Researchers and NSO staff could teach students topics related to official statistics and thus equip them with the needed knowledge to use and understand data and produce statistics. E.g. an employee at the NSO could teach economics students the details of national accounts to better understand how a country’s economy is measured. Also, employees at the NSO could teach journalism students how to find stories in published statistics and how to use statistics when writing articles [6] [Norway].
Training staff within the National Statistical System
Academia could train staff within the National Statistical System (NSS) in specialized fields e.g in the use of probabilistic record linkage or satellite imagery [7]. Shorter- and longer-term training could be considered depending on the complexity of the field. University students might even participate in the training and thus benefit from the collaboration [Colombia].
Researchers co-train in capacity building projects
Projects, whether international, regional or national, that seek to build the capacity of the NSO, could include academia and let them participate as co-trainers in training courses. E.g. a researcher could act as a co-facilitator together with an international expert in a series of training in the use of geospatial data as part of a project on capacity building financed by an international donor.
Internships
The NSO and institutions within the national statistical system could offer internships to university students [8, 9]. This would allow students to work on concrete tasks related to statistics and contribute to tangible outputs relevant to the NSO or the NSS. The student would potentially benefit from the expertise of NSO/NSS staff and could also bring some of this back to his/her engagement at the university. An internship might have different lengths ranging from weeks to months and take place during the semester or summer break [Colombia and Norway].
Topics for master’s and Ph.D. thesis
Providing an opportunity for Master’s and Ph.D. students to research real data problems in their thesis could be an excellent way to create an interest in official statistics. Students could e.g. get access to anonymized data for their analysis or examine methodological challenges which the NSO could benefit from [Colombia and Norway].
Organising competition for students
The NSO might organise competitions or hackathons for students on a selected topic to increase awareness of the production and use of official statistics [10, 11]. Such competitions can be organised in many different ways and include groups of students who compete to develop the most user-friendly story on a selected dataset or a two-day hackathon to produce statistics on a selected dataset by using R [Ghana].
Curriculum for official statistics
The NSO and universities could work together to develop a curriculum for teaching official statistics to students either through a semester-long course or a comprehensive postgraduate programme [12, 13]. The latter would require recognition from government agencies responsible for accrediting education acknowledging the value of such an education.
Research
Collaborating on research is a long-term and substantial collaboration that requires dedicated time and resources.
Research projects
Universities and NSOs could engage in national or cross-national research projects on the use of innovative data sources and methodologies to generate new knowledge, explain trends and develop tools to support real-life problems [14, 15]. The opportunity for the two partners to combine different skills and access different data sources seems to be the competitive edge. Other partners could join the research project including the private sector as well as other government agencies and academic partners [Colombia and Norway].
Think tanks
An NSO could contribute to a university Think Tank in the role of end-user to provide feedback on practicallevel implementation of cutting-edge methodologies and knowledge generated by the university Practicallevel implementation could be considered a competitive advantage of the NSO, as it often has to navigate challenges of incorporating new methodologies and tools into actual practice.
Promote the use of data in society
Promoting the use of statistics and scientific results in informed decision-making and policy formulation is something both NSOs and academic institutions are tasked with and have a keen interest in doing. By collaborating they could send a strong message and might become more visible to decision-makers and other stakeholders.
Decision-making and policy formulation
The end products of the NSO and academia (statistics and research respectively) provide valuable information on society and should be used in policy formulation and decision-making [16]. The two institutions could benefit from working together to promote the use of data by legislators, policymakers and decision-makers. Collaboration could e.g. consist of the organisation of an awareness-raising event for policymakers to present different examples of how data use makes a difference in understanding policy-related problems and designing solutions [17] [Colombia, Ghana].
Statistical and scientific literacy
Improving statistical and scientific literacy in-country could contribute to a better understanding of societal problems and possible solutions and thus help informed decision-making by rooting policies and decisions in statistics and science. The NSO and academia have an interest in promoting statistical literacy and scientific literacy respectively and provide society with the skills needed to use the valuable knowledge, which the institutions are producers of [Brazil].
