Open science

Science is, by its very nature, an open activity. Research results are intended to be publicly assessed and utilised, both within the scientific community and in all other contexts in which research-based knowledge is used.
What is open science?

Open science refers to the mechanisms by which the findability, accessibility and use of scientific knowledge is promoted in the digital era. This encompasses open research outcomes, including:

  • Publications
  • Research data
  • Research methods
  • Open source code

In addition to the above, open science refers to:

  • Open research infrastructures
  • Research-based open learning material
  • Inclusive research processes such as citizen science
Open science supports the University’s goals

The University’s operations are based on the aim of responding to societal and global information needs. Openly accessible research-based knowledge is an important tool for achieving this goal. Open science promotes the quality and impact of research, and offers new cooperation opportunities.

Open learning environments and content create new models of learning and support continuous learning. Open science enables the University to promote the UN’s sustainable development goals, particularly democracy, justice, global responsibility and the general public’s and society’s trust in science.

Openness is one of the overarching themes of the University of Helsinki strategic plan for 2021–2030. Read the strategic plan on our website.

Declaration for open science and research 2020–2025

The University of Helsinki has signed the Finnish declaration for open science and research.

The declaration for open science and research 2020–2025 represents the shared goals and recommendations of the Finnish research community for the consolidation of open science. The Finnish research community includes

  • Universities
  • Universities of applied sciences
  • Research institutes
  • Funders
  • Libraries
  • Archives

The declaration includes four areas, for which separate guidelines will be drawn up:

By signing the declaration, the University of Helsinki has committed to promoting its goals as part of its strategic work and guidance and to supporting the achievement of the goals in the daily life of University community members.

University of Helsinki Principles of open publishing

The Rector has confirmed the University of Helsinki's Open Access Principles 20.1.2025 (HY/722/00.00.06.00/2025). The decision replaces the principles from 2017. 

The implementation of the principles is monitored and assessed regularly. The principles of open access will next be updated no later than 2030.

  • Academic publications and theses produced at the University of Helsinki are openly available by default. The requirement for open access publication applies to all members of the University community.
  • The University promotes equal opportunities for members of the University community to publish their research openly, regardless of the research field, funding source or career stage. • The research results are published in high-quality academic publication channels.
  • The University has an overall picture of the costs of open access publishing and aims to ensure that they remain reasonable.
  • The University aims to promote the immediate open access of new publications.
  • The University develops its own assessment practices in such a way that activities to promote open access to research publications and, more generally, open publication culture, are considered when evaluating researchers and research.
  • The University guarantees the long-term availability of research publications produced within its scope. 
  • The University offers a long-term stable repository for research publications. The comprehensive archiving of publications in the Helda open digital repository guarantees the documentation, long-term preservation and further use of research.
  • All academic publications produced at the University are self-archived in Helda in accordance with publisher terms. To ensure the long-term availability of research publications, open publications are also self-archived. The University Library offers a selfarchiving service and support with licensing issues. University of Helsinki, Principles of open publishing 2025 20.1.2025 •
  • The University of Helsinki’s theses are published in Helda. The University offers support for responsibly opening access to theses.
  • The University of Helsinki Library offers centralised support for the open access publication fees of articles. However, this support excludes hybrid publishing, where the same publisher is paid both a subscription fee for its closed access material and the fee to publish an article openly.
  • The University offers remarkable and high-quality publishing platform and publisher services for the open publishing of journals and books.
  • Academic publications will also be opened for access retrospectively in Helda, taking into account the needs of the disciplines and copyright.
  • The University offers training, support and guidelines for open access publishing. Training is offered both as part of curricula and as personnel training.
  • The University supports the development of open publishing practices, publishing infrastructures and new funding models. 
  • Open publication must be considered when planning the research process. In the plan, the researcher will explain how the resulting publications will be openly available. The plan must also consider the terms and conditions of the research funders for open publication.
  • It is the responsibility of the author to assess the quality and reliability of the publication channel already when planning the publication. The University Library will provide support when needed.
  • The issues of authorship for joint publications must be agreed upon even before the manuscript is sent to the publisher. The University recommends that the authors familiarise themselves in this regard with the guidelines of the Finnish National Board on Research Integrity (TENK, pdf in Finnish).
  • The University of Helsinki must be named in the publication as an affiliation of the author. • The author must check the terms of the publication channel for self-archiving before submitting the manuscript. The University Library will provide support when necessary. The University discourages publishing in publication channels that do not allow parallel recording.
  • The author must add information on the research publication to the research information system Tuhat and ensure that the publication or its latest approved manuscript version is self-archived in accordance with the terms and conditions of the publisher. If needed, the University Library self-archiving service will take care of the archiving.
  • Further use of academic publications should not be unnecessarily restricted, and the terms and conditions of use should be clearly stated. The University recommends using a current version of the Creative Commons license unless the publisher or funder requires otherwise.
  • It is recommended that researchers use the international ORCID identifier and add it to their publications. The identifier enables reliable identification of the researchers and their publications. The University also recommends that researchers create a public profile using the ORCID identifier in the national Research.fi service, where publication data can be collected from different sources. University of Helsinki, Principles of open publishing 2025 20.1.2025
  • The University of Helsinki’s theses (bachelor’s, master’s, and doctoral dissertations) are published openly in the Helda publication archive unless there are reasonable grounds for doing otherwise. It is recommended to use the Creative Commons CC BY or CC BY-NC-ND licence. At the minimum, the summarising report of the article-based dissertations must be openly available. The University Library provides support on licensing the theses. 

