Librarians can contribute significantly to the success of citizen science projects by providing expertise in data management. Here are some important considerations for librarians involved in data management for citizen science projects:

  • Data collection: Librarians can help citizen science projects by providing guidance on data collection tools and methodologies. They can also help to ensure that data collection instruments are standardized and fit for purpose.
  • Data storage: Librarians can assist with data storage by providing guidance on data storage platforms and systems. They can help to ensure that data is stored securely and that access to the data is appropriately controlled.
  • Data curation: Librarians can assist with data curation by providing guidance on data formatting, metadata creation, and data quality control. They can also help to ensure that data is preserved and made available for future use.
  • Data sharing: Librarians can assist with data sharing by providing guidance on data sharing policies, licenses, and platforms. They can help to ensure that data is shared in a manner that is open, accessible, and sustainable.
  • Data preservation: Librarians can assist with data preservation by providing guidance on data preservation strategies, policies, and best practices. They can also help to ensure that data is archived and made available for long-term access.

In order to successfully involve librarians in data management activities through cooperation in citizen science projects, librarians require knowledge of the important features of data management. In particular, they should know what the data lifecycle is, the principles of FAIR data, how to ensure the quality of the data, and how to develop a data management plan. 

  • Data lifecycle

The data lifecycle is the sequence of stages through which a particular piece of data moves from initial generation, capture or reuse to final archiving and/or deletion at the end of its useful life. The data lifecycle is difficult to determine. Depending on your industry or profession, the data lifecycle can vary greatly. Data management experts often identify six or more stages in the data lifecycle. The following research data lifecycle model is appropriate for citizen science research:


    • Creating (Designing research, locate existing data, data collection and management, capturing and creating metadata)
    • Accessing (Distributing data, sharing data, controlling access, establish copyright, promoting data)
    • Processing (Entering, transcribing, checking, validating, cleaning data, anonymising data, describing data, manage and store data)
    • Analysing (Interpreting and deriving data, producing outputs, authoring publications, preparing for sharing)
    • Preserving (Data storage, back-up and archiving, migrating to best format and medium, creating metadata and documentation)
    • Re-using (Follow up research, new research, undertake research reviews, scrutinising findings, teaching and learning)

Why research data lifecycle management in CS projects is needed? Management enables to find and understand data when you need it; to continue project through researcher, staff, student turnover; to save time due the organized data; to reduce risk of data loss, theft or misuse; to share data, that lead to collaboration and high impact; to validate research results in an easier way. Data lifecycle management is also necessary to establish and maintain compliance with key data security and privacy legislation.

  • FAIR principles

The FAIR Principles provide guidelines for good data management practices, aiming to make data FAIR: discoverable, accessible, interoperable and reusable. "Data" in this context refers to all kinds of digital objects that are produced in the course of research: research data in the strictest sense, code, software, presentations etc. FAIR principles can be applied within all research disciplines.


    • Findable (data are easy to find for humans and computers, data described with rich metadata)
    • Accessible (meta)data are accesible by their identifier using an open protocol)
    • Interoperable (data need to interoperate with applications or workflows for analysis, storage, and processing)
    • Reusable (clear and accessible data usage license)

It should be emphasized that not only machines but also humans are intended as digesters of data. Also necessary to know that FAIR principes apply to both data and metadata. Finally, FAIR principles are not some kind of standarts or rules to evaluate data or tools. As a result, the FAIR principles will quickly become outdated and will not be applied in all areas of research. The Fair principles have no expiry date and the worldwide research community must ensure their routine application in research. 

No data set can be claimed to be 100% FAIR-compliant, as situations may be very different (for example, what tools are available, what regulatory documents must be followed, what metadata standards are available or not, etc.). Research projects can only approach FAIR principles as closely and comprehensively as possible.  It should be remembered that FAIR is not the same as open data.



  • Data quality assurance
    • Accuracy (known accuracy, automated checks for quality)
    • Completeness (no missing values)
    • Validity (data matches the rules)
    • Uniqueness (no duplicated data)
    • Consistency (data consistent across various data stores)
    • Timeliness (data represent reality from the required point in time)


  • Data marginalization problem

Data marginalization means that part of the data is missing because some citizen groups are ignored or discriminated against (e.g. LGBT, homeless people, women in the Middle East etc.). So there is simply no information about some (illegal), some are impossible to contact or collect data from (they are illiterate, so they cannot respond in writing, do not have a permanent place of residence), others cannot participate due to their age, health condition etc. That is one of the possible risk speaking about the citizen generated data issue.

  • Recommendations for data quality assurance   
    • providing training/close supervision
    • cross-checking for consistency with existing literature
    • cross-checking for consistency with participants own observations
    • simplifying the tasks asked of the public and/or adapting the research questions

  • Data management plan
    • Data description (what type of data is collected)
    • Documentation and data quality (describing why and how data was collected, methods used to collect data, data provenance, quality-assurance steps taken, how consistency is guaranteed)
    • Storage and backup (institutional storage, infrastructures)
    • Legal and ethical requirements, codes of conduct (consent, regulation, institutional agreements, restricted access, embargoes)
    • Data sharing and long-term preservation
    • Data management responsibilities and resources.



References:

Danish National Forum for Research Data Management. (n.d.). How to FAIR. https://howtofair.dk/what-is-fair/

Shwe, K.M. (2020). Study on the data management of citizen science: from the data life cycle perspective. Data and information management, 4(4): 279–296. doi:10.2478/dim-2020-0019

Costa, M. (2022). How to ensure data quality in field research. SurveyCTO. https://www.surveycto.com/blog/how-to-ensure-data-quality/

Smith, J. (2022). Extreme citizen science gives a voice to the marginalised in remote communities. Horizon : the EU Research & Innovation Magazine. https://ec.europa.eu/research-and-innovation/en/horizon-magazine/extreme-citizen-science-gives-voice-marginalised-remote-communities

Hansen, J. S., Gadegaard, S., Hansen, K. K., Larsen, A. V., Møller, S., Thomsen, G. S., & Holmstrand, K. F. (2021). Research data management challenges in citizen science projects and recommendations for library support services. A scoping review and case study. Data Science Journal, 20(1), 1-29. [25]. https://doi.org/10.5334/dsj-2021-025

Last modified: Monday, 12 June 2023, 8:37 AM