Increase interest and demand for statistics and data skills
Society has an increasing demand for statistics and data skills [18]. Teaching school-aged children about statistics and data skills and increasing their interest in how data can be used, could benefit society in the long term. Creating an early interest in working with data and statistics might help foster the creation of skilled citizens who can contribute to meeting society’s need for analysts and statisticians. Further, some students will grow up and become policymakers, legislators and decision-makers and would thus already have some level of understanding of how data could be used for social and economic decisions [Brazil].
Educating teachers on statistics and data skills is equally important as they are responsible for teaching children. The South African “Maths4Stats” is an example of an innovative programme to help primary and high school mathematics teachers to teach basic statistics and increase the interest in data and statistics among students [19]. The programme has trained hundreds of teachers across South Africa and is a collaboration between the University KwaZulu-Natal, Statistics South Africa and the Department of Education.
Providing services to each other
NSOs and academia can support each other by offering services in areas where they possess distinct expertise and strengths.
Meeting data needs of advanced users
Understanding and meeting data needs is a key role for an NSO. If the NSO categorises its users to ensure it meets different user needs, academia would rank as an “advanced user”. Academia uses lots of data and would want access to details including metadata and if possible microdata. Giving researchers access to anonymized microdata under strict confidentiality and security rules to avoid misuse and disclosure of individual data could be considered a valuable service provided by the NSO (though in some countries academia has to pay for the service) [2, 21].
Establishing a mechanism where the NSO can learn from academia as an advanced user and where academia can provide feedback to the NSO on their data needs (easy access to (anonymized) micro-data and metadata, open and user-friendly data platforms, coordination of publication within the national statistical system etc.) makes official statistics more relevant for all advanced users. Such a mechanism could involve that the NSO meets with members of academia, potentially formalized as an expert user group a few times a year [Colombia].
Delivery of specialized services
Academia could deliver specialized assignments to the NSO in instances where academia already has the needed expertise rather than the NSO investing in developing those unique skills themselves. The NSO can e.g. agree with a university to deliver a base map of the country with geocoded land cover types every 2–3 years to be used for different statistics [22].
Advisory role in statistical methods
NSOs need to keep abreast of new ways of producing statistics. Academia could support NSOs in investigating new statistical methods and turning experimental (or pilot-stage) statistics into regular official statistics. Some countries have experience with establishing an advisory committee on statistical methods, which has provided valuable contributions to official statistics [23, 24, 25] e.g. by examining the potential for small area estimation and then later reviewing the newly developed statistics before it is published by the NSO [26] [Colombia].
Sharing staff
Academia and the NSO could agree to share an employee, so the person works part-time at both institutions. The staff can thus contribute with specialized knowledge at both institutions and be a natural link between the two partners [Norway].
Joint capacity building
Statistical capacity-building projects financed by international and bilateral donors often concentrate on building the capacity of the NSO and other statistics-producing authorities. However, staff at both the NSO and universities could jointly benefit from participating in the same capacity-building activities. It would not require much extra effort or cost for the project owners to invite researchers to participate in training workshops, where they can benefit as learners together with NSO staff.
Experience from Colombia, Norway, Brazil and Ghana
Many NSOs have experience with academia. In this section, some of the above areas of collaboration will be illustrated with experiences from Colombia Norway Brazil, and Ghana which were presented at the 64th ISI World Statistics Congress 2023 [3, 27].
Colombia
The example from Colombia presents experience in many different areas of collaboration between the National Statistical Office, DANE, and both national and foreign academic institutions.
Internships
DANE’s institutional agreements with academia aim to facilitate student engagement. To streamline this process, DANE has developed a standardized agreement to involve undergraduate and graduate students from partner institutions in internships at the organisation.