Academic publication

A publication in an academic publication channel containing new research knowledge, which other researchers must be able to verify, evaluate and use. The publication may be produced by a single or multiple researchers and its main target group is other research experts. According to the publication type classification of the Ministry of Education and Culture (OKM), the publication is peer-reviewed, meaning the entire manuscript to be published must be reviewed by independent experts. The publisher of an academic publication must always be external, and its editorial board must consist of research experts. Academic publications include journal articles (OKM classification A1 and A2), compilation articles (A3), conference articles (A4), separate works (C1) and edited works (C2). Sources and more information about publication types at Publication Forum website.

Affiliation

The organisation or organisations of the researcher in which the research or expert work is mainly carried out. The author may have more than one affiliation. Entering affiliation information in the publication metadata ensures the publications are attributed to the right organisation and person, for example in research information systems, reference databases and statistics. See also: ORCID 

Hybrid publishing

Publishing in a subscription-based publishing channel, where it is possible to publish a single article or book chapter openly. In hybrid publications, the same publisher is paid both a subscription fee for the right to read closed access publications and the publication fee for an article’s open access. Hybrid publishing has been found to lead to double dipping, thereby increasing the cost of open publishing. 

Immediate open access

The open availability of an academic publication (e.g. article, book, chapter) without publication delay (embargo). The publication delays set by publishers vary, but University of Helsinki, Principles of open publishing 2025 20.1.2025 usually range from 6 to 24 months. Various open science projects seek immediate openness because embargos are an obstacle to open access. See also: Open access

Licence

The author’s permission for open distribution of the work (e.g. article, book, chapter of the book) under certain conditions of further use. Creative Commons licenses are usual in open publishing. More information on CC licenses.

Long-term availability

Ensuring access to information in digital form and keeping it understandable and accessible. 

Manuscript versions

Pre-publication versions of an article, book or book chapter. There are usually two manuscript versions: (1) a pre-print, submitted version, and (2) a peer-reviewed and f inal draft (Author’s Accepted Manuscript, AAM). This is followed by the publisher’s published and finished version (Version of Record). The last version of the approved manuscript is often the version that the publisher allows to self-archive in the publication archive. (For more information on manuscript versions on Think Open blog.) See also: Self-archiving  

Open publication, open access

Free access to research publications digitally via an online connection. There are different ways to publish openly; the main routes are: (1) publishing in the open access publishing channel (gold model / gold open access), where the author can be charged a publication fee; (2) parallel publishing in an open publication archive independent of the original publisher (green model / green open access); and (3) publication in a hybrid open access channel (hybrid model / hybrid open access). See also: Publication fee 

Open publication archive

A digital platform for storing and publishing academic research outputs, and the technical system and services built around it. Publication archives may be disciplinespecific or maintained by an organisation. The open publication archive of the University of Helsinki is Helda. 