Topics for master’s and Ph.D. thesis
Currently, two Ph.D. students are conducting their research using data from DANE. One student from the University of Oxford is proposing a methodology to build a register-based census utilising information from the 2018 National Population and Housing Census and the Statistical Register of Population. The other student, from the University of Southampton, is focusing on small-area estimation using linked data to derive indicators on victims of the armed conflict at the municipal level.
Training staff within the NSS
The Leverhulme Centre for Demographic Science at the University of Oxford and DANE have obtained the University of Oxford’s “Economic and Social Research Council (ESRC) Impact Acceleration Account (IAA)” award. As a result of this award, the “Workshop on Statistical Population Modelling for Census Support” was carried out at DANE in March 2023. The workshop’s objective was to teach a methodology to determine census omission using geospatial information and Bayesian Models. Participants included representatives from the national statistical offices of Brazil, Costa Rica, Ecuador, and Colombia.
In June 2023, WorldPop, from the University of Southampton, conducted a workshop at DANE on “Exploring smart and flexible approaches (preEA tool package) to support pre- and post-census and survey activities”, which was attended by staff from the Geostatistical and Collection divisions of DANE. The workshop aimed to teach tools to design customised national sampling frames to serve census and various socio-economic surveys.
Research projects
DANE, in collaboration with the University of Southampton, Profamilia, and the Saldarriaga-Concha Foundation, have established the Colombian Healthy Ageing Research Partnership (CHARP). CHARP addresses the lack of information on ageing and the elderly population. Efforts are underway to secure funding from the UK Research Institute (UKRI) and the Medical Research Council (MRC) to strengthen this cross-sector partnership. The CHARP partnership commenced in February 2023 with WorldPop, from the University of Southampton, and DANE as initial members. Through consistent meetings with other partners, the collaboration has evolved and expanded.
Additionally, DANE, along with Carleton University and the Universidad Industrial de Santander, has applied for a grant from the Canadian Institutes of Health Research (CIHR) for the project “Colombian Census Health Cohort: The Creation of a Population-Based Cohort to Investigate Social, Environmental, and Occupational Determinants of Health”. This initiative will catalyse the creation of a longitudinal population database with health information, facilitating high-quality research and providing valuable insights for health policymakers.
Meeting data needs of advanced users
The External Specialized Processing Room (SPEE) is a service provided by DANE so that specialized researchers, national or foreign, can access and process anonymized microdata from DANE’s statistical operations, not available on the website.
DANE signed a Data Sharing Agreement with the universities of Southampton and Oxford. The data included in the agreement is being used to develop and validate new statistical methods to estimate the population in data-scarce environments, develop novel demographic methods, and generate population estimates in intercensal periods. This last topic is the Ph.D. thesis of the student at the University of Oxford.
In addition, monthly meetings are held between the
Under the cooperation agreement with the National University of Colombia, data centres have been established for remote access to DANE’s statistical information at the university’s different locations throughout the country. DANE’s staff can also utilise these centres to transmit field-collected information to the Collection Division at DANE’s headquarters.
Advisory role in statistical methods
DANE counts on the support of the Technical Advisory Council of the National Statistical System (CASEN), which is a coordinating body within the National Statistical System (SEN) responsible for advising DANE and the Government on strategically important matters for the country’s official statistics. The council provides advice and evaluates the development of SEN concerning coordination, methodologies, production, relevance, and exchange of official statistical information. Similarly, it endorses the formulation and updates of the National Statistical Plan presented by DANE. Five specialised groups, which include academics and experts, conform to the council: a) Technological modernisation of statistical production; b) Health, social welfare and demography; c) Government, security and justice; d) Geography, environment and territorial planning; and e) Economics [28].
Two key activities included in the cooperation agreement between DANE and the National University of Colombia are to work on projects of revision and strengthening of the quality of data and information related to official statistics and to promote research in fields of common interest that allow optimising processes, designing, and implementing statistics. The Data for Now Initiative has also supported DANE in bringing in academic experts to progress work on natural language processing and probabilistic record linkage.
Decision-making and policy formulation
Two projects that will benefit health policymakers are the Colombian Healthy Ageing Research Partnership (CHARP) and the project “Colombian Census Health Cohort: The Creation of a Population-Based Cohort to Investigate Social, Environmental and Occupational Determinants of Health”. The first seeks updated well-being indicators for the elderly, proposing interventions for improvement, and projecting future needs. The second seeks to combine national census data with health registries enabling top-notch research and capturing valuable information for policymakers.
Lessons learned
Active participation and consideration of all partners’ perspectives are essential to fostering trust and confidence. Building a partnership is a time-consuming process that requires ingenuity to put together the project puzzle ensuring inclusivity and satisfaction for everyone involved.
To support collaborative efforts with academia, DANE is actively pursuing international grants from organizations such as the Canadian Institutes of Health Research (CIHR), the UK Research Institute (UKRI), the Bill and Melinda Gates Foundation, and the Inter-American Development Bank.
Norway
The experience from Norway discusses the evolving landscape of collaboration between Statistics Norway and academic institutions. Experiences in three main areas of collaboration are presented: teaching and supervision, internships, and research and innovation.
One driving force behind the evolvement of collaboration between Statistics Norway and academia is the challenge of recruiting individuals with data science and programming expertise. While collaboration has traditionally been the result of personal initiatives, a notable shift towards institutionalization has occurred in recent years. This transformation has been facilitated by the involvement of the Human Resources (HR) department at Statistics Norway, which now plays a coordinating role in initiatives spanning the entire organization.
Teaching and supervision
In the realm of academia, one of Statistics Norway’s most traditional forms of collaboration has been through teaching and supervision. Employees from various departments in Statistics Norway, hold part-time positions at universities in Norway and abroad, teaching both bachelor and master-level courses. These courses, while more generic, incorporate examples and cases from official statistics, primarily in the fields of economics and mathematical statistics. Furthermore, some of these positions involve the supervision of master’s or Ph.D. students.
As an example, Statistics Norway initiated a collaboration with the Norwegian University of Life Sciences (NMBU) three years ago. A range of student projects at the master’s level were identified and specified inspired by ongoing projects and work at Statistics Norway. In subsequent semesters, these projects were offered to master students with a data science profile as thesis topics. This collaboration resulted in two successful master projects addressing classification challenges within the Household Budget Survey (HBS).
Internships
A relatively new type of collaboration between Statistics Norway and academia is internships of various types.
Statistics Norway has offered students paid internships, either in the form of summer jobs or as part-time positions. A summer job will typically be an 8–week temporary engagement, allowing a student to work intensively on one or several projects for the designated period of internship. As most other employees will be on vacation for parts of this period, this type of project, therefore, requires quite a bit of planning. The tasks must be well-defined, and the students must be enabled to work independently, at times.
Statistics Norway recently introduced part-time positions for bachelor and master students at the Division for Methods. These positions, paid and working one day a week, have allowed students to apply the skills they have acquired through their studies in a more practical context. It has also provided them with the opportunity to learn more about topical issues and give them valuable insight into the day-to-day work at a statistical agency.
It seems to be a current trend that education which traditionally has been focused on theory and academic perspectives, is exploring ways to give students an experience of working life. Educational institutions are searching for collaborations with private and public sector companies, that allow students to acquire practical knowledge and apply their knowledge to real-world problems.
Statistics Norway has collaborated with a master’s course offered by the Department of Economics at the University of Oslo. In the course, named “Internship in economics” students received credits in exchange for an 8–week internship in a private or public sector enterprise. In the spring of 2023, Statistics Norway hosted such internships for 8 students, involving 5 different divisions and a wide variety of tasks. Statistics Norway is currently considering a repetition and is also discussing several similar initiatives with other educational institutions.