ORCID

Open Researcher and Contributor ID is an international digital researcher identifier. ORCID helps to identify researchers, for example, in situations where a researcher changes name, or several researchers have the same name. ORCID reduces the need to enter publication metadata multiple times into different systems, for example, if the researcher transfers between organisations. The ORCID identifier allows the researcher to create a public profile in the national Research.fi service, where publication metadata is collected from different sources. More information on ORCID.org.

Publication channel

Publishers as well as journals and series are called publication channels. ‘Academic publication channel’ refers to book publishers, conferences, and printed and electronic publication series that specialise in publishing academic research results (monographs, compilation articles, articles of a periodical or of a series). The channel has an editorial board of experts and follows peer review practice. The channel may also include unreviewed publications, such as comments, reviews and manuals. In Finland, only academic publication channels can receive the Publication Forum category (levels 1, 2 and 3). In the open publishing channel, all content is immediately available to readers at the time of publication. Source: Publication Forum - Classification Criteria. University of Helsinki, Principles of open publishing 2025 20.1.2025. 

Publication fee

A fee for open publication, paid by the author or organisation. Articles are subject to APC (article processing charge) and books to BPC (book processing charge). Author fees may be charged by fully open publication channels or hybrid publication channels, where only part of the content is open. If the author fee is not charged, the channel is called diamond open access. See also: Open publication

Self-archiving, parallel publishing

 A way to publish openly, alongside the original publication. Selfarchiving also ensures the long-term availability of the publication. In self-archiving, a specific version of the article – often peer-reviewed and approved for publication, the last manuscript version – is stored in a discipline-specific or organizational publication archive, under the terms of the publisher, either immediately or after the publication delay (embargo). Open publications are also self-archived to ensure long-term storage. (For more information see our page for Self-archiving). See also: Open Publishing, Manuscript Versions. 

Research data policy

The research data policy expresses the principles and goals that guide research data management at the University of Helsinki. It also describes the services supporting these goals, and defines the mutual responsibilities of different actors. The research data policy applies to all members of the University community. The University of Helsinki’s research data policy has been updated in 2021. You can also download the research data policy as a pdf file from Helda.

  • Introduction
    • Scope of the research data policy
    • Other policies pertaining to research data management at the University of Helsinki
    • Research agreements and agreeing on rights
    • Updating and monitoring the research data policy
  • Principles and goals for research data management
    • General principles
    • Planning and preparation of data management
    • Documentation and processing of research data
    • Publication and accessibility of research data
    • Commercialisation of research data
    • Destruction or long-term preservation of research data
  • Responsibilities related to research data management
    • Researchers’ responsibilities
    • Responsibilities of faculties and independent institutes involved in research
    • The University’s responsibilities
  • Glossary

Research data are central to science and research. The production of data requires long-term efforts as well as technical and financial resources. In fact, the reusability of research data has become an increasingly important question in terms both of science and research, and of the impact of research.

For the University of Helsinki as well as its units and researchers, research data constitute a strategic resource and an international competitive factor. The goal of the University is to promote responsible research data management, which is of crucial significance for the findability, accessibility and reuse of research-based knowledge (for the definition of ‘research data’ and ‘research data management’, see the glossary).

The principles of open science and open research data play a prominent role in the strategic plan of the University of Helsinki for 2021–2030. The University’s strategic goals include open research infrastructures and open research data, the implementation of the FAIR Principles in research data management, and the advancement of competencies in the analysis of large and open datasets.

A key goal of the research data policy is to guide everyone involved in research data management to familiarise themselves with current data management requirements as well as to adopt good and responsible practices as part of everyday research activities. Another goal is to make research data management increasingly clear for individual researchers.