Research and innovation
The Norwegian research institutes serve as intermediaries between academia and business/government. One such program, the “Centers for Research-Based Innovation”, was launched by the Research Council of Norway in 2015 to finance long-term innovations over 4–5 years. Statistics Norway partnered with one such center called “Big Insight”, which focuses on developing statistical and machine learning methodologies to address data-driven challenges.
The collaboration has involved researchers from Big Insight joining ongoing statistical production teams at Statistics Norway, creating an interdisciplinary environment that included subject matter experts, IT developers, and methodologists. This collaboration has resulted in advanced machine learning and classification methodologies tailored to specific data challenges.
Lessons learned
Statistics Norway’s collaboration with academia has been very beneficial but not without its challenges. This type of collaboration requires resources. Naturally, there are administrative preparations, but what is most demanding is identifying issues that are suitable for the various types of collaboration. Tasks must be well-defined to fit the individuals who will be executing or collaborating with the statistical agency. For students, the task must align with their skill level. An important factor for success is that the work provides value for both parties.
However, the returns include useful results, inspiration for employees, and potential recruitment opportunities, in the short and long term. Statistics Norway considers collaboration with academia, particularly in terms of teaching, supervision, and providing experiences for students, as part of Statistics Norway’s mission. The experience demonstrates that collaborations between statistical agencies and academia can yield fruitful outcomes, making them well worth the investment.
Brazil
The Brazilian Institute of Geography and Statistics (IBGE) has a peculiarity compared to most other NSOs as the National School of Statistical Sciences (hereafter ENCE) is part of its functional structure. Thus, the partnership between the NSO and academia is easier facilitated, though it still requires ongoing efforts to establish the bridge between professors at ENCE and specialists at IBGE.
The project on statistical literacy which is illustrated here shows the benefits of that partnership.
Increasing statistical literacy
The Project “Statistical education: statistics as an ally in building critical thinking” was launched in 2023 by a professor from ENCE and conjoins the participation of ENCE graduating students and professors, and specialists from the dissemination department of IBGE. Communication specialists of IBGE bring knowledge on how to speak about data to different audiences and offer dissemination channels. Academia brings knowledge on how to develop competencies in statistics and the best ways to work with theoretical concepts and practical application to promote statistical literacy.
The project includes the production of educational material and activities for schools. IBGE offers a web portal to communicate with the schools: IBGE Educa (
ENCE offers educational activities for school students (on its premises, in an “open door” fashion) to learn about statistics and the graduation course on statistics – which is typically something young people are not aware of when they make their career decisions. The communication with students is in plain language using resources such as storytelling and examples from the reality of the students.
The project has educated public school students to use evidence for decision-making. E.G., students participate in challenges to decide if the information is fake or fact or to respond to puzzles considering statistical information. The facilitators of these exercises are graduating students of ENCE, who developed the activities and elaborated materials.
The project has also developed events for school students based on specific interests from schools. E.G., about 200 students have been trained in practices to develop questionnaire skills as their school had a project to develop an internal school Census. The theme was particularly interesting for IBGE as it coincided with the release of the results of the Populational Census. Besides the partnership with the dissemination department of IBGE, ENCE also partnered with the census department, which participated by offering specialists to develop the conception of the event and to present the process of Census operation to the students.
The benefits of this multilateral partnership are the promotion of statistical literacy, as well as of IBGE and its surveys. The project is an investment in user engagement to both enhance the use of data and the response to surveys. As a social impact, awareness is built about the importance of evidence-based thinking and statistics as a career choice. The social impact is not only limited to the students or teachers directly involved in the project but also has a spill-over effect on their families. Further, there is a direct benefit to the quality of the education of graduating students, as they learn about the importance of statistics through the promotion of social inclusion and contribution to citizenship. The graduating students also develop their professional capability of talking about data with different people and adapting statistical language to a variety of audiences.