From an organisational perspective, the research data policy describes the goals that guide the development of research data services at the University of Helsinki. The goal is for University of Helsinki researchers to have at their disposal infrastructures and services that enable responsible research data management, developed in an economically sustainable manner while taking researchers’ needs into account.

In terms of its fundamental goals, the new research data policy of the University of Helsinki does not significantly differ from the previous policy published in 2015. The new policy has been amended to comply with current legislation as well as national and international guidelines and recommendations. In contrast to the previous policy, whose more limited scope only encompassed digital data, the new research data policy applies to all research data and related management principles. Another important change is the further clarification and highlighting of responsibilities.

1.1 Scope of the research data policy

This research data policy covers all research conducted at the University of Helsinki, as well as the research data collected and produced in conjunction with it. This denotes both digital data as well as physical and analogue data, which are all referred to as research data in the research data policy (for the definition of ‘research data’, see the glossary).

The research data policy applies to all members of the University community involved in research, including University employees and students as well as those who conduct research on behalf of the University.

1.2 Other policies pertaining to research data management at the University of Helsinki

The principles and goals presented in the research data policy supplement other University of Helsinki policies relevant to the management of research data, including the principles of open science, the data protection principles (requires login to Flamma), the information security policy and the guidelines of the Finnish National Board on Research Integrity for the responsible conduct of research. The research data policy is also in line with the University’s data management practices and principles.

Policies related to research data management at the University of Helsinki are guided primarily by EU and Finnish law. By employing policies and guidelines, the University strives to ensure the legality of research activities. In international research and other cooperation, research data management can also be regulated by legislation outside the European Union.

External research funders and partners may also have specific requirements for research data management.

The general objectives for the processing of data stored in University of Helsinki research infrastructures or those on the national or international level are described in the University of Helsinki Research Infrastructure Programme (for the definition of ‘research infrastructure’, see the glossary). In addition, national and international research infrastructures may have their own jointly agreed policies for the processing of the data that they produce. When research infrastructures draw up or update their data management principles, their alignment with the University’s research data policy must be ensured.

1.3 Research agreements and agreeing on rights

Research agreements

Concluding agreements and undertakings is an important part of responsible data management, regardless of the source of research funding. Agreements are always needed when research is conducted in a collaboration or other partnership with external parties, such as other universities, businesses, research organisations and hospital districts. Agreements help to safeguard the interests of researchers and the University of Helsinki, manage risks related to research and ensure the legality of activities.

Research agreements must be concluded as early as possible, preferably before the collection or use of research data commences. Agreements are used to ensure that the research data collaboratively collected by research groups is available to all of the participating researchers. Agreements safeguard the continuity of research and the further use of research data, for example, after the conclusion of research projects.

Agreeing on rights related to research data

Many international and national research funders require that the research data and findings that they have funded are open access. Obligations imposed by funders on researchers and the University make it necessary to transfer rights to research data to the extent necessary to fulfil the funding terms. In addition, the sharing of rights is needed, for example, to enable the opening, further use and archiving of research data. As a rule, researchers retain a concurrent right that enables them to continue using the research data.

The rights related to research data and their sharing should be agreed when researchers’ employment at the University begins.

Agreeing on the rights related to research data is in line with legislation and the responsible conduct of research. Instead of altering researchers’ responsibilities in the management of their research datasets, such agreements are part of it. Sharing rights does not alter researchers’ right to be referred to as the collectors or producers of the research data in question.

1.4 Updating and monitoring the research data policy

This document supersedes the current University of Helsinki research data policy that was approved in 2015. An implementation plan for the updated data policy will be drawn up in 2022.

The implementation of the research data policy will be monitored by carrying out assessments at regular intervals, the first in 2023. 

The research data policy will be updated by 2025.

2.1 General principles

Below, the principles and goals for responsible research data management are described in relation to the lifespan of research projects, beginning from the planning of data management and ending in the long-term preservation or destruction of data. The potential for the further use of data must be taken into consideration during research projects.

In general, the responsible management of research data is guided by the FAIR Principles, according to which research data must be findable, accessible, interoperable and reusable (for the definition of the ‘FAIR Principles’, see the glossary). Another guiding principle pertains to the openness of research data: as open as possible, as closed as necessary.