Lessons learned
The project could benefit from a more active engagement of the schoolteachers in the development of activities and materials. Although schoolteachers were encouraged, they mostly passively agreed with the proposals that the graduating students made. Another lesson is that pedagogic advice to graduating students is relevant in the process of developing activities and the implementation of the project. Also, the partnership of an educational specialist and the professor of statistics leading the project had gains for the graduating students’ performance. Further, during the events, more time on hands-on activities and less time on lectures could be more appealing to the school students. Alternatively, mixing presentations with quizzes or fast games could reduce attention load.
The biggest challenge is scaling up the project as the practical activities themselves can only be performed in an in-presence modality. The project leaders have started conversations with stakeholders in academia and teacher and science associations to gain traction for reaching the national curriculum so other universities can build their capacity to engage schoolteachers around the country.
Another way to scale up is to benefit from IBGE’s presence in Brazil’s 27 states. Each state unit has a local dissemination staff of which some already visit schools to promote IBGE and its production. However, they do not yet engage school students in capacity-building activities. Training those possible promoters of the project and building models that can be replicated may help in scaling up.
Ghana
The experience from Ghana presents how Ghana Statistical Service (GSS) has collaborated with academia to promote data use and understanding among parliamentarians to pave the way for more informed and effective governance, ensuring that decisions made by the Parliament align with the needs and aspirations of the Ghanaian people.
Decision-making and policy formulation
In 2021, the Africa Centre for Parliamentary Affairs (ACEPA) and GSS with crucial input from academia organized the Data Fair, which was specifically designed to engage members of the Ghanaian Parliament. The objective was clear: to shift the parliamentary decision-making process from being guided by chance and perception to one grounded in data and evidence. This transition sought to empower Parliament to fulfill its core functions of oversight, representation, and legislation with greater precision and effectiveness.
The Data Fair was conceived as a strategic response to the challenge, which Ghana had grappled with for some time regarding a dearth of readily available and accessible data that could be comprehended and used effectively. This data deficit presented a significant obstacle to informed decision-making and policy formulation. The Data Fair therefore aimed to stimulate interest among Parliamentarians in the transformative power of data in their work.
The Data Fair provided an unprecedented opportunity for Ghanaian Parliamentarians to familiarize themselves with the wealth of data and evidence at their disposal. Workshops, seminars, and interactive sessions served as a platform for education and engagement, enabling Members of Parliament to navigate the vast landscape of data and harness it more effectively in their roles.
The University of Energy and Natural Resources (UENR), and the Digital Earth Africa programme engaged with parliamentarians showcasing e.g. how Earth Observation can be used for evidence-based decision-making.
Organising competition for students
GSS has worked throughout 2023 to ensure maximum access to all the statistical products it produces, with the launch of its StatsBank. This new platform provides an opportunity to access all the 2021 Population and Housing Census data as well as macro-economic indicators for Ghana, ensuring access to and uptake of data for inclusive development. Data from the StatsBank provided a unique opportunity for GSS in 2023 to collaborate with five public universities to organize hackathon competitions across each of these universities and pitch the winners from the five institutions in a grand finale in October 2023 [10].
Lessons learned
The Data Fair is seen as a pioneering step in transforming the landscape of parliamentary decision-making in Ghana. As Ghana continues its path to data-driven governance, the collaboration between GSS and academia proves crucial to showcase how the use of data, both traditional data sources and those in the new data ecosystem, plays an important role in the development of Ghana. The Data Fair and other programmes, including the recent hackathon with university students, serve as inspiring and transformative milestones in this journey.
Conditions for good collaboration
Establishing good and fruitful collaboration between academia and the NSO would require an agreement where each party can see the benefits of the collaboration (the win-win scenario).
Experience from other government and university partnerships brings forth four factors for good collaboration [29]. Many of these factors have also been brought forward in the country examples above.