2.2 Planning and preparation of data management

  • A data management plan is to be drawn up for all research projects. The plan is to be updated during the project whenever substantial changes are made to the content (for the definition of ‘data management plan', see the glossary).
  • Questions related to legislation and research ethics are taken into consideration in the data management plan.
  • When processing personal data, the University of Helsinki data protection guidelines (in Flamma) on informing data subjects and ensuring the necessary safety measures are to be taken into account. When necessary, a data protection impact assessment (in Flamma) must be carried out (for the definition of ‘personal data’, see the glossary).
  • The rights (in Flamma) and responsibilities related to research are to be agreed within research groups, while the relevant agreements (in Flamma) are to be concluded with external parties prior to commencing the collection or use of research data.
  • Preparations must be made for the protection and, when necessary, anonymisation of sensitive data as well as the destruction of confidential data, with sufficient resources allocated at the planning stage of research projects (for the definition of ‘sensitive and confidential data', see the glossary).
  • Agreements, consents, undertakings and other documents related to research must be archived so that they are always available to researchers and support services.
  • Research infrastructures extensively involved in service provision in particular should draw up a data management policy describing the principles of processing, storing, distributing, preserving and destroying data produced by the infrastructure, as well as the questions of responsibility and ownership related to the processing of data.

2.3 Documentation and processing of research data

  • The documentation and metadata associated with research data should follow discipline-specific standards to enable the reuse and further enrichment of the data in future research projects (for the definition of ‘metadata’, see the glossary).
  • Appropriate information security (in Flamma) must be taken into consideration in the storage and processing of research data.
  • The University provides researchers with basic services in the storage and processing of research data. If the quantity of data is high or special computing power is needed, the costs of storage and processing must be taken into consideration at the planning stage.

2.4 Publication and accessibility of research data

  • Research data produced at the University of Helsinki and linked to published research results are, as a rule, open and available for shared use. The principle of ‘as open as possible, as closed as necessary’ is observed when making research data openly available.
  • Research data are published in data archives that safeguard the findability of data and enable references to them (for the definition of ‘data archive’, see the glossary).
  • Metadata associated with research data must be published whenever possible, either in national or international metadata services.
  • Identifiers and licences enabling the further use of data should always be available for open research data (for the definition of ‘persistent identifier’, see the glossary).
  • If possible, sensitive data are also made available to other researchers, taking into consideration legal, ethical and contractual restrictions. In this, services that enable safe storage and restricted access with a research permit are to be employed.
  • The University has practices and services for data collection and monitoring pertaining to previously collected, produced and opened research data.

2.5 Commercialisation of research data

  • In the case of commercially valuable data, measures must be taken to ensure that their use does not endanger commercialisation. For example, research results must be published in compliance with terms and conditions set on confidentiality by the funder as well as the University’s guidelines for commercialisation.
  • In the case of research that generates commercially valuable research data, the University provides researchers with support and a streamlined service process.

2.6. Destruction or long-term preservation of research data

  • The part of research data that is valuable in the long term and that will be preserved, as well as the part that will be destroyed at the end must be specified during research projects.
  • Research data that have become useless will be destroyed after the required storage period. Research data that contain sensitive data will be destroyed with particular care.
  • The University has a process for the curation and transfer to long-term digital preservation of valuable research data (for the definition of ‘long-term preservation’, see the glossary).