Shared objectives and effective transfer of knowledge and experience (Output factors). Availability of sufficient resources (i.e. financial, human resources, time), acknowledgment of cultural and structural differences, and a willingness of the partners to adapt to each other (Institutional factors). A good relationship between the persons involved (e.g., trust, good communication, engagement) (Relationship factors). Clear framework conditions (e.g. clarification on who can access data and when, ownership as well as confidentiality and data protection) (Framework factors).
Experience from NSO and academia collaboration shows that it takes time to build a common understanding of shared goals and that acknowledging different cultures is important: academia has a culture of doing things once, competitiveness and creativity, while the NSO has a culture of repeating the same tasks over and over again, collaborating and following manuals [14]. However, the culture of academia and public institutions are not as different as often perceived [30].
Other aspects that would potentially need attention when establishing collaboration are ownership of the products developed as well as confidentiality and data protection depending on the type of collaboration.
Funding is a crucial element in collaboration. Some of the collaboration areas presented above will require external funding before academia (and the NSO) is likely to be interested, however, other areas might fit into existing funding structures of academia or the NSO and will likely only require recognition of mutual benefits and a decision to engage without additional funding.
Some universities might be cash-poor, however they can be rich in the time of researchers and students, why some of the collaboration areas can be easier to establish. A university might also be interested in cooperating if the collaboration attracts more students and thus a greater income. Students might be attracted to a university programme if they see possibilities for a job once they complete their studies.
If external funding is needed for the collaboration to materialize, it should be planned carefully and applied for well in advance from relevant sources which could include the Government, foundations and research institutions, the private sector and wealthier universities.
Government and research grants
Contribution to research can already be an integrated part of the annual state budget. Getting access to those funds will require planning and close collaboration with academia. State institutions and municipalities might be interested in contributing to collaboration if the benefit is visible and a valuable outcome realistic.
Foundations and research institutions
There are many foundations and research institutions, which provide funding to projects that will have a social or environmental impact; also known as impact funding. This type of funding might be less familiar to the NSO than academia but might provide a great opportunity to establish collaboration where NSO and academia bridge their different strengths.
Private sector
The private section has research needs that they could be willing to pay for. The private sector could also have an interest in the benefits that come with cross-sectoral collaboration with NSO and academia.
Wealthier universities
Wealthier universities might be interested in contributing to a collaboration if it fits into existing work, research and programmes or if it attracts new funds. Wealthier universities can contribute with additional resources (not only financial) as they more likely will have in-house capacity, administrative muscles, and relevant contacts.
Countries that have little experience with establishing a partnership between the NSO and academia could start with mapping possible areas of collaboration where both partners could have an interest. Acknowledging the difference in the partner’s culture and main tasks – with academia focusing on innovation and publication of research results, while the NSO focuses on the regular production of official statistics and society’s access and use of data – seems crucial in this process. Reaching out to countries where experience already exists is another step to get inspiration and learn from others. In general, thinking of partnerships that are long-term and nationally anchored could be a way forward – even if funding is international.
UNSD will continue working on strengthening collaboration between NSOs and academia within the Data For Now Initiative, including discussing with the project countries how this can be further expanded at the country level.
Footnotes
Acknowledgments
Thanks to the following persons for providing valuable input during the early phase of gathering information and lessons learned on the topic:
David Stern, Innovations in Development, Education and the Mathematical Sciences (IDEMS) International; Laust Hvas Mortensen, Statistics Denmark, Data Science Lab; Maryam Rabiee, Sustainable Development Solutions Network’s (SDNS’s) Thematic Research Network on Data and Statistics (TReNDS); Faryal Ahmed, United Nations Statistics Division/DESA; Haoyi Chen, United Nations Statistics Division/DESA, Inter-Secretariat Working Group on Household Surveys; And many thanks to Professor Delia North, University KwaZulu-Natal, South Africa, for sharing information on the Maths4Stats programme.