 

3.1 Researchers’ responsibilities

  • Familiarising themselves with guidelines related to responsible research data management and complying with them
  • Planning and implementing data management for their research
  • Ensuring that agreements, undertakings, and consents required for research are drawn up and concluded
  • Familiarising the members of their research group with responsible data management when serving as principal investigators
  • Ensuring that data that have been agreed to be shared by the group or the collaborative research project are accessible to others
  • Conveying the principles and good practices of responsible data management when serving as thesis supervisors
  • Updating their skills in research data management on a regular basis

3.2 Responsibilities of faculties and independent institutes involved in research

  • Maintaining an overview of research data and their management in the unit, including the unit’s agreements and obligations
  • Taking the research data policy into consideration in the planning of operations and finances, as well as allocating the necessary resources
  • Integrating University-level guidelines and policies into unit operations, taking research field–specific differences into account
  • Ensuring that the academic staff and students are familiar with research data management as part of the responsible conduct of research
  • Offering and allocating resources for on-site support for research data management together with research support services
  • Carrying out preventive risk management and anticipating potential information security incidents (for the definition of ‘risk management’ and ‘information security incident’, see the glossary).
  • Ensuring the implementation of responsible research data management in research infrastructures

3.3 The University’s responsibilities

  • Engendering the necessary preconditions for the implementation of responsible research data management at the University
  • Ensuring that the University-level research data infrastructure is fit for purpose, up to date and sufficiently funded so that services are available at all stages of research projects’ lifespans (For the definition of ‘research data infrastructure’, see the glossary.)
  • Establishing, in collaboration with academic units, assessment practices and incentives to consider researchers’ efforts to promote the sharing and further use of research data as well as skills in research data management as academic merits
  • Establishing incentives for University units for responsible research data management
  • Promoting opportunities for researchers and support services specialists to specialise in the management of research data by consolidating related specialist roles and developing career paths
  • Coordinating development efforts related to research data management
  • Drawing up University-level policies for research data management and processes that support their implementation
  • Drawing up the necessary guidelines for research data management
  • Carrying out and supporting preventive risk management and anticipating potential information security incidents
  • Offering researchers and other staff training and orientation related to research data management
  • Offering researchers and academic units support in research data management

Data archive (data repository)

A virtual, typically discipline-specific archive or database where researchers can transfer their research data for sharing, reporting and reuse. Data repositories store research data, make it available and organise it in a logical manner. Data repositories also make it easier to cite research data through the use of persistent identifiers.

Data management plan

A document that describes the research data that will be acquired or produced in the research project, commonly abbreviated as DMP. In connection with data management plans, the term ‘data’ is understood in broad terms, meaning that it encompasses all of the data and resources on which the research results are based. The plan also encompasses codes, software and other methodological descriptions. 

In addition, the plan describes how rights related to research data are managed, which agreements are needed, how data protection is ensured, how research data are stored, how research data are opened, or how their findability and use for the verification of the research results and further research is otherwise enabled. While the data management plan is drawn up at the planning stage, it is a living document that must be updated as the research project progresses.

FAIR Principles

European principles for the quality of research data and associated metadata. The acronym FAIR stands for findable, accessible, interoperable and reusable. The FAIR Principles guide the drawing up of metadata in particular. Findable means that research data has a unique persistent identifier that functions as a link to the data that can always be found even if the storage location changes. Findability can also be implemented for non-digital research data whose metadata are openly available. Accessible means that research data and the associated metadata are accessible via web browsers. Interoperable means that data are stored using open file formats and common standards. Reusable means that research data has rich metadata and a licence that specifies the terms of reuse.

Information security incident

An event or circumstances that diverge from normal, which may, for example, delay, prevent or harm the conduct of research. The nature of such incidents varies by research field.

Information security incident related to data protection refer to circumstances due to which research data are destroyed, lost, altered or disclosed without authorisation, or due to which an unauthorised party gains access to the data.

Long-term preservation

The preservation of digital data in understandable and usable form for dozens or even hundreds of years. Long-term preservation is designed for valuable research data. The goal of long-term preservation is to ensure the accessibility, authenticity, understandability and completeness of digital objects, also regardless of the eventual obsolescence of or changes to hardware, software and file formats. Long-term preservation ensures the long-term availability of research data.

Metadata

Metadata is data about data. There are several types of metadata, including descriptive, structural, administrative, statistical and legal as well as reference and citation metadata. Metadata ensures the findability and reusability of research data. When research data is described and documented appropriately, other users can trace and understand specific elements of research. Metadata makes it easier to search for and find research data stored in data repositories.

Persistent identifier

A unique and unambiguous, machine-readable name for research output, commonly abbreviated as PID. Identifiers constitute permanent links that always lead to, for example, publications or metadata pages associated with research data. Persistent identifiers enable the long-term findability of digital research data.

Personal data

Any data that can be linked to living natural persons, that is, data associated with or linkable to an identified or identifiable person. Any data that can be used to indirectly identify a person, for example, by linking a specific detail to another detail that would enable identification (pseudonymised personal data) also constitute personal data. Personal data can be stored in, for example, digital files and databases, on paper, in card indexes, in document files and survey forms, or in audio or video recordings.

According to data protection regulation, specific personal data constitute what is known as special categories of personal data, or sensitive personal data (see also ‘sensitive and confidential data’):

  • Data revealing racial or ethnic origin, political opinions, religious or philosophical beliefs or trade union membership 
  • Genetic data 
  • Biometric data processed for the purpose of uniquely identifying a person
  • Data concerning health
  • Data concerning a person’s sex life or sexual orientation
  • Data concerning criminal convictions or offences

In addition to the data above, the processing of personal data can be sensitive if the processing may pose risks to the data subjects (see ‘risk management’).

Research data

Data that have been collected, observed, measured or created to verify research results, or that are otherwise considered necessary in the research community for the confirmation of results. The context turns data into research data. Any data can become research data when it is analysed for research purposes.

Research data can include measurement results, test results, survey results, audio and video recordings as well as samples and specimens. While research data is often in digital form, it can also include physical or analogue data. Research data can be raw data, processed data, data in the possession of a third party, shared data or published data. The degree of openness of research data varies from confidential and sensitive data to open data.

Research data infrastructure

Processes, technical solutions and services through which research data management is carried out in practice. Research data infrastructures involve organisation, an operating culture and long-term social networks that enable the realisation of technical and administrative solutions and services.

Research data management

Commonly referred to using the abbreviation RDM. A process encompassing the lifespan of the research project that includes the collection or acquisition, organisation, curation, storage, (long-term) preservation, protection, quality assurance, licensing and distribution of research data as well as the use of persistent identifiers and other metadata in compliance with the rules and procedures of the relevant discipline (European Commission/Horizon Europe).

Research infrastructure

Research infrastructures are instruments, equipment, information networks, databases, materials and services that serve to facilitate research, promote research collaboration and reinforce research and innovation capacity and know-how (Academy of Finland).

Risk management

Proactive anticipation of events that have negative consequences. Risk management constitutes coordinated activities used to guide, manage and monitor the actions of the University, its units or individual researchers in relation to risks. The purpose of risk management is to help the University, its units and researchers to attain their goals and make decisions.

When processing personal data (see the definition of ‘personal data’), the risks associated with the processing must be assessed while ensuring the implementation of the relevant data protection principles. In the case of personal data, the risk assessment must be carried out from the perspective of data subjects.

Sensitive and confidential data

Research data whose storage, use and sharing are restricted on ethical, legal, contractual or commercial grounds. Such data must be processed and protected with particular care.

Sensitive data are associated with, for example, specific personal data (see special categories of personal data in the definition of ‘personal data’), endangered species, biosecurity or national defence. Confidential information on patents and trade secrets are also to be protected, and their exposure can result in claims for damages.

Open Science Award

Each year, the University of Helsinki presents the Open Science Award in recognition of significant work to promote open science. The aim of this award is to highlight active open science advocates and increase information on good practices at the University.

Recipients of the award have included individual researchers and teachers, projects and units.

Award recipients

2024: Species learning environment Pinkka

2023: Helsinki Urban Rat Project

2022: HiLIFE and Helsinki Term Bank for the Arts and Sciences

2021: Research Coordinator Kati Lassila-Perini and the Language Bank of Finland

2020: Postdoctoral Researcher Laura Riuttanen and open teaching at the Department of Computer Science

2019: Finnish Museum of Natural History Luomus

2018: Professor Jaana Bäck

2017: Professor Tuuli Toivonen (in Finnish only)

You might also be interested in
Read more about Open